Fischer-Tropsch synthesis on a ruthenium catalyst in two-phase systems: An excellent opportunity for the control of reaction rate and selectivity

Article abstract of DOI:10.1039/c4cy00803k

The activity and selectivity of Fischer-Tropsch synthesis over hydrophobic Ru/C catalysts were efficiently controlled in the reaction medium consisting of organic and aqueous phases. A higher reaction rate was observed in two-phase systems compared to Fischer-Tropsch synthesis in the organic phase; however, catalyst localization in the organic phase leads to higher and tuneable selectivity to long-chain hydrocarbons. the Partner Organisations 2014.

Full text of DOI:10.1039/c4cy00803k

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
COMMUNICATION
Fischer–Tropsch synthesis on a ruthenium catalyst
in two-phase systems: an excellent opportunity
for the control of reaction rate and selectivity†
Cite this: Catal. Sci. Technol., 2014,
4, 2896
Received 20th June 2014,
Accepted 25th June 2014
*
Vitaly V. Ordomsky, Andrei Y. Khodakov, Benoit Legras and Christine Lancelot
DOI: 10.1039/c4cy00803k
www.rsc.org/catalysis
The activity and selectivity of Fischer–Tropsch synthesis over
hydrophobic Ru/C catalysts were efficiently controlled in the
reaction medium consisting of organic and aqueous phases.
A higher reaction rate was observed in two-phase systems
compared to Fischer–Tropsch synthesis in the organic phase;
however, catalyst localization in the organic phase leads to higher
and tuneable selectivity to long-chain hydrocarbons.
It was found that at low reaction temperature (100–150 °C)
oxygenates such as aldehydes and alcohols were predominant
products, while at higher temperature (150–220 °C) light
hydrocarbons were the major products. This phenomenon is
probably due to the higher probability of chain termination
in the presence of large amounts of water.
In contrast to aqueous phase FT synthesis, higher selec-
tivity to long-chain hydrocarbons is usually observed in
organic phase FT synthesis on ruthenium catalysts. At
the same time, the overall FT reaction rate on Ru catalysts
in the organic phase is somewhat lower than in the aque-
ous phase.
It would be interesting therefore to combine the advan-
tages of FT synthesis in aqueous and organic phases. It can
be expected that a combination of aqueous phase and organic
phase FT synthesis would lead to the enhancement of the
catalytic performance: higher reaction rate and better selec-
tivity to the desired products. Recently, the authors applied a
Ru catalyst supported over a composite constituted by carbon
nanotubes and MgO−Al2O3 in two-phase FT synthesis.14
Because of its hydrophilic properties, the catalyst was princi-
pally located in the aqueous phase. Consequently, only slight
modifications of the catalytic performance and, in particular,
hydrocarbon selectivity were observed. Quek et al.15 studied
the effect of organic capping agents (polar organic solvents
and polymers) on the performance of Ru nanoparticles in
aqueous phase FT. Strong interaction of the functional groups
with Ru resulted in a lower catalytic activity and significant
catalyst deactivation.
The present paper shows that the two-phase reaction
medium with a hydrophobic ruthenium catalyst provides an
excellent opportunity for efficient control of both reaction
rate and hydrocarbon selectivity in FT synthesis. Higher
reaction rates were observed in the two-phase systems and
in aqueous phase FT synthesis, coinciding with high and
controllable hydrocarbon selectivity. The catalytic results
obtained in the two-phase medium are compared to the
catalytic results of FT synthesis obtained either in the aqueous
phase or in the organic phase.
Low temperature Fischer–Tropsch (FT) synthesis converts
syngas (mixture of carbon monoxide and hydrogen) into
valuable aliphatic linear long-chain hydrocarbons. Methane,
light paraffins and carbon dioxide are undesirable products.
Their production should be reduced for better reaction
efficiency. The syngas for FT synthesis can be produced
from fossils and renewable resources (natural gas, coal,
carbon residues, and biomass), making this reaction suitable
for manufacturing alternative sustainable liquid fuels.
Iron-, cobalt- or ruthenium-based catalysts are common
catalysts for FT synthesis.1–3 Although more expensive than
cobalt and iron, Ru possesses a number of advantages for FT
synthesis compared to Co and Fe. These advantages are
higher catalytic activity, higher selectivity to long-chain
hydrocarbons, higher stability compared to any other FT
metal and the capacity to operate in the presence of large
amounts of water.4–8 Addition of water during FT synthesis
over Ru-based catalysts leads to a significant increase in the
reaction rate9,10 with major modifications in hydrocarbon
selectivity. Recently Xiao et al.11 showed higher FT activity of
Ru nanoparticles in the aqueous phase with the reaction rate
more significant than for conventional supported catalysts.
Quek12 and Pendyala13 studied the effect of reaction tempera-
ture on FT synthesis in the aqueous phase over Ru catalysts.
Unité de catalyse et de chimie du solide (UMR 8181 CNRS), Université Lille 1,
Sciences et Technologies de Lille, Bat. C3, Cité Scientifique, 59655 Villeneuve d'Ascq,
France. E-mail: Vitaly.Ordomsky@univ-lille1.fr; Fax: +33 0320336236;
Tel: +33 0320436953
† Electronic supplementary information (ESI) available: Preparation of the
catalyst, catalytic tests, TEM, SSITKA, selectivity to methane and CO2, and
hydrocarbon distribution. See DOI: 10.1039/c4cy00803k
2896 | Catal. Sci. Technol., 2014, 4, 2896–2899
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Communication
The Ru/C catalyst with a Ru loading of 5 wt.% was
prepared by aqueous impregnation of the carbon support.
Further experimental details are given in the ESI.† Analysis by
TEM showed that the size of Ru nanoparticles varied in the
range of 1 to 10 nm (Fig. 1S, ESI†). The average ruthenium
dispersion estimated from SSITKA experiments was about
28% (Fig. 2S, ESI†). The FT synthesis over the Ru/C catalyst in
the aqueous phase, organic (dodecane) phase and two-phase
systems was performed in a batch autoclave reactor.
The carbon monoxide hydrogenation data are shown in
Fig. 1 and Table 1.
can be attributed to the water-gas shift (WGS) reaction,
CO + H2O = CO2 + H2, which may proceed rapidly in the
aqueous phase. Analysis of the gas phase product distribu-
tion suggests that WGS took place mainly at low carbon
monoxide conversions (Fig. 3S, ESI†). At higher conversions,
the CO2 selectivity decreased and CH4 selectivity significantly
increased (Fig. 3S, ESI†). The high amount of hydrogen
formed during WGS seems to favour methanation. Finally, at
the CO conversion of 100%, the selectivity to CH4 and CO2
was 8 and 17%, respectively.
FT synthesis in the organic phase
FT synthesis in the aqueous phase
Table 1 shows a much lower FT catalytic activity on the
Ru/C catalyst in the organic phase (dodecane). The initial
carbon monoxide conversion rates were several times lower
compared to those in the aqueous phase FT synthesis
(4–10 molCO molRu−1 h−1, Table 1). However, the catalyst exhibited
a much more significant selectivity to long-chain hydrocarbons
The Ru/C catalyst exhibited a high initial FT reaction rate in
the aqueous phase (25 molCO molRu−1 h−1, Table 1). The aque-
ous phase FT synthesis resulted in the production of mostly
light hydrocarbons. The selectivity to C8+ products was 7%,
while C6–C7 hydrocarbons were the major products of FT syn-
thesis (Table 1). The reaction products also contained trace
amounts of higher alcohols (C1–C6). The obtained catalytic
results were consistent with the previous report by Xiao et al.11
In addition, aqueous phase FT synthesis with the Ru cata-
lyst produced significant amounts of CO2. CO2 formation
(SC = 30–60%) relative to FT synthesis in the aqueous phase.
8+
Interestingly, the catalyst exposed to syngas at room
temperature before starting the reaction (run 2, Table 1)
showed a much lower catalytic activity compared to the
catalyst which was exposed to syngas in dodecane at 180 °C
(run 3). Previous reports16,17 indicate a very low activation
energy for carbon monoxide dissociation on ruthenium nano-
particles compared to that for carbon monoxide desorption or
hydrogenation. This suggests that carbon monoxide adsorp-
tion and dissociation could be fast and irreversible even at
low temperatures. It can be suggested that strong carbon
monoxide adsorption followed by its rapid dissociation could
block13 the active sites for hydrogen adsorption. This could
lead to low activity in FT synthesis. Lower FT reaction rates
after the catalyst's exposure to syngas at room temperature
can therefore be explained by “poisoning” of the Ru catalyst
by CO under these conditions. The reaction under these
conditions also yielded significant amounts of CO2 (selectivity
of 36%). Note that in the aqueous phase, addition of syngas
at room temperature before starting the reaction resulted in a
high reaction rate similar to that obtained when syngas was
Fig. 1 Carbon monoxide conversion as a function of time in different
systems (run numbers according to Table 1).
Table 1 CO hydrogenation over Ru/C in the aqueous phase, organic phase and two-phase systems (T = 220 °C, 0.3 g Ru/C, p(CO) = 12.5 bar,
P(H2) = 25 bar, stirring rate = 700 rpm)
Reaction medium composition, g
Selectivity, mol. C%
Run
Initial reaction rate, CO
molCO molRu h−1
conversion, % H2O Surf. CO2 CH4 C2–C4
α
C5–C7 C8+
−1
number C12H26
FT synthesis in the aqueous phase
1
—
40
—
25
83
17.7 8.2
21.2
5.7
41
7
—
FT synthesis in the organic phase
2
3
30
—
—
—
—
3.9
10
23
86
36
8.1
4.0
7.0
33.7
57.0
—
0.87
30a
1.8
11.2 5.8
FT synthesis in two-phase systems
4
10
10
10
10
30
40
40
40
40
10
—
—
0.5
0.2
—
31
34
54
41
21
93
81
87
91
91
7.8
3.9 6.1
23.8 49
0.88
—
0.84
0.89
0.90
5b
6
14.9 15.1 16.0
11.1 3.6
9.2
5.5
19.6 29.8
24.4 51.7
22.5 50.8
6.4
5.6
6.0
7
8
3.7
9.8
8.2
58
a
b
H2 and CO were added in a preheated reactor (180 °C) with subsequent heating up to 220 °C. Stirring rate: 200 rpm.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 2896–2899 | 2897

View Article Online
Communication
Catalysis Science & Technology
added to the catalyst at the reaction temperature. Continuous
removal of CO from the surface by the WGS reaction and com-
petition for surface sites between CO and water seem to pro-
tect the catalyst surface area from blocking by strongly
adsorbed carbon species which may form from CO dissocia-
tion on ruthenium at low temperatures. A similar mechanism
of initiation of the FT reaction over Ru catalysts by CO2 forma-
tion was previously proposed by Jacobs et al.18
Fig. 1 also displays carbon monoxide conversion versus
reaction time when syngas was added to the ruthenium
catalyst in dodecane at 180 °C (run 3). After 23 h of reaction,
carbon monoxide conversion reached 86% with about 57%
selectivity to C8+ (Fig. 1, Table 1). Fig. 2 shows the distribu-
tion of the hydrocarbons produced during FT synthesis in
the organic phase. FT synthesis in dodecane leads to a wide
range of hydrocarbons (up to C60–C65). The distribution
curve follows Anderson–Schulz–Flory (ASF) statistics with an
apparent chain growth probability (α) of 0.87 (Table 1).
Thus, examination of the catalytic results obtained by FT
synthesis in aqueous and organic phases with the Ru/C cata-
lyst suggests that the aqueous phase leads to higher reaction
rates, although with lower selectivity to higher hydrocarbons.
In the organic phase the FT reaction rate was lower but
significantly higher hydrocarbon selectivity to long-chain
hydrocarbons was observed.
Fig. 3 Distribution of dodecane with the catalyst relative to the
aqueous phase before reaction (a), during stirring at 700 rpm without a
surfactant (b) and after addition of a surfactant (0.5 g of 1-hexanol) (c).
and the presence of larger droplets are indicative of insuffi-
cient mixing between organic and aqueous phases. In order
to intensify the mixing and to form emulsions, 1-hexanol was
added as a surfactant. Fig. 3c shows that in the presence of
1-hexanol, the organic phase is uniformly distributed in the
aqueous phase with formation of small droplets. In compari-
son with other surfactants, 1-hexanol was stable under the
reaction conditions.
The catalytic results obtained for two-phase systems are
shown in Table 1 and Fig. 1. Carbon monoxide conversion
and hydrocarbon selectivity were strongly affected by the
presence of both organic and aqueous phases. The FT reac-
tion rate in the two-phase system (run 4) with the addition of
10 g of dodecane was higher in comparison with that of FT
synthesis conducted either in the aqueous phase or in the
organic phase. Addition of 0.2 and 0.5 ml of 1-hexanol as a
surfactant (runs 6 and 7) resulted in an increase in the activity
FT synthesis in two-phase systems
Table 1 displays FT results obtained in an autoclave reactor
which simultaneously contained the aqueous and organic
phases and the Ru/C catalyst. The important issue is catalyst
repartition between the two phases. Fig. 3 shows that hydro-
phobic Ru/C in the two-phase system was totally localized in
the organic phase before and after the reaction. The top layer
contained mostly dodecane, while the lower layer was mostly
constituted by the aqueous phase with larger droplets of
dodecane (Fig. 3b). The observed distinct phase separation
−1
−1
from 31 molCO molRu h−1 to 41 and 54 molCO molRu h−1
.
The distribution of reaction products was also affected by the
composition of the reaction medium. Interestingly, the
amounts of produced methane and CO2 were almost two
times lower in the two-phase systems compared to those
obtained from FT synthesis in the aqueous phase. At the same
time the selectivity to long-chain hydrocarbons is significantly
higher in comparison with that obtained from the aqueous
phase experiments. Note that the selectivity to methane and
long-chain hydrocarbons was close to the selectivities in the
two-phase systems without addition of a surfactant. This
suggests that the catalyst is still in the organic phase, whereas
a close contact between the catalyst and the aqueous phase
leads to a higher reaction rate. A lower autoclave stirring rate
(run 5) resulted in a significant increase in the selectivity
to methane and CO2 due to localization of the catalyst mainly
in the interface in closer interaction with the aqueous phase.
At the same time a higher amount of dodecane (run 8) resulted
in a lower reaction rate with a higher selectivity to methane
due to the more prolonged catalyst contact with the organic
phase.
Similar to that in the organic phase, the distribution of
long-chain hydrocarbons in the two-phase medium also
followed the Anderson–Schulz–Flory (ASF) curve up to C35–C40
Fig. 2 ASF hydrocarbon distributions obtained in the organic phase
and in two-phase systems.
2898 | Catal. Sci. Technol., 2014, 4, 2896–2899
This journal is © The Royal Society of Chemistry 2014

View Article Online
Catalysis Science & Technology
Communication
hydrocarbons (Fig. 2 and 4S, ESI†). The high fraction of the
organic phase in the two-phase system (run 8) leads to higher
chain growth probability. At the same time, the selectivity and
chain growth probability for heavy C35+ hydrocarbons were
somewhat lower in the two-phase system compared to those
obtained from organic phase FT synthesis (Fig. 2).
3 G. P. van der Laan and A. A. C. M. Beenackers, Catal. Rev.:
Sci. Eng., 1999, 41, 255.
4 E. Iglesia, S. L. Soled, R. A. Fiato and G. H. Via, J. Catal.,
1993, 143, 345.
5 L. Fan, K. Yokota and K. Fujimoto, AIChE J., 1992, 38, 1639.
6 M. A. Vannice and R. L. Garten, J. Catal., 1980, 63, 255.
7 J. Kang, K. Cheng, L. Zhang, Q. Zhang, J. Ding, W. Hua,
Y. Lou, Q. Zhai and Y. Wang, Angew. Chem., Int. Ed., 2011,
50, 5200–5203.
Conclusion
8 L. Liu, G. Sun, C. Wang, J. Yang, C. Xiao, H. Wang, D. Ma
and Y. Kou, Catal. Today, 2012, 183, 136.
9 M. Claeys and E. van Steen, Catal. Today, 2002, 71, 419.
10 D. D. Hibbitts, B. T. Loveless, M. Neurock and E. Iglesia,
Angew. Chem., 2013, 125, 12499.
11 C. Xiao, Z. Cai, T. Wang, Y. Kou and N. Yan, Angew. Chem.,
Int. Ed., 2008, 47, 746.
12 X. Quek, Y. Guan, R. A. van Santen and E. J. M. Hensen,
ChemCatChem, 2011, 3, 1735.
13 V. R. R. Pendyala, W. D. Shafer and B. H. Davis, Catal. Lett.,
2013, 143, 895.
14 D. Shi, J. A. Faria, A. A. Rownaghi, R. L. Huhnke and
D. E. Resasco, Energy Fuels, 2013, 27, 6118.
In summary, conducting FT synthesis with a supported ruthe-
nium catalyst in a two-phase medium consisting of aqueous
and organic phases can overcome major drawbacks of FT
synthesis in the aqueous phase and in the organic phase. A
much higher FT reaction rate is observed in the two-phase
medium compared to FT synthesis in the organic phase
which is probably due to the intensive cleavage of the surface
by water molecules. At the same time, the hydrophobic catalyst
localised in the organic phase in the two-phase system exhibits
high selectivity to long-chain hydrocarbons. The hydrocarbon
selectivity in two-phase systems is controlled by the composi-
tion of the reaction medium and operating conditions.
15 X. Quek, R. Pestman, R. A. van Santen and E. J. M. Hensen,
ChemCatChem, 2013, 5, 3148.
Notes and references
1 A. Y. Khodakov, W. Chu and P. Fongarland, Chem. Rev.,
2007, 107, 1692.
2 F. J. Pérez-Alonso, T. Herranz, S. Rojas, M. Ojeda,
M. López Granados, J. L. J. Fierro and J. R. Gancedo,
Green Chem., 2007, 9, 663.
16 S. B. Vendelbo, M. Johansson, D. J. Mowbray, M. P. Andersson,
F. Abild-Pedersen, J. H. Nielsen, J. K. Nørskov and I. Chorkendorff,
Top. Catal., 2010, 53, 357.
17 J. Chen and Z.-P. Liu, J. Am. Chem. Soc., 2008, 130, 7929.
18 H. H. Nijs and P. A. Jacobs, J. Catal., 1980, 66, 401.
This journal is © The Royal Society of Chemistry 2014
Catal. Sci. Technol., 2014, 4, 2896–2899 | 2899