Fuel Processing Technology 91 (2010) 18391844

Contents lists available at ScienceDirect

Fuel Processing Technology

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / f u p r o c

Investigation of FischerTropsch synthesis performance and its intrinsic reaction

behavior in a bench scale slurry bubble column reactor
Heon Jung, Jung-Il Yang , Jung Hoon Yang, Ho-Tae Lee, Dong Hyun Chun, Hak-Joo Kim
Clean Fossil Energy Research Center, Korea Institute of Energy Research, Daejeon 305-343, Korea

a r t i c l e i n f o a b s t r a c t

Article history: A bench scale slurry bubble column reactor (SBCR) with active-Fe based catalyst was developed for the
Received 9 March 2010 FischerTropsch synthesis (FTS) reaction. Considering the highly exothermic reaction heat generated in the
Received in revised form 23 July 2010 bench scale SBCR, an effective cooling system was devised consisting of a U-type dip tube submerged in the
Accepted 6 August 2010
reactor. Also, the physical and chemical properties of the catalyst were controlled so as to achieve high
activity for the CO conversion and liquid oil (C5+) production. Firstly, the FTS performance of the FeCuK/SiO2
Keywords:
FischerTropsch synthesis
catalyst in the SBCR under reaction conditions of 265 C, 2.5 MPa, and H2/CO = 1 was investigated. The CO
Bench scale conversion and liquid oil (C5+) productivity in the reaction were 88.6% and 0.226 g/gcat-h, respectively,
Slurry bubble column reactor corresponding to a liquid oil (C5+) production rate of 0.03 bbl/day. To investigate the FTS reaction behavior
Space velocity in the bench scale SBCR, the effects of the space velocity and supercial velocity of the synthesis gas and
Supercial velocity reaction temperature were also studied. The liquid oil production rate increased up to 0.057 bbl/day with
Reaction temperature increasing space velocity from 2.61 to 3.92 SL/h-gFe and it was conrmed that the SBCR bench system
developed in this research precisely simulated the FTS reaction behavior reported in the small scale slurry
reactor.
2010 Elsevier B.V. All rights reserved.

1. Introduction plant with a capacity of 12,500 bbl/day in Bintulu, Malaysia, which

Synthetic fuels produced by the FischerTropsch synthesis (FTS)
process are a realistic alternative to replace conventional oils. Thus,
the need for the FTS technology is globally on the rise. However, only a
limited number of companies are available to supply the commercial
FTS technology. This technology barrier has spurred many institutions
on to development efforts of FTS technology. Among them, Chinese
company has been very active in proving their FTS technology to a
demonstration scale of 3500 bbl/day [1].
In the FTS reaction, the slurry bubble column reactor (SBCR) and
multi tubular xed-bed reactor (FBR) are currently used in industry.
The advantages and disadvantages of SBCR and FBR technologies in
the FTS reaction are summarized in Table 1 [2]. Sasol began the
development of an iron (Fe)-based SBCR and, after a 10-year
research program, a semi-commercial scale SBCR with a capacity
of 2500 bbl/day was put into operation at Sasol in 1993 [3].
Especially, in the FTS reaction, the commercialization of slurry
phase reactor technology was made possible by the development of
Sasol's intrinsic solution for catalyst/wax separation. Shell started to
develop the Shell Middle Distillate Synthesis (SMDS) process in
1973 using an FBR with cobalt (Co) catalysts and began building a
Corresponding author. Tel.: + 82 42 860 3795; fax: + 82 42 860 3134.
E-mail address: yangji@kier.re.kr (J.-I. Yang).
was nally commissioned in 1993 [4].
The metals, Fe, Ni, Co, and Ru, have FischerTropsch synthesis
activity [5], but only Fe and Co catalysts are in industrial use, as
described above. Fe and Co catalysts share similarities in the FTS
reaction, being very active and producing a broad range of straight
chain hydrocarbons [6]. Furthermore, the product distribution
afforded by these catalysts follows the ASF distribution. However, Fe
and Co catalysts are also quite different in many respects. Fe catalysts
are active in the form of Fe carbides, whereas the active form of Co
catalysts is Co metal. Also, from the point of view of the technological
preferences, a precipitated Fe catalyst is preferable to a slurry bed
reactor with a synthesis gas (H2/CO) ratio of 2/3, mainly due to its low
cost, high tolerance to contaminants, and water-gas shift reaction
activity, but a supported Co catalyst is more suitable for a xed-bed
reactor with a synthesis gas ratio of 2/1, mostly because of its high
cost, high activity, and long catalyst life (i.e., the lifetime of Co is 10
times that of Fe) [7]. Furthermore, the amount of S compounds in the
synthesis gas feed for both Fe and Co catalysts has to be very low, but
Co catalysts require an additional step for the purpose of lowering the
S contaminant level from the low 100's ppb (e.g., ~ 200 ppb) for Fe
catalysts to the low 10's ppb (e.g., ~20 ppb) [7].
CTL/GTL processes utilize 3 distinct steps to convert coal or natural
gas into synthetic transport fuels, viz. synthesis gas generation, the
FischerTropsch synthesis reaction, and product upgrading [8]. As a
key step in the CTL process, the FischerTropsch synthesis reaction
that converts the synthesis gas generated by coal gasication into

0378-3820/$ see front matter 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.fuproc.2010.08.008
1840 H. Jung et al. / Fuel Processing Technology 91 (2010) 18391844

Table 1 basicities of the catalysts were analyzed by means of temperature-

Advantages (+) and disadvantages () of SBCR and FBR technologies in FTS reaction programmed desorption using CO2 as an adsorbent (CO2-TPD). The
(partly from [2]).
catalysts were heated up to 400 C for several tens of minutes until the
SBCR FBR TCD baseline leveled off. After cooling the catalysts to 50 C, CO2 was
Catalyst effective factor + (close to unity) (below unity) introduced into the sample cell for 30 min, followed by purging with
Catalyst pore diffusion + He for 1 h to remove the weakly adsorbed species. The CO2-TPD was
Catalyst content in reactor + carried out at temperature of up to 400 C in owing He at a heating
Catalyst concentration gradient +
rate of 2 C/min.
Gas-liquid mass transfer +
Isothermal behavior +
Catalyst exchange +
2.2. Reaction system
Catalyst attrition +
Mixing behavior +
Distribution of reactants + The FTS reaction was carried out in a stainless steel bubble column
Liquid-solid separation + reactor (0.05 m in diameter and 2.5 m in height with an effective
Gas-liquid separation a + volume of 3.6 l) equipped with a sintered metal lter to allow the
Scale-up +
Reactor costs +
removal of waxy products from the reactor. The ltered wax was
a
collected in a wax reservoir. A schematic diagram of the experimental
Foam formation in a slurry bubble column reactor.
apparatus is shown in Fig. 1. The level of the slurry in the reactor was
continuously monitored by measuring the temperature across the
height of the reactor. As the level of slurry within the reactor went up
liquid oil continues to be an important research eld. Therefore, in this
due to the formation of waxy products in the slurry during the FTS,
study, a highly effective FTS reaction system was developed,
these waxy products were regularly sampled from the lter depend-
consisting of an SBCR with Fe catalyst, for CTL applications. Especially,
ing on the level of slurry and the maximum level in the reactor
a bench scale SBCR with active Fe-based catalyst was invented and the
remained near the top of the reactor, as shown in Fig. 1.
liquid oil (C5+) productivity in the FTS reaction was evaluated.
The reactor temperature was controlled using 11 external jacket
Furthermore, the FTS reaction behavior in the bench scale SBCR was
furnaces whose temperatures were automatically controlled by a PID
investigated as a function of the space velocity and supercial velocity
controller (Hanyoung Science, Korea). To prevent the sudden increase
of the synthesis gas and reaction temperature.
of the reaction temperature during the FTS reaction, a cooler system
consisting of a U-type dip tube submerged in the slurry was designed.
2. Experimental
Furthermore, considering the severe axial temperature gradient in the
reactor, two U-type dip tubes were installed across the height of the
2.1. Catalysts
reactor. Therefore, the reaction temperature was automatically
controlled to the initial set point by allowing the coolant (cold
A catalyst with a nominal composition of 100Fe/5Cu/5 K/18SiO2
water at 25 C) to ow through the dip tube whenever there was a
(in parts per weight) was prepared by the conventional co-
sudden increase of the reaction temperature caused by the severe
precipitation technique (FeCuK/SiO2). In brief, the requisite amount
exothermic reaction heat generated in the FTS. Prior to the FTS
of a SiO2 suspension was added to a solution containing both Fe(NO3)3
reaction, the bubble column reactor was charged with a slurry
and Cu(NO3)2 at the desired ratio. Then, the solution was precipitated
consisting of the catalyst and liquid parafn using a slurry pump
at 80 1 C using a sodium carbonate solution until the pH reached
(BRAN + LUEBBE, Germany). The feed gas was preheated and then
8.0 0.1. The precipitate slurry was ltrated, washed with distilled
introduced into the reactor from the sparger placed at the bottom
water, and subsequently re-slurried in distilled water. After repeating
with a corresponding free plate area of 0.14%. The ow rate of the inlet
the washing process several times, the ltrated cake was withdrawn
gas was controlled and monitored using a pre-calibrated mass ow
and then dried at 95 C for at least 8 h to completely remove the
controller (MKP, Seoul, Korea). The vent gas was passed through a hot
residual moisture. The dried precipitate was crushed and sieved to
trap (150 C) to separate the high-boiling products, and then a cold
less than 75 m (200 mesh), and the requisite amount of K2CO3
trap to collect the liquid products via a condenser maintained at 0 C
solution was added via the incipient wetness pore-lling technique.
to condense the remaining liquid products. Finally, the ow rate of the
The nal precipitate powder was calcined in air at 400 C for 8 h. A
tail gas was measured using a dry gas ow meter (Shinogawa, Tokyo,
commercial catalyst was also provided for the sake of comparing the
Japan). All instruments were connected to a PC via LabVIEW program
FTS performance of the two catalysts.
(LabVIEW 7.1, NI Co.) to record the experimental data automatically
Prior to the FTS reaction, 720 g of fresh catalyst was loaded in the
and to precisely measure the FTS reaction behavior at each part.
reactor and suspended in 2880 g of liquid parafn (Mineral oil,
Aldrich Co.). Thus, the concentration of the catalyst in the slurry was
xed at 20 wt.% and the catalyst was activated in-situ in the slurry 2.3. Reactant and product analysis
phase. The catalysts were activated using an in-situ reduction
procedure with a ow rate of synthesis gas (H2/CO = 1) of 3.0 l/min The reactant and product gases in the FTS reaction were analyzed
at 0.1 MPa and 265 C for 25 h. online by a gas chromatograph (DS6200; DONAM INSTRUMENTS,
The surface area, pore size, and pore volume of the FeCuK/SiO2 and INC., Korea) equipped with a thermal conductivity detector (TCD) and
commercial catalysts were measured by BET using a Micromeritics a downstream ame ionization detector (FID). H2, CO, CH4, and CO2
2010 system. The results are summarized at Table 2. The surface were analyzed with a stainless steel packed column (Carboxen-1000,
2 m 0.3 mm) and a TCD. Also, hydrocarbons (C2 C4) were analyzed
Table 2 with a capillary column (GS-GasPro, 30 m 0.32 mm) and an FID. The
The physical properties of the FeCuK/SiO2 and commercial catalysts. liquid product distribution was analyzed by the simulated distillation
Catalyst BET surface Ave. Pore Ave. Pore (SIMDIS) technique using the ASTM D2887 method and HPLC grade
carbon disulde (CS2, 99.9 + %, Aldrich Co.) as the solvent. A
area/m2 g 1 diameter / volume / cm3 g 1
calibration mix (C5C44, Analytical Controls) and reference mix
FeCuK/SiO2 181 82 0.369 (140400 C, Analytical Controls) were used as calibration standards
Commercial 195 55 0.268
for the SIMDIS.
 H. Jung et al. / Fuel Processing Technology 91 (2010) 18391844 1841

PG

T/C

Needle Valve
Coolant
in T11 out

T10 Condenser
Filter
T9
Chiller
T8
Hot
T7 Chamber P

T6 Vent

T5 Back Pressure Dry Gas

Wax Regulator Flow Meter
T4 Reservoir
Cold
T3 Chamber
T2 On/Off
Valve
T1
Pressure
Relief Vent
Heating
Jacket Preheater MFC H2

MFC CO

MFC N2

Fig. 1. Schematic diagram of slurry bubble column reactor for FTS reaction.

3. Results and Discussion
3.1. Slurry bubble column reactor activity
The FischerTropsch synthesis (FTS) reaction was carried out in
the slurry bubble column reactor for the two Fe based catalysts, viz.
FeCuK/SiO2 catalyst and commercial catalyst. Compared to the
commercial catalyst, the FeCuK/SiO2 catalyst was prepared with the
goal of improving its surface basicity. C.-H. Zhang et al. observed that
basic sites existing on the catalyst surface were responsible for the
suppression of methane and other light-hydrocarbon products and
the enhancement of the heavy products [9]. The slurry bubble column
reactor (SBCR) was designed to take into consideration the gas
supercial velocity (cm/s) and gas space velocity (SL/h-gFe) in the
reactor for the selective production of liquid oil in the FTS. In the FTS
reaction, the CO conversion and liquid oil selectivity are closely
related to the gas space velocity and gas supercial velocity,
respectively [10,11]. Therefore, not only the SBCR size (Diameter,
Height), but also the ow rate of synthesis gas (H2/CO) is precisely
the CO conversion obtained using the FeCuK/SiO2 catalyst was
superior to the previously reported activity. However, the result was
considered to be quite reasonable, because space velocity of the
synthesis gas in this reaction was much lower than that in the
previous study. The effect of the space velocity of the synthesis gas on
the CO conversion in the FTS reaction will be discussed in more detail
below. Furthermore, it was conrmed that the FTS reaction was
performed in the homogeneous ow regime because the supercial
velocity of the synthesis gas was controlled to 1.5 cm/s as described
Table 3. Many researchers have reported that the transition from
homogeneous ow regime to heterogeneous occurred at the
supercial velocity of synthesis gas from 3.0 to 8.0 cm/s [13,14].
Especially, Daly et al., have studied on hydrodynamics using the
similar liquid media and reactor geometry to our experimental
100
80
CO conversion [ % ]

controlled to provide a good gas space velocity and gas supercial

velocity for the selective production of liquid oil in the FTS reaction.
60
Fig. 2 shows the FTS performance of the FeCuK/SiO2 catalyst in the
SBCR under reaction conditions of 265 C, 2.5 MPa, and H2/CO = 1.
Furthermore, the gas supercial velocity and gas space velocity were 40
controlled to 1.5 cm/s and 2.61 SL/h-gFe, respectively. As shown in
Fig. 2, the CO conversion reached 88.6% after a run time of 20 h, which FeCuK/SiO2 catalyst
is close to the steady-state value. Also, the liquid oil (C5+) was 20 commercial catalyst
collected in the wax reservoir, hot trap, and cold trap during the
reaction and its productivity reached 0.226 g/gcat-h. J.K. Neathery et al.
also reported the performance of a precipitated Fe/K catalyst in the 0
0 5 10 15 20 25 30 35 40 45
enhanced SBCR, for which the CO conversion reached a maximum of
Time [ h ]
78% after a run time of 72 h [12]. Table 3 shows a comparison of the
operation conditions and FTS performance for this investigation and Fig. 2. CO conversion of the FeCuK/SiO2 and commercial catalysts in the SBCR; reaction
the previous study done by the Neathery's group. As shown in Table 3, conditions: Cat. 720 g, 265 C, 2.5 MPa, H2/CO = 1.
1842 H. Jung et al. / Fuel Processing Technology 91 (2010) 18391844

Table 3 Table 4
Comparison of the operations conditions and FTS performance in this investigation Comparison of FTS performance by the FeCuK/SiO2 and commercial catalysts (catalyst
(case 1) and the previous study done by Neathery's group (case 2). load (w); 720 g, temperature (T); 265 C, pressure (P); 2.5 MPa, H2/CO = 1).

FTS reaction Case 1 Case 2 Catalyst FeCuK/SiO2 Commercial

SBCR size Diameter [cm] 5.08 5.08 CO conversion [%] 88.6a 73.8b
Height [m] 2.5 2 Product selectivity [%] CO2 50.6 41.2
Operation conditions H2/CO ratio 1 0.7 CH4 2.20 3.50
Catalyst concentration [wt.%] 20 20 C2C4 10.4 11.0
Gas space velocity [SL/h-gFe] 2.61 5.3 C5+ 36.8 44.3
Gas supercial velocity [cm/s] 1.5 3 Total HC (C1+) yield [%] 43.8 43.4
Temperature [C] 265 270 Liquid oil (C5+) productivity [g/gcat-h] 0.226 0.196
Pressure [MPa] 2.5 1.21 a
An almost steady-state value during the run from 20 to 40 h.
FTS performance max. CO conversion [%] 88.6 78 b
An averaged value of the activities from 25 to 45 h.
CO2 selectivity [%] 50.6 45
CH4 selectivity [%] 2.20 4

the liquid oil (C5+) productivity obtained by the FeCuK/SiO2 catalyst

conditions, and they also reported that the transition to heteroge-
neous ow regime occurred above the supercial velocity of 2.0 cm/s
[15].
Meanwhile, it was generally assumed that the high CO conversion
of 88.6% in the FTS reaction obtained in this work resulted in the
irreversible deactivation of the catalyst due to the highly exothermic
reaction heat that was generated, but the activity obtained by the
SBCR operation was very stable for as long as 40 h. This was ascribed
to the fact that the cooling system of the U-type dip tube submerged
in the slurry reactor effectively controlled the highly exothermic
reaction heat in the FTS reaction. Finally, it was conrmed that the
SBCR bench system described in this work was quite suitable for the
production of liquid oil in the FTS reaction and the liquid oil (C5+)
production rate afforded by the bench system reached 0.03 bbl/day in
this operation. Furthermore, it was also conrmed that the two zone
cooling system consisting of the two U-type dip tubes was suitable to
effectively remove the highly exothermic reaction heat generated in the
slurry bubble column reactor scaled for the production of 0.03 bbl/day
of liquid oil. Fig. 3 shows the reaction temperature and pressure
excursions in the SBCR during the FTS reaction.
3.2. Catalyst activities
The FTS performance of a commercial catalyst was also investigated
in the SBCR under the same reaction conditions. CO conversion
obtained by the commercial catalyst in the SBCR under reaction
conditions of 265 C, 2.5 MPa, H2/CO = 1 is displayed in Fig. 2. Table 4
shows a comparison of the FTS reaction activities of the commercial
catalyst and FeCuK/SiO2 catalysts. As described in Table 4, compared to
the commercial catalyst, the FeCuK/SiO2 catalyst showed a high CO
conversion, low CH4 selectivity, and high CO2 selectivity. Furthermore,
300
T9
T8
Reaction pressure
290 T7
T6
Temperature [oC ]
was 0.226 g/gcat-h, which was higher than that of the commercial
catalyst, viz. 0.196 g/gcat-h. Therefore, it was found that the FeCuK/SiO2
catalyst was more suitable for the production of liquid oil in the FTS
reaction than the commercial catalyst. As shown in Table 2, the pore
diameter and pore volume of the FeCuK/SiO2 catalyst were much
higher than those of the commercial catalyst, thus allowing the
produced long-chained hydrocarbons, viz. liquid oil (C5+), to more
readily diffuse outside of the catalyst pellets [16]. In addition, the good
activity of the FeCuK/SiO2 catalyst was also attributed to its enhanced
surface basicity. According to A.N. Pour et al., Mg, La, and Ca promoters
in an Fe/Cu/SiO2 catalyst can improve the CO conversion, promote
the WGS reaction, suppress the formation of methane, and enhance the
selectivities to olen and higher molecular weight products due to
the increase in the surface basicity of the catalyst [17]. To investigate the
surface basicity of the catalysts, a CO2-TPD experiment was performed.
The CO2-TPD curves of the FeCuK/SiO2 and commercial catalysts are
shown in Fig. 4.
In the TPD proles, the CO2 uptake and peak position can be used
as indicators of the intensity and strength of the basic sites on the
surface of the catalysts, respectively [9]. As shown in Fig. 4, an intense,
broad peak in the temperature range of 80190 C is present in the
TPD prole of the FeCuK/SiO2 catalyst, corresponding to the
desorption of CO2 that interacts moderately with the surface basic
sites. However, in the TPD prole of the commercial catalyst, no
apparent peak appears in this temperature range. These results
strongly indicate that the surface basicity of the FeCuK/SiO2 catalyst
was higher than that of the commercial catalyst, as shown by the
intensity and strength of the basic sites on its surface, resulting in its
relatively high liquid oil productivity of 0.226 g/gcat-h.
Therefore, it was concluded that the physical (pore size, pore
volume) and chemical (surface basicity) properties of the FeCuK/SiO2
catalyst were superior to those of the commercial catalyst, thereby
3.0
2.5

T5
Pressure [ MPa ]

T4
Intensity [ a.u. ]

280 T3 2.0
T2
T1 FeCuK/SiO2 cat.
270 1.5

260 1.0 commercial cat.

Reaction temperature
Tave.; 265 oC

250 0.5
0 5 10 15 20 25 30 35 40
Time [ h ] 50 100 150 200 250 300 350 400
Temperature [ oC ]
Fig. 3. Reaction temperature and pressure excursions in the SBCR during the FTS
reaction; reaction conditions: Cat. 720 g, 265 C, 2.5 MPa, H2/CO = 1. Fig. 4. CO2-TPD proles of the FeCuK/SiO2 and commercial catalysts.
 H. Jung et al. / Fuel Processing Technology 91 (2010) 18391844 1843

providing a high CO conversion, low CH4 selectivity, and high liquid a

oil (C5+) productivity in the FTS reaction. 100

CO conversion / CO2 & HC yield [ % ]

3.3. Effect of synthesis gas space velocity 80
CO conversion
CO2 yield
The effect of the space velocity of the synthesis gas on the FTS
HC yield
performance of the FeCuK/SiO2 catalyst in the SBCR was investigated. 60
Table 5 shows the effect of the space velocity on the FTS performance at
265 C, 2.5 MPa, and H2/CO= 1. The CO conversion decreased from 88.6
to 81.7% as the space velocity was increased from 2.61 to 3.92 SL/h-gFe. 40
It is well known that the CO conversion in the FTS reaction, particularly
with iron based catalysts, is the summation of the conversion of CO to
20
hydrocarbons (via the FischerTropsch synthesis; FTS) and the
conversion of CO to CO2 (via the water-gas shift reaction; WGS) [17].
These two reactions are described below; 0
240 245 250 255 260 265 270
CO 2H2  CH2  H2 O; FischerTropsch synthesis 1 Temperature [oC ]

b 60 18
CO H2 OCO2 H2 ; Watergas shift reaction 2
As shown in Table 5, the decrease in the CO conversion with 50 15

CH4 & C2~C4 selectivity [ % ]

CO2 & C5+ selectivity [ % ]
increasing space velocity resulted entirely from the decrease in the CO2 selectivity
conversion of CO to CO2 via the WGS. The conversion of CO to C5+ selectivity
40 12
hydrocarbons via the FTS did not vary with the space velocity.
Furthermore, the liquid oil (C5+) productivity signicantly increased
from 0.226 to 0.430 g/gcat-h with increasing space velocity. Increasing 30 9
the space velocity gives rise not only to a decrease in the residence
time of the reactants, but also to an increase in their supercial 20 CH4 selectivity 6
velocity. As the CO conversion via the WGS is greatly inuenced by the C2~C4 selectivity

effect of the residence time, the conversion of CO to CO2 decreased

10 3
with increasing space velocity. However, as the FTS reaction is
typically an external and internal mass transfer limited reaction, the
CO conversion via the FTS is considerably affected by not only the 0 0
240 245 250 255 260 265 270
effect of the residence time, but also the effect of the supercial
Temperature [oC ]
velocity [10,11]. As a result, there is a compensation effect of the
residence time and supercial velocity on the CO conversion via the Fig. 5. (a) Effect of reaction temperature on CO conversion, CO2 yield, and HC yield in
FTS with the space velocity from 2.61 to 3.92 SL/h-gFe. Therefore, there the FTS reaction; reaction conditions: Cat. 720 g, 2.5 MPa, H2/CO = 1. (b) Effect of
was no variation in the conversion of CO to hydrocarbons (C1+) with reaction temperature on CO2, C5+, CH4, and C2C4 selectivities in the FTS reaction;
reaction conditions: Cat. 720 g, 2.5 MPa, H2/CO = 1.
the space velocity and the liquid oil (C5+) productivity rather increased
with increasing space velocity. Finally, the liquid oil (C5+) productivity
at a space velocity of 3.92 SL/h-gFe was 0.430 g/gcat-h and, furthermore,
the liquid oil (C5+) production rate reached 0.057 bbl/day. reaction temperature, while the CH4 and C2C4 selectivities increased
from 1.44 to 2.20% and from 8.93 to 10.4%, respectively, and the C5+
3.4. Effect of reaction temperature selectivity decreased from 42.2 to 36.8%. From these results, it can be
concluded that although a higher temperature favors the CO
The reaction temperature was increased from 245 to 265 C to conversion and CH4 and C2C4 selectivities in the FTS reaction, it is
investigate the effect of temperature on the FTS performance of the unfavorable for the liquid oil (C5+) selectivity. The results were also
FeCuK/SiO2 catalyst in the SBCR. As shown in Fig. 5 (a) and (b), the CO comparable to those obtained by Y. Liu et al., who reported that a
conversion gradually increased from 61.2 to 88.6% with increasing higher temperature was preferential for the chain termination
required to produce light hydrocarbons, while a lower temperature
was preferential for chain growth and the production of heavy
hydrocarbons [18]. Moreover, H.W. Pennline et al., in their studies of
Table 5
Effect of gas space velocity on the FTS performance of the FeCuK/SiO2 catalyst in the
Fe-Mn catalysts with various manganese contents in a slurry reactor,
SBCR (catalyst load (w); 720 g, temperature (T); 265 C, pressure (P); 2.5 MPa, H2/ showed that in the temperature range of 250290 C, the CO
CO = 1). conversion and CH4 and C2C4 selectivities increased with increasing
temperature [19].
Space velocity [SL/h-gFe] 2.61 3.26 3.92
Supercial velocity [cm/s] 1.49 1.86 2.23 The CO2 selectivity remained at about 50% over the whole
CO conversion [%] 88.6 85.5 81.7 temperature range and was only slightly inuenced by the temper-
Product selectivity [%] CO2 50.6 48.7 46.8 ature [18]. Thus, the conversion of CO to CO2 steadily increased with
CH4 2.20 1.93 1.80
increasing temperature, as shown in Fig. 5(a). In addition, the WGS
C2C4 10.4 9.29 8.80
C5+ 36.8 40.1 42.6 reaction equilibrium which was calculated in the FTS reaction with
CO2 yield [%] 44.8 41.9 38.2 the CO conversion (the summation of CO conversion via the FTS and
Total HC (C1+) yield [%] 43.8 43.9 43.5 CO conversion via the WGS), also showed that the CO2 selectivity was
Liquid oil (C5+) productivity 0.226 0.331 0.430 not greatly changed in the temperature range of 245265 C and it
[g/gcat-h]
was almost 50%.
1844 H. Jung et al. / Fuel Processing Technology 91 (2010) 18391844
Therefore, it was conrmed that the SBCR bench system developed
in this study precisely simulates the FTS reaction behavior reported in
the small scale slurry reactor [1820].
4. Conclusions
In the FischerTropsch synthesis reaction, a bench scale slurry
bubble column reactor with an active-Fe based catalyst shows a high
CO conversion of 88.6% and high liquid oil (C5+) productivity of
0.226 g/gcat-h at 265 C , 2.5 MPa, H2/CO = 1, and catalyst concentra-
tion of 20 wt.% in the slurry. The effective cooling system of the slurry
reactor and enhanced basicity of the catalyst were responsible for the
excellent activity of the FTS system, resulting in a liquid oil (C5+)
production rate of 0.03 bbl/day. The liquid oil productivity increased
from 0.226 to 0.430 g/gcat-h with increasing space velocity from 2.61
to 3.92 SL/h-gFe, because the conversion of CO to hydrocarbons via the
FTS reaction was characteristically inuenced by the effect of the gas
supercial velocity. The reaction temperature also had the anticipated
effect on the FTS performance, with the result that the CO conversion
and CH4 and C2C4 selectivities increased, but the C5+ selectivity
decreased with increasing reaction temperature. Therefore, it was
conrmed that the bench scale SBCR described in this research not
only showed superior liquid oil productivity, but also simulated the
FTS reaction behavior precisely.
Acknowledgements
Financial support by the Grant-in-aid from the Ministry of Science
and Technology and the Korea Institute of Energy Research (Basic
Project Program B0-2415) is highly appreciated.
References
[1] Y.-W. Li, CTL in Sustainable Energy System, http://www.nrcce.wvu.edu/cleanenergy/
docs/12-Li.pdf.
[2] R. Guettel, U. Kunz, T. Turek, Reactors for FischerTropsch synthesis, Chem. Eng.
Technol. 31 (5) (2008) 746754.
[3] R.L. Espinoza, A.P. Steynberg, B. Jager, A.C. Vosloo, Low temperature Fischer
Tropsch synthesis from a Sasol perspective, Appl. Catal., A Gen. 186 (1-2) (1999)
1326.
[4] M.M.G. Senden, A.D. Punt, A. HoekShell, Gas-to-liquids process: current status &
future prospects, Stud. Surf. Sci. Catal. 119 (1998) 961966.
[5] M.E. Dry, The FischerTropsch process: 19502000, Catal. Today 71 (2002)
227241.
[6] I. Wender, Y. Zhang, L. Hou, J. Tierney, Chain Growth in FischerTropsch: cobalt
versus iron, CFFS Annual Meeting, Roanoke, West Virginia, Aug. 1-4, 2004.
[7] R. Espinoza, Gas to Liquids GTL, www.transpacpart.com/assets/gtl.ppt.
[8] Introducing Sasol's unique coal-to-liquids technology, www.sasol.com/sasol_
internet/downloads/CTL_Brochure_1125921891488.pdf.
[9] C.-H. Zhang, Y. Yang, B.-T. Teng, T.-Z. Li, H.-Y. Zheng, H.-W. Xiang, Y.-W. Li, Study of
an iron-manganese FischerTropsch synthesis catalyst promoted with copper,
J. Catal. 237 (2) (2006) 405415.
[10] H.-J. Kim, J.-H. Ryu, H. Joo, J. Yoon, H. Jung, J.-I. Yang, Mass- and heat-transfer
enhanced catalyst system for FischerTropsch synthesis in xed-bed reactors, Res.
Chem. Intermed. 34 (89) (2008) 811816.
[11] J.H. Yang, H.-J. Kim, D.H. Chun, H.-T. Lee, J.-C. Hong, H. Jung, J.-I. Yang, Mass
transfer limitations on xed-bed reactor for FischerTropsch synthesis, Fuel
Process. Technol. 91 (3) (2010) 285289.
[12] J.K. Neathery, R.L. Spicer, D.E. Sparks, B.H. Davis, A comparison of FischerTropsch
synthesis in a slurry bubble column reactor and a continuous stirred tank reactor,
Stud. Surf. Sci. Catal. 139 (2001) 407414.
[13] R.G. Rice, M.A. Littleeld, Dispersion coefcients for ideal bubbly ow in truly
vertical bubble columns, Chem. Eng. Sci. 42 (8) (1987) 20452053.
[14] C.L. Hyndman, F. Larachi, C. Guy, Understanding gas-phase hydrodynamics in
bubble columns: a convective model based on kinetic theory, Chem. Eng. Sci. 52
(1) (1997) 6377.
[15] J.G. Daly, S.A. Patel, D.B. Bukur, Measurement of gas holdups and sauter mean
bubble diameters in bubble column reactors by dynamic gas disengagement
method, Chem. Eng. Sci. 47 (1314) (1992) 36473654.
[16] J.-I. Yang, J.H. Yang, H.-J. Kim, H. Jung, D.H. Chun, H.-T. Lee, Highly effective cobalt
catalyst for wax production in FischerTropsch synthesis, Fuel 89 (1) (2010)
237243.
[17] A.N. Pour, S.M.K. Shahri, H.R. Bozorgzadeh, Y. Zamani, A. Tavasoli, M.A. Marvast,
Effect of Mg, La and Ca promoters on the structure and catalytic behavior of iron-
based catalysts in FischerTropsch synthesis, Appl. Catal. A Gen. 348 (2) (2008)
201208.
[18] Y. Liu, B.-T. Teng, X.-H. Guo, Y. Li, J. Chang, L. Tian, X. Hao, Y. Wang, H.-W. Xiang,
Y.-Y. Xu, Y.-W. Li, Effect of reaction conditions on the catalytic performance of
FeMn catalyst for FischerTropsch synthesis, J. Mol. Catal. A Chem. 272 (12)
(2007) 182190.
[19] H.W. Pennline, M.F. Zarochak, R.E. Tischerr, R.R. Schehl, Slurry phase Fischer
Tropsch synthesis with iron-manganese catalysts, Appl. Catal. A Gen. 21 (1986)
313328.
[20] B. Wu, L. Bai, H. Xiang, Y.-W. Li, Z. Zhang, Bing Zhong, An active iron catalyst
containing sulfur for FischerTropsch synthesis, Fuel 83 (2) (2004) 205212.