Sodium-Containing Spinel Zinc Ferrite as a Catalyst Precursor for the Selective Synthesis of Liquid Hydrocarbon Fuels

Article abstract of DOI:10.1002/cssc.201701437

A microwave-assisted hydrothermal synthesis produces ZnFe₂O₄ containing Na residue as a precursor to a CO₂ hydrogenation catalyst that displays high CO₂ conversion and high selectivity to liquid hydrocarbon prod

Full text of DOI:10.1002/cssc.201701437

DOI: 10.1002/cssc.201701437
Full Papers
Sodium-Containing Spinel Zinc Ferrite as a Catalyst
Precursor for the Selective Synthesis of Liquid
Hydrocarbon Fuels
Yo Han Choi,[a] Eun Cheol Ra,[b] Eun Hyup Kim,[b] Kwang Young Kim,[b] Youn Jeong Jang,[a]
Kyeong-Nam Kang,[b] Sun Hee Choi,[a] Ji-Hyun Jang,[b] and Jae Sung Lee*[b]
A
microwave-assisted hydrothermal synthesis produces
persed iron particles with Zn serving as a structural promoter.
A profound effect of the residual Na as an electronic promoter
is also observed, which improves the selectivity for C5+ hydro-
carbons and olefins. The ZnFe2O4-derived catalyst exhibits ex-
cellent performance in the CO2 Fischer–Tropsch reaction as it
forms the active Hꢀgg iron carbide (c-Fe5C2) phase readily
through the in situ carburization of iron.
ZnFe2O4 containing Na residue as a precursor to a CO2 hydro-
genation catalyst that displays high CO2 conversion and high
selectivity to liquid hydrocarbon products in the gasoline and
diesel range with high olefin-to-paraffin ratios. Compared to
reference catalysts derived from Fe2O3 and a ZnO–Fe2O3 physi-
cal mixture, the ZnFe2O4-derived catalyst contains well-dis-
Introduction
CO2 þ H2 Ð CO þ H2O DH298 ¼ þ41 kJ molꢀ1
CO þ 2 H2 ! ðCH2Þn þ H2O DH298 ¼ ꢀ152 kJ molꢀ1
ð1Þ
ð2Þ
Global warming caused by the greenhouse effect of anthropo-
genic CO2 in the atmosphere is a serious issue that threatens
the future of humankind.[1] Carbon capture and storage (CCS)
has been considered as an effective and aggressive measure to
counter the problem.[2] However, the storage of CO2 in geologi-
cal or ocean reservoirs has serious drawbacks such as possible
leakage, long-term liability, and the availability of sufficient
storage capacity in many regions of the world. By contrast, in
carbon capture and utilization (CCU), the captured CO2 is con-
verted into useful products such as fuels, chemicals, plastics,
and alternative building materials.[3] Among the many possible
CCU products, transportation fuels are the most attractive be-
cause of their huge market size, which could accommodate
the large amount of CO2 released from industrial vent
streams.[4–6] In particular, liquid hydrocarbon fuels such as gaso-
line, diesel, and jet fuel are convenient for transportation and
storage and readily adaptable to the current distribution infra-
structure.
Compared to the conventional CO FT reaction with synthesis
gas, CO2 hydrogenation requires one more H2 molecule per
molecule of CO2 and produces more of the water byproduct,
which is a deactivating agent for FT catalysts. In addition, the
thermodynamic barrier of the RWGS reaction limits the CO2
conversion below 3008C and 10 bar.[8] Thus, the major chal-
lenges in the hydrogenation of CO2 to liquid fuels are the acti-
vation of thermodynamically and kinetically stable CO2 mole-
cules and the control of the product selectivity toward heavy
hydrocarbons instead of undesired methane. Owing to the
need for RWGS activity, catalyst selection is confined to Fe in-
stead of other common FT catalysts such as Co, Ni, and Ru.[1]
The hydrogenation of CO2 over iron-based FT catalysts under
typical reaction conditions produces mainly methane. There
have been extensive studies to produce more useful heavier
hydrocarbon products. For example, an iron catalyst derived
from Fe2O3 synthesized by a template-assisted method yielded
a high CO2 conversion and a good C2–C4 hydrocarbon selectivi-
ty in CO2 hydrogenation at 15 bar and 627 K with a H2/CO2
ratio of 3:1.[9] In another recent study, MnFeOx nanocomposites
were active and selective toward C2–C4 synthesis.[10] To improve
CO2 conversions and heavy-hydrocarbon selectivity, iron-based
catalysts are commonly modified with a metal oxide (a-Al2O3
or TiO2), carbon supports such as N-doped materials, and alka-
line promoters.[8,11] In spite of these efforts, selectivity to liquid
products heavier than C5 (C5+) remained below 40%. Recently,
our group reported that a CuꢀFe catalyst derived from delafos-
site CuFeO2 could produce C5+ products with selectivity great-
er than 60%.[12] Since then, the direct conversion of CO2 to gas-
oline fuel over Na-Fe3O4/HZSM-5 as catalyst was reported.[13] As
these catalysts result in the selective formation of heavy hydro-
The synthesis of hydrocarbon fuels by CO2 hydrogenation in-
volves two consecutive reactions, namely, the reverse water–
gas shift (RWGS) and Fischer–Tropsch (FT) reactions [Eqs. (1)
and (2), respectively].[7]
[a] Y. H. Choi, Dr. Y. J. Jang, Dr. S. H. Choi
Division of Advanced Nuclear Engineering
Pohang Accelerator Laboratory
Pohang University of Science and Technology (POSTECH)
Pohang, 790-784 (Korea)
[b] E. C. Ra, E. H. Kim, K. Y. Kim, K.-N. Kang, Prof. J.-H. Jang, Prof. J. S. Lee
School of Energy and Chemical Engineering
Ulsan National Institute of Science and Technology (UNIST)
Ulsan, 44919 (Korea)
E-mail: jlee1234@unist.ac.kr
Supporting Information and the ORCID identification number(s) for the
author(s) of this article can be found under https://doi.org/10.1002/
cssc.201701437.
ChemSusChem 2017, 10, 1 – 8
1
ꢁ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
carbons in a similar fashion to the CO FT catalysts, they could
be called “CO2 FT catalysts”. The catalysts form Hꢀgg iron car-
bide (c-Fe5C2) readily through the in situ carburization of Fe
during CO2 hydrogenation, and c-Fe5C2 is the known active
phase of iron CO FT catalysts.
In the presented work, we synthesize Na-containing ZnFe2O4
as a new Fe–Zn catalyst precursor through a simple micro-
wave-assisted hydrothermal method. During the pre-reduction
step with H2, ZnFe2O4 is reduced to metallic Fe decorated with
Zn nanoparticles containing Na residue derived from the syn-
thesis step. The Zn serves as a structural promoter to suppress
the growth of Fe particles, whereas the Na serves as an elec-
tronic promoter to modify the catalytic activity and selectivity
of the Fe particles for CO2 hydrogenation. The beneficial effect
of the Zn and Na promoters results in the facile in situ forma-
tion of the c-Fe5C2 phase. As a result, our new Fe–Zn catalyst
derived from ZnFe2O4 (ex-ZnFe2O4) overcomes the common
barriers of CO2 FT catalysts, and high CO2 conversions as well
as high selectivities to liquid fuels can be achieved.
Results and Discussion
Figure 1. a) Powder XRD patterns of Fe2O3, ZnO–Fe2O3, ZnFe2O4, and Na-free
ZnFe2O4. b) XRD patterns of the catalysts after the CO2 FT reaction for 48 h.
SEM images of ZnFe2O4 c) before and d) after the reaction.
Physicochemical properties of the catalysts
We synthesized ZnFe2O4 by a microwave-assisted hydrothermal
method to obtain small particle sizes and high surface areas,
as described in the Experimental Section. As reference catalyst
precursors, Fe2O3, a ZnO–Fe2O3 physical mixture (Zn/Fe molar
ratio 1:2), and Na-free ZnFe2O4 were also synthesized. The
powder XRD patterns (Figure 1a) exhibit intense diffraction
peaks at 2q=34 and 368 for Fe2O3 and broad peaks at 2q=30
and 358 for the spinel ZnFe2O4 phase. The physically mixed
Fe2O3–ZnO sample shows the peaks of both ZnO and Fe2O3.
All catalyst precursors were pre-reduced in a H2 flow at 4008C
for 2 h immediately before the reactions (Figure S1, Supporting
Information). They showed different degrees of reduction to
metallic Fe and Zn phases. The ex-ZnFe2O4 and ZnO–Fe2O3-de-
rived (ex-ZnO–Fe2O3) catalysts both reveal metallic Fe and FeO
at 2q=41.7 and 44.68, respectively. The XRD pattern of the re-
duced Fe2O3 sample (ex-Fe2O3) indicates the formation of pure
iron metal. An interaction between iron and zinc apparently
suppresses the complete reduction of the oxide precursors
under the reduction conditions.
maintained after the reduction (Figure 4c). This microwave-as-
sisted hydrothermal method allows the swift synthesis of small
and stable ZnFe2O4 particles at a much lower temperature of
1808C compared with those for other methods involving a
high-temperature calcination step.[14]
Catalytic CO2 hydrogenation
The CO2 hydrogenations over the prepared catalysts were per-
formed in a fixed-bed reactor with a continuous feed of H2/
CO2 (3:1) and
a gas hourly space velocity (GHSV) of
1800 mLgꢀ1 hꢀ1 at 3408C and 10 bar. ZnO alone as a reference
catalyst caused no CO2 conversion. As shown in Table 1 and
Figure 2, reduced Fe2O3 shows a performance typical of iron FT
catalysts in CO2 hydrogenation with a good CO2 conversion
and fair selectivity to C2–C4 hydrocarbons. The catalyst derived
from the ZnO–Fe2O3 mixture exhibits a 36% increase in CO2
conversion from that for bare Fe2O3. The increased CO2 conver-
sion accompanies increased methane formation at the expense
of heavier hydrocarbons. Our new catalyst derived from
ZnFe2O4 synthesized by the microwave-assisted hydrothermal
method exhibits a further improved CO2 conversion (by 86%
relative to that of bare Fe2O3). In terms of the reaction rates
[expressed as Fe time yield (FTY), i.e., moles of CO2 converted
per g of Fe in the catalyst per second], the ZnFe2O4-derived
The textures catalyst were examined by scanning electron
microscopy (SEM) and N2 sorption (Figure S2 and Table S1).
The Fe2O3 and ZnO–Fe2O3 catalysts are composed of small
nanoparticles of approximately 200–300 nm (SEM images) with
low surface areas (8 and 20.4 m2 gꢀ1, respectively) and small
pore volumes (0.17 and 0.19 cm3 gꢀ1, respectively; Figure S3).
The SEM image of reduced Fe2O3 shows severe agglomeration
(Figure S4a). As a result of the addition of ZnO to Fe2O3, the
Fe2O3 particle size becomes smaller and the agglomeration
caused by reduction is not significant (Figures S3b and S4b).
The ZnFe2O4 synthesized by a microwave-assisted hydrother-
mal method has a much higher surface area (119 m2 gꢀ1) and
pore volume (0.33 cm3 gꢀ1). Its SEM image (Figure 1c) shows
clusters of small ZnFe2O4 particles, and the small size is almost
catalyst shows a significantly higher FTY (3.1 mmolgFeꢀ1 sꢀ1
)
than those of the catalysts derived from bare Fe2O3
(1.3 mmolgFeꢀ1 sꢀ1) and ZnO–Fe2O3 (2.7 mmolgFeꢀ1 sꢀ1). However,
the most conspicuous characteristic of the ZnFe2O4-derived
catalyst is the selectivity pattern of the hydrocarbon products
&
ChemSusChem 2017, 10, 1 – 8
www.chemsuschem.org
2
ꢁ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Table 1. Steady-state FTY, CO2 conversion, and product selectivity of various catalysts in CO2 hydrogenation.[a]
Precursor
FTY[b]
CO2 conversion
[%]
CO selectivity
[%]
CO-free HC selectivity [%]
O/P
C1
C2
C3
C4
C5+
Fe2O3
ZnO–Fe2O3
ZnFe2O4, Na-free
1.3
2.7
3.0
3.7
18.2
24.7
27.8
34.0
20.5
23.2
21.9
11.7
40.5
53.1
43.6
9.7
21.4
26.0
18.6
8.0
19.4
13.5
17.8
13.0
10.1
4.5
10.5
10.8
8.6
2.9
9.5
0.16
0.09
0.4
[c]
ZnFe2O4
58.5
11.3
[a] Reaction conditions: 3408C, 10 bar, 1800 mLgꢀ1 hꢀ1, H2/CO2 =3. [b] FTY in mmolCO gFeꢀ1 sꢀ1. [c] ZnFe2O4 containing 0.08 wt% of Na.
2
Another unique characteristic of our ex-ZnFe2O4
catalyst in CO2 hydrogenation is the high olefin/par-
affin ratio (O/P) of the C2–C4 products of 11.3, which
is 70 times higher than that of the ex-Fe2O3 catalyst
(0.16). For FT synthesis over iron-based catalysts, the
alkaline content of the catalyst is generally the most
important parameter for the determination of the O/
P ratio.[16] The hydrothermal reaction solution for the
preparation of the ZnFe2O4 catalyst contained NaOH
to produce basic conditions, and the prepared cata-
lyst contains Na residue. To study the effect of Na,
the residual Na content was varied systematically
through the variation of the amount of washing after
the hydrothermal step. Thus, inductively coupled
plasma optical emission spectroscopy (ICP-OES) anal-
ysis of the prepared catalyst indicate that the Na con-
tent decreased from 0.7 wt% (mild washing with
500 mL of water) to an undetectable level (0%, thor-
ough washing with 6 L of water). As demonstrated in
Figure 2b, considerable changes to the O/P ratio and
CO selectivity occurred at 0.08 wt% Na. The results
are consistent with the well-established roles of Na
as (i) an electronic promoter for Fe FT catalysts to
shift product distributions to heavier hydrocarbons,
(ii) a promotor of CO2 adsorption through the provi-
sion of basic sites,[17] and (iii) an inhibitor of surface
hydrogenation.[16,18,19] The Na-free (Na<0.03 wt%)
ZnFe2O4-derived catalyst shows a decreased CO2 con-
version, a lower C5+ selectivity, and a smaller O/P
ratio than the Na-containing ex-ZnFe2O4 (Table 1 and
Figure 2. a) FTY versus time on stream for Fe2O3, ZnFe2O4, and Na-free ZnFe2O4 catalysts.
Reaction conditions: 3408C, 10 bar, 1800 mLgꢀ1 hꢀ1, H2/CO2 =3. Inset: stability test of
ZnFe2O4 up to 95 h. b) O/P ratios and CO selectivity versus sodium content in ZnFe2O4.
c) CO-free hydrocarbon selectivity of Fe2O3, ZnO–Fe2O3, ZnFe2O4, and Na-free ZnFe2O4
catalysts. d) Carbon-number distribution of the liquid products of CO2 hydrogenation
with the ZnFe2O4-derived catalyst, as determined by a simulated distillation method
(ASTM D2887).
from CO2 hydrogenation. Methane formation is suppressed
greatly (40.5!9.7%), and the formation of C5+ liquid products
increases dramatically (8.6!58%). The collected liquid prod-
ucts were analyzed by a GC simulated distillation method
(ASTM D2887) and a detailed hydrocarbon analysis method
(ASTM D6729). The products were mostly in the gasoline and
diesel range of hydrocarbons (Figure 2d). The product distribu-
tion is almost the same as that for typical CO Fischer–Tropsch
synthesis over iron-based catalysts.[15] The detailed analysis of
the hydrocarbon distribution of the liquid (C5–C11) products
shows that aromatic components represent the largest fraction
(ꢁ40.5 wt%), and the olefin fraction is also high (ꢁ26.2 wt%,
Figure S5). The performance of the ZnFe2O4-derived catalyst re-
mained stable in terms of both CO2 conversion and selectivity
over a 95 h stability test (Figure 2a, inset).
Figure 2c). The CO selectivity also increases if the Na is re-
moved. The RWGS reaction poses a thermodynamic barrier for
CO2 hydrogenation. Alkaline promoters such as Na can donate
electrons to Fe metal to promote CO adsorption[20] and acceler-
ate the subsequent CO conversion to hydrocarbons.
The ex-ZnFe2O4 catalyst was tested for CO hydrogenation
(H2/CO=3, 3408C, 10 bar) to determine its performance
under typical CO FT conditions. As shown in Table S2, the
ex-ZnFe2O4 catalyst shows 100% CO conversion (FTY>
11.1 mmolCO gFeꢀ1 sꢀ1), 45.7% of C5+ selectivity, and a high O/P
2
ratio (7.6). The slightly lower C5+ selectivity and O/P ratio rela-
tive to those for CO2 hydrogenation seem to originate from
the almost complete CO2 conversion.
Thus, our ex-ZnFe2O4 catalyst is much more active in CO hy-
drogenation but similarly selective for C5+ hydrocarbons and
ChemSusChem 2017, 10, 1 – 8
www.chemsuschem.org
3
ꢁ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
olefins. The lower rates of CO2 hydrogenation are attributed to
the reversible RWGS reaction and the competitive adsorption
of water.
separate rod-type ZnO particles and spherical Fe3O4 particles
can be observed (Figures S3d and S7). However, the ex-
ZnFe2O4 catalyst exhibits a dramatically changed morphology;
the ZnFe2O4 precursor decomposes and is reduced into ZnO
and metallic Fe particles after the reduction steps (Figures 1c,
d and S4). After the CO2 hydrogenation reaction, the high-reso-
lution transmission electron microscopy (HRTEM) and electron
energy loss spectroscopy (EELS) mapping images (Figure 4)
show that small ZnO particles decorate the surfaces of spheri-
cal Fe or Fe3O4 particles (<120 nm). The images demonstrate
how ZnO works as a structural promoter of Fe; it decorates
the surface of Fe or Fe3O4 particles to suppress the crystal
growth of the Fe particles. If the ZnO and Fe are derived from
ZnFe2O4, the structural promotion effect is more effective than
that for the ZnO–Fe2O3 physical mixture because of the atomic
mixing of Zn and Fe in the ZnFe2O4 precursor. The XRD pattern
in Figure 1b shows that the main phases of the ex-Fe2O3 and
ex-ZnO–Fe2O3 catalysts are Fe3O4 and metallic Fe. Fe3O4 is an
active phase of the RWGS reaction[21] and must be formed
through the partial oxidation of Fe metal by the byproduct
water. In contrast, the ZnFe2O4 catalyst forms the c-Fe5C2
phase through the in situ carburization of iron during the CO2
hydrogenation, and c-Fe5C2 is the active phase of iron FT cata-
lysts for hydrocarbon synthesis. However, Na-free ZnFe2O4
shows only Fe3O4 and metallic iron phases without the carbide
phase after the reaction, and this result highlights the critical
role of the small amount of Na residue (ꢁ0.08 wt% in this
case) as an electronic promoter.
Role of Na-containing ZnFe2O4 as efficient CO2 hydrogena-
tion catalyst precursor
To understand the unique performance of the Na-containing
ZnFe2O4-derived catalyst in the reaction of CO2 and H2, the re-
duction processes of the catalyst precursors were studied
through temperature-programmed reduction with H2 (H2-TPR).
For bare Fe2O3, the reduction starts at 2708C (Figure 3). This
Figure 3. H2-TPR of Fe2O3, ZnO–Fe2O3, and ZnFe2O4.
The postreaction X-ray photoelectron spectroscopy (XPS)
analysis (Figure 5) shows that all samples reveal Fe2p3/2 and
Fe2p1/2 peaks at binding energies (BEs) of 711.0 and 723.5 eV,
indicating that oxidized Fe species form on the surface be-
cause of the water byproduct.[22] This observation matches well
with the XRD patterns in Figure 1b. However, the ZnFe2O4-de-
rived catalysts show an additional well-defined peak at BE=
707 eV, which could be assigned to iron carbide.[23] The Na-free
ZnFe2O4 also shows the carbide XPS peak but only as a small
shoulder peak, which indicates that a much smaller quantity is
present than that in the regular ZnFe2O4 catalyst containing
Na.
temperature is substantially lower than that for Fe2O3 synthe-
sized by a precipitation method (ꢁ3308C), probably because
of the smaller particle size and the omission of the calcination
step. If Fe2O3 is combined with ZnO, the reduction tempera-
ture increases from 330 to 3608C. Compared with those for
bare Fe2O3, the reduction-peak intensities of ZnO–Fe2O3 de-
crease and the peaks are shifted to higher temperatures above
7008C, because ZnO retards the reduction of Fe2O3. The reduc-
tion of iron in ZnO–Fe2O3 and Fe2O3 produces Fe3O4 at low
temperatures and then metallic Fe at high temperatures. The
reduction of ZnFe2O4 begins at a lower temperature
than those for the two reference samples and occurs
over a wider temperature range up to 8008C. These
results indicate that a strong interaction between Fe
and ZnO retards the reduction of Fe2O3. This behavior
is also consistent with the XRD pattern in Figure S1,
which shows that both ZnO–Fe2O3 and ZnFe2O4 con-
tain the FeO phase after H2 treatment at 4008C. Thus,
the reduction behavior of the catalysts does not have
a significant effect on the catalytic performance.
Next, we performed postreaction analyses of the
catalysts after reactions at 3408C and 10 bar for 48 h.
After the reaction, the ex-Fe2O3 catalyst shows severe
particle aggregation (as shown by the SEM and TEM
images in Figures S3c and S6), which results in a low-
surface-area catalyst. The ex-ZnO–Fe2O3 catalyst does
Figure 4. a) HRTEM images of the ZnFe2O4 catalyst after CO2 hydrogenation for 48 h.
not show significant morphological changes, and b) Fe, C) O, d) Zn, and e) Na EELS mapping images.
&
ChemSusChem 2017, 10, 1 – 8
www.chemsuschem.org
4
ꢁ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Figure 5. Fe2p XPS spectra for a) ex-Fe2O3, b) ex-ZnO–Fe2O3, c) ex-ZnFe2O4 (Na 0.08 wt%), and d) Na-free ex-ZnFe2O4 after 48 h of reaction. The iron carbide
peaks are highlighted in (c) and (d).
X-ray absorption near-edge structure (XANES) spectroscopy
was applied to Fe2O3, ZnO–Fe2O3, ZnFe2O4, and Na-free
ZnFe2O4 used for 48 h of CO2 hydrogenation. The Fe K-edge
XANES spectra of the used catalysts as well as that of the refer-
ence Fe5C2 prepared by a previously reported method are
shown in Figure 6a.[24,25] All samples show different shapes for
the edge-rising portion in the energy range from 7114 to
7130 eV owing to their different electronic structures. In partic-
ular, ZnFe2O4 and Na-free ZnFe2O4 present significantly differ-
ent spectra, which indicate their different chemical states. Only
the spectrum of regular (Na-containing) ZnFe2O4 has a similar
pre-edge shape and position to those of the Fe5C2 reference at
energies of 7114 and 7117 eV. To show the characteristics of
the XANES spectra more clearly, the derivative spectra of nor-
malized absorbance are provided in Figure 6b. Again, only ex-
ZnFe2O4 has a spectrum similar to that of the Fe5C2 reference.
As mentioned above, ZnO is a structural promoter that re-
tards the sintering of the Fe particle size and, thus, increases
the exposure of the active Fe metal surface. Usually, ZnO has a
strong interaction with iron particles under FT or CO hydroge-
Figure 6. a) Fe K-edge XANES profiles of Fe2O3, ZnO–Fe2O3, ZnFe2O4, and Na-free ZnFe2O4 with the reference Hꢀgg iron carbide. b) Fe K-edge XANES derivative
spectra.
ChemSusChem 2017, 10, 1 – 8
www.chemsuschem.org
5
ꢁ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
nation conditions.[26] In contrast, Na is an electronic promoter
that provides electrons to form reduced iron species as the
active catalytic sites. Iron metal and Fe3O4 are active sites for
the formation of hydrocarbons and RWGS reactions, respec-
tively, in FT synthesis.[21,27] Both ZnO–Fe2O3 and Na-free
ZnFe2O4 contain Fe and Fe3O4 in addition to ZnO (as shown by
XRD), which results in the formation of the undesired products
methane and short-chain light hydrocarbons. However,
ZnFe2O4 promoted with residual Na as well as Zn can form c-
Fe5C2 readily through the in situ carburization of iron during
CO2 hydrogenation. This Hꢀgg iron carbide is the known active
phase of iron-based FT catalysts for CO hydrogenation. Our re-
sults demonstrate that this phase is also the active phase in
CO2 hydrogenation for the formation of C5+ liquid hydrocar-
bons. Thus, the catalyst presents interesting selectivity for C5+
liquid products, high O/P ratios, and high hydrocarbon yields
in CO2 hydrogenation at 3408C and 10 bar. The evolution of
the catalyst structure and the mechanistic idea of a doubly
promoted c-Fe5C2 catalyst with electronic (Na) and structural
(ZnO) promoters are illustrated in Scheme 1.
Experimental Section
Catalyst preparation
ZnFe2O4 was synthesized by a microwave-assisted hydrothermal
method. Fe(NO3)3·9H2O (Kanto, 2.02 g) and Zn(NO3)2·6H2O (Zn/Fe=
1:1 molar ratio, 1.47 g) were dissolved in distilled water (40 mL),
and NaOH (0.1 mol) was added to the mixture to ensure that it
was basic. An excess of the Zn source over the stoichiometry (1:2)
was needed to prepare the pure-phase ZnFe2O4. After a few mi-
nutes, propionaldehyde (0.5 mL) was added. The solution was
transferred to a 100 mL Teflon tube, which was then placed in a
microwave oven (CEM MARS-5, 300 W) for 2 h at 1808C for the syn-
thesis reaction. After the sample cooled to ambient temperature, it
was washed with distilled water (0.5–1.5 L) to control the amount
of residual Na. If the samples were washed with more than 6 L of
water, the amount of Na residue was negligible, and Na-free
ZnFe2O4 samples were obtained. The Fe2O3 reference was also syn-
thesized by a similar microwave-assisted hydrothermal method
with only Fe(NO3)3·9H2O. For the ZnO–Fe2O3 physical mixture, com-
mercial ZnO (Sigma–Aldrich) and Fe2O3 (Kanto) were mixed in a
mortar with a Zn/Fe molar ratio of 1:2.
Catalytic CO2 hydrogenation
The CO2 hydrogenation was performed in a fixed-bed reactor at a
CO2/H2 ratio of 1:3. All samples were pre-reduced in a pure H2 flow
[100 mLminꢀ1, standard temperature and pressure (STP)] at 4008C
for 2 h. After the reduction step, the reactor was purged with Ar to
remove the H2, and then CO2, H2, and N2 (7.69 vol% as an internal
reference for GC analysis) entered the reactor at 10 bar and 3408C
with a gas hourly space velocity (GHSV) of 1800 mLgꢀ1 hꢀ1. The
CO2, H2, and N2 products were analyzed by online GC (Agilent
7890A) with a thermal conductivity detector (TCD, Carboxen 1000
packed column). Hydrocarbons from C1 to C6 were analyzed contin-
uously using the same GC instrument with a flame ionization de-
tector (FID, alumina sulfate PLOT capillary column). The heavy-hy-
drocarbon liquid products were collected in a cold trap. The prod-
ucts were analyzed through simulated distillation (SIMDIS) and de-
tailed hydrocarbon analysis (DHA) based on their carbon-number
distributions (ASTM D2887 and D6729).[28]
Scheme 1. Structural evolution and mechanistic idea of ZnFe2O4-derived cat-
alyst for the selective synthesis of C5+ liquid fuels by CO2 hydrogenation.
Characterization and analysis
Conclusions
Powder XRD was performed with a PANalytical X’pert diffractome-
ter with CuKa radiation (40 mA, 30 kV). H2-TPR was performed with
a Micromeritics AutoChem II apparatus (model 2920). HRTEM and
SEM were performed with JEOL JEM-2200FS and Philips Electron
Optics XL30S FEG instruments, respectively. The BET surface areas
and pore-size distributions were determined from the N2 sorption
isotherms measured at liquid N2 temperature (Mirae SI, Nanoporos-
ity-XQ). The X-ray absorption fine structure (XAFS) measurements
were performed at the 7D beamline of the Pohang Accelerator
Laboratory (PLS-II, 3.0 GeV, 400 mA). The synchrotron radiation was
monochromatized with Si(111) double crystal monochromators.
Under ambient conditions, the Fe K-edge (E0 =7112 eV) spectra
were collected in transmission mode with He- and N2-filled IC SPEC
ionization chambers. The incident beam was detuned by 30% for
the Fe K-edge to reduce contamination by higher harmonics. The
spectrum of the Fe foil reference was recorded to enable the cali-
bration of the energy in the spectrum of the sample to the K-edge
energy of Fe metal. The AHENA program in the IFEFFIT suite of
programs was used to analyze the obtained data to determine the
Sodium-containing ZnFe2O4 synthesized by a microwave-assist-
ed hydrothermal method becomes an effective catalyst precur-
sor that gives rise to a high CO2 conversion, an improved
liquid-fuel selectivity (ꢁ58%), and a high olefin-to-paraffin
ratio (ꢁ11) in CO2 hydrogenation with the assistance of Zn as
a structural promoter and Na as an electronic promoter. The
combination of Zn and Na with Fe increases the CO2 adsorp-
tion properties and promotes the in situ formation of Hꢀgg
iron carbide, which is the active phase for the formation of
heavy hydrocarbons in CO2 hydrogenation.
&
ChemSusChem 2017, 10, 1 – 8
www.chemsuschem.org
6
ꢁ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
[12] Y. H. Choi, Y. J. Jang, H. Park, W. Y. Kim, Y. H. Lee, S. H. Choi, J. S. Lee,
Appl. Catal. B 2017, 202, 605–610.
[13] J. Wei, Q. Ge, R. Yao, Z. Wen, C. Fang, L. Guo, H. Xu, J. Sun, Nat.
Commun. 2017, 8, 15174.
local structures of the Fe atoms in ZnFe2O4, Fe2O3, and ZnO–
Fe2O3.
[29]
[14] M. Zhang, R. Lan, J. Liu, X. Chen, W. Zhou, J. Chem. Soc. Faraday Trans.
1992, 88, 637–644.
Acknowledgements
[15] J. C. Park, D. H. Chun, J.-I. Yang, H.-T. Lee, S. Hong, G. B. Rhim, S. Jang, H.
Jung, RSC Adv. 2015, 5, 44211–44217.
[16] H. M. Torres Galvis, J. H. Bitter, C. B. Khare, M. Ruitenbeek, A. I. Dugulan,
K. P. de Jong, Science 2012, 335, 835–838.
[17] R. Satthawong, N. Koizumi, C. Song, P. Prasassarakich, Catal. Today
2015, 251, 34–40.
[18] C. G. Visconti, M. Martinelli, L. Falbo, A. Infantes-Molina, L. Lietti, P. For-
zatti, G. Iaquaniello, E. Palo, B. Picutti, F. Brignoli, Appl. Catal. B 2017,
200, 530–542.
This work was supported by the Climate Change Response proj-
ect (2015M1A2A2074663, 2015M1A2A2056824), the Basic Science
Grant (NRF-2015R1A2A1A10054346), Korea Center for Artificial
Photosynthesis (KCAP, No. 2009-0093880) funded by MSIP, and
MOTIE of Republic of Korea (Project Nos. 10050509 and KIAT
N0001754).
[19] M. Amoyal, R. Vidruk-Nehemya, M. V. Landau, M. Herskowitz, J. Catal.
2017, 348, 29–39.
[20] B. Graf, M. Muhler, Phys. Chem. Chem. Phys. 2011, 13, 3701–3710.
[21] T. Riedel, H. Schulz, G. Schaub, K.-W. Jun, J.-S. Hwang, K.-W. Lee, Top.
Catal. 2003, 26, 41–54.
Conflict of interest
The authors declare no conflict of interest.
[22] A. M. Hilmen, D. Schanke, K. F. Hanssen, A. Holmen, Appl. Catal. A 1999,
186, 169–188.
[23] K. Cheng, V. V. Ordomsky, M. Virginie, B. Legras, P. A. Chernavskii, V. O.
Kazak, C. Cordier, S. Paul, Y. Wang, A. Y. Khodakov, Appl. Catal. A 2014,
488, 66–77.
Keywords: carbon dioxide · fischer–tropsch synthesis · fuels ·
hydrogenation · spinel phases
[1] R. W. Dorner, D. R. Hardy, F. W. Williams, H. D. Willauer, Energy Environ.
Sci. 2010, 3, 884–890.
[24] C. Yang, H. Zhao, Y. Hou, D. Ma, J. Am. Chem. Soc. 2012, 134, 15814–
15821.
[2] J. Patricio, A. Angelis-Dimakis, A. Castillo-Castillo, Y. Kalmykova, L.
Rosado, J. CO2 Util. 2017, 17, 50–59.
[25] H. Park, D. H. Youn, J. Y. Kim, W. Y. Kim, Y. H. Choi, Y. H. Lee, S. H. Choi,
J. S. Lee, ChemCatChem 2015, 7, 3488–3494.
[3] Y. J. Jang, I. Jeong, J. Lee, J. Lee, M. J. Ko, J. S. Lee, ACS Nano 2016, 10,
6980–6987.
[4] J. Wang, S.-M. Lu, J. Li, C. Li, Chem. Commun. 2015, 51, 17615–17618.
[5] T. Witoon, T. Permsirivanich, N. Kanjanasoontorn, C. Akkaraphataworn,
A. Seubsai, K. Faungnawakij, C. Warakulwit, M. Chareonpanich, J. Limtra-
kul, Catal. Sci. Technol. 2015, 5, 2347–2357.
[26] M. K. Gnanamani, H. H. Hamdeh, G. Jacobs, D. Qian, F. Liu, S. D. Hopps,
G. A. Thomas, W. D. Shafer, Q. Xiao, Y. Hu, B. H. Davis, RSC Adv. 2016, 6,
62356–62367.
[27] H. Schulz, T. Riedel, G. Schaub, Top. Catal. 2005, 32, 117–124.
[28] C. Wang, R. Firor, Simulated Distillation System for ASTM D2887 Based on
the Agilent 6890n GC, 2005, Agilent Technologies publication 5989-
[6] S. Saeidi, N. A. S. Amin, M. R. Rahimpour, J. CO2 Util. 2014, 5, 66–81.
[7] W. Wang, S. Wang, X. Ma, J. Gong, Chem. Soc. Rev. 2011, 40, 3703–3727.
[8] M. D. Porosoff, B. Yan, J. G. Chen, Energy Environ. Sci. 2016, 9, 62–73.
[9] M. Albrecht, U. Rodemerck, M. Schneider, M. Brçring, D. Baabe, E. V.
Kondratenko, Appl. Catal. B 2017, 204, 119–126.
2726EN,
2726EN.pdf.
https://www.agilent.com/cs/library/applications/5989-
[29] B. Ravel, M. Newville, J. Synchrotron Radiat. 2005, 12, 537–541.
[10] M. Al-Dossary, A. A. Ismail, J. L. G. Fierro, H. Bouzid, S. A. Al-Sayari, Appl.
Catal. B 2015, 165, 651–660.
[11] L. M. Chew, P. Kangvansura, H. Ruland, H. J. Schulte, C. Somsen, W. Xia,
G. Eggeler, A. Worayingyong, M. Muhler, Appl. Catal. A 2014, 482, 163–
170.
Manuscript received: August 3, 2017
Version of record online: && &&, 0000
ChemSusChem 2017, 10, 1 – 8
www.chemsuschem.org
7
ꢁ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

FULL PAPERS
Y. H. Choi, E. C. Ra, E. H. Kim, K. Y. Kim,
Y. J. Jang, K.-N. Kang, S. H. Choi,
J.-H. Jang, J. S. Lee*
It’s a gas: A catalyst derived from spinel
ZnFe2O4 containing Na residue displays
high CO2 hydrogenation activity and
liquid-fuel selectivity in the gasoline and
diesel range with high olefin-to-paraffin
ratios.
&& – &&
Sodium-Containing Spinel Zinc Ferrite
as a Catalyst Precursor for the
Selective Synthesis of Liquid
Hydrocarbon Fuels
&
ChemSusChem 2017, 10, 1 – 8
www.chemsuschem.org
8
ꢁ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!