Enhanced Production of C5+ Hydrocarbons from CO2 Hydrogenation by the Synergistic Effects of Pd and K on Γ-Fe2O3 Catalyst

Article abstract of DOI:10.1007/s10562-018-2622-y

Abstract: Pure γ-Fe₂O₃ was prepared by two steps. L(+)-Tartaric acid and Fe(NO₃)₃·9H₂O were sequentially dissolved into water, then the stick liquid was dried and calcined in air. Promoter Pd and K we

Full text of DOI:10.1007/s10562-018-2622-y

Catalysis Letters
https://doi.org/10.1007/s10562-018-2622-y
Enhanced Production of C Hydrocarbons from CO Hydrogenation
5
+
2
by the Synergistic Effects of Pd and K on γ-Fe O Catalyst
2
3
Wensheng Ning1 · Bei Li · Biao Wang · Xiazhen Yang · Yangfu Jin
1
1
1
2
Received: 16 August 2018 / Accepted: 19 November 2018
©
Springer Science+Business Media, LLC, part of Springer Nature 2018
Abstract
Pure γ-Fe O was prepared by two steps. L(+)-Tartaric acid and Fe(NO ) ·9H O were sequentially dissolved into water,
2
3
3 3
2
then the stick liquid was dried and calcined in air. Promoter Pd and K were added to Fe O by impregnation. The influ-
2
3
ence of Pd or K on Fe O was different in view of the reducibility of Fe O , H and CO adsorption. But synergistic effects
2
3
2
3
2
2
were observed from the catalyst co-promoted by Pd and K (Catalyst PKF). The improving effect of Pd on Fe O reduction
2
3
was further stimulated by K, and the CO quantity desorbed from catalyst PKF at the temperature lower than 300 °C was
2
increased. The synergistic effects resulted in PKF having the highest C selectivity for CO hydrogenation at 235 °C among
5
+
2
the studied catalysts.
Graphical Abstract
Keywords γ-Fe O catalyst · CO hydrogenation · Promoter Pd · Promoter K · Synergistic effect
2
3
2
1
Introduction
Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s10562-018-2622-y) contains
supplementary material, which is available to authorized users.
To hydrogenate CO into C hydrocarbons is a novel
2 5+
method to produce liquid fuels. It can overcome the
worry on the shortage of crude oil, because the crude oil
is main source for liquid fuels at present. The method is
also helpful to mitigate the greenhouse effect originated
*
Wensheng Ning
wenshning@sohu.com
from anthropogenic CO emission. The concept and trial
Extended author information available on the last page of the article
2
Vol.:(0
1
123456789)
3

W. Ning et al.
using seawater as starting materials brings an applicable
2 Experimental
and profitable scene to CO hydrogenation [1, 2]. With
2
the concept breakthrough where and how to perform CO
2.1 Catalyst Preparation
2
hydrogenation, the active catalysts are the key factor to
commercialize CO hydrogenation [3].
18.576 g L(+)-Tartaric acid was dissolved with 15.0 g dis-
tilled water in a 250 mL flask, then 40.0 g Fe(NO ) ·9H O
2
In the last year, some catalysts were reported with high
3
3
2
selectivity of C hydrocarbons. Na–Fe O /HZSM-5 cata-
was added into the solution and dissolved by stirring at room
temperature for about 40 min to form stick liquid. The liquid
was put into an oven of 90 °C for 12 h, and the formed black
solid was calcined at 350 °C for 6 h. The final powder was
pressed into particles of 150–300 µm as catalyst precursor.
The precursor was impregnated with Pd and (or) K in 2 wt
% relative to the mass of precursor. After the impregnated
precursors were dried at 110 °C for 12 h, they were calcined
at 350 °C for 6 h. The obtained catalysts were expressed as
F, PF, KF and PKF according to the promoter(s) added to the
precursor. The catalysts were named as UF, UPF, UKF and
5
+
3
4
lyst can directly convert CO to gasoline-range (C –C )
2
5
11
hydrocarbons with the selectivity up to 78% of all hydro-
carbons at a CO conversion of 22% under the condi-
2
tions of H /CO = 3, 320 °C, 3 MPa and 4 L/(h·gcat) [4].
2
2
A catalyst composed of reducible indium oxides (In O )
2
3
and zeolites yields a selectivity to gasoline-range hydro-
carbons (78.6%) in hydrocarbon distribution at a CO
2
conversion of 13.1% under the conditions of H /CO /
2
2
N = 73/24/3, 340 °C, 3.0 MPa and 9 L/(h·gcat) [5]. The
2
high C selectivity is achieved by the necessary compo-
5
+
nent of zeolite in these bifunctional catalysts. Because
the working temperature for zeolite is high [6–8], the
bifunctional catalysts have to be used at temperature
UPKF after they experienced CO hydrogenation reaction.
2
higher than 320 °C [4, 5]. However, CO hydrogenation
2.2 Activity Test and Product Analysis
2
is an exothermic reaction [9]. It is thermodynamically
favorable to operate CO hydrogenation at lower tempera-
The reactivity of catalysts was tested in a stainless steel fixed
bed reactor of inter-diameter of 8 mm [17]. 1.0 g catalyst
(150–300 µm) was filled into the reactor. After the catalyst
was reduced in CO of 50 mL/min at 300 °C for 6 h, it was
cooled to room temperature. Then, the feed gas was changed
into pre-mixed gases of H :CO :N =69:23:8 of 1.6 MPa and
2
ture, with additional benefits including less costs for the
reactor construction and operation. The low temperature
operation would make zeolite useless in the CO hydro-
2
genation. It is necessary to look for novel composition
of catalyst.
2
2
2
CO is the final oxidized product of element C. Its
100 mL/min. The catalyst was heated to 235 °C in about 3 h
and kept at it for activity evaluation of 21 h. The condensa-
ble products were collected in a cold trap of 0 °C at system
pressure. After the system pressure was released through a
backpressure regulator, the exited gas was analyzed by GC
A90 (Shanghai Yimeng LTD.) on line. The relative contents
of CO, CH , CO and N were supplied with TCD detector
2
4
+
utilization is only available by reducing the C in CO to
2
low valence. CO hydrogenation leverages the reduction
2
of CO by H . Jun et al. [10] reported the influence of H
2
2
2
content in feed gas on the conversion of CO and CO to
2
hydrocarbons. The reaction with H -deficient feed (CO/
2
CO = 0.33, H /(2CO + 3CO ) = 0.44) showed that only
2
2
2
4
2
2
CO was converted to hydrocarbons, while in H -enriched
and TDX-01 column. N was used as an internal standard for
2
2
feed (CO/CO = 0.33, H /(2CO + 3CO ) = 1), CO was
the quantification of CO, CH and CO . C –C hydrocarbons
2
2
2
2
4
2
1
5
converted to hydrocarbons as well as CO. The high con-
were analyzed with FID detector and Porapak Q column.
centration of H was thought to promote the conversion of
2
CO into CO by reverse Water–Gas shift (WGS) reaction,
2
2.3 Characterization
followed by Fischer–Tropsch synthesis (FTS) reaction in
which CO was further hydrogenated to hydrocarbons.
Such effect may be realized in the method of enhancing
Surface area and pore structure of the catalysts were meas-
ured by ASAP-2020 of Micromeritics at liquid nitrogen tem-
perature. The crystal structure of the catalysts was acquired
by X-ray diffraction (XRD, PNAlytical X’Pert Pro diffrac-
tometer) with a Cu Kα radiation source (λ= 0.15406 nm)
in reflection mode. X-ray tube was operated at 40 kV and
H adsorption on catalyst which would result in higher
2
H /CO ratio on catalyst surface than the feed gas. Pd can
2
2
enhance the hydrogenation activity of Fe catalyst [11–13],
while K increase CO adsorption on Fe-based catalyst
2
[
9, 14, 15]. Our previous works have found that Fe-based
4
0 mA. Temperature-programmed reduction (TPR) was
catalyst in γ-Fe O phase is better than α-Fe O phase for
2
3
2
3
carried out in PX200 (Tianjin Pengxiang LTD.) with TCD
detector. 0.03 g catalyst was fixed into U-type quartz tube
and pretreated with 25 mL/min He at 300 °C for 1 h. After
the sample was cooled to room temperature in He stream,
CO hydrogenation [3, 16]. Therefore, the combination
2
of Pd and K was added to γ-Fe O in order to prepare
2
3
Fe-based catalyst active for CO hydrogenation at low
2
temperature.
1
3

Enhanced Production of C5+ Hydrocarbons from CO Hydrogenation by the Synergistic Effects…
2
the feedgas was shifted into 5% H /N of 30 mL/min and
chain growth, which resulted in more paraffinic and light
2
2
the sample was heated in 10 °C/min to 800 °C for H -TPR.
hydrocarbons in the products from CO hydrogenation [13].
2
For temperature-programmed desorption (TPD) of H and
According to the data in Table 1, this kind of adsorbed H
2
2
CO measurement, 0.5 g catalyst was fitted into the U-type
only hydrogenated the re-adsorbed olefins without any dis-
2
quartz tube and pretreated by 20 mL/min H from room tem-
turbance on the chain growth in CO hydrogenation.
2
2
perature to 260 °C in 1 h and to 320 °C in another 1 h and
Unlike catalyst PF, KF’s behavior is completely opposite
kept at 320 °C for 2 h. Then, the sample was flushed with
to that of catalyst F. Its CO conversion and CH selectiv-
2
4
2
5 mL/min N at 320 °C for 0.5 h and cooled to 50 °C.
ity are decreased, but CO selectivity, C selectivity and the
2
5+
The reduced sample was contacted with 20 mL/min H or
olefin/paraffin ratio in C –C hydrocarbons are increased.
2
2
5
5
mL/min CO at 50 °C for 0.5 h and subsequently heated
It has been reported that K inhibited H adsorption on Fe
2
2
to 800 °C at the rate of 10 °C/min in 25 mL/min N . XPS
catalyst [22] which brought forth the observed change of
hydrocarbons. The increased CO selectivity is related to the
promotion of K on the rate of WGS reaction [23–25].
2
analysis was done by ESCALAB 250Xi (ThermoFischer
Ltd.) using Al Kα radiation (1486.6 eV). The binding energy
of C 1 s (284.8 eV) was used to calibrate the peak position
of other elements.
For the doubly promoted catalyst PKF, its CO conversion
2
and CO selectivity is between that of PF and KF, reflect-
ing the effects of Pd and K on the two items are mediated
each other. Notably, PKF has the highest C selectivity and
5
+
3
Results
olefin/paraffin ratio in C –C hydrocarbons, and the lowest
3 5
CH selectivity among the four catalysts. The reactive per-
4
3.1 Catalytic Performance
formance of PKF does obey neither the trend from F to PF,
nor the trend from F to KF according to the data in Table 1.
Some synergies happened between Pd and K on the doubly
promoted PKF. The physical mixture of equal PF and KF
(PF+KF) was evaluated, too. All the items of its catalytic
performance are between that of PF and KF. Because the
performances of CH selectivity, C selectivity and olefin/
Table 1 lists the reactive performance of four catalysts in
CO hydrogenation. Catalyst F and PF present similar CO
2
2
conversion, CO selectivity and hydrocarbon selectivity, but
the olefin/paraffin ratio of C –C hydrocarbons is different
2
5
for the two catalyst. The C –C hydrocarbons produced by
2
5
4
5+
PF has less olefinic content than F. We had observed that the
paraffin ratio in C –C hydrocarbons of PF+KF are contrary
3 5
selectivity of CH and paraffin in C –C hydrocarbons were
to PKF, a close contact of the two promoters is the prerequi-
site to the synergies between Pd and K which can be realized
by co-impregnation of Pd and K to Fe O precursor.
4
2
4
simultaneously increased by Pd-promoted Fe catalyst in CO
hydrogenation [13]. Promoter Pd can enhance H adsorp-
2
2
3
tion on catalyst, but it seems that the route of adsorbed H
The optimum amounts of Pd and K to γ-Fe O were
2
2 3
induced by Pd to react with the intermediate might depend
explored and the results are tabulated in Table S1 in Sup-
porting Information. PKF is the best one to synthesize
on the feed gas composition (CO and H , or CO and H ).
2
2
2
The synthesized hydrocarbons in FTS reaction (CO hydro-
genation) are divided into primary products and secondary
products [18, 19]. Olefins are primary products. They desorb
from active sites and then diffuse through catalyst pores to
form products. In the diffusing process, olefins are easily
re-adsorbed on catalyst for chain growth, or hydrogenated
C hydrocarbons from CO and H . It worked stably during
5+ 2 2
100 h on stream as shown in Figure S1.
3.2 Texture Structure
Figure 1 is the nitrogen adsorption/desorption isotherms of
the four catalysts. It belongs to the Type IV isotherm with
capillary condensation taking place in meso-pores [26].
into paraffins [18, 20, 21]. The adsorbed H induced by Pd
2
hydrogenated the re-adsorbed olefins and interrupted the
Table 1 The catalytic performance in CO hydrogenation
2
ꢀ
ꢀ
ꢀ
ꢀ
0
=
0
C2 C2
=
0
C3 C3
=
0
C4 C4
=
C5 C5
Catalyst
CO conv. (%)
CO selec. (%)
CH selec. (%)
C5+ selec. (%)
2
4
F
PF
KF
PKF
PF+KF
23.4
23.8
16.0
20.7
22.0
6.2
7.1
24.1
18.2
10.6
40.0
39.9
13.5
11.1
30.8
15.8
15.2
40.6
47.7
22.3
0.2
0.1
2.1
1.3
0.2
0.7
0.4
2.8
3.2
1.0
0.6
0.4
2.7
3.1
0.9
0.2
0.1
2.3
2.6
0.5
a
Reaction condition: H :CO :N =69:23:8, 1.6 MPa, 6.0 L/(h·g-cat), 235 °C
2
2
2
a
PF+KF is the physical mixture of equal amount of catalyst PF and KF
1
3

W. Ning et al.
Fig. 1 N adsorption and
2
desorption isotherms of the
catalysts
Table 2 Texture structure of the catalysts
Catalyst
Surface area
Pore volume
(mL/g)
Average pore
diameter (nm)
2
(
m /g)
F
64.5
45.3
39.1
36.0
0.19
0.15
0.15
0.13
10.2
11.2
11.5
11.2
PF
KF
PKF
of 3.7 nm disappeared from catalyst PF, KF and PKF due
to the blocking of promoter to the pore entrance, which
induced pore volume shrunk and pore diameter enlarged
for the promoted catalysts [3].
Fig. 2 Pore size distribution of the catalysts
The surface area of catalyst F is higher than other non-
promoted Fe catalysts [3, 27–30]. Especially, the raw
materials for catalyst F preparation is as same as the P-3
Figure 2 presents the distribution of pore size in the four
catalysts. There are two pore sizes in catalyst F which cen-
tered on 3.7 nm and 11.4 nm, respectively. The small one
disappears from the mono-promoted and double-promoted
catalysts, and the area beneath these curves, i.e. the pore
volume, is shrunk.
2
(55.1 m /g) in our previous work [3], but high surface
area is obtained for catalyst F by changing the prepar-
ing method from solidly grinding Fe(NO ) ·9H O and
3
3
2
L(+)-Tartaric acid to mixing the two chemicals sequen-
tially in water, and decreasing the calcination tempera-
ture. Recently, the specific surface area of pure γ-Fe O
Table 2 compares the surface area, pore volume and
average pore diameter of the four catalysts. After catalyst
F was impregnated with promoter Pd or (and) K, the sur-
face area and pore volume are decreased, while the pore
diameter is enlarged. Figure 2 has disclosed that the pore
2
3
2
materials is increased to 75.8 m /g by simultaneously dis-
solving Fe(NO ) ·9H O and L(+)-Tartaric acid in water.
3
3
2
There is potential to make high dispersive γ-Fe O from
2
3
Fe(NO ) ·9H O and L(+)-Tartaric acid which is beneficial
3
3
2
to increase the activity of Fe O -based catalyst.
2
3
1
3

Enhanced Production of C5+ Hydrocarbons from CO Hydrogenation by the Synergistic Effects…
2
3.3 Crystal Structure
The XRD patterns of catalyst F, PF, KF and PKF are shown
in Fig. 3. The main peaks in F are assigned to γ-Fe O
2
3
(
PDF: 39-1346), except the very weak one at 33.1° related
to α-Fe O (PDF: 33-0664). After catalyst F was impreg-
2
3
nated with Pd and K, the peak at 33.1° becomes high only
in PF and PKF, i.e. the catalysts promoted by Pd. In contrast,
3
+
there is no change around 33.1° in KF. The radius of Fe ,
2
+
1+
Pd and K is 0.64 Å, 0.86 Å and 1.33 Å, respectively. The
impregnated K ion mainly distributed on the catalyst surface,
while Pd ion probably entered into the cation vacancy or
3
+
replace Fe in crystal Fe O . Such entrance is able to trans-
2
3
fer some γ-Fe O into α-Fe O , because γ-Fe O is meso-
2
3
2
3
2
3
stable phase of Fe O .
2
3
The crystalline of the four catalysts experienced CO
2
hydrogenation are displayed in Fig. 4. Fe C (PDF: 51-0997)
5
2
is detected in the four catalysts, but Fe C (PDF: 17-0333)
7
3
is only found in UKF. Iron carbides are the active phase for
carbon chain growth [4, 7, 9, 12]. The C/Fe atom ratio in
Fe C is higher than Fe C . Because K can increase carburi-
7
3
5 2
zation of Fe-based catalysts [9, 31], the appearance of Fe C
7
3
in UKF is related to promoter K. Although the four catalysts
were not stored in inert atmosphere after they were taken
out of the reactor, there is no iron oxide found in them. The
Fig. 4 The XRD patterns of catalysts experienced reaction
hydrocarbons of long carbon chain remained on the catalyst
surface may protect the catalysts from oxidation.
3.4 Reducibility in H2
Figure 5 compares the reducibility of four catalysts in H
2
stream. Two groups of H consumption peak can be found
2
for catalyst F. The peak around 290 °C is from the reduction
Fig. 3 The XRD patterns of catalyst F, PF, KF and PKP
Fig. 5 The H -TPR profiles of the four catalysts
2
1
3

W. Ning et al.
of γ-Fe O to Fe O , and the peak around 570 °C is the
2
3
3
4
reduction of Fe O to α-Fe. There are two shoulder peaks
3
4
at 330 °C and 630 °C, respectively. The shoulder peak at
3
30 °C is the reduction of α-Fe O which was produced from
2
3
meso-stable γ-Fe O during the heating progress for H -TPR
2
3
2
[
3], and the peak at 630 °C is the reduction of the rest Fe O
3 4
to α-Fe with Fe O as intermediate [32]. For catalyst PF, the
X
first peak is shifted down to 96 °C. It contains the reduction
2
+
0
of Pd to Pd and γ-Fe O to Fe O [13]. It was reported
2
3
3
4
that the reduction of PdO happened before 72 °C for Pd/C
[
33], 150 °C for Pd/ZnO [34] and 174 °C for Pd/Al O cata-
2 3
lyst [35]. The improvement of promoter Pd on the reduction
of catalyst is disclosed not only by the much low peak tem-
peratures of both γ-Fe O and α-Fe O to Fe O , but also the
2
3
2
3
3
4
downshifted starting and ending temperatures for Fe O to
3
4
α-Fe. Figure S2 in Supporting Information gives clear evi-
dence of Pd improving the reduction of catalyst. According
to the peak temperatures of catalyst KF, promoter K inhibits
the reduction of catalyst. However, the inhibition of K disap-
pears in the doubly promoted catalyst PKF. Promoter K even
stimulated the improving effect of Pd on the catalyst reduc-
tion, because the first peak area of PKF is bigger than PF.
There are two separate peaks for PF below 300 °C, but
they merge into one for PKF. It is the result of simultaneous
reduction of PdO and Fe O . This phenomenon indicates
2
3
Fig. 6 XPS spectrum of element Fe on the catalysts after CO hydro-
that Pd atom distributed homogeneously with Fe atom in
PKF. The homogeneous distribution of Pd with Fe was pro-
moted by the co-existed K. It has been deducted that Pd ion
2
genation
3
+
probably entered into the cation vacancy or replace Fe in
crystal Fe O based on the results in Fig. 3.
distributed on catalyst surface. It has been pointed out that
Pd ion probably entered into the cation vacancy or replace
2
3
3
+
Fe in crystal Fe O by analyzing the results in Figs. 3, 5
2
3
3.5 XPS Result
and Figure S2.
The catalysts experienced CO hydrogenation were meas-
2
ured by XPS. Figure 6 is the detected Fe 2p signal, and
Fig. 7 is K 2p signal. The binding energy of Fe 2p in the
Pd-promoted catalysts (PF and PKF) is lower than the cata-
lysts without Pd (F and KF). The binding energy of K 2p in
PKF is lower than KF, too. The shift may be resulted from
the electron-supply of Pd to Fe. Pd 3d signal was also col-
lected, but it was too weak to be separated from the noisy
background. Although it has been speculated that there is
hydrocarbons remained on the catalyst surface according to
Fig. 4, the signal of Pd could not be completely weakened
by the remained hydrocarbons. The assertion is supported
by comparing KF and PKF. Catalyst PKF is more active than
KF to form C hydrocarbons (Table 1). The hydrocarbons
3.6 H ‑TPD and CO ‑TPD
2 2
Figure 8 discloses the adsorption strength of H on the cata-
2
lysts. For catalyst F, the initial rising started from 50 °C is
related to physically adsorbed H , and the main desorption
2
happened around 400 °C. For catalyst PF, the quantity of
desorbed H situated at 160 °C is more than 388 °C. Pd
2
created sites for H adsorption at low temperature on PF.
2
Consisted with the finding that K inhibited H adsorption on
2
Fe catalyst [22, 25, 31, 36], there is no physical adsorption
of H on catalyst KF, and only one peak around 396 °C. The
2
peak temperature of catalyst PKF is at 105 °C and 386 °C,
respectively. In general, the desorption pattern of PKF is
5
+
remained on PKF is correspondingly more than KF. How-
ever, the strengths of Fe 2p and K 2p in PKF are approximate
to KF. Because the remained hydrocarbons do not weaken
XPS detection on Fe and K, it could not only influence
Pd signal. The possible reason for the lack of Pd signal is
that Pd inserted into the bulk of iron carbide, rather than
similar to PF, but the adsorption strength of H on PKF,
2
which desorbed below 250 °C, is weakened by K. Compar-
ing the pattern of F with PF, Pd enhanced H adsorption at
2
low temperature, while inhibited that at high temperature. It
reflects that the catalyst surface is modified by Pd through
the electron-supply of Pd to Fe according to Figs. 6 and 7.
1
3

Enhanced Production of C5+ Hydrocarbons from CO Hydrogenation by the Synergistic Effects…
2
Figure 9 contrasts the adsorption strength of CO on the
2
catalysts. The physically adsorbed CO molecular on catalyst
2
F desorbed from 50 °C and a peak centered at 405 °C was
originated from chemical adsorption of CO . There are two
2
peaks around 173 °C and 389 °C on catalyst PF. For catalyst
KF, besides the strong adsorption located at 388 °C, there
are physically adsorbed CO in the range of 50 °C–110 °C
2
and one peak situated at 202 °C. The desorption peaks are at
1
99 °C and 387 °C for PKF. The ability of the site to adsorb
2
CO is further increased by Pd as reflected by comparing the
peak area of PKF and KF below 300 °C.
For catalyst F, the pattern (indexed by the number and
temperature of the desorption peak) of H desorption is as
2
same as that of CO desorption. It means that the adsorption
2
site for H on catalyst F is the same one for CO adsorp-
2
2
tion. Therefore, there is competition to the adsorption site
between H and CO during CO hydrogenation. This con-
2
2
2
clusion can be applied to catalyst PF, too. Because H con-
2
tent was three times of CO in the feed gas during the hydro-
2
genation reaction, the quantity of adsorbed H was more than
2
the adsorbed CO on catalyst F and PF under the assump-
2
tion that the adsorbed quantity was proportional to its partial
pressure in gaseous phase. The high ratio of adsorbed H to
2
CO on catalyst F and PF resulted in the high content of
2
paraffin in the products (Table 1). Promoter K can enhance
Fig. 7 XPS spectrum of element K on the catalysts after CO hydro-
2
genation
Fig. 8 The H -TPD profiles of the four catalysts
2
Fig. 9 The CO -TPD profiles of the four catalysts
2
1
3

W. Ning et al.
CO adsorption [9, 14, 15, 25, 36], and CO adsorbs on both
Because CO hydrogenation was tested at 235 °C, the CO
2
2
2
2
Fe and K [15, 22, 36]. There is site only adsorbing CO
and H desorbed from catalysts at the temperature higher than
2
2
on catalyst KF as indicated by the physically adsorbed CO
300 °C would be too stable to take part in the reaction. In
2
and very small peak at 202 °C, whereas no H is desorbed
other words, the CO and H desorbed from catalysts below
2
2
2
in this temperature range. An apparent difference between
300 °C were the main participator for CO hydrogenation.
2
the adsorption strength of CO (199 °C) and H (105 °C) is
With this knowledge, the data in Table 1, especially why PKF
2
2
observed on catalyst PKF. Based on the results in Figs. 8 and
synthesized hydrocarbons with the highest C selectivity and
5+
9
, it can be concluded that Pd created sites for both H and
olefin/paraffin ratio can be understood. On PKF, not only the
2
CO adsorption, but K created sites only for CO adsorption.
activated degree of CO is higher than H , but also the relative
2
2
2
2
ratio of adsorbed CO to H is also high. The above adsorp-
2
2
tion state is beneficial to chain growth of hydrocarbons and
decrease the probability of the primary olefin hydrogenated
into paraffin.
4
Discussion
Although promoter Pd and K were added to γ-Fe O by the
2
3
same method of impregnation, they existed differently in the
5 Conclusions
catalysts. There is K signal and none of Pd in XPS analysis
(
Fig. 7). Therefore, K mainly distributed on catalyst surface,
Pure γ-Fe O was prepared by a simple method that was to
2
3
but Pd entered into the bulk of Fe O . The appearance of
dissolve L(+)-Tartaric acid and Fe(NO ) ·9H O in water by
2
3
3 3 2
α-Fe O in PF and PKF in Fig. 3 and the merged reduction
sequence, followed by drying and calcination. After Fe O was
2 3
2
3
peak in Fig. 5 support the deduction that Pd ion probably
impregnated with promoter Pd and K, the crystal phase was not
influenced. Pd ion probably entered into crystal Fe O . Fe C
entered into crystal Fe O .
2
3
2
3
5 2
Contrasting the results in Fig. 5, synergistic effects hap-
was formed in the four catalysts experienced CO hydrogena-
2
pened on PKF which was doubly promoted by Pd and K.
tion, but Fe C was also found in KF. For the mono-promoted
7
3
According to the temperature of H consumption peaks, Pd
catalyst PF and KF, the reduction of Fe O was inhibited by
2
2 3
improves the reduction of Fe O in PF, but K inhibits Fe O
K, and improved by Pd, respectively. However, the effect of
K on the reduction was reversed in the doubly promoted cata-
lyst PKF. The improving effect of Pd was further stimulated
2
3
2
3
reduction in KF. Surprisingly, not only the inhibition of K is
disappeared from the doubly promoted catalyst PKF, but also
the improving effect of Pd on the reduction are enhanced. The
area of the first peak in PKF is evidently bigger than PF. The
synergistic effects due to the co-existed Pd and K also presents
in the manner to influence H and CO adsorption on catalyst
by K. According to the analysis to the results of H -TPD and
2
CO -TPD, Pd created sites for both H and CO adsorption,
2
2
2
while K created sites only for CO adsorption. The quantity of
2
adsorbed CO below 300 °C was further increased by Pd on
2
2
2
PKF. By discussing the results in Figs. 8 and 9, it is concluded
that Pd created sites for both H and CO adsorption, and K
PKF. The synergistic effects from co-existed Pd and K resulted
in PKF possessing the highest C selectivity and olefin/paraf-
2
2
5+
created sites only for CO adsorption. Relative to the adsorbing
fin ratio in C –C hydrocarbons, and the lowest CH selectivity
2
3
5
4
strength (indexed by the desorption temperature) of H and
among the four catalysts for CO hydrogenation at 235 °C.
2
2
CO on PF below 300 °C, CO adsorption is intensified and
2
2
Acknowledgements This work was supported by the Zhejiang
Provincial Natural Science Foundation of China [LY14B030003]
and the National Ministry of Science and Technology of China
[2014BAD02B05].
H adsorption is weaken on PKF. The desorbing temperature
2
for CO and H from PKF is 199 °C and 105 °C, respectively.
2
2
Correspondingly, the activated degree of CO is higher than
2
H on PKF. Figures 8 and 9 are the result of mono-adsorption
2
of H or CO on catalysts, respectively. During the reaction
2
2
Compliance with Ethical Standards
of CO hydrogenation, H and CO contacted with catalysts
2
2
2
simultaneously. Therefore, there is competition for H and CO
Conflict of interest On behalf of all authors, the corresponding author
2
2
states that there is no conflict of interest.
to occupy the adsorption site. The adsorbed quantities of H
2
and CO are determined by two factors. One is the adsorbate
2
content in the feed gas, and another is the interaction between
References
the adsorbate and adsorption site. Because H content is three
2
times of CO in the reactants, the quantity of adsorbed H
2
2
1
.
Drab DM, Willauer HD, Olsen MT et al (2013) Hydrocarbon
synthesis from carbon dioxide and hydrogen: a two-step process.
Energy Fuels 27:6348–6354. https://doi.org/10.1021/ef4011115
would be more than the adsorbed CO on catalysts. This state
2
is changed on catalyst PKF as the result of CO adsorption was
2
intensified and H adsorption was weakened on the catalyst.
2. Willauer HD, DiMascio F, Hardy DR et al (2014) Feasibility of
2
CO extraction from seawater and simultaneous hydrogen gas
The relative ratio of adsorbed CO to H on PKF is probably
2
2
2
generation using a novel and robust electrolytic cation exchange
the highest among the four catalysts during reaction.
1
3

Enhanced Production of C5+ Hydrocarbons from CO Hydrogenation by the Synergistic Effects…
2
module based on continuous electrodeionization technology. Ind
Eng Chem Res 53:12192–12200. https://doi.org/10.1021/ie502
20. Schulz H (1999) Short history and present trends of Fis-
cher–Tropsch synthesis. Appl Catal A 186:3–12. https://doi.
org/10.1016/S0926-860X(99)00160-X
1
28x
3
4
5
6
7
.
.
.
.
.
Ning W, Wang T, Chen H et al (2017) The effect of Fe O crys-
21. Liu J, Zhang A, Jiang X et al (2018) Selective CO hydrogenation
2
3
2
tal phases on CO hydrogenation. PLoS ONE 12(8): e0182955–
to hydrocarbons on Cu-promoted Fe-based catalysts: dependence
on Cu–Fe interaction. ACS Sustain Chem Eng 6:10182–10190.
https://doi.org/10.1021/acssuschemeng.8b01491
2
e0182912. https://doi.org/10.1371/journal.pone.0182955
Wei J, Ge Q, Yao R et al (2017) Directly converting CO into
2
a gasoline fuel. Nat Commun 8:15174–15178. https://doi.
org/10.1038/ncomms15174
22. Dry ME, Shingles T, Boshoff LJ et al (1969) Heats of chem-
isorption on promoted iron surfaces and the role of alkali in
Fischer-Tropsch synthesis. J Catal 15:190–199. https://doi.
org/10.1016/0021-9517(69)90023-2
Gao P, Li S, Bu X et al (2017) Direct conversion of CO into liquid
2
fuels with high selectivity over a bifunctional catalyst. Nat Chem
9
:1019–1024. https://doi.org/10.1038/NCHEM.2794
23. Anderson RB, Seligman B, Schultz JF et al (1952) Fischer-
Tropsch Synthesis. Some important variables of the synthesis on
iron catalysts. Ind Eng Chem 44:391–397. https://doi.org/10.1021/
ie50506a052
Guan N, Liu Y, Zhang M (1996) Development of catalysts for the
production of aromatics from syngas. Catal Today 30:207–213.
https://doi.org/10.1016/0920-5861(96)00014-4
Yoneyama Y, He J, Morii Y et al (2005) Direct synthesis of isopar-
affin by modified Fischer–Tropsch synthesis using hybrid catalyst
of iron catalyst and zeolite. Catal Today 104:37–40. https://doi.
org/10.1016/j.cattod.2005.03.031
24. Krishnamoorthy S, Li A, Iglesia E (2002) Pathways for CO
2
formation and conversion during Fischer-Tropsch synthe-
sis on iron-based catalysts. Catal Lett 80:77–86. https://doi.
org/10.1023/A:1015382811877
8
9
.
.
Cheng K, Gu B, Liu X et al (2016) Direct and highly selective
conversion of synthesis gas into lower olefins: design of a bifunc-
tional catalyst combining methanol synthesis and carbon–car-
bon coupling. Angew Chem Int Ed 55:4725–4728. https://doi.
org/10.1002/anie.201601208
25. Boreriboon N, Jiang X, Song C et al (2018) Fe-based bimetallic
catalysts supported on TiO for selective CO hydrogenation to
2
2
hydrocarbons. J CO2 Util 25:330–337. https://doi.org/10.1016/j.
jcou.2018.02.014
26. Sing KSW, Everett DH, Haul RAW et al (1985) Reporting phys-
isorption data for gas/solid systems with special reference to the
determination of surface area and porosity (Recommendations
1984). Pure Appl Chem 57:603–619. https://doi.org/10.1351/
pac198557040603
Riedel T, Claeys M, Schulz H et al (1999) Comparative study of
Fischer–Tropsch synthesis with H /CO and H /CO syngas using
2
2
2
Fe- and Co-based catalysts. Appl Catal A 186:201–213. https://
doi.org/10.1016/S0926-860X(99)00173-8
1
0. Jun KW, Roh HS, Kim KS et al (2004) Catalytic investigation
for Fischer–Tropsch synthesis from bio-mass derived syngas.
Appl Catal A 259:221–226. https://doi.org/10.1016/j.apcat
a.2003.09.034
27. Bian G, Oonuki A, Koizumi N et al (2002) Studies with a pre-
cipitated iron Fischer-Tropsch catalyst reduced by H or CO.
2
J Mol Catal A 186:203–213. https://doi.org/10.1016/S1381
-1169(02)00186-3
1
1
1. Luo M, O’Brien R, Davis BH (2004) Effect of palladium on iron
Fischer–Tropsch synthesis catalysts. Catal Lett 98:17–22. https://
doi.org/10.1007/s10562-004-6442-x
28. Li T, Yang Y, Zhang C et al (2007) Phase transformation and
textural properties of an unpromoted iron Fischer–Tropsch cata-
lyst. Colloids Surf A 302:498–505. https://doi.org/10.1016/j.colsu
rfa.2007.03.022
2. Minnermann M, Pokhrel S, Thiel K et al (2011) Role of palla-
dium in iron based Fischer-Tropsch catalysts prepared by flame
spray pyrolysis. J Phys Chem C 115:1302–1310. https://doi.
org/10.1021/jp106860d
29. Ning W, Yang S, Chen H et al (2013) Influences of K and Cu on
coprecipitated FeZn catalysts for Fischer–Tropsch reaction. Catal
Commun 39:74–77. https://doi.org/10.1016/j.catcom.2013.05.013
30. Albrecht M, Rodemerck U, Schneider M et al (2017) Unex-
1
1
1
3. Ning W, Yang X, Yamada M (2012) Influence of palladium
on the hydrocarbon distribution of Fischer-Tropsch reaction
over precipitated iron catalyst. Curr Catal 1:88–92. https://doi.
org/10.2174/2211544711201020088
pectedly efficient CO hydrogenation to higher hydrocarbons
2
over non-doped Fe O . Appl Catal B 204:119–126. https://doi.
2
3
org/10.1016/j.apcatb.2016.11.017
4. Xu L, Wang Q, Liang D et al (1998) The promotions of MnO and
31. Bukur DB, Mukesh D, Patel SA (1990) Promoter effects on pre-
cipitated iron catalysts for Fischer-Tropsch synthesis. Ind Eng
Chem Res 29:194–204. https://doi.org/10.1021/ie00098a008
32. Zhang CH, Wan HJ, Yang Y et al (2006) Study on the iron–silica
K O to Fe/silicalite-2 catalyst for the production of light alkenes
2
from CO hydrogenation. Appl Catal A 173:19–25. https://doi.
2
org/10.1016/S0926-860X(98)00141-0
5. Yan SR, Jun KW, Hong JS et al (2000) Promotion effect of Fe–Cu
interaction of a co-precipitated Fe/SiO Fischer–Tropsch synthe-
2
catalyst for the hydrogenation of CO and application to slurry
sis catalyst. Catal Commun 7:733–738. https://doi.org/10.1016/j.
catcom.2006.03.018
2
reactor. Appl Catal A 194–195:63–70. https://doi.org/10.1016/
S0926-860X(99)00354-3
33. Nag NK (2001) A study on the formation of palladium hydride in
a carbon-supported palladium catalyst. J Phys Chem B 105:5945–
5949. https://doi.org/10.1021/jp004535q
1
1
1
1
6. Chen HX, Ning WS, Chen CH et al (2015) Influence of Fe O
2
3
2
crystal phase on the performance of Fe-based catalysts for CO
hydrogenation. J Fuel Chem Technol 43:1387–1392
34. Iwasa N, Arai S, Arai M (2008) Selective oxidation of CO with
7. Ning W, Koizumi N, Yamada M (2009) Researching Fe catalyst
modified Pd/ZnO catalysts in the presence of H : effects of addi-
2
suitable for CO -containing syngas for Fischer-Tropsch synthesis.
tives and preparation variables. Appl Catal B 79:132–141. https
://doi.org/10.1016/j.apcatb.2007.10.001
2
Energy Fuels 23:4696–4700. https://doi.org/10.1021/ef900428t
8. Iglesia E (1997) Design, synthesis, and use of cobalt-based Fis-
cher-Tropsch synthesis catalysts. Appl Catal A 161:59–78. https
35. Gaspar AB, Santos GR, Costa RS et al (2008) Hydrogenation of
synthetic PYGAS—Effects of zirconia on Pd/Al O . Catal Today
2
3
:
//doi.org/10.1016/S0926-860X(97)00186-5
133–135:400–405. https://doi.org/10.1016/j.cattod.2007.12.058
36. Choi PH, Jun KW, Lee SJ et al (1996) Hydrogenation of car-
bon dioxide over alumina supported Fe-K catalysts. Catal Lett
40:115–118. https://doi.org/10.1007/BF00807467
9. Overett MJ, Hill RO, Moss JR (2000) Organometallic chemistry
and surface science: mechanistic models for the Fischer–Tropsch
synthesis. Coordin Chem Rev 206–207:581–605. https://doi.
org/10.1016/S0010-8545(00)00249-6
1
3

W. Ning et al.
Affiliations
Wensheng Ning1 · Bei Li · Biao Wang · Xiazhen Yang · Yangfu Jin
1
1
1
2
1
2
College of Chemical Engineering, Zhejiang University
of Technology, Hangzhou 310032, China
College of Materials Science and Technology, Zhejiang
University of Technology, Hangzhou 310032, China
1
3