Effect of Iron Precursor on Catalytic Performance of Precipitated Iron Catalyst for Fischer–Tropsch Synthesis Reaction

Article abstract of DOI:10.1007/s10562-020-03158-3

Abstract: In this study, we synthesized three precipitated iron catalysts using continuous co-precipitation method with three different iron precursors: ferric nitrate Fe(NO₃)₃, ferrous sulfate FeSO₄, ferrous oxalate FeC₂O₄. The catalysts were characterized with XRD, SEM, XPS, N₂ sorption and H₂-TPR, respectively. The results indicated that the phases of catalysts using ferrous salts as iron precursor were composed of Fe₂O₃ and Fe₃O₄. Especially, the catalyst sample prepared by ferrous oxalate (Fe–C) exhibited the largest crystallite size, the highest crystallinity (99.78%), and the easiest reducibility. Fischer–Tropsch synthesis reaction was carried out to study the catalytic performance of three precipitated iron catalysts. In comparison with the other two catalysts, Fe–C has showed ability of the highest catalytic activity, the best stability and the highest selectivity of C₅₊ hydrocarbons. All the results lead us to draw a conclusion that the properties of iron precursors can notably affect morphology, phase structure, reduction behaviors and catalytic performances of the catalyst. Graphic Abstract: [Figure not available: see fulltext.].

Full text of DOI:10.1007/s10562-020-03158-3

Catalysis Letters
https://doi.org/10.1007/s10562-020-03158-3
Efect of Iron Precursor on Catalytic Performance of Precipitated Iron
Catalyst for Fischer–Tropsch Synthesis Reaction
Zuoxing Di1 · Xuleng Feng1 · Zhi Yang1 · Mingsheng Luo1
Received: 2 September 2019 / Accepted: 21 February 2020
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
In this study, we synthesized three precipitated iron catalysts using continuous co-precipitation method with three diferent
iron precursors: ferric nitrate Fe(NO3)3, ferrous sulfate FeSO4, ferrous oxalate FeC2O4. The catalysts were characterized
with XRD, SEM, XPS, N2 sorption and H2-TPR, respectively. The results indicated that the phases of catalysts using fer-
rous salts as iron precursor were composed of Fe2O3 and Fe3O4. Especially, the catalyst sample prepared by ferrous oxalate
(Fe–C) exhibited the largest crystallite size, the highest crystallinity (99.78%), and the easiest reducibility. Fischer–Tropsch
synthesis reaction was carried out to study the catalytic performance of three precipitated iron catalysts. In comparison with
the other two catalysts, Fe–C has showed ability of the highest catalytic activity, the best stability and the highest selectivity
of C5+ hydrocarbons. All the results lead us to draw a conclusion that the properties of iron precursors can notably afect
morphology, phase structure, reduction behaviors and catalytic performances of the catalyst.
Graphic Abstract
Keywords Precursor · Precipitated iron catalyst · Fischer–tropsch synthesis · Heterogeneous catalysis
Extended author information available on the last page of the article
Vol.:(0123456789)
1 3

Z. Di et al.
above two solutions were then poured together into a
beaker which contains 40 ml of distilled water by drop-
wise with stirring at 80 °C. The rate of dropping was pre-
cisely controlled to keep the precipitation solution pH at
8.57 0.02. After precipitation reaction, the mixture was
stirred for a further 1 h and aged for 3 h. Then the resulted
precipitate was fltered, followed by washing for several
times to neutral. The above precipitate, 0.1962 g K2CO3
(99.9%, Aldrich) and 1.1088 g silica sol (Ludox: 30 wt%,
HUIXIN Chemical Co., Ltd) were added into 500 ml
deionized water under vigorous stirring for 6 h. Finally,
the catalyst was obtained after dried at 100 °C for 12 h and
calcined at 500 °C for 6 h.
1 Introduction
Fischer–Tropsch synthesis (FTS) is a crucial technology
to produce a wide-ranging hydrocarbons from syngas that
obtained from coal, biomass or nature gas [1–6]. Com-
pared with the other FT catalysts (Ru, Ni and Co), the
iron-based catalyst is more attention-getting because of its
capability to process low H2/CO ratio syngas, high activity
for water gas shift reaction, low cost and fexible product
distribution [7–12]. Many researchers have made attempt
to improve the catalytic performance of iron-based FTS
catalysts.
As we all know, the classic commercial precipitated
iron catalyst was usually prepared from Fe(NO3)3 as iron
precursor on account of its low cost and high water solu-
bility. Nevertheless, a large amount of harmful gases (such
as NOx) that causes great environmental damage were
produced during industrial preparation. For this reason, a
few of studies have been reported [13–20]. Based on these
previous studies, it can be concluded that controlling iron
precursors plays a critical role in the modifcation of the
precipitated iron catalyst. Whereas, those explorations are
not systematic enough. Iron (II) oxalate (FeC2O4) has not
been used as iron precursor in precipitated iron catalyst
preparation. To this end, it is very desirable to study the
efect of diferent iron precursors on catalyst structure and
catalytic performance of FTS reaction.
The other two catalysts were prepared with
20 g FeSO4·7H2O (99.9%, Sino pharm), 0.6082 g
Cu(NO3)2·3H2O, 0.285 g K2CO3, 2.1484 g silica sol and 20 g
FeC2O4·3H2O (99.9%, Aldrich), 0.9401 g Cu(NO3)2·3H2O,
0.4406 g K2CO3, 3.3205 g silica sol, respectively. The three
resulted catalysts after the procedures as described above
with mass ratio of 100Fe/4Cu/4 K/16SiO2 were named as
Fe–N, Fe–S and Fe–C, respectively.
2.2 Characterization
X-ray powder diffraction was measured at room tem-
perature on Shimadzu XRD-7000 using Cu Kα radiation
(λ = 0.154 nm) at 30 mA and 40 kV. The patterns were
recorded with a step size of 4°/min from 10° to 80°. Accord-
ing to the Scherrer equation, the crystallite diameters were
calculated using the most intense refexion at 2θ=33.404°
and 30.340°.
In this work, three precipitated iron catalysts with three
diferent iron precursors [Iron(III) nitrate, Iron(II) sulfate,
iron(II) oxalate] were prepared by continuous co-precipita-
tion method. The infuence of diferent iron precursors on
the phase structure, morphology, reduction behaviors and
catalytic performance were investigated. These samples
were systematically characterized with XRD, SEM, XPS,
N2 sorption and H2-TPR. The catalytic performance was
tested via FTS reaction.
The surface morphology of samples was investigated
by SEM (XL 30 S-FEG, FEI) at an accelerating voltage of
20 kV. In order to improve conductivity, all the catalyst sam-
ples were pre-coated with a flm of gold before analysis.
The pore structure and specifc surface area of the sam-
ples were measured by N2 adsorption–desorption at−196 °C
on a Japanese BELSORP-max analyzer. Prior to measure-
ments, the samples were degassed at 300 ℃ for 4 h. And
then, the pore structure was estimated using BJH method,
the BET method was used to calculate the specifc surface
area.
2 Experimental
2.1 Catalyst Preparation
XPS spectra were collected on Thermo Fischer,
ESCALAB 250Xi X-ray photoelectron spectroscopy
equipped with a Al Kα ray (hν = 1253.6 eV) as the X-ray
source at a base pressure of 8×l0−10 Pa.
Three catalyst samples used in this paper with three difer-
ent iron precursors [Fe(NO3)3, FeSO4, FeC2O4] were pre-
pared by continuous co-precipitation method [13, 21]. Tak-
ing Iron(III) nitrate as an example, 20 g Fe(NO3)3·9H2O
(99.9%, Aldrich) and 0.4186 g Cu(NO3)2·3H2O (99.9%,
Sino pharm) was added to 400 ml deionized water with
stirring and kept at 75 ℃ for 10 min. In addition, the
sodium carbonate solution (Na2CO3, 99.9%, Aldrich) as
precipitating agent was kept at 75 °C for 10 min. The
Temperature programmed reduction with hydrogen
(H2-TPR) was completed on a Japanese BELCAT-B Chem-
ical adsorption instrument. Approximately 50 mg of the
catalyst was frst placed in a quartz reactor to pretreat in
Ar atmosphere at 300 ℃ for 1 h with the aim of removing
physically adsorbed impurities. After cooling the samples to
50 ℃, the TPR experiments were performed with mixture
1 3

Efect of Iron Precursor on Catalytic Performance of Precipitated Iron Catalyst for Fischer–…
of H2 and Ar (30 sccm) from 50 to 800 ℃ at a ramp rate of
10 ℃/ min and then maintained for 30 min. The signal for
the consumption of H2 was collected by a thermal conduc-
tivity detector.
2.3 Fischer–Tropsch Synthesis Reaction Tests
FTS reaction was conducted in a 12 mm diameter stain-
less steel reactor at reaction temperature 270 ℃, total pres-
sure 1.5 MPa and H2/CO ratio 1:1. About 1.0 g catalyst was
sieved to 60–80 mesh in size and then placed in the mid-
dle of the reactor tube. Prior to FTS reaction, each sample
was pretreated in-situ at 270 ℃ and 0.1 MPa for 24 h with
pure CO. After the reactor cooled to 50 ℃, the syngas was
Fig. 1 XRD patterns of Fe–N, Fe–S and Fe–C
introduced into it at a gas velocity of 50 ml/min. The reac-
tion products passed through two traps kept at 150 and 0 ℃,
respectively. Then, those product samples were divided
into three parts (oil, wax and aqueous) for quantifcation
analysis. Gas volatilized from the separator was analyzed
using an on-line gas chromatograph (Micro GC490, Agilent)
equipped with four fame ionization detectors (FID) and four
columns (PoraPlot 10 m, Molsieve 5 Å, CP-Sil 5CB and
PLOT Al2O3/KCl). CO conversion and reaction products
selectivity were calculated using the following equations:
Table 1 Crystalline structure of Fe–N, Fe–S and Fe–C
Iron precursor Phase structure Crystal parti- Crystallinity (%)
cle size (nm)
Fe–N
Fe–S
Fe2O3
Fe2O3
Fe3O4
Fe2O3
Fe3O4
12.0
19.9
27.8
27.9
22.4
93.8
97.39
Fe–C
99.78
C[CO]in − C[CO]out
XCO
=
× 100%
C[CO]in
57.715°, 62.538°, 64.127°, 72.094°, 77.999° correspond to
α-Fe2O3 (JCPDS No.06-0502). This reveals that the main
phase of Fe–N sample is only Fe2O3 under present condi-
tions. Whereas, unlike Fe–N sample, the XRD patterns of
Fe–S and Fe–C manifest that the phases of the two catalysts
are composed of Fe2O3 and Fe3O4 (JCPDS No. 01-1111).
Hiroshi Hayakawa and his group [13] have reported that
Fe3+ from Fe(NO3)3 formed the phase of Fe2O3, while Fe2+
[
]
out
[
C CO2
]
SCO
=
× 100%
[
]
2
C CO2 − C CO2
in
out
Shydrocarbon = 1 − S
CO2
x × C[Cx]out
C[CO]in − C[CO]out − C[CO2]out
SCx(x = 1,2,3,4) =
× 100%
SC = 1 − SCH − SC − SC − SC
4
5+
4
2
3
where XCO, SCO2, SCx and Shydrocarbon denote the conversion
of CO, the selectivity of CO2 and CxHy, respectively. C rep-
resents the concentration of CO, CxHy and CO2, respectively.
3 Results and Discussion
3.1 Catalyst Characterization
The crystalline type of the three catalysts, Fe–N, Fe–S
and Fe–C, were identifed by XRD. All of the XRD pat-
terns are shown in Fig. 1. For Fe–N, the peaks around
2θ=24.164°, 33.279°, 35.743°, 40.990°, 49.583°, 54.196°,
Fig. 2 XPS spectra of the catalysts
1 3

Z. Di et al.
from FeSO4 to FeC2O4 formed the mixture of Fe2O3 and
Fe3O4. Moreover, the Fe3O4 pattern of the Fe–S catalyst
presented a higher intensity than that of the Fe–C catalyst.
Correspondingly, the average crystallite sizes of the iron
oxide calculated using Scherrer equation and the crystallin-
ity of the three catalysts were shown in Table 1. The Fe–C
catalyst has the largest crystallite size (Fe2O3 27.9 nm and
Fe3O4 22.4 nm) and the highest crystallinity (99.78%). The
above phenomenon indicated that the diference between
iron precursors must have strong relations to the phases of
the catalysts.
material surface composition, as shown in Fig. 2. Two obvi-
ous peaks at 724.5 eV (Fe 2p1/2) and 710.9 eV (Fe 2p3/2) of
Fe–N catalyst could be ascribed to Fe2O3due to the partial
oxidation of Fe0 [22–28]. The XPS spectrums of Fe–S and
Fe–C show two series of peaks (Fe2O3 and Fe3O4) in the
range of Fe 2p1/2 and Fe 2p3/2 orbital signals. The ratio of
Fe3O4 in Fe–S (IFe2O3/IFe3O4 =0.7883) is higher than that in
Fe–C (IFe2O3/IFe3O4 =1.2935). The result is in good conform-
ity to XRD.
The morphology and microstructure of the catalysts are
researched by SEM. As can be seen in Fig. 3, three cata-
lysts prepared from three diferent iron precursors have three
totally diferent structures. Fe–N catalyst (Fig. 3a) exhib-
its spherical particles with a uniform size distributed at
To further investigate the structure of the three catalysts,
the samples were investigated using XPS analysis which is
a powerful characterization method for qualitative study of
Fig. 3 SEM and particle size distribution images of a Fe–N, b Fe–S and c Fe–C
1 3

Efect of Iron Precursor on Catalytic Performance of Precipitated Iron Catalyst for Fischer–…
31–83 nm. Fe–S catalyst (Fig. 3b) has irregular block struc-
tures with a wide particle size ranging from 16 to 270 nm.
Fe–C catalyst (Fig. 3c) possesses a long rod-like structure
(Width: 18–54 nm; Length: 78–229 nm) which are stacked
together to form 2–6 μm block structure. This diference of
structures is ascribed to the diference of the iron precursors
used in the preparation.
The surface area and porous structure of diferent sam-
ples were investigated using N2 adsorption analysis. The
results are shown in Fig. 4 and Table 2. The N2 isotherms
of three catalyst samples could all be classifed as classical
irreversible type IV adsorption–desorption isotherm [29,
30]. The hysteresis loop ranging from 0.7 to 1.0 relative
pressure implies textural mesoporous structure. Compared
with the distinct H1 hysteresis loop of Fe–N catalyst, the
Fe–S and Fe–C display the H3 hysteresis loop without obvi-
ous saturated adsorption platform. It can be concluded that
Fe–N catalyst has spherical particles with relatively nar-
row uniform size, while Fe–S and Fe–C have irregular pore
structure. As can be seen in Table 2, the BET surface area
and pore volume for Fe–N is signifcantly larger than that of
Fe-Sand Fe–C, whereas the average pore diameter of Fe–N
is smaller than the other two catalysts, as reported elsewhere
[13]. SEM images (Fig. 3) revealed that Fe–N catalyst has
the spherical particles with a uniform and the smallest size.
It appears that the small size of particles can give rise to a
signifcant increase in BET surface area and pore volume.
Figure 5 presented the profles of H2-TPR for the three
catalysts in order to investigate their reducibility. As seen
from H2-TPR, all profles of the catalysts exhibit two distinct
reduction stages. According to previous literature reports
[31–34], the frst peaks from 200 to 450 °C is attributed to
the reduction of α-Fe2O3 to Fe3O4. The other correspond-
ing peaks from 600 to 800 °C is ascribed to the reduction
of Fe3O4 to FeO or α-Fe. Compared with the tested three
catalyst samples, the TPR peaks of Fe–S and Fe–C catalysts
were obviously shifted to lower temperature. The reduction
peak of Fe–N catalyst is at 300 °C, whereas the correspond-
ing values of Fe–S and Fe–C are 294 °C and 275 °C, respec-
tively. Thus, it concludes that the Fe–C catalyst exhibits the
best reduction performance.
Fig. 4 N2 adsorption–desorption isotherms of three catalysts
Table 2 The textural properties of diferent catalysts
Iron precursor
BET surface area Pore volume
Mean pore
diameter
(nm)
(m2 g−1
)
(cm3 g−1
)
Fe–N
Fe–S
Fe–C
65.2
32.2
34.6
0.31
0.19
0.18
19.3
24.1
20.9
3.2 Catalytic Performance
Fig. 5 H2-TPR profles of the catalysts
The catalytic performance and product selectivity of the
three catalysts prepared from diferent iron precursors were
studied on a fxed-bed reactor. CO conversion rate and prod-
ucts selectivity as a function of time on stream (TOS) were
presented in Fig. 6. At the beginning of the reaction, CO
conversion of the three catalysts was all approximately 60%
at 3 h. As the reaction proceeded, it reached stable state at
about 20 h. This suggested that the catalysts have to undergo
an induction period to form the active component. As can be
seen in Fig. 6, the Fe–C catalyst generated the highest CO
conversion 98.3% at TOS=81 h on stream which far better
than the other two catalysts. However, CO conversion of
Fe–N catalyst started to decrease after 60 h on stream. From
the results, it is clear that Fe–C catalyst exhibited the highest
catalytic activity (~96.2%) among the tested three catalysts.
1 3

Z. Di et al.
Fig. 6 The catalytic performance of diferent catalysts
In the selectivity data, Fe–S and Fe–C prepared from fer-
ric iron salt showed lower selectivity of CO2, CH4 and C2–C4
hydrocarbons, but higher C5+ hydrocarbons. Meanwhile, the
results of carbon number distribution are presented in Fig. 7.
The peak center of hydrocarbons weight distribution for
Fe–S and Fe–C catalysts increased from C5–C18. As shown
in Table 3, the C5–C18 weight percent of Fe–S and Fe–C
were 85.5% and 91.0%, respectively, suggesting that it was
benefted the production of gasoline and diesel.
The iron catalytic activity was compared with FTY, an
abbreviation of Fe time yield value, which means mole num-
ber of CO converted to hydrocarbons per gram of iron per
1 3

Efect of Iron Precursor on Catalytic Performance of Precipitated Iron Catalyst for Fischer–…
Fig. 7 Carbon number distribu-
tion of hydrocarbons in liquid
product for diferent catalysts
Table 3 Carbon number distribution of C5+
4 Conclusions
Iron precursor
C5–C18 Weight percent
C18+ Weight
percent (%)
In summary, three precipitated iron catalysts with three
diferent iron precursors [Fe(NO3)3, FeSO4, FeC2O4] were
prepared by continuous co-precipitation method and the
efect of the iron precursor on the catalytic performance
of FTS reaction were studied. The characterization results
proved that the phase of Fe–N catalyst sample is only
Fe2O3, while the phases of catalysts Fe–S and Fe–C are
composed of hematite Fe2O3 and magnetite Fe3O4. This is
attributed to the efect of using ferrous salts as iron precur-
sor. And that, the Fe–C catalyst has the largest crystallite
size (Fe2O3 27.9 nm and Fe3O4 22.4 nm), the highest crys-
tallinity (99.78%), and the easiest reducibility. The three
precipitated iron catalysts were tested to investigate their
catalytic ability on Fischer–Tropsch synthesis reaction. All
reaction results confrmed Fe–C catalyst yield the highest
catalytic activity, the highest selectivity of C5+ hydrocar-
bons and the best stability. It was found that the properties
of iron precursors can signifcantly afect the phase struc-
ture, morphology, reduction behaviors and catalytic per-
formances of the catalyst. Further efort to explore more
properties of ferrous salts as iron precursor are ongoing.
(%)
Fe–N
Fe–S
Fe–C
73.8
85.5
91.0
26.2
14.5
9.0
Fig. 8 FTY results on TOS in the FT reaction
Compliance with Ethical Standards
second. From Fig. 8, Fe–C catalyst presented the highest
FTY value on stream in comparison with Fe–S and Fe–N
catalysts. Compared with the downward trend of Fe–N,
the activity of Fe–C catalyst nearly remained unchanged
state after 20 h. It can be concluded that Fe–C has the best
stability and the most active sites existed in it, which is in
good conformity to H2-TPR results. This provides further
evidence that the diference of catalytic activity is closely
related with the properties of iron precursors, which can
signifcantly afect the phase structure, morphology, reduc-
tion behaviors and catalytic performances of the catalyst.
Conflict of interest There is no confict of interest for each contribut-
ing author.
References
1. Luo M, Li H (2018) Reac Kinet Mech Catal 124:279–291
2. Di Z, Zhao T, Feng X et al (2019) Catal Lett 149:279–291
3. Luo M, Davis BH (2003) Appl Catal A 246:171–181
4. Keyvanloo K, Hecker WC, Woodfeld BF et al (2014) J Catal
319:220–231
5. Xue Y, Ge H, Chen Z et al (2018) J Catal 358:237–242
6. Hans S (1999) Appl Catal A 186:3–12
7. Saheli S, Rezvani AR (2017) J Mol Struct 1144:166–172
8. Zhang Q, Kang J, Wang Y (2010) Chem Cat Chem 2:1030–1058
1 3

Z. Di et al.
9. Ojeda M, Granados ML, Rojas S (2003) J Mol Catal A Chem
202:179–186
24. Graat PCJ, Somers MAJ (1996) Appl Surf Sci 100–101:36–40
25. Graat PCJ, Somers MAJ (1998) Surf Interface Anal 26:773–782
26. Ruby C, Humbert B, Fusy J (2000) Surf Interface Anal
29:377–380
10. Luo M, Hamdeh H, Davis BH (2009) Catal Today 140:127–134
11. Dry ME (2004) Appl Catal A 276:1–3
12. Fraser I, Rabiu AM, van Steen E (2016) Energy Procedia
100:210–216
27. Mekki A, Holland D, McConville CF et al (1996) J Non Cryst
Solids 208:267–276
13. Hayakawa H, Tanaka H, Fujimoto K (2007) Catal Commun
8:1820–1824
28. Yamashita T, Hayes P (2006) J Electron Spectrosc Relat Phenom
152:6–11
14. Sirimanothan N, Hamdeh HH, Zhang Y et al (2002) Catal Lett
82:181–191
29. Cheng K, Virgnie M, Ordomsky VV (2015) J Catal 328:139–150
30. Brunauer S, Deming LS, Deming WS et al (1940) J Am Chem
Soc 62:1723–1732
15. Bukur DB, Sivaraj C (2002) Appl Catal A 231:201–204
16. Bian G, Oonuki A, Kobayashi Y et al (2001) Appl Catal A
219:13–14
31. Munteanu G, Ilieva L, Andreeva D (1997) Thermochim Acta
291:171–177
17. Wu B, Bai L, Xiang H (2004) Fuel 83:205–212
18. Xiong H, Moyo M, Motchelaho MAM (2010) Appl Catal A Gen
388:168–178
32. Jin Y, Datye AK (2000) J Catal 196:8–17
33. Bukur DB, Lang X, Mukesh D, Zimmerman WH, Rosynek MP,
Li C (1990) Ind Eng Chem Res 29:1588–1599
34. Zhang CH, Yang Y, Teng BT et al (2006) J Catal 237:405–415
19. Torres Galvisa HM, Koekena ACJ, Bitter JH (2013) Catal Today
215:95–102
20. Wei Y, Luo D, Zhang C (2018) Catal Sci Technol 8:2883–2893
21. Ma W, Jacobs G, Sparks DE (2015) J Catal 326:149–160
22. Yamashita T, Hayes P (2008) Appl Surf Sci 254:2441–2449
23. Roosendaal SJ, van Asselen B, Elsenaar JW et al (1999) Surf Sci
442:329–337
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional afliations.
Afliations
Zuoxing Di1 · Xuleng Feng1 · Zhi Yang1 · Mingsheng Luo1
1
* Mingsheng Luo
Center for Applied Energy and Environmental Research,
luoms9297@163.com
Beijing Institute of Petrochemical Technology, 19 Qing-Yuan
North Road, Beijing 102600, China
Zuoxing Di
dizuoxing@bipt.edu.cn
1 3