Industrial catalyst for the selective Fischer–Tropsch synthesis of long-chain hydrocarbons

Article abstract of DOI:10.1134/S0023158417010062

The results of development of an industrial supported cobalt–silica gel catalyst for the Fischer–Tropsch synthesis are reported. The studies included the selection of a support and the determination of an optimum active component content, a calcination temperature, and the effect of doping with aluminum oxide on the physicochemical and catalytic properties of the Co–SiO₂ system. The catalyst samples were characterized by a set of physicochemical methods. The on-stream stability of the supported cobalt–silica gel catalyst was tested in the continuous mode for 1000 h. In the course of the entire test cycle, the catalyst exhibited stable operation under varied synthesis temperature and gas space velocity, and it can be recommended for industrial applications. The experimental results were used for the preparation of a pilot batch of the catalyst.

Full text of DOI:10.1134/S0023158417010062

ISSN 0023-1584, Kinetics and Catalysis, 2017, Vol. 58, No. 1, pp. 81–91. © Pleiades Publishing, Ltd., 2017.
Original Russian Text © A.P. Savost’yanov, R.E. Yakovenko, G.B. Narochnyi, V.G. Bakun, S.I. Sulima, E.S. Yakuba, S.A. Mitchenko, 2017, published in Kinetika i Kataliz, 2017,
Vol. 58, No. 1, pp. 86–97.
Industrial Catalyst for the Selective Fischer–Tropsch Synthesis
of Long-Chain Hydrocarbons
A. P. Savost’yanova, R. E. Yakovenkoa, G. B. Narochnyia, *, V. G. Bakuna, S. I. Sulimaa,
E. S. Yakubaa, and S. A. Mitchenkoa, b, **
a Platov South-Russian State Polytechnic University (NPI), Novocherkassk, Rostov oblast, 346128 Russia
b Litvinenko Institute of Physical Organic Chemistry and Coal Chemistry, National Academy of Sciences of Ukraine,
Donetsk, 83114 Ukraine
*e-mail: narochgb@bk.ru
**e-mail: samit_RPt@mail.ru
Received May 30, 2016
Abstract—The results of development of an industrial supported cobalt–silica gel catalyst for the Fischer–
Tropsch synthesis are reported. The studies included the selection of a support and the determination of an opti-
mum active component content, a calcination temperature, and the effect of doping with aluminum oxide on
the physicochemical and catalytic properties of the Co–SiO2 system. The catalyst samples were characterized
by a set of physicochemical methods. The on-stream stability of the supported cobalt–silica gel catalyst was
tested in the continuous mode for 1000 h. In the course of the entire test cycle, the catalyst exhibited stable oper-
ation under varied synthesis temperature and gas space velocity, and it can be recommended for industrial appli-
cations. The experimental results were used for the preparation of a pilot batch of the catalyst.
Keywords: Fischer–Tropsch synthesis, Co/SiO2 catalyst, Al2O3 promoter, physicochemical properties, selec-
tivity for C5+
DOI: 10.1134/S0023158417010062
INTRODUCTION
Secunda in South Africa), Escravos (Nigeria), Shell
(Bintulu, Malaysia, and Qatar), Sasol/QP (Ras Laf-
fan, Qatar), and PetroSA (Mossel Bay, South Africa).
Building of such plants is planned by ExxonMobil,
Conoco, and Syntroleum [5, 6]. In Russia, a factory of
this kind with an annual output of 50000 t occurred at
the Novocherkassk plant of synthetic products (Ros-
tov oblast). It was launched in 1952 with the use of
technology, equipment, and catalyst developed in
Germany. Subsequently, the German deposited
cobalt–thorium (Co–Th) catalyst was replaced by an
original cobalt–magnesium–zirconium (Co–Mg–
Zr) catalyst [7, 8]. Initially, coal (anthracite) from the
Donets Basin served as raw material; then, natural gas
was used. In addition to gasoline and diesel fractions,
the products of the plant included C35+ hydrocarbons
(ceresin of brand 100) and individual high-purity
hydrocarbons, including α-olefins with odd hydrocar-
bon numbers. This factory was in operation up to
1992, and it is currently lost.
The exhaustion of world oil reserves increases
interest in technologies for the production of hydro-
carbons from alternative sources (natural gas, associ-
ated petroleum gas, coal, peat, biomass, carbon-con-
taining wastes, etc.), which can be converted into liq-
uid motor fuel or raw materials for chemical and
petrochemical industries [1]. Among them are the
most promising technologies based on the conversion
of carbon-containing matter into synthesis gas with
the subsequent transformation of this latter into
hydrocarbons by the Fischer–Tropsch (FT) process
[2]. The synthetic hydrocarbons obtained by this pro-
cess are in demand on the Russian and world markets.
Thus, liquid hydrocarbons can be used as high-quality
synthetic motor fuels (almost free from nitrogen, sul-
fur, and aromatic hydrocarbons) or for the production
of reagents and standard reference substances, and
α-olefins can be used as raw materials in the petro-
chemical branch of industry. Solid hydrocarbons
(waxes) find use in the cosmetic, food, and defense
branches of industry [3, 4], or they can be converted
into liquid products by hydrocracking.
It is obvious that these technologies should be
renewed for the further development of the industry of
synthetic hydrocarbons in the Russian Federation.
Industrial technology for the production of syn- The industrial production of synthetic hydrocarbons is
thetic hydrocarbons according to this method was planned to be implemented in Dzerzhinsk (Nizhny
implemented at the plants of Sasol (at Sasolburg and Novgorod oblast) [9, 10]; for this purpose, a pilot plant
81

82
SAVOST’YANOV et al.
for the synthesis of hydrocarbons from natural gas was the aid of the PDF-2 database [13] and the Crystallo-
preliminarily developed. The main engineering solu- graphica software. Crystallite sizes were calculated
tions used in this case were described elsewhere [11]. A from the Scherrer equation [14].
new supported cobalt–silica gel catalyst whose pro-
duction process was developed under industrial condi-
tions [12] was proposed as a catalyst for the synthesis
of long-chain hydrocarbons.
Here, we report the studies performed in the devel-
opment of a commercial cobalt–silica gel (Co/SiO2)
catalyst with an increased selectivity for C35+ hydro-
carbons.
The studies of the catalysts by temperature-pro-
grammed reduction (TPR) were carried out on a
ChemiSorb 2750 chemisorption analyzer (Micromer-
itics, United States) with a thermal conductivity
detector. The sample (0.1–0.2 g) was preliminarily
kept in a flow of helium (20 mL/min) at a temperature
of 200°C for 2 h; then, it was cooled to 20°C, blown
with a mixture of 10% H2 in nitrogen (20 mL/min),
and reduced in a temperature range of 20–800°C at a
rate of 5 K/min.
The ChemiSorb 2750 analyzer (Micromeritics)
with a thermal conductivity detector was used for the
study of catalysts by the temperature-programmed
desorption of CO (CO TPD). The catalyst sample was
reduced in accordance with the above procedure. The
pulse adsorption of CO until sample surface saturation
was performed in a flow of helium (20 mL/min), and
physically adsorbed gas was removed by blowing for
1 h at 100°C. CO was desorbed by increasing the tem-
perature from 25 to 800°C at a rate of 20 K/min.
EXPERIMENTAL
Catalyst Preparation
The catalyst samples were prepared by the impreg-
nation of a support with an aqueous solution of cobalt
nitrate in a single step. Commercial KSKG silica gel
(GOST [State Standard] 3956-76, Salavat Catalyst
Factory, Russia) was used as the support. The concen-
tration of cobalt nitrate in the impregnating solution
was varied in a range of 20–55 wt %. The support was
impregnated at a temperature of 80°C for 0.5 h. The
catalyst was dried under the following conditions: at
80°C for 6 h and at 100, 120, and 140°C for 2 h at each
particular temperature; then, it was calcined at 400°C
for 4 h. The effect of final heat treatment regimes on
the properties of catalysts was studied with the samples
calcined at final temperatures of 200, 250, 300, 400,
and 500°C. It was preliminarily found that an opti-
mum rate of sample heating is 5 K/min. The cobalt
content of all of the catalyst samples with different cal-
cination temperatures was 20 0.5 wt %.
For the determination of the effect of an Al2O3
additive on the properties of cobalt–silica gel catalysts,
aluminum was introduced in the form of nitrate
together with cobalt nitrate at the stage of impregna-
tion. The catalyst samples were designated as Co–
xAl2O3/SiO2, where x = 0.4, 1, 2, and 3 wt % is the alu-
minum oxide content of the catalyst. The cobalt con-
tent of the test catalyst samples was varied in a range of
20.9–22.8 wt %.
The micrographs of catalysts were obtained on a
Tecnai G2 Spirit BioTWIN transmission electron
microscope (FEI, United States) at an accelerating
voltage of 120 kV. The samples of the reduced catalysts
were ground in a flow of CO2 and supported on copper
grids as suspensions in toluene.
Catalytic Tests
The Fischer–Tropsch synthesis was carried out in a
steel flow reactor (inside diameter, 28 mm) with a fixed
catalyst bed (70 cm3) at atmospheric or elevated pres-
sure (1.5 MPa). The catalyst samples were preliminarily
reduced in a flow of hydrogen (GHSV = 1000 h–1) at
400°C for 1 h.
Before the catalytic tests, all of the catalysts were
activated at GHSV = 100 h–1 in synthesis gas contain-
ing 33 vol % CO and 67 vol % H2. The activation was
carried out under a pressure of 0.1 MPa with a stepwise
increase in the temperature (2.5 K/h) from 150°C to a
temperature at which ~50% CO conversion was
reached. Then, the temperature was decreased to
160°C and the specified values of pressure and gas
space velocity were adjusted; the temperature was
increased once again to the specified value at a rate of
2.5 K/h. After catalyst activation with synthesis gas, the
synthesis products were removed from collectors and a
balance experiment was performed for 90–120 h.
Characterization of Physicochemical Properties
The cobalt content was determined by X-ray fluo-
rescence elemental analysis (XRFEA) on an ARL
QUANT’X EDXRF X-ray spectrometer (Thermo
Electron Corporation, United States) in an atmo-
sphere of air with the use of a Teflon substrate with an
effective irradiation area of 48.99 mm2.
The X-ray diffraction (XRD) analysis of the cal-
cined catalysts was carried out on an ARL X’TRA
powder diffractometer (Thermo Fisher Scientific,
United States) with monochromatic CuKα radiation
by point-by-point scanning (step size, 0.01°; counting
time per data point, 2 s) in the 2θ range from 15° to 70°.
The activity of the catalyst samples with different
active component contents and final heat treatment
temperatures was determined at GHSV = 100 h–1 and
a pressure of 0.1 MPa.
The effect of the Al2O3 promoter on the properties of
Qualitative phase composition was determined with cobalt–silica gel catalysts was studied in a synthesis
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

INDUSTRIAL CATALYST FOR THE SELECTIVE FISCHER–TROPSCH SYNTHESIS
83
temperature range of 213–220°C at GHSV = 1000 h–1
СCo, wt %
and a pressure of 2.0 MPa.
22
20
18
16
14
12
10
8
The on-stream stability tests of the Co–
1%Al2O3/SiO2 catalyst were conducted at a pressure
of 1.5 MPa with the use of synthesis gas containing
20 vol % CO, 40 vol % H2, and 40 vol % N2.
The composition of the synthesis gas and gaseous
synthesis products was analyzed by gas-adsorption
chromatography on a Kristall 5000 chromatograph
(Khromatek, Russia) with a thermal conductivity
detector and two columns. The first Haysep R col-
umn was used for the analysis of C1–C5 hydrocar-
bons and CO2 (carrier gas, helium; flow rate,
15 mL/min), and the second column (molecular
sieve NaX) was used for the determination of CO, H2,
and N2 (carrier gas, argon; flow rate, 15 mL/min).
Temperature-programmed conditions (80–240°C;
heating rate, 8 K/min) were used.
20
25
30
35
40
45
50
55
C
Co(NO ) , wt %
)
2
2
3
Fig. 1. Dependence of the cobalt content (C ) of the cat-
alyst on the cobalt nitrate concentration
the impregnating solution.
Co
in
)
(ССо(NO )
3
2
The composition of C5+ hydrocarbons was deter-
mined by capillary gas chromatography–mass spec-
trometry on an Agilent GC 7890 gas chromatograph
(Agilent, United States) equipped with an
MSD 5975C mass-selective detector and an HP-5MS
capillary column.
alyst on cobalt nitrate concentration in the impregnat-
ing solution (Fig. 1) is described by the following
equation (correlation coefficient, 0.99):
ССо
where CCo is the cobalt content of the catalyst (wt %),
and is the cobalt nitrate concentration in the
=
+ 1.06,
0.39С
Со(NO3)2
С
Со(NO3)2
RESULTS AND DISCUSSION
Selection of a Support and Determination
of the Optimum Cobalt Content
impregnating solution (wt %). This equation is correct
in the range of cobalt concentrations of 8.0–22.0 wt %
in the catalyst.
With increasing the cobalt content of the Co/SiO2
catalyst, the yield of C5+ hydrocarbons increased
(Fig. 2) and reached a maximum at a cobalt content of
~20.0 wt %. A further increase in the cobalt content of
the catalyst led to a decrease in selectivity and pro-
ductivity. It is obvious that the above decrease in pro-
ductivity with respect to the C5+ hydrocarbons is
related to a decrease in the dispersity of the active
component [2, 22].
The technology of supported cobalt-containing
catalysts differs from that of precipitated ones in a
number of advantages: the simplicity of technological
operations, the possibility of active component regen-
eration, and a smaller amount of waste [15]. The main
function of supports in catalysts, including those in the
FT process, is to afford a high dispersity and specific
surface area of the supported active metal [16]. The
support should have a developed polymodal porous
structure and a large and accessible specific surface
area [17]. On the supporting of an active component,
it can interact with the support and form new com-
pounds and phases [18]. The support frequently man-
ifests independent activity in the course of synthesis to
catalyze the subsequent transformations of the initial
products of the FT synthesis; this is characteristic of
zeolites [19–21].
The following commercially available supports
were studied previously [10, 11]: silica gel, synthetic
aluminosilicate, and aluminum oxides of different
grades. It was found that the Co/SiO2 catalytic system
is promising for the synthesis of long-chain hydrocar-
bons. In order to determine the effect of the active
component content on the catalytic properties of a
supported Co/SiO2 catalyst and an optimum catalyst
Thus, the optimum cobalt content of the catalyst
on the KSKG support is ~20.0 wt %.
Effect of the Thermal Pretreatment of a Catalyst
on Its Activity and Selectivity in the FT Synthesis
It is well known [23] that the optimum size of
cobalt crystallites in a reduced form on the support
surface should be 6–8 nm. The rate of heating and the
final calcination temperature at the stage of the prepa-
ration of the oxide form of the catalyst exert a consid-
erable effect on the dispersity of active component
particles [24]. At a high rate of thermal treatment, the
active component is comparatively evenly distributed
over the support surface: in this case, rapid evaporation
excludes the entering of new solution portions because
composition, we prepared samples with different of capillary flow or film transfer. On the contrary, com-
cobalt contents of 8.5–22.0 wt %. We found that the paratively coarse particles are formed at a low rate of
dependence of the cobalt content of the supported cat- heating (because of the gradual additional supply of new
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

84
SAVOST’YANOV et al.
Intensity
−3
Productivity, kg m h−1
Cat
10
500°С
400°С
9
300°С
250°С
200°С
8
7
6
15
25
35
45
55
65
75
2θ, deg
5
8
10
12
14
16
18
20
22
Co content, %
Fig. 2. Effect of the cobalt content of the Co/SiO catalyst
Fig. 3. X-ray diffraction patterns of the Co/SiO catalyst
2
2
on the productivity for C hydrocarbons.
samples with different final calcination temperatures.
5+
solution portions to the supported cobalt particles phous, and cobalt occurred in the structure of catalysts
formed) [25].
in the form of a number of reflections from the crystal-
line phase of the Co3O4 oxide. The sizes of the Co3O4
crystallites estimated from the Scherrer equation were
in a range of 13–16 nm.
The experimental results (Table 1) are consistent
with the above concepts. The samples calcined at low
final temperatures of 200 and 250°C exhibited lower
activity in the FT synthesis. Thus, the CO conversion
on the Co/SiO2 catalyst (200°C) at a temperature of
190°C and on Co/SiO2 (250°C) at 195°C was 48.8 and
Figure 4 shows the results of the TPR studies. The
spectra of the Co/SiO2 catalyst samples with final cal-
cination temperatures of 200, 250, and 300°C exhib-
only 45.8%, respectively. As the synthesis tempera- ited weak peaks in a range of 150–200°C, which can be
ture was increased, the increased formation of gases attributed to the reductive decomposition of residual
cobalt nitrate. These peaks are absent from the spectra
of the Co/SiO2 (400°C) and Co/SiO2 (500°C) sam-
ples; this fact indicates the almost complete decompo-
sition of cobalt nitrate at final treatment temperatures
of 400 and 500°C. In addition to the low-temperature
peak noted, two intense reduction peaks were present
in a range of 280–650°C with maximums at 300–330
and 370–470°C, which will be referred to as signal 1
and signal 2, respectively (Table 2). These peaks of
hydrogen absorption can be attributed to the stepwise
reduction of a Co3O4 cobalt oxide phase [26, 27]
according to the following reaction equations:
(C1–C4) was observed; as a consequence, selectivity
for C5+ hydrocarbons decreased. The calcination of
catalyst samples in a range of 300–500°C increased
their catalytic activity: the CO conversion on these
samples was higher than 50% even at 180°C. Selectiv-
ity for the C5+ hydrocarbons was also higher on the
catalysts calcined at a temperature of 400–500°C. This
fact is indicative of the formation a larger number of
polymerization sites on the surface of these catalytic
systems.
The samples with different calcination temperatures
were studied by XRD analysis and TPR. The XRD
analysis of Co/SiO2 catalyst samples in oxide form
showed that all of the samples exhibited diffraction pat-
terns of the same type (Fig. 3): SiO2 was X-ray amor-
Со3О4 + Н2 → 3СоО + Н2О,
СоО + Н2 → Со0 + Н2О.
(I)
(II)
Table 1. Catalytic properties of Co/SiO2 catalyst samples with different final calcination temperatures
Synthesis
temperature,
°С
Selectivity, %
С2–С4
Calcination
temperature, °С
CO conversion,
%
CH4
С5+
CO2
200
250
300
400
500
190
195
180
180
180
48.8
45.8
52.1
52.2
52.5
14.5
16.8
7.5
14.5
18.0
7.1
69.5
63.6
83.8
88.7
87.0
1.5
1.6
1.6
1.5
2.2
5.2
5.6
5.4
5.4
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

INDUSTRIAL CATALYST FOR THE SELECTIVE FISCHER–TROPSCH SYNTHESIS
85
The S1/S2 ratio between the areas of signals 1 and
2 increased with catalyst calcination temperature
(Table 2) to approach a value of 3, which is theoreti-
cally expected from the stoichiometry of cobalt oxide
reduction (Eqs. (I) and (II)). Obviously, the smaller
values of the S1/S2 ratio for the samples with final
calcination temperatures lower than 400°C can be
attributed to the partial oxidation of cobalt at low cal-
cination temperatures.
Detector signal
500°С
400°С
300°С
The calcination temperature has an essential effect
on the physicochemical and catalytic properties of
supported catalysts; for the catalyst developed, it
should be in a range of 400–500°C, which is consis-
tent with published data [28].
250°С
200°С
100
200
300
400
500
600
700
800
Temperature,°С
Effect of the Al2O3 Promoter on the Properties
of Cobalt–Silica Gel Catalysts
Fig. 4. TPR spectra of the Co/SiO catalyst samples with
2
different final calcination temperatures.
Promotion with noble metals [2, 29–32] substan-
tially facilitates the reduction of cobalt and increases
the productivity of catalysts. In spite of the strong pro-
moting effect of noble metals on catalysts for the FT
synthesis, the industrial application of these metals is
limited because of their high cost [32, 33]. The doping
of cobalt catalysts with inexpensive and readily avail-
able promoters such as metal oxides is a more accept-
able method for improving the properties of FT cata-
lysts [2, 34, 35]. Recently, Zhang et al. [36] reported
that the introduction of aluminum oxide into a 10 wt %
Co/SiO2 catalyst considerably increased activity and
selectivity for C5+ hydrocarbons in the FT synthesis in
a slurry reactor at 240°C and a pressure of 1 MPa. Pei
et al. [37] obtained similar results in the FT synthesis
at 220°C and 3 MPa in a tubular fixed-bed reactor
with the 15 wt % Co/activated carbon catalyst. They
Table 3 summarizes the catalytic characteristics of
cobalt–silica gel catalyst samples promoted with alu-
minum oxide in the FT synthesis over a temperature
range of 213–220°C. The characteristics were mea-
sured at a pressure of 2.0 MPa and a gas space velocity
of 1000 h–1 because, under the conditions of the
intense FT process, the promoting effect of the intro-
duction of Al2O3 became more pronounced: the for-
mation of CH4 decreased by a third; the conversion of
CO increased, and the productivity of catalysts with
respect to a C5+ fraction also increased by 5–13% with
an increase in the aluminum oxide content to 1 wt %.
Table 4 shows the effect of aluminum oxide addi-
tives to the Co/SiO2 catalyst on the distribution of
C5+ products. Alkanes are the main products
carried out the reactions on a laboratory scale with a obtained on all of the catalysts. Olefins and alcohols
relatively short time of catalyst testing (to 24 h) [37]. In were also formed in trace amounts on the catalyst not
this work, we studied the effects of promotion with promoted with aluminum oxide. The low yields of
aluminum oxide in prolonged catalytic experiments these products increased by a factor of almost two on
(150 h) at a pilot plant, which simulated the conditions the catalyst with an optimum 1 wt % content of alu-
of the industrial FT synthesis.
minum oxide, and they increased considerably as the
Table 2. TPR spectra of the catalyst samples depending on calcination temperature
Signal 1
Signal 2
Calcination
temperature, °C
S1/S2
area S1, %
area S2, %
maximum, °C
maximum, °C
200
250
300
400
500
303
317
318
295
328
28
30
27
26
26
474
440
430
370
416
72
70
73
74
74
2.5
2.4
2.7
2.9
2.9
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

86
SAVOST’YANOV et al.
Table 3. Catalytic properties of the promoted Co–xAl2O3/SiO2 catalysts
Selectivity,
Productivity
with respect to C5+
hydrocarbons,
%
CO
Catalyst
conversion, %
CH4
С2–С4
С5+
СО2
kg m–3 h–1
Co/SiO2
49.0
51.1
53.9
46.2
22.7
15.4
10.8
10.7
17.6
23.5
6.6
9.3
77.4
79.5
80.1
74.2
62.0
0.6
0.4
0.7
0.6
0.8
90.4
95.0
Co–0.4Al2O3/SiO2
Co–1Al2O3/SiO2
Co–2Al2O3/SiO2
Co–3Al2O3/SiO2
8.4
102.4
79.5
7.7
13.7
33.2
Table 4. Proportions of C5+ products of the FT synthesis over the Co–xAl2O3/SiO2 catalysts
Product distribution, wt %
Al2O3 content,
paraffins
С19–С35
olefins
alcohols
C5–С18
wt %
C5–С18
С35+
C5–С18
0
1.0
3.0
49.6
54.9
48.1
47.6
42.1
37.5
2.4
2.3
2.2
0.4
0.7
10.1
0.04
0.07
1.2
Table 5. Average size (d) of the Co0 nanoparticles in the
reduced Co–xAl2O3/SiO2 catalysts
tallites for reduced catalysts with different concentra-
tions of Al2O3 additives were studied by TEM (Fig. 6).
An increase in the aluminum oxide content to 1.0 wt %
facilitated a decrease in the particle sizes of cobalt (an
increase in the dispersity) and a narrowing of particle-
size distribution. A further increase in the concentration
of the promoter to 2.0 wt % led to a coarsening of Co
nanoparticles and a broadening of the particle-size dis-
tribution. Table 5 summarizes the average sizes of Co0
nanoparticles.
x, wt %
d, nm
0
12
8
8
2
2
2
2
0.4
1.0
2.0
11
amount of Al2O3 was further increased. Figure 5
shows the histograms of the molecular-weight distri-
butions of C5+ paraffins. The promotion of the cata-
lysts with aluminum oxide (1 wt %) changed the
molecular-weight distribution to make it narrower
due to an increase in the fraction of C8–C25.
It is believed that the Al2O3 additives in a specific
amount inhibit the aggregation of cobalt metal
nanoparticles; this manifests itself in a decrease in the
average particle size. The total number of active cobalt
atoms on the external surface of the nanoparticles
increases with the dispersity of Co0; because of this,
the catalyst activity in the FT synthesis increases. The
above effect of the average size of cobalt nanoparticles
on the selectivity of the FT synthesis on Co/SiO2 can
also be a consequence of the activity of the cobalt
nanoparticles in the hydrogenolysis of the hydrocar-
bons formed. Recently, Peng et al. [38] found that high
activity in the selective hydrogenolysis of hexadecane
to lighter hydrocarbons (C10–C15) under the Fischer–
Tropsch synthesis conditions was observed on cobalt
Thus, the doping of Co/SiO2 with aluminum oxide
exerted a promoting effect on the catalytic activity and
selectivity of the FT synthesis. Under the test condi-
tions, the greatest promoting effect was observed in cat-
alyst samples with aluminum oxide additives to 1 wt %.
A further increase in the aluminum oxide content led
to an opposite effect.
Promotion with Al2O3 did almost not change the
texture of catalysts, which is caused by the nature of the catalysts with the average size of cobalt nanoparticles
support (the specific surface area was 230–239 m2/g,
and the mean pore diameter and the total pore volume
were 8–11 nm and 0.6–0.7 cm3/g, respectively); how-
ever, it substantially changed the particle size of the
active component. The size distribution of cobalt crys-
of precisely 8 nm. Therefore, it is possible to assume
that the more intensive subsequent hydrogenolysis of
long-chain hydrocarbons on the Co0 nanoparticles is
responsible for the observed increase in the fraction of
C8–C25 hydrocarbons due to heavier products (Fig. 5)
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

INDUSTRIAL CATALYST FOR THE SELECTIVE FISCHER–TROPSCH SYNTHESIS
87
in the FT synthesis on the Co–1Al2O3/SiO2 catalyst,
as compared with that on Co/SiO2.
(а)
Concentration, wt %
On the other hand, the doping of Co/SiO2 with
aluminum oxide can create a certain acidity of the sup-
port [39]; therefore, the possibility of acid catalysis in
the hydrogenolysis of the resulting hydrocarbons can-
not be excluded [38, 40]. In particular, the observed
sharp increase in selectivity for CH4 and olefins on the
Co–3Al2O3/SiO2 catalyst, as compared with that on
Co–1Al2O3/SiO2 (Table 1), can be a consequence of
the expected higher concentration of acid sites on the
surface of the former catalyst and the consecutive
demethylation of hydrocarbons on them.
6
4
2
0
5
10
15
20
25
30
35
Number of carbon atoms
(b)
Concentration, wt %
For studying the effect of the aluminum oxide con-
tent on the adsorption of CO on the reduced catalysts,
we used the CO TPD method. Figure 7 shows the CO
TPD profiles for the reduced Co–xAl2O3/SiO2 sam-
ples. A hardly distinguishable desorption peak at
~70°C can be attributed to physically adsorbed CO. In
addition to this low-intensity peak, two main desorp-
tion signals were present: a low-temperature peak with
a maximum at ~220°C, which can be attributed to the
desorption of chemisorbed CO, and a high-tempera-
ture peak with a maximum at ~750°C, which can be
attributed to the desorption of CO2 formed from the
chemisorbed CO according to the Boudoir reaction.
The addition of aluminum oxide did not change
the position of the low-temperature peak. At the same
time, the area of the low-temperature signal mono-
tonically increased as the Al2O3 content was increased
to 1 wt % (Table 6). However, a further increase in the
Al2O3 load over 1 wt % produced an opposite effect:
the low-temperature signal area decreased. Obviously,
this dependence of the chemisorption of CO on the
introduced amount of aluminum oxide with a maxi-
mum to 1 wt % is a consequence of the maximum dis-
persity of cobalt metal, which is reached exactly at this
amount of Al2O3 added.
6
4
2
0
5
10
15
20
25
30
35
Number of carbon atoms
Fig. 5. Molecular-weight distributions of the C paraffins
5+
obtained on the (a) Co/SiO and (b) Co–1Al O /SiO
2
2
2
3
catalysts.
tion of products in the FT synthesis. Peng et al. [38]
assumed that the predominance of superficially
adsorbed hydrogen leads to an increase in the process
of CO hydrogenation and, correspondingly, in selec-
tivity for methane. Conversely, a higher concentration
of CO on the surface facilitates the growth of a carbon
chain with the formation of longer chain products. It
is likely that these effects can explain better selectivity
for the C5+ hydrocarbons on the Co/SiO2 catalysts
promoted with aluminum oxide.
Note that, in addition to the above structure fac-
tors, electronic effects can also play a specific role. It
is well known that aluminum oxide possesses electron-
donor properties [36]. The presence of donor particles
on the catalyst surface will facilitate the adsorption of
CO because carbon monoxide is a π acceptor. Then,
aluminum oxide can be considered not only as a struc-
tural promoter, which leads to an increase in the dis-
persity of cobalt, but also as an electronic promoter,
which modifies the chemisorption properties of the
active metal. Both of these functions of aluminum
oxide enhance the adsorption of CO. On the other
hand, Zhang et al. [36] demonstrated that Al2O3 as a
promoter inhibits the adsorption of H2; this also leads
to a decrease in the H2/CO ratio on the surface of the
promoted catalysts.
On-Stream Stability of the Co–Al2O3/SiO2 Catalyst
For the industrial use of the supported cobalt–sil-
ica gel catalyst, it was necessary to estimate the stabil-
ity of its operation in the continuous mode under
Table 6. Characteristics of the low-temperature peak of CO
TPD for the reduced catalysts
Тmax, °C
Catalyst
Area, %
Co/SiO2
207
207
220
219
221
12
15
18
8
Co–0.4Al2O3/SiO2
Co–1Al2O3/SiO2
Co–2Al2O3/SiO2
Co–3Al2O3/SiO2
It is necessary to emphasize that the H2/CO ratio
between the surface concentrations of adsorbed gases
was related [25, 28, 38] to the selectivity and distribu-
7
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

88
SAVOST’YANOV et al.
(а)
Concentration, %
40
30
20
10
0
0
2
4
6
8 10 12 14 16 18 20
Co particle size, nm
100 nm
(b)
Concentration, %
40
30
20
10
0
0
2
4
6
8
10 12 14 16 18 20
Co particle size, nm
100 nm
(c)
Concentration, %
40
30
20
10
0
0
2
4
6
8
10 12 14 16 18 20
Co particle size, nm
100 nm
(d)
Concentration, %
40
30
20
10
0
0
2
4
6
8
10 12 14 16 18 20
Co particle size, nm
100 nm
Fig. 6. TEM images of the reduced catalysts and cobalt particle size distribution histograms: (a) Co/SiO , (b) Co–
2
0.4Al O /SiO , (c) Co–1Al O /SiO , and (d) Co–2Al O /SiO .
2
3
2
2
3
2
2
3
2
long-term service conditions. The process parameters
The on-stream stability of the cobalt–silica gel
(temperature, gas space velocity, etc.) can be changed catalyst was studied at a pressure of 1.5 MPa in a flow
under industrial conditions, and this can affect the system in the continuous mode for 1000 h in cycles
catalytic properties of the catalyst.
(Table 7). The results of the studies were represented
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

INDUSTRIAL CATALYST FOR THE SELECTIVE FISCHER–TROPSCH SYNTHESIS
89
In the second cycle, the gas space velocity and tem-
perature were decreased (to 60 h–1 and 165°C, respec-
tively). The CO conversion changed only slightly in the
course of this cycle, and it varied over a range of 55–
60%. In the third cycle of tests, the temperature was ele-
vated to 183°C at a constant gas space velocity (60 h–1)
in order to increase the CO conversion. In the course of
this cycle, the conversion of CO was 71–72%.
In the fourth cycle of tests, the gas space velocity
was increased to 100 h–1 at a fixed temperature; this led
to a small increase in the conversion of CO (to 76%) at
the beginning of work, whereas these values reached a
steady level of 72–73% at the end of the cycle. In the
fifth cycle of tests, we returned to the conditions of the
first cycle (GHSV = 100 h–1, 170°C). After 30 h of
operation in this mode, the CO conversion asymptot-
ically approached a value of 46%, which coincided
with the CO conversion in the first cycle. At the final
sixth stage of the studies, the conversion of CO was
increased to 55% due to an increase in the temperature
to 175°C. In the first 200 h of continuous catalyst
operation, the conversion monotonically decreased
from 55 to ~48% and then remained almost unchanged
within the next 350 h of operation.
Detector signal
733°С
5
4
3
773°С
772°С
773°С
744°С
221°С
219°С
220°С
207°С
2
1
207°С
0
150
300
450
600
750
900
Temperature, °С
Fig. 7. CO TPD spectra of the reduced catalysts:
(1) Co/SiO , (2) Co–0.4Al O /SiO , (3) Co–1Al O /SiO ,
2
2
3
2
2
3
2
(4) Co–2Al O /SiO , and (5) Co–3Al O /SiO .
2
3
2
2
3
2
As follows from the data given in Table 9, the catalyst
that operated for a significant time under the conditions
of varied gas space velocity and synthesis temperature
exhibited stable catalytic properties. Selectivity for C5+
hydrocarbons, including long-chain C35+ hydrocar-
bons, and the fractional composition of the resulting
hydrocarbons did not substantially change. The molec-
ular-weight distribution of hydrocarbons corresponded
to a chain growth probability (α) of 0.94.
Thus, the Co–1Al2O3/SiO2 catalyst exhibited sta-
ble operation under varied FT process parameters for
1000 h of the entire period of tests; this fact evidences
that it is promising for applications in industrial tech-
nologies.
as a diagram of CO conversion (xCO)–synthesis tem-
perature (T)–catalyst on-stream time (τ) (Fig. 8).
In the initial period of tests (first cycle, Fig. 8),
~52% CO conversion was reached at a gas space veloc-
ity (GHSV) of 100 h–1 and a temperature of 170°C; the
conversion decreased to 45% after 30 h of continuous
catalyst operation and remained almost unchanged
during the entire period of the first cycle. Table 8 sum-
marizes the group and fractional compositions of the
condensed hydrocarbons obtained in the first cycle.
Long-chain (C19+) unbranched saturated hydrocar-
bons were predominantly formed.
CO conversion, %
100
Temperature, °С
210
205
90
80
70
60
50
40
1
2
cycle 1 cycle 2
cycle 3 cycle 4 cycle 5
cycle 6
200
195
190
185
180
175
170
165
160
0
100
200
300
400
500
600
700
800
900
1000
On-stream time, h
Fig. 8. The x –T–τ time diagram obtained by experimental testing of the supported cobalt–silica gel catalyst: (1) CO conversion
CO
and (2) temperature.
KINETICS AND CATALYSIS Vol. 58 No. 1 2017

90
SAVOST’YANOV et al.
Table 7. Operation cycles of the Co–1Al2O3/SiO2 catalyst Table 8. Group and fractional compositions of the hydro-
at 1.5 MPa
carbons synthesized in the first cycle
Temperature, On-stream
Fractional composition,
GHSV, h–1
Cycle
Group
composition
°С
time, h
wt %
First
100
60
60
100
100
100
170
165
183
183
170
175
80
120
80
100
80
С5–С10
С11–С18
С19+
Second
Third
Fourth
Fifth
n-Alkanes
Isoalkanes
Alkenes
Total
8.9
0.1
0.7
9.7
31.8
0.2
57.7
0.1
0.5
0.0
32.5
57.8
Sixth
540
Table 9. Catalytic properties of the Co–1Al2O3/SiO2 catalyst in the first and sixth cycles
Selectivity, %
С5+
Fractional composition, wt %
Temperature,
CO
°С
conversion, %
СН4
С35+
С5–С10
С11–С18
С19+
170
175
45
48
4.7
5.2
89.0
88.4
33.1
31.8
9.7
11.5
32.5
33.6
57.8
54.9
2. Khodakov, A.Y., Chu, W., and Fongarland, P., Chem.
Rev., 2007, vol. 107, p. 1692.
CONCLUSIONS
The set of studies on the development of a cobalt-
containing supported catalyst for the selective synthe-
sis of long-chain hydrocarbons on an industrial scale
allowed us to make the following conclusions:
(1) We found that KSKG silica gel is an effective
support for the cobalt-containing catalyst for the syn-
thesis of hydrocarbons with an optimum cobalt con-
tent of 20.0 wt %.
(2) We found that the maximum productivity and
selectivity for C35+ hydrocarbons were reached upon
catalyst calcination at a temperature of 400–500°C at
the stage of the preparation of the oxide form.
3. Bekker, M., Louw, N.R., Rensburg, J.V., and Potgiete, J.,
Int. J. Cosmet. Sci., 2013, vol. 35, p. 99.
4. www.sasolwax.com/paraffin_wax.html.
Accessed
August 29, 2016.
5. Mordkovich, V.Z., Sineva, L.V., Kul’chakovskaya, E.V.,
and Asalieva, E.Yu., Katal. Prom–sti., 2015, no. 5, p. 23.
6. Ellepola, J., Thijssen, N., Grievink, J., Baak, G.,
Avhale, A., and Schijndel, J.V., Comput. Chem. Eng.,
2012, vol. 42, p. 2.
7. Krylova, A.Yu. and Kozyukov, E.A., Solid Fuel Chem.,
2007, vol. 41, no. 6, p. 335.
8. Grigor'ev, D.A., Neft’, Gaz Biznes, 2011, no. 3, p. 51.
(3) We observed a considerable promoting effect of
aluminum oxide in an amount to 1 wt % on the size of
cobalt crystallites and, as a consequence, on the activity
and selectivity of the catalyst.
(4) We tested the on-stream stability of the Co–
1Al2O3/SiO2 catalytic system at a pilot plant for 1000 h
of continuous operation; this made it possible to
obtain long-chain hydrocarbons with a chain growth
probability of 0.94.
9. Savost'yanov, A.P., Narochnyi, G.B., Yakovenko, R.E.,
Bakun, V.G., and Zemlyakov, N.D., Catal. Ind., 2014,
vol. 6, no. 4, p. 292.
10. Yakovenko, R.E. and Narochnyi, G.B., Bakun. V.G.,
Astakhov A.V., Izv. Vyssh. Uchebn. Zaved,. Severo-
Kavkazskii Region, Tekh. Nauki, 2014, no. 6, p. 92.
11. Savost'yanov, A.P., Narochnyi, G.B., Yakovenko, R.E.,
Astakhov, A.V., Zemlyakov, N.D., Merkin, A.A., and
Komarov, A.A., Catal. Ind., 2014, vol. 6, no. 3, p. 212.
12. Narochnyi, G.B., Yakovenko, R.E., Savost’yanov, A.P.,
and Bakun, V.G., Catal. Ind., 2016, vol. 8, no. 2, p. 139.
ACKNOWLEDGMENTS
13. PDF-2, The Powder Diffraction File, International
Center for Diffraction Data (ICCD), PDF-2 Release
2012. http://www.icdd.com.
This work was supported by the Russian Science
Foundation (project no. 14-23-00078) and was per-
formed with the use of laboratory equipment of the
Nanotechnology Shared Facilities at the Platov South-
Russian State Polytechnic University (NPI).
14. The Rietveld Method, Young, R.A., Ed., Oxford: Oxford
Univ. Press, 1995.
15. Satterfield, C., Heterogeneous Catalysis in Practice,
New York: McGraw-Hill, 1980.
16. Boreskov, G.K., Geterogennyi kataliz (Heterogeneous
REFERENCES
Catalysis), Moscow: Nauka, 1986.
1. David, A., Wood, A., and Chikezie, N., J. Nat. Gas Sci. 17. Krylov, O.V., Geterogennyi kataliz (Heterogeneous
Eng., 2012, vol. 9, p. 196. Catalysis), Moscow: Akademkniga, 2004.
KINETICS AND CATALYSIS
Vol. 58
No. 1
2017

INDUSTRIAL CATALYST FOR THE SELECTIVE FISCHER–TROPSCH SYNTHESIS
91
18. Slivinskii, E.V., Kliger, G.A., Kuz’min, A.E., Abram- 31. Girardon, J.-S., Quinet, E., Griboval-Constant, A.,
ova, A.V., and Kulikova, E.A., Ross. Khim. Zh., 2003,
vol. 47, no. 6, p. 12.
Chernavskii, P.A., Gengembre, L., and Khodakov, A.Y.,
J. Catal., 2007, vol. 248, p. 143.
19. Kibby, C., Catal. Today, 2013, vol. 215, p. 131.
20. Sun, J., Catal. Sci. Technol., 2014, vol. 4, p. 1260.
21. Yao, M., Catal. Sci. Technol., 2015, vol. 5, p. 2821.
22. Girardon, J.S., J. Catal., 2005, vol. 230, p. 339.
23. Chernavskii, P.A., Chu, Vei., Khodakov, A.Yu.,
Pankina, G.V., and Peskov, N.V., Russ. J. Phys. Chem. A,
2008, vol. 82, no. 6, p. 951.
32. Beer, M., Kunene, A., Nabaho, D., Claeys, M., and
Steen, E., J. South Afr. Inst. Min. Metall., 2014, vol. 114,
no. 2, p. 157.
33. París, L., Lopez, J., Barrientos, F., and Pardo, M., in
Catalysis, vol. 27, Spivey, J.J., Han, Y.-F., and
Dooley, K.M., Eds., Cambridge: R. Soc. Chem.,
2015, ch. 3, p. 62.
24. Dzis'ko, V.A., Karnaukhov, A.P., and Tarasova, D.V.,
Fiziko-khimicheskie osnovy sinteza okisnykh kataliza-
torov (Physicochemical Foundations of the Synthesis of
Oxide Catalysts), Novosibirsk: Nauka, 1978.
25. Kheifets, L.I. and Neimark, A.V., Mnogofaznye prot-
sessy v poristykh sredakh (Multiphase Processes in
Porous Media), Moscow: Khimiya, 1982.
34. Jahangiri, H., Bennett, J., Mahjoubi, P., Wilson, K.,
and Gu, S., Catal. Sci. Technol., 2014, vol. 4, p. 2210.
35. Johnson, G.R. and Bell, A.T., ACS Catal., 2016, vol. 6,
p. 100.
36. Zhang, Y., Nagamori, S., Hinchiranan, S., Vitidsant, T.,
and Tsubaki, N., Energy Fuels, 2006, vol. 20, p. 417.
26. Chu, W., Xu, J., Hong, J., Lin, T., and Khodakov, A., 37. Pei, Y., Ding, Y., Zhu, H., Zang, J., Song, X., Dongm, W.,
Catal. Today, 2015, vol. 256, p. 41.
Wang, T., and Lu, Y., Catal. Lett., 2014, vol. 144, p. 1433.
27. Guo, P., Saw, W., Eyk, P., Ashman, P., Nathan, G.,
38. Peng, X., Cheng, K., Kang, J., Gu, B., Yu, X., Zhang, Q.,
and Stechel, E., Energy Procedia, 2015, vol. 69, p. 1770.
and Wang, Y., Angew. Chem., 2015, vol. 127, p. 4636.
28. Prieto, G., Concepción, P., Murciano, R., and
39. Sedran, U.A. and Figoli, N.S., Appl. Catal., 1985,
Martínez, A., J. Catal., 2013, vol. 302, p. 37.
vol.19, p. 317.
29. Tsubaki, N., Sun, S., and Fujimoto, K., J. Catal., 2001,
40. Kruse, N., Machoke, A.G., Schwieger, W., and Gut-
vol. 199, p. 236.
tel, R., ChemCatChem, 2015, vol. 7, p. 1018.
30. Chu, W., Chernavskii, P.A., Gengembre, L.,
Pankina, G.A., Fongarland, P., and Khodakov, A.Y.,
J. Catal., 2007, vol. 252, p. 215.
Translated by V. Makhlyarchuk
KINETICS AND CATALYSIS Vol. 58 No. 1 2017