Direct hydrogenation of CO2 on deposited iron-containing catalysts under supercritical conditions

Article abstract of DOI:10.1016/j.mencom.2018.03.012

The hydrogenation of CO₂ in the supercritical state on supported Fe/TiO₂ catalysts (0.5–1% iron) at 350–550 °C affords carbon monoxide and methane, the conversion of CO₂ approaching 24%. The work demonstrates the possibili

Full text of DOI:10.1016/j.mencom.2018.03.012

Mendeleev
Communications
Mendeleev Commun., 2018, 28, 147–149
Direct hydrogenation of CO2 on deposited iron-containing
catalysts under supercritical conditions
Nikolay D. Evdokimenko,a,b Alexander L. Kustov,*a,b,c Konstantin O. Kim,c
Maria S. Igoninac and Leonid M. Kustova,b,c
a N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 119991 Moscow, Russian Federation.
E-mail: kyst@list.ru, nikolayevdokimenko@bk.ru
b National University of Science and Technology ‘MISIS’, 119049 Moscow, Russian Federation
c Department of Chemistry, M. V. Lomonosov Moscow State University, 119991 Moscow, Russian Federation
DOI: 10.1016/j.mencom.2018.03.012
1st step
2nd step
CO2 + H2
CO + H2O
The hydrogenation of CO2 in the supercritical state on
supported Fe/TiO2 catalysts (0.5–1% iron) at 350–550 °C
affords carbon monoxide and methane, the conversion of CO2
approaching 24%. The work demonstrates the possibility of
significant increase in the productivity of CO2 hydrogenation
process and prolongation of the working time of the catalyst.
Iron catalyst
CO + H2 CnH2n + 2 + H2O
Iron catalyst
The constant significant growth of carbon dioxide in the Earth’s
atmosphere is a challenge to develop processes of utilization and
conversion of CO2. The main problem of CO2 conversion is the
inertness of its molecule.1 One of the most promising methods
of CO2 conversion is its direct catalytic heterogeneous hydro-
genation into valuable products such as hydrocarbons.2 The
possible routes for this process are presented by reactions (1)–(3).
catalysts, where CO adsorption rate is higher than that of CO2. The
conversion of CO2 on iron-containing catalysts reaches 40%.9–12
The common drawbacks of the most processes described are
their low productivity, as well as the short lifetime of the catalyst.
In the course of the reaction, the solid products are deposited
on the surface of the catalyst and block its active centers, followed
by formation of inactive carbide phases of iron such as Fe3C.4
In the gas phase, the productivity is limited by slow diffusion of
reactants to the surface of the active sites and the diffusion of
products from these centers. In liquid solvents, the productivity is
limited by the small concentration of dissolved components in
the reaction medium. To change the situation, the process can be
conducted under supercritical conditions for CO2 known as the
safe and environmentally friendly solvent, which corresponds to
the concept of ‘green’ chemistry.13–16 Any quantities of nonpolar
substances, such as hydrogen, can be dissolved in the supercritical
CO2. Under these conditions, the life of the catalyst is prolonged
due to improvement of the diffusion of solid reaction products
from active sites of the catalyst, and the productivity of the process
is significantly increased by higher density of the reaction
medium.17–19
CO2 + H2 ® CO + H2O, DRH573K = 38 kJ mol–1
CO + 2H2 ® CnHm + H2O, DRH573K = –166 kJ mol–1
CO2 + 3H2 ® CnHm + 2H2O, DRH573K = –128 kJ mol–1
(1)
(2)
(3)
In this way, CO2 can be converted into the products directly or
indirectly, for example, through the formation of CO or methanol.
Nowadays the process of CO2 methanation (the Sabatier reaction)
is known. The target product can be selected based on economic
reasoning. The most simple and economically profitable is the
conversion of CO2 into synthesis gas by reaction (1), defined as
the reverse water shift reaction,2 largely on using iron-based
catalysts. In this case, the conversion of CO2 is a two-stage
reaction sequence proceeding through the step of CO formation.3
The most probable mechanism for formation of hydrocarbons on
these catalysts4 involves CO formation followed by the Fischer–
Tropsch reaction. It is also known that the metallic phase of iron
does not emerge under the conditions of CO2 conversion and
iron remains in an oxidized form.5,6 Obviously, the Fischer–
Tropsch principles may be applied to the problem in question.
For example, if the CO2 conversion is aimed at producing hydro-
carbons, the selectivity factor will be the ratio of the various
phases of iron responsible for various stages of this process on
the surface of catalyst. It can be speculated that iron carbides are
responsible for chain growth, and iron oxides and metallic iron
are responsible for hydrogenation processes.7,8 When synthesis
gas is the target reaction product, iron-containing catalysts also
proved to be effective due to a high rate of CO2 adsorption as
compared with that of CO, in contrast to, for example, cobalt
Therefore, to investigate direct hydrogenation of CO2 under
supercritical conditions on iron-containing catalysts with different
iron contents is a good challenge.
The samples of catalysts were prepared by incipient wetness
impregnation of the support† with an iron nitrate solution in
distilled water of the appropriate concentration. The wet impregnated
samples were left to dry at room temperature for 48 h and then
†
For the catalyst support, granular titanium(iv) oxide was prepared by
the modification of the anatase (Degussa AG). The support was ground in
a ceramic mortar and the fraction of 0.25–0.5 mm was selected, ideally
suitable for carrying out the catalytic experiment. Then, the support was
dried at 120°C for 2 h. The surface of the resulting fraction, measured in
air, was 71 m2 g–1, and the water capacity was 1.3 ml per gram of the
support.
© 2018 Mendeleev Communications. Published by ELSEVIER B.V.
on behalf of the N. D. Zelinsky Institute of Organic Chemistry of the
Russian Academy of Sciences.
– 147 –

Mendeleev Commun., 2018, 28, 147–149
they were calcined in a quartz reactor in air flow of 50 ml min–1
4
at a temperature of 500° C for 4 h. The activation of the catalysts
was carried out in situ in the catalytic unit in a hydrogen flow of
60 ml min–1 at 500°C and the pressure of 40 atm for 4 h. Using
this method, a series of iron-containing catalysts with the
different mass concentrations of iron atoms (0.1, 0.25, 0.5, 1, 2.5
and 5%) was prepared.
3
2
It was established that catalytic activity of the samples was
not affected by the activation of the catalyst carried out before
the start of the experiment.‡ Both activated and non-activated
samples differed only in the initial time of reaching the equilibrium.
For the activated samples, the equilibrium was reached in 3–4 h
from the beginning at 300°C, while for the non-activated ones
this parameter was 6–8 h. This allowed us to assume that the
iron phase on the catalyst was not metallic it and changed on
movement to the point of equilibrium state between the reactants
and the products. The fact that iron remains oxidized and does
not transform to metal finds analogies in the literature.6,7
The main gaseous products formed on all samples in the entire
temperature range were CO and methane. The selectivity of the
conversion of CO2 to CO on each sample was similar under the
same contact time (~55 s) and amounted to 67% calculated on
the gaseous products at 350°C, and decreased slightly to 60%
when the temperature was raised to 550°C (Figure 1). The other
gaseous products of the reaction contain mainly methane. At the
temperatures above 400°C, ethane and propane were detected in
low quantities (£3%). Since the hydrogenation of CO2 on iron-
containing catalysts is described as a two-stage process involving
the step of CO formation, it may be assumed that, with increasing
temperature, the formed CO is converted to hydrocarbons, which
explains the decrease in the selectivity of CO formation and the
appearance of low amounts of C2–C3 hydrocarbons.3 However,
there is little information on the reaction mechanism under
supercritical conditions on iron-containing catalysts of this type,
that is why this question requires further study. It may only be
concluded that the selectivity of CO formation does not depend
on the concentration of the deposited phase.
1
20 24 28 32 36 40 44 48 52 56 60
2q/deg
Figure 2 XRD catalyst diagram 10% Fe/TiO2: (1) TiO2 anatase, (2) catalyst
before reaction, (3) catalyst after reaction, (4) catalyst after reaction and DTA.
Figure 2 shows the typical XRD pattern of the 10% Fe/TiO2
samples and the support. Initially, anatase lines predominate on
the pattern (2q = 25.23°).§ After preparing the catalysts and their
air calcination at 500°C, the intensity of anatase peaks decreases
in respect to that of rutile peaks (2q = 27.45°). One can expect
increase in intensities of iron(iii) oxide modification hematite
(2q = 33.20° and 35.68°) which can form at high calcination
temperatures. However, these characteristic peaks are absent and
start to appear only after several hours of the reactor work, with
negligible intensities. DTA study of air calcined spent catalyst
does not reveal significant changes of these peaks. Unfortunately,
characteristic peaks of other iron oxides (magnetite 2q = 35.58°
and 43.25°, maghemite 2q = 45.211° and 41.715°) overlap with
those of titanium oxides. Thus, it can be assumed than during
preparation of the catalysts, amorphous iron oxides are formed
which are responsible for the reverse water shift reaction and the
formation of CO. In the course of catalysis process, a crystalline
phase of iron oxide-hematite starts to form.
The contact time influenced significantly the dependence of
the gaseous product pattern on the catalyst composition. Thus,
on the sample of 2.5%, three regimes of the reaction were
investigated at different contact times maintaining the amount
and composition of the initial flow. The contact time was
controlled by the amount of the catalyst loading. Three points
with the contact times of 40, 55 and 70 s were thus examined.
The experiments showed that the conversion did not change when
the contact time was varied, but the composition of the catalytic
gases changed significantly. The selectivity of CO formation
calculated on the gaseous products of the reaction decreased
from 72 to 60% with raising the contact time from 40 to 70 s.
Thus, the selectivity of hydrocarbon formation increased. This
observation was expected, since it is known that the adsorption
and, accordingly, the activation of CO2 molecule on iron-con-
taining catalysts proceeds faster than those of CO molecule.7 It
can be assumed that during the formation of hydrocarbons, the
limiting step is the adsorption and the activation of CO molecule
produced in the first stage of the hydrogenation of CO2.
75
70
65
1
60
2
55
3
50
300
400
T/°C
500
600
Figure 1 Selectivity of CO formation at different contact times on Fe/TiO2
catalysts with an iron content of 0.1% to 5% and contact time of (1) 40, (2)
55 and (3) 70 s.
The conversion of CO2 (which was close to the total yield of
the products) depended on the concentration of the deposited
phase only in the concentration range from 0.1 to 0.5% of iron
and then remained approximately constant in the examined range
of concentrations (Figure 3). On the samples containing 0.5,
1, 2.5 and 5% of iron, the maximum of CO2 conversion reached
22–23%, while on the sample containing 0.25 and 0.1% of iron
the maximum conversion was 20 and 17%, respectively. This is
‡
The reaction of CO2 with H2 was carried out in a flow catalytic unit
with the fixed-bed catalyst. The H2 :CO2 ratio was 1:1. The catalyst
loading was 1 g. The reaction was investigated in the temperature range
of 300 to 600°C. Carbon dioxide in the liquid state was supplied by the
liquid syringe pump at the pressure of 90 atm, hydrogen was supplied at
the pressure of the experiment by a mass flow controller (Bronhorst).
The pressure in the reactor (90 atm) was controlled by the valve manu-
factured in N. D. Zelinsky Institute of Organic Chemistry of the Russian
Academy of Sciences. Gas products were analyzed online on a Crystal-5000
chromatograph using thermal conductivity detectors and packed columns
Heysep Q and Zeolite CaA. The analysis of liquid products was performed
on a chromatograph with the mass detector and on a chromatograph with
the flame ionization detector and capillary column OV-101.
§
Powder X-ray diffraction data were obtained on a DRON3 diffracto-
meter using CuKa radiation. The samples were scanned in the region
of 2q = 35–65° with a scanning rate 1 deg min–1. The crystallite size was
calculated based on the broadening of the diffraction maxima.
– 148 –

Mendeleev Commun., 2018, 28, 147–149
DTA plots of 10% Fe/TiO2. The total weight loss of 1.1% is
22
18
14
10
6
5
small enough, however it is possible to distinguish two peaks on
DTG curve at 80 and 290°C. These weight losses may be assigned
to the removal of adsorbed water and probably to combustion of
residual deposited carbon. On account of insignificant effects
DTA curve is not informative. Thus we do not observe coking of
the catalyst under the reaction conditions.
Another advantage of carrying out the process under super-
critical conditions of CO2 is prolongation of operation time of
the catalyst through the prevention of the coke formation on the
catalyst surface by increasing the diffusion of solid products of
the reaction from the active centers of the catalyst.
4
3
2
1
2
0
2
4
6
Iron content in the catalysts (wt%)
Figure 3 Conversion of CO2 into gaseous products on Fe/TiO2 catalysts
with an iron content of 0.1% to 5% at different temperatures: (1) 350, (2) 400,
(3) 450, (4) 500 and (5) 550°C.
We are grateful to I. V. Mishin for assistance in X-ray phase
analysis of catalyst samples and V. D. Nissenbaum for the TG-DTA
studies (N. D. Zelinsky Institute of Organic Chemistry).
This work was supported by the Ministry of Education and
Science of the Russian Federation (project RFMEFI61615X0041).
probably caused by the morphology and the number of the active
catalyst centers. One may also assume that at lower iron con-
centration in the catalyst (below 0.5%), the conversion is lower
mainly due to the decrease in the number of the active centers of
CO2 hydrogenation. At the same time, the variation of the contact
time does not affect the conversion of CO2 but changes the
product distribution.
References
1 T. E. Müller, W. Leitner, P. Markewitz and W. Kuckshinrichs, in Carbon
Capture, Storage and Use, eds. W. Kuckshinrichs and J.-F. Hake,
Springer, 2015, pp. 67–100.
The investigated catalysts showed high productivity towards
CO, namely, 0.66 g per pram of the catalyst per hour (for the
catalysts with the iron concentration of 0.5 to 5% at 500°C).
This value is higher nearly an order of the productivity of CO
formation as compared with the experiments without super-
critical CO2.3,13 For the catalysts with the iron concentration
less than 0.5%, the conversion is lower. The productivity on
CO significantly depends on the temperature. The amount of
produced CO is extremely low at 300°C, and grows as the
temperature approaches 500 °C. With further raising the tem-
perature, there is a tendency to reduce CO productivity and
to increase the yield of hydrocarbons. Therefore, performing the
hydrogenation of CO2 under supercritical conditions significantly
increases the productivity of the process, whilst the catalyst
activity remains practically unchanged. After the completion of
the experiment, there was practically no coking of the catalyst
surface.
2 S. Tada, I. Thiel, H. K. Lo and C. Copéret, Chimia, 2015, 69 (12), 759.
3 W. Wang, S. Wang, X. Ma and J. Gong, Chem. Soc. Rev., 2011, 40, 3703.
4 B. Hu, C. Guild and S. L. Suib, J. CO2 Utilization, 2013, 1, 18.
5 S. Saeide, A. S. Amin and M. R. Rahimpour, J. CO2 Utilization, 2014,
5, 66.
6 D. H. Kim, S. W. Han, H. S. Yoon and Y. D. Kim, J. Ind. Eng. Chem.,
2015, 23, 67.
7 J.-F. Lee, W.-S. Chern, M.-D. Lee and T.-Y. Dong, Can. J. Chem. Eng.,
1992, 70, 511.
8 A. A. Shesterkina, O. A. Kirichenko, L. M. Kozlova, G. I. Kapustin,
I. V. Mishin, A. A. Strelkova and L. M. Kustov, Mendeleev Commun.,
2016, 26, 228.
9 C. G. Visconti, M. Martinelli, L. Falbo, L. Fratalocchi and L. Lietti, Catal.
Today, 2016, 227, 161.
10 M. Niemelä and M. Nokkosmäki, Catal. Today, 2005, 100, 269.
11 M. Albrecht, U. Rodemerck, M. Schneider, M. Bröring, D. Baabe and
E. V. Kondratenko, Appl. Catal., B, 2017, 204, 119.
12 L. M. Kustov and A. L. Tarasov, Mendeleev Commun., 2014, 24, 349.
13 N. Utsis, R. Vidruk-Nehemya, M. V. Landau and M. Herskowitz, Faraday
Discuss., 2016, 188. 545.
14 M. N. da Ponte, J. Supercrit. Fluids, 2009, 47, 344.
15 R. Liu, P. Zhang, S. Zhang, T. Yan, J. Xin and X. Zhang, Rev. Chem.
Eng., 2016, 32, 587.
The sample discharged after several hours of work was
examined by TG-DTA method.¶ Figure 4 shows the TG-DTG-
0.5
exo
16 E. Ramsey, Q. Sun, Z. Zhang, C. Zhang and W. Gou, J. Environ. Sci.,
2009, 21, 720.
17 A. Kruse and H. Vogel, Chem. Eng. Technol., 2008, 31, 23.
18 V. I. Bogdan, A. E. Koklin, S. A. Nikolaev and L. M. Kustov, Top Catal.,
2016, 59, 1104.
DTA
0.0
DTG
–0.5
19 V. P.Ananikov, D. B. Eremin, S.A.Yakukhnov,A. D. Dilman,V.V. Levin,
M. P. Egorov, S. S. Karlov, L. M. Kustov, A. L. Tarasov, A. A. Greish,
A. A. Shesterkina, A. M. Sakharov, Z. N. Nysenko, A. B. Sheremetev,
A. Yu. Stakheev, I. S. Mashkovsky, A. Yu. Sukhorukov, S. L. Ioffe,
A. O. Terent’ev, V. A. Vil’, Yu. V. Tomilov, R. A. Novikov, S. G. Zlotin,
A. S. Kucherenko, N. E. Ustyuzhanina, V. B. Krylov, Yu. E. Tsvetkov,
M. L. Gening and N. E. Nifantiev, Mendeleev Communications, 2017,
27, 425.
TG
–1.0
0
200
400
600
800
T/°C
Figure 4 TG-DTG-DTA curves of sample 10% Fe/TiO2 after operation.
¶
Thermogravimetric differential thermal analysis (TG-DTA) of dried
samples was performed using a thermoanalytical Derivatograph-C instrument
(MOM). The sample (15 mg of the bulk or 20 mg of the supported sample)
was placed in an alund crucible and heated in air from 20 to 500°C at a
heating rate 10 K min–1.
Received: 25th October 2017; Com. 17/5380
– 149 –