A durable nanocatalyst of potassium-doped iron-carbide/alumina for significant production of linear alpha olefins via Fischer-Tropsch synthesis

Article abstract of DOI:10.1016/j.apcata.2018.07.037

Improvement of activity, selectivity, and stability of the catalyst used in Fischer-Tropsch synthesis (FTS) to produce targeted hydrocarbon products has been a major challenge. In this work, the potassium-doped iron-carbide/alumina (K-Fe₅C₂/Al₂O₃), as a durable nanocatalyst containing small iron-carbide particles (～ 10 nm), was applied to high-temperature Fischer-Tropsch synthesis (HT-FTS) to optimize the production of linear alpha olefins. The catalyst, suitable under high space velocity reaction conditions (14–36 N L g_cat^?1 h^?1) based on the well-dispersed potassium as an efficient base promoter on the active iron-carbide surface, shows very high CO conversion (up to ～90%) with extremely high activity (1.41 mmol_CO g_Fe^?1 s^?1) and selectivity for C₅–C₁₃ linear alpha olefins.

Full text of DOI:10.1016/j.apcata.2018.07.037

Accepted Manuscript
Title: A Durable Nanocatalyst of Potassium-Doped
Iron-Carbide/Alumina for Significant Production of Linear
Alpha Olefins via Fischer-Tropsch Synthesis
Authors: Ji Chan Park, Sanha Jang, Geun Bae Rhim, Jin Hee
Lee, Hyunkyoung Choi, Heon-Do Jeong, Min Hye Youn,
Dong-Wook Lee, Kee Young Koo, Shin Wook Kang, Jung-Il
Yang, Ho-Tae Lee, Heon Jung, Chul Sung Kim, Dong Hyun
Chun
PII:
S0926-860X(18)30378-8
DOI:
Reference:
https://doi.org/10.1016/j.apcata.2018.07.037
APCATA 16763
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
31-5-2018
24-7-2018
26-7-2018
Please cite this article as: Park JC, Jang S, Rhim GB, Lee JH, Choi H, Jeong H-Do,
Youn MH, Lee D-Wook, Koo KY, Kang SW, Yang J-Il, Lee H-Tae, Jung H, Kim
CS, Chun DH, A Durable Nanocatalyst of Potassium-Doped Iron-Carbide/Alumina for
Significant Production of Linear Alpha Olefins via Fischer-Tropsch Synthesis, Applied
Catalysis A, General (2018), https://doi.org/10.1016/j.apcata.2018.07.037
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

A Durable Nanocatalyst of Potassium-Doped Iron-
Carbide/Alumina for Significant Production of Linear
Alpha Olefins via Fischer-Tropsch Synthesis
Ji Chan Park,a,b,* Sanha Jang,a Geun Bae Rhim,a Jin Hee Lee,a Hyunkyoung Choi,c Heon-Do Jeong,a Min
Hye Youn,a Dong-Wook Lee,d Kee Young Koo,e Shin Wook Kang,a Jung-Il Yang,a Ho-Tae Lee,a Heon
Jung,a Chul Sung Kim,c,* Dong Hyun Chuna,b,*
a Clean Fuel Laboratory, Korea Institute of Energy Research, Daejeon 34129, Republic of Korea.
b
Advanced Energy and System Technology, University of Science and Technology, Daejeon 34113,
Republic of Korea.
c Department of Physics, Kookmin University, Seoul 02707, Republic of Korea
d Separation and Conversion Materials Laboratory, Korea Institute of Energy Research, Daejeon 34129,
Republic of Korea.
e Hydrogen Laboratory, Korea Institute of Energy Research, Daejeon 34129, Republic of Korea.
*Corresponding authors. E-mail: cdhsl@kier.re.kr, cskim@kookmin.ac.kr, jcpark@kier.re.kr
1

Graphical Abstract
A potassium-doped iron-carbide/alumina nanocatalyst shows very high CO conversion (~90%) and
significant productivity for C5−C13 linear alpha olefins in Fischer-Tropsch synthesis under high space
velocity conditions.
Highlights

The active iron-carbide nanoparticles (~10 nm) embedded on gamma-alumina with well-dispersed
potassium promoters showed high activity and productivity for linear alpha olefins in high-
temperature Fischer-Tropsch synthesis.
ABSTRACT
Improvement of activity, selectivity, and stability of the catalyst used in Fischer-Tropsch synthesis (FTS)
to produce targeted hydrocarbon products has been a major challenge. In this work, the potassium-doped
iron-carbide/alumina (K-Fe5C2/Al2O3), as a durable nanocatalyst containing small iron-carbide particles
(~ 10 nm), was applied to high-temperature Fischer-Tropsch synthesis (HT-FTS) to optimize the
2

production of linear alpha olefins. The catalyst, suitable under high space velocity reaction conditions
-1·h-1) based on the well-dispersed potassium as an efficient base promoter on the active
−36 NL·g
(14
cat
iron-carbide surface, shows very high CO conversion (up to ~ 90%) with extremely high activity (1.41
mmolCO∙gFe-1∙s-1) and selectivity for C5−C13 linear alpha olefins.
KEYWORDS
Fischer-Tropsch synthesis, linear alpha olefins, high durability, iron-carbide, alkali promoter
1. INTRODUCTION
The linear alpha olefins (LAOs) obtained by oligomerization of ethylene or by Fischer-Tropsch synthesis
(FTS), are very important chemical intermediates used in such as detergents, plasticizers, synthetic
lubricants, and polymers (polyethylene) [1, 2]. Because a wide variety of customer demands involve the
LAO market, high profits can be expected if LAO production can be increased [3].
The process of oligomerization of ethylene to LAOs is well-known, but the FTS process for effective
LAO production is not well studied [4, 5]. Liu et al. reported the use of pretreated CoMn/SiO2 as a
selective catalyst for alpha olefins [6]. Recently, bio-promoted iron-carbide catalysts to convert carbon
dioxide to LAOs directly were developed, but the long term stability of the catalysts was not discussed
[7].
Commercially, gasoline and lower-olefin products can be obtained efficiently using the high-temperature
FTS (HT-FTS) process. This process is conducted at temperatures of 300–350 °C using an iron-based
catalyst [8-10]. However, conventional HT-FTS catalysts remain inadequate for production of significant
quantities of LAO [11]. Typically, iron-based-support catalysts (e.g., Fe/-Al2O3, Fe/CNT, Fe/SiO2,
Fe@C, and Fe/zeolite) have been used in HT-FTS reactions to produce a variety of hydrocarbon products
[12-18]. In recent years, more selective and efficient ways for production of high value-added
3

hydrocarbons have been developed for FTS. For example, high selectivity (~60%) for C2−C4 lower olefins
was achieved using Fe/-Al2O3 and Fe/carbon-nanofiber (CNF) catalysts [19].
Gamma-phase alumina provides a robust support for the high-temperature catalytic reactions. Its use led
to high metal dispersion based on strong metal-support interaction, which resulted in more active and
stable sites [20, 21]. In addition, alkali metals used as a base-additive donated electrons to active sites.
These were also exploited to enhance the FTS catalyst [22-25]. For instance, potassium (K) and cesium
(Cs) showed dramatic effects in HT-FTS by facilitating CO dissociation and increasing the growth of
carbon chains [26-29]. The doped-K in Fe catalysts also enhanced alkene production [30].
On the other hand, elaborate control of the reaction- conditions could also affect FTS products. For
instance, increasing the space velocity in FTS is one effective way to achieve selective formation of the
alkene products that result from reduced secondary reactions [31]. The hydrocarbon products spend less
time in contact with the catalyst and the probability of re-adsorption and reaction becomes less, thereby
reducing alkene hydrogenation.
Herein, we introduce for the first time an extremely active K-Fe5C2/Al2O3 nanocatalyst with high
durability. It is prepared through melt-infiltration of an iron salt and sequential impregnation of a
potassium solution, for the efficient production of high value-added alkene hydrocarbons. The catalyst
shows extremely high activity and productivity for C5−C13 LAOs. The high stability of the catalyst was
monitored as it was recycled two times under harsh reaction conditions (340 °C and 15 bar).
2. EXPERIMENTAL SECTION
2.1. Synthesis of potassium doped iron-carbide/alumina nanocatalyst
Details for the preparation of iron-carbide/alumina were reported in our previous publication [20].
Typically, 1.57 mmol Fe(NO3)3·9H2O (Aldrich, ≥98%) was physically ground with 1.0 g of alumina
4

powder (Strem Chemicals, Inc.) in a mortar for 5 min under ambient conditions. Next, the mixed powder
was placed in a polypropylene bottle and aged at 50 °C in a tumbling oven. After aging for 24 h, the
sample was allowed to cool to room temperature, and then it was transferred to an alumina boat in a tube-
type furnace. The powder mixture was slowly heated at the rate of 2.7 °C·min-1 up to 350 °C under a flow
of CO (0.2 L·min-1); then was thermally treated at the same temperature for 8 h. After that, the resulting
black powder was cooled to room temperature, and then submerged into 30 mL of ethanol under an N2
flow of 1.0 L·min-1. The powder immersed in ethanol was separated simply using a magnet and dried in
a vacuum oven at 60 °C. A portion (0.376 mmol) of K2CO3 (Aldrich, ReagentPlus, 99%) dissolved in
distilled water (0.5 mL) was impregnated on the dried iron-carbide/alumina powder. The K-incorporated
sample was thermally treated again at 350 °C for 8 h under a continuous CO flow at 0.2 mL·min-1. Then,
the resulting powder was cooled to room temperature and submerged again in anhydrous ethanol under
flowing nitrogen. Finally, the potassium doped iron-carbide/alumina was separated and dried in a vacuum
oven.
2.2. High-temperature Fischer-Tropsch synthesis reaction
Fischer-Tropsch (F-T) reactions were performed in a fixed-bed stainless steel reactor tube (inner diameter:
5 mm). The catalyst sample (0.3 g) was diluted with glass beads (425−600m, 3.5 g) for prevention of
hot-spot generation by the severe exothermic reaction. Before reaction, the catalyst was reduced in situ at
atmospheric pressure by passing a flow of CO (40 mL·min-1) over it at 350 °C for 8 h. After reduction,
the reaction was carried out at 340 °C and 15 bar for 90 h using reactant gas (H2/CO = 1.0, GHSV = 14
NL·gcat-1·h-1). After reaction for 90 h, the cooled iron-carbide/alumina catalyst was recycled two times at
340 °C and 15 bar for an additional 48 h using the same reactant gas (H2/CO = 1.0, GHSV = 20 and 36
NL·gcat-1·h-1) with no pre-treatment. The composition of the outlet gases was analyzed using a gas
chromatograph (Agilent, 3000 Micro GC) equipped with TCD and FID. The flow rates of the outlet gases
were measured using a wet-gas flow meter (Shinagawa Co.). After reaction, the solid-phase hydrocarbons
were collected in a hot trap (240 °C) and the liquid-phase hydrocarbons and water were collected in a cold
5

trap (−2 °C), respectively. The isolated solid (wax) and liquid (oil) products were analyzed using an offline
gas chromatograph (Agilent, 6890 N) and the simulated distillation method (ASTM D2887). Detailed
hydrocarbon analysis (DHA) of the LAO component and mass percent in the liquid oils was determined
using ultra-high resolution capillary gas chromatography and ASTM D6730.
2.3. Characterization
The catalyst samples were characterized using a Tecnai TF30 ST and a Titan Double Cs corrected TEM
instrument (Titan cubed G2 60-300). Energy-dispersive X-ray spectroscopy (EDS) elemental mapping
data were obtained using a higher-efficiency detection system (Super-X detector) and high power powder
XRD (Rigaku, D/MAX-2500, 18 kW). The TEM samples were prepared by putting a few drops of the
corresponding colloidal solutions on carbon-coated copper grids (Ted Pellar, Inc). XPS studies were
carried out using a Sigma Probe (Thermo VG Scientific, Inc.) with a micro-focused monochromator X-
ray source. The XPS sample was prepared by placing a few drops of the colloidal solution on small pieces
of gold wafer. A Mössbauer spectrometer of the electromechanical type, with a 50 mCi 57Co source in a
rhodium matrix, was used in constant-acceleration mode. Nitrogen sorption isotherms were measured at
−196 °C with a Tristar II 3020 surface area analyser. Before measurements, the samples were degassed
in a vacuum at 300 °C for 4 h.
3. RESULTS AND DISCUSSION
3.1. Synthesis of potassium doped iron-carbide/alumina nanocatalyst
Scheme 1 illustrates the overall procedure for synthesis of the K-Fe5C2/Al2O3 nanocatalyst. The active
iron-carbide nanoparticles promoted by potassium were obtained through a melt-infiltration process and
the impregnation method [28, 32]. First, an iron nitrate hydrate salt was melt-infiltrated into mesopores
of the alumina support by aging at 50 °C. By a thermal treatment at 350 °C under a CO condition, small
iron-carbide nanoparticles in alumina pores could be generated. K-Fe5C2/Al2O3 nanocatalyst was obtained
using a wetness impregnation method with aqueous K2CO3 solution as a K source.
6

The transmission electron microscopy (TEM) image shows well-dispersed iron-carbide nanoparticles
around 10 nm (Fig. 1a and b). The high-angle annular dark-field scanning transmission electron
microscopy (HAADF-STEM) image also presents bright dots, indicating Fe-rich regions within the
porous alumina support (Fig. 1c). The elemental mapping of iron (red) and aluminum (green)
demonstrates small and uniform nanoparticles (Fig. 1d). Elemental potassium (cyan) is also well-
dispersed in the K-Fe5C2/Al2O3 sample. The high-resolution TEM (HR-TEM) image and the
corresponding Fourier-transform (FT) pattern indicated the formation of iron-carbide crystals with a
distance of 0.21 nm between adjacent fringes that match with the (510) planes of Hägg iron-carbide (Fig.
1e−f).
In the X-ray diffraction (XRD) patterns of pristine alumina and K-Fe5C2/Al2O3 nanocatalyst, alumina
peaks were mainly observed at 2 = 37.6°, 45.9°, and 67.0°; corresponding to the reflections of the (311),
(400), and (440) planes of gamma-phase alumina (Fig. 2a, JCPDS no. 10-0425). From the XRD data, the
exact crystal phase of the incorporated nanoparticles was difficult to be assigned due to the broad peaks
of the alumina, but the shoulder (2 = 41°–45º) of the main peak showed a relatively increased intensity
compared to that of pristine alumina, indicating the presence of iron-carbide phases.
To investigate the surface states of K-Fe5C2/Al2O3 nanocatalyst, core-level X-ray photoelectron spectra
(XPS) of iron and potassium (K) were recorded (Fig. 2b and c). The XPS spectra of the energy region of
the Fe band exhibited peaks assigned to iron-carbide (Fe(0)) at 707 eV and iron-oxide (Fe(3+)) at 711 eV
(Fig. 2b). The surfaces of the iron-carbide nanoparticles were mainly observed as oxidized forms, because
oxidation ofthe fresh catalyst powders occurs readily. In the K 2p core-level XPS spectrum, a peak around
293 eV was observed, which is low binding energy compared to the peak (294.4 eV) of metallic K (Fig.
2c).
The quantitative analysis of iron-carbides and iron-oxides could be accessed by Mössbauer
spectroscopy (Fig. 3). The Mössbauer spectrum at 295 K could be fitted with the superposition of the 3
sextets for Fe sites corresponding to Fe5C2 crystallography (Fig. 3a). In the spectrum, the deconvoluted
peaks (1 doublet and 1 singlet) are analyzed to be Fe2O3 and Fe3O4 phase. The Mossbauer spectrum at 4.2
7

K was fitted with 3 sextets for Fe5C2, 1 sextet for Fe2O3, and 1 singlet for Fe3O4 with different hyperfine
parameters (Fig. 3b). From the fitted Mössbauer parameters, the relative areas of -Fe5C2, Fe2O3, and
Fe3O4 were calculated to be approximately 38%, 48% and 14% (Table S1 in the Electronic Supplementary
Material: ESI).
N2 sorption experiments for the K-Fe5C2/Al2O3 catalyst exhibited type IV adsorption-desorption
hysteresis (Fig. 4a). The Brunauer-Emmett-Teller (BET) surface area and total pore volume calculated
was 176 m2∙g-1 and 0.28 cm3∙g-1, respectively. The pore size of the catalyst was found to be ca. 8.5 nm on
the adsorption branch by the Barrett-Joyner-Halenda (BJH) method (Fig. 4b). The Fe content in the K-
Fe5C2/Al2O3 catalyst was calculated to be ~ 8 wt% on the basis of Fe converted from the used Fe salt after
the thermal treatment.
3.2. Fischer-Tropsch Synthesis reaction
To investigate the performance of the catalyst for high-temperature Fischer-Tropsch synthesis, reaction
tests with K-Fe5C2/Al2O3 were performed at 15 bar, 340 °C, and a H2/CO ratio of ‘1’. The CO conversion
and selectivity of the catalysts were measured over time-on-stream (TOS) by gas chromatography (GC)
of the outlet gases containing the unreacted CO, H2, CH4, C2–C4 hydrocarbons, and CO2. Liquid oil and
solid wax recovered in a cold trap and a hot trap, respectively, were further analyzed using simulated
distillation (SIMDIS).
During FTS, multiple reactions can proceed as indicated below:
nCO + (2n+1)H2 → CnH2n+2 + nH2O (paraffins)
nCO + 2nH2 → CnH2n + nH2O
Significant amounts of CO2 are generated by the accompanying water-gas shift (WGS) reaction as
follows:
CO + H2O ↔ CO2 + H2
(olefins)
8

The K-Fe5C2/Al2O3 nanocatalyst exhibited very high CO conversions even under very high gas hourly
-1·h-1. The total CO conversion rates of the
−30 NL·g
space velocity (GHSV) conditions in the range 14
cat
-
K-Fe5C2/Al2O3 nanocatalyst were observed to be 94.1% at GHSV = 14 NL∙gcat−1∙h−1, 92.9% at 20 NL·gcat
1·h-1, and 89.8% at 36 NL·gcat-1·h-1 (Fig. 5a, c, and e, respectively). The data for the K- Fe5C2/Al2O3
nanocatalyst showed selectivity for CO2 (41.5%), CH4 (7.6%), C2−C4 (15.6%), and C5+ (35.3%) at GHSV
= 14 NL·gcat-1·h-1 (Fig. 5b). At GHSV = 20 NL·gcat-1·h-1, selectivity for CO2 (43.4%), CH4 (7.8%), C2−C4
(15.8%) and C5+ (33.0%) was observed (Fig. 5d). With increase of GHSV, the selectivity for CH4 and
C2−C4 slightly increased. The C5+ selectivity with GHSV at 20 and 36 NL·gcat-1·h-1 was observed to be
33.0 % in both cases (Fig. 5d and e).
On the other hand, the CO conversion rates were significantly lower (58.0% at GHSV = 20 NL·gcat-1·h-
1 and 41.9% at 36 NL·gcat-1·h-1) with the K-free Fe5C2/Al2O3 catalyst than with the K-Fe5C2/Al2O3 catalyst
under the same GHSV conditions (Fig. 6a and c). The selectivity data for K-free Fe5C2/Al2O3 catalyst at
GHSV = 20 NL·gcat-1·h-1 showed selectivity for CO2 (42.9%), CH4 (8.3%), C2−C4 (20.5%), and C5+
(28.3%) (Fig. 6b).
The FT activity can be represented by iron-time-yield (FTY) indicating the number of CO moles
converted to hydrocarbons per gram of iron per second. At GHSV = 20 NL·gcat-1·h-1, the K-Fe5C2/Al2O3
nanocatalyst showed a higher FTY value (8.3×10-4 molCO∙gFe-1∙s-1) than did the K-free Fe5C2/Al2O3
catalyst (5.3×10-4 molCO∙gFe-1∙s-1) (Fig. 7a). With Fe5C2/Al2O3 catalyst, increasing the space velocity to
GHSV = 36 NL·gcat-1·h-1 resulted in the highest FTY value (14.1×10-4 molCO∙gFe-1∙s-1) (Fig. 7b). This value
is extremely high, comparable to the data for the recently reported Fe catalysts for FTS (Table S2). In
particular, the K-Fe5C2/Al2O3 nanocatalyst at GHSV = 36 NL·gcat-1·h-1 showed a very high liquid oil
productivity of 2.15 gliq∙gcat-1∙h-1 after the FTS reaction for 48 h (Table S3). This is a very high value
compared to those of other catalysts such as Fe5C2 nanoparticles on porous carbon and silica [20, 33].
To determine the hydrocarbon product distributions and specific hydrocarbon productivity, the detailed
composition of the liquid and solid hydrocarbons was analyzed by the simulated distillation method. For
the hydrocarbon products recovered after the FTS reactions using the K-Fe5C2/Al2O3 nanocatalyst, higher
9

GHSV conditions led to more narrower carbon distributions in the C5−C13 range (Fig. 8). The measured
C5−C13 portions in all the syncrude and liquid oils was 78.9% (Fig. 8a), 85.1% (Fig. 8b), 80.6% (Fig 8c),
86.6% (Fig. 8d), 82.0% (Fig. 8e), and 86.6% (Fig. 8f).
With increase of space velocity, heavy hydrocarbons (C19+) were slightly decreased from 4.47% at
GHSV = 14 NL·gcat-1·h-1 to 3.27% at GHSV= 36 NL·gcat-1·h-1 (Fig. 9). On the other hand, the portion of
C2−C4 olefins was increased from 15.7% at GHSV = 14 NL·gcat-1·h-1 to 17.4% at GHSV = 36 NL·gcat-1·h-
1. These trends were supported by the carbon chain growth probability () of the hydrocarbons, using the
Anderson-Schulz-Flory (ASF) chain growth mechanism [34]. The equation used is as follows:
log(Wn/n) = log(ln2) + n∙log
where Wn is the weight fraction of hydrocarbons with carbon number n. The value of the K-
Fe5C2/Al2O3 nanocatalyst under the different reaction conditions was 0.816 at GHSV = 14 NL·gcat-1·h-1,
0.804 at GHSV= 20 NL·gcat-1·h-1, and 0.799 at GHSV = 36 NL·gcat-1·h-1.
The specific hydrocarbon (HC) productivity data (gram of generated hydrocarbons per gram of iron per
second) for reactions with the K-Fe5C2/Al2O3 and K-free Fe5C2/Al2O3 nanocatalysts were obtained (Fig.
10). The K-Fe5C2/Al2O3 nanocatalyst at GHSV = 36 NL·gcat-1·h-1 showed the highest total HC productivity,
calculated to be 17.1×10-3 gHC∙gFe-1∙s-1. In particular, regarding the production of C5−C12, the K-
Fe5C2/Al2O3 catalyst at GHSV = 36 NL·gcat-1·h-1 showed a 6.2×10-3 gHC∙gFe-1∙s-1 product yield, which is
-
3.4 times higher than the yield (1.8×10-3 gHC∙gFe-1∙s-1) of the K-free Fe5C2 catalyst at GHSV= 20 NL·gcat
1·h-1.
The specific C5−C13 LAO contents in the liquid oils recovered after the HT-FTS reactions were obtained
by detailed hydrocarbon analysis (Table S4-8). The K-doped catalyst showed a higher C5−C13 LAOs/C5-
C13 n-paraffin ratio than did the K-free catalyst under the same space velocity conditions (Table S5-8).
As expected, with increase of the space velocity in the FTS reaction using K-Fe5C2/Al2O3 catalyst, the
10

ratio of C5−C13 LAOs/C5−C13 n-paraffins was much increased. These ratios were measured and found to
be 0.59 at GHSV = 14 NL·gcat-1·h-1, 0.62 at GHSV = 20 NL·gcat-1·h-1, and 0.72 at GHSV = 36 NL·gcat-1·h-
1, calculated by the contents of C5-C13 LAOs and n-paraffins in isolated liquid products after reaction
using the K-Fe5C2/Al2O3 catalyst (Table S4, S5, S7). The maximum C5−C13 LAOs content (17.46 wt%)
in the recovered liquid oil was obtained with the K-Fe5C2/Al2O3 nanocatalyst at GHSV= 36 NL·gcat-1·h-1
(Table S7).
The C5−C13 LAO productivity data (the number of CH2 micromoles assigned to LAOs per gram of
catalyst per second) of reactions with K-Fe5C2/Al2O3 and K-free Fe5C2/Al2O3 nanocatalysts were also
calculated, based on the obtained C5−C13 LAO contents (Fig. 11). Figure 11a, b, and c show the detailed
productivity values of each catalyst for C5−C7, C8−C10, and C11−C13 LAOs, respectively. The K-
Fe5C2/Al2O3 nanocatalyst at GHSV = 36 NL·gcat-1·h-1 showed the highest total C5−C13 LAO productivity
(7.45 CH2 mol∙gcat-1∙s-1, Fig. 11d). This value is five times higher than the 1.45 CH2 mol∙gcat-1∙s-1 of K-
free Fe5C2/Al2O3 nanocatalyst at GHSV = 20 NL·gcat-1·h-1. In particular, 1-octene (a valuable chemical
intermediate) was effectively obtained (0.87 CH2 mol∙gcat-1∙s-1) using the K-Fe5C2/Al2O3 nanocatalyst at
GHSV = 36 NL·gcat-1·h-1 (Fig. 11b). The high productivity for LAOs mainly originated from the less H*
around the methylene monomer (CH2*) on the active surfaces, and from high catalytic conversion of the
adsorbed carbon monoxide under the increased high space velocity conditions (Fig. 12).
After the long-term FTS reaction, the recovered K-Fe5C2/Al2O3 nanocatalyst showed slightly increased
particle sizes, but any significant agglomeration or sintering from the original one was not observed (Fig.
S1).
To compare the catalytic performance of the K-Fe5C2/Al2O3 nanocatalyst with other supported
nanocatalysts for high-temperature Fischer-Tropsch synthesis, Fe5C2/SBA-15 and Fe5C2/activated carbon
(AC) were employed. The catalysts could be obtained via melt infiltration of hydrated iron nitrate salt
with the same iron loading (8 wt%) of the K-Fe5C2/Al2O3 nanocatalyst. Under the same reaction condition
at GHSV = 14 NL·gcat-1·h-1, 15 bar, and a H2:CO ratio of ‘1’, the K-Fe5C2/Al2O3 nanocatalyst showed a
much higher FTY value (6.0×10-4 molCO∙gFe-1∙s-1) at TOS = 90 h than those of Fe5C2/SBA-15 and
11

Fe5C2/activated carbon (Table S9). The C5−C13 LAOs contents in the recovered liquid oil were measured
to be 15.70% at K-Fe5C2/Al2O3, 13.93% at Fe5C2/SBA-15, and 12.43% at Fe5C2/activated carbon,
respectively (Table S10). In the C5−C13 LAO productivity data, the K-Fe5C2/Al2O3 catalyst showed much
a higher value (2.63 CH2 mol∙gcat-1∙s-1) than those of Fe5C2/SBA-15 (0.41 CH2 mol∙gcat-1∙s-1) and
Fe5C2/activated carbon (0.48 CH2 mol∙gcat-1∙s-1) under the same space velocity condition (Fig. S2). At
the K-Fe5C2/Al2O3 nanocatalyst among the hydrocarbon distribution data of the catalysts, the highest
C5−C12 selectivity (36.5%) was observed due to the appropriate hydrogenation reactions based on the
suitable surface electronic state, apt for the selective C5−C13 LAO production (Fig. S3).
Conclusions
The K-Fe5C2/Al2O3 nanocatalyst was prepared simply through a melt-infiltration of hydrated iron salt
and subsequent impregnation of aqueous potassium solution. The catalyst showed high catalytic
performance (FT activity = 1.41 mmolCO∙gFe-1∙s-1, liquid oil productivity = 2.15 gliq∙gcat-1∙h-1, and C5−C13
LAO productivity = 7.45 CH2 mol∙gcat-1∙s-1) under the optimum FTS reaction condition (GHSV condition
= 36 NL·gcat-1·h-1), based on the well-dispersed Fe5C2 nanoparticles with potassium promoter. The doped-
K on the active Fe5C2 surface under the higher space velocity conditions suppressed the hydrogenation of
alkenes and led to greater production of alkenes. It is anticipated that the optimum reaction condition and
active iron-carbide nanoparticles promoted by potassium could be effectively applied to commercial FTS
to produce high value-added LAO products.
Acknowledgements
This research was supported by the Research and Development Program of the Korea Institute of
Energy Research (KIER) (B8-2432-02). C. S. Kim and H. Choi at Kookmin University were supported
by the National Research Foundation of Korea (NRF) grant funded by Korea government (No.
2017R1A2B2012241).
12

References
[1] E. F. J. Lutz, Chem. Educ. 63 (1986) 202-203.
[2] A. Chatterjee, S. H. Hopen Eliasson, K. W. Törnroos, V. R. Jensen, ACS Catal. 6 (2016) 7784-7789.
[3] G. P. Belov, P. E. Matkovsky, Petroleum Chemistry 50 (2010) 296-302.
[4] A. K. Rausch, L. Schubert, R. Henkel, E. van steen, M. Claeys, F. Roessner, Catal. Today 275 (2016)
94-99.
[5] X. Liu, M. Tokunaga, ChemCatChem 2 (2010) 1569-1572.
[6] J. H. Gregor, Catal. Lett. 7 (1990) 317-332.
[7] L. Guo, J. Sun, X. Ji, J. Wei, Z. Wen, R. Yao, H. Xu, Q. Ge, Nat. Commun. Chem. 11 (2018) 1-8.
[8] A. P. Steynberg, R. L. Espinoza, B. Jager, A. C. Vosloo, Appl. Catal. A: Gen. 186 (1996) 41-54.
[9] H. M. Torres Galvis, K. P. de Jong, ACS Catal. 3 (2013) 2130-2149.
[10] Q. Zhang, K. Cheng, J. Kang, W. Deng, Y. Wang, ChemSusChem 7 (2014) 1251-1264.
[11] D. Leckel, Energy Fuels 23 (2009) 2342-2358.
[12] J.-Y. Park, Y.-J. Lee, P. K. Khanna, K.-W. Jun, J. W. Bae, Y. H. Kim, J. Mol. Catal. A Chem. 323
(2010) 84-90.
[13] 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: Gen. 488 (2014) 66-77.
[14] W. Chen, Z. Fan, X. Pan, X. Bao, J. Am. Chem. Soc. 130 (2008) 9414-9419.
[15] V. P. Santos, T. A. Wezendonk, J. J. D. Jaén, A. I. Dugulan, M. A. Nasalevich, H.-U. Islam, A.
Chojecki, S. Sartipi, X. Sun, A. A. Hakeem, A. C. J. Koeken, M. Ruitenbeek, W. Davidian Chen, Z. Fan,
X. Pan, X. Bao, J. Am. Chem. Soc. 130 (2008) 9414-9419.
[16] T. A. Wezendonk, V. P. Santos, M. A. Nasalevich, Q. S. E. Warringa, A. I. Dugulan, A. Chojecki,
A. C. J. Koeken, M. Ruitenbeek, G. Meima, H.-U. Islam, G. Sankar, M. Makkee, F. Kapteijn, J. Gascon,
ACS Catal. 6 (2016) 3236-3247.
[17] S.-H. Kang, J. W. Bae, P. S. Sai Prasad, K.-W. Jun, Catal. Lett. 125 (2008) 264-270.
13

[18] B. Zhao, P. Zhai, P. Wang, J. Li, T. Li, M. Peng, M. Zhao, G. Hu, Y. Yang, Y.-W. Li, Q. Zhang, W.
Fan, D. Ma, Chem 3 (2017) 1-11.
[19] H. M. Torres Galvis, J. H. Bitter, C. B. Khare, M. Ruitenbeek, A. I. Dugulan, K. P. de Jong, Science
335 (2012) 835-838.
[20] S. Jang, S. W. Kang, D. H. Chun, H.-T. Lee, J.-I. Yang, H. Jung, H.-D. Jeong, K. M. Nam, J. C. Park,
New J. Chem. 41 (2017) 2756-2763.
[21] H. Dong, M. Xie, J. Xu, M. Li, L. Peng, X. Guo, W. Ding, Chem. Commun. 47 (2011) 4019-4021.
[22] J. Gaube, H.-F. Klein, Appl. Catal. A: Gen. 350 (2008) 126-132.
[23] W. Ngantsoue-Hoc, Y. Zhang, R. J. O’Brien, M. Luo, B. H. Davis, Appl. Catal. A: Gen. 236 (2002)
77-89.
[24] K. Cheng, V. V. Ordomsky, B. Legras, M. Virginie, S. Paul, Y. Wang, A. Y. Khodakov, Appl. Catal.
A: Gen. 502 (2015) 204-214.
[25] M. C. Ribeiro, G. Jacobs, B. H. Davis, D. C. Cronauer, A. J. Kropf, C. L. Marshall, J. Phys. Chem.
C 114 (2010) 7985-7903.
[26] G. Zhao, C. Zhang, S. Qin, H. Xiang, Y. Li, J. Mol. Catal. A: Chem. 286 (2008) 137-142.
[27] Graf, B.; Muhler, M. Phys. Chem. Chem. Phys. 2011, 13, 3701-3710.
[28] J. C. Park, S. C. Yeo, D. H. Chun, J. T. Lim, J.-I. Yang, H.-T. Lee, S. Hong, H. M. Lee, C. S. Kim,
H. Jung, J. Mater. Chem. A 2 (2014) 14371-14379.
[29] J. C. Park, D. H. Chun, J.-I. Yang, H.-T. Lee, S. Hong, G. B. Rhim, S. Jang, H. Jung, RSC Advances
5 (2015) 44211-44217.
[30] M. V. Cagnoli. N. G. Gallegos, A. M. Alvarez, J. F. Bengoa, A. A. Yeramián, M. Schmal, S. G.
Marchetti, Appl. Catal. A: Gen. 230 (2002) 169-176.
[31] A. De Klerk, Fischer-Tropsch Refining; Wiley-Vch Verlag GmbH & Co. KGaA, (2011) 83–84.
[32] T. M. Eggenhuigen, J. P. Den Breejen, D. Verdoes, P. E. De Jongh, K. P. De Jong, J. Am. Chem.
Soc. 132 (2010) 18318.
14

[33] J. C. Park, D. H. Chun, J.-I. Yang, H.-T. Lee, S. Hong, G. B. Rhim, S. H. Jang, H. Jung, RSC Adv.
5 (2015) 44211.
[34] W.-P. Ma, Y.-L. Zhao, Y.-W. Li, Y.-Y. Xu, J.-L. Zhou, React. Kinet. Catal. Lett., 66 (1999) 217.
15

FIGURE CAPTIONS
Fig. 1. (a, b) TEM images, (c) HADDF-STEM image, (d) Elemental mapping image, (e) HRTEM image
of a single particle with (f) corresponding FT pattern of K-Fe5C2/Al2O3 nanocatalyst. The bars indicate
200 nm (a), 40 nm (b-d) and 2 nm (e).
Fig. 2. (a) XRD spectra of the pristine Al2O3 and K-Fe5C2/Al2O3 nanocatalyst and (b, c) XPS spectra of
K-Fe5C2/Al2O3 nanocatalyst.
Fig. 3. Mössbauer spectra at (a) 295 K and (b) 4.2 K of K-Fe5C2/Al2O3 nanocatalyst.
Fig. 4. (a) N2 adsorption/desorption isotherms, (b) Pore size distribution diagram of K-Fe5C2/Al2O3
nanocatalyst.
Fig. 5. CO conversion (a, c, e) and hydrocarbon product selectivity (b, d, f) graphs of K-Fe5C2/Al2O3
nanocatalyst by GHSV variation (a−b: GHSV = 14 NL∙gcat-1∙h-1, c-d: GHSV = 20 NL∙gcat-1∙h-1, e-f: GHSV
= 36 NL∙gcat-1∙h-1). The total CO conversion is the sum of the CO conversion to hydrocarbons (CO to HC)
and the CO conversion to CO2 (CO to CO2). All reaction tests were conducted at 340 °C, 15 bar, and H2
: CO ratio of ‘1’.
Fig. 6. CO conversion (a, c) and hydrocarbon product selectivity (b, d) graphs of K-free Fe5C2/Al2O3
-1 -1
NL∙gcat-1∙h-1). All
−d: GHSV = 36
−b: GHSV = 20
nanocatalyst by GHSV variation (a
NL∙gcat ∙h , c
reaction tests were conducted at 340 °C, 15 bar, and H2 : CO ratio of ‘1’.
Fig. 7. Catalytic performance for K-doped and K-free Fe5C2/Al2O3 catalysts by GHSV conditions. (a:
GHSV=20 NL∙gcat-1∙h-1, b: GHSV=36 NL∙gcat-1∙h-1). All reaction tests were conducted at 340 ºC, 15 bar,
and ‘1’ in the H2 : CO ratio.
Fig. 8. Hydrocarbon product distribution graphs of total syncrude (a,c,e) and liquid oil products (b,d,f)
obtained using the K-Fe5C2/Al2O3 catalyst. The samples were recovered after FTS reactions conducted at
three different GHSV conditions of (a,b) 14 NL∙gcat-1∙h-1; (c,d) 20 NL∙gcat-1∙h-1; and (e,f) 36 NL∙gcat-1∙h-1.
The total syncrude means the sum of recovered liquid oil and solid wax.
Fig. 9. The hydrocarbon distribution graphs of the K-Fe5C2/Al2O3 catalyst. The samples were recovered
after FTS reactions conducted at three different GHSV conditions.
Fig. 10. Specific hydrocarbon productivity data for K-Fe5C2/Al2O3 and K-free Fe5C2/Al2O3 nanocatalysts
in relation to GHSV variation.
16

Fig. 11. Productivity data of (a) C5−C7 LAOs, (b) C8−C10 LAOs, (c) C11−C13 LAOs, and (d) total C5−C13
LAOs for the catalysts by GHSV variation.
Fig. 12. Brief scheme for LAO formation over K-Fe5C2/Al2O3 and K-free Fe5C2Al2O3 nanocatalysts: The
combination of high space velocity condition and potassium as an alkali promoter on iron-carbide leads
to decreased secondary hydrogenation of olefins and increased carbon coupling reaction.
Scheme 1. Synthetic scheme of K-Fe5C2/Al2O3 nanocatalyst.
17

Fig. 1
18

Fig. 2
19

Fig. 3
20

Fig. 4
21

Fig. 5
22

Fig. 6
23

Fig. 7
24

Fig. 8
25

Fig. 9
26

Fig. 10
27

Fig. 11
28

Fig. 12
29