Catalytic Performance of Fe-Mn / SiO 2 Nanocatalysts for COHydrogenation

A series of xx(Fe, Mn)/SiO2 nanocatalysts (xx x x, 10, 15, 20, 25, and 30wt.%) were prepared by sol-gel method and studied for the light ole�ns production from synthesis gas. It was found that the catalyst containing 20wt.% (Fe, Mn)/SiO2 is an optimal nano catalyst for production of C2–C4 ole�ns. Effects of sulfur treatment on the catalyst performance of optimal catalyst have been studied by espousing different volume fractions of H2S in a �xed bed stainless steel reactor. e results show that the catalyst treated with 6 v% of H2S had high catalytic performance for C2–C4 light ole�ns production. e best operational conditions were H2/CO = 3/2 molar feed ratio at 260 C and GHSV = 1100 h under 1 bar total pressure. Characterization of catalysts was carried out using X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and surface area measurements.


Introduction
Since initial introduction of Fischer-Tropsch synthesis (FTS), increasing concentration has been made for the increasing of advantages as well as drawback reduction of this potentially commercial process.One of the best approaches to improve the selectivity toward viably more important products involves the use of a supported bimetallic catalyst [1].Using this approach, selective production of petrochemical feed stocks such as ethylene, propylene, and butylenes (C 2 -C 4 light ole�n) directly from syngas is hoped to be attainable [2,3].Due to opinion of high activity and selectivity as well as the cost problems, iron-based catalysts are the catalysts of choice.e other metal partner is typically manganese or cobalt.A high ole�n selectivity for Fe-Mn catalysts has been reported due to formation of iron-manganese oxides and carbide phases [3,4].Recent studies show that primary -ole�n products of the FTS were subjected to aerward reactions on iron surface [5][6][7][8].Sulfur treatment is the subject of contrast results.Barrault et al. [9] studied the poisoning of cobalt and iron catalysts by sulfur and observed a decrease in catalytic activity and a propensity for forming lower ole�ns.Similar results are reported by Kitzelmann and Vielstich [10] who used a K 2 S to achieve a light poisoning of Fe and Co catalysts.e methane selectivity was reduced by 50% and the lower ole�n selectivity was increased by ≈10%.Due to accumulation and deactivating times, most of the researches on sulfur poisoning were done using exaggerated sulfur levels of the syngas.In contrast, there are some recent reviews [11,12] and reports highlighting the advantageous effects of sulfur on both iron [13,14] and other active metals [15][16][17][18][19][20].It has been demonstrated that a small amount of sulfur species on the catalyst surface could be associated with improved FTS activity and enhanced ole�n selectivity.It was also reported that the activity was only slightly lowered by this treatment.Li and Coville have observed a decrease in methanation for Co/TiO 2 catalyst [21,22].Anderson et al. found that selectivity toward light hydrocarbons products increased with increasing sulfur content of alkali-promoted iron catalysts and [23][24][25][26][27]. Stenger and Satter�eld [26] reported a 60% increase in the activity of a fused magnetite catalyst aer exposure to synthesis gas containing H 2 S.
Herein, we investigated the role of H 2 S-treated catalyst on decreasing of methane and increasing the C 2 -C 4 ole�ns in products.We also reported the optimization process and effects of operational conditions on the catalytic performance of an optimal catalyst.Characterization of catalysts was carried out using X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and surface area measurements.

Experimental
All chemical reagents and solvents were analytical grade and purchased from Fluka and Merck.e speci�c surface area, the total pore volume, and the mean pore diameter were measured using a NOVA 2200 instrument.e XRD patterns of the precursor and calcined samples were recorded on a Philips X' Pert (40 kV, 30 mA) X-ray diffractometer using a Cu K radiation source (  2 �) and a nickel �lter.e TEM investigations were carried out using an H-7500 (120 kV).e morphology of catalyst and precursor was observed by means of S-360 scanning electron microscopy.2).e meshed catalyst (0.5 g) was diluted with similar granules of quartz beads (1.0 g) and held in the middle of the reactor (30 cm length and internal diameter 6 mm).All catalysts were activated on-line (reduced) for a 10 h period in pure hydrogen (1 bar) at a temperature of 400 ∘ C and space velocity of 800 h − .Reactant and stream products were analyzed on-line using a Varian Star 3600CX gas chromatograph equipped with a thermal conductivity detector (TCD) and a chromosorb column.e heavy hydrocarbon products were analyzed off-line using a Varian CP 3800 gas chromatograph with a Petrocol Tm DH100 fused silica capillary column and a �ame ionization detector (F�D).e conversion percentage of CO was based on the fraction of CO that formed carboncontaining products according to where   is the number of carbon atoms in product ,   is the weight percentage of product , and  CO is the percentage of CO in the syngas feed.e selectivity () toward product  is based on the total number of carbon atoms in the product and is therefore de�ned as (the time required to attain steady-state condition was about 10 h) were summarized in Table 1.According to the results, the catalyst containing 20 wt.% (Fe, Mn)/SiO 2 has the highest selectivity towards C 2 -C 4 ole�ns and the lowest selectivity with respect to methane and CO 2 .erefore, this catalyst was chosen as the optimal catalyst for the conversion of synthesis gas to light ole�ns.Characterization studies were carried out using various techniques for both the precursors and calcined catalysts.SEM images show a few differences in morphology of precursor and calcined catalysts (Figure 3).e catalyst precursor seems to have relatively larger agglomerations of particles than the calcined catalyst.Characterization studies were carried out using XRD technique for the 20 wt.% (Fe, Mn)/SiO 2 calcined catalysts (Figure 4).e actual identi�ed phases for this catalyst were Fe 3 O 4 (cubic), SiO 2 (hexagonal), Mn 2 O 3 (cubic), and Fe 2 O 3 (rhombohedral).e particle size was determined from the half width of the most intense peak of the diffraction pattern around 2  334 using the Scherrer equation [29],     , where  is the mean crystallite diameter,  has an assumed value of 0.9,  is the X-ray wave length (1.54 Å), and  is the width of the diffraction peak at half maximum.e catalysts containing 20 wt.% (Fe,Mn)/SiO 2 have particle sizes of about 35 nm.e catalyst containing 20 wt.% (Fe, Mn)/SiO 2 was characterized with TEM [30] (Figure 5).As shown in Figure 5, the particle sizes are from 25-40 nm.is result conform with obtained results that was studied by using the Scherrer equation.

Process for H
Partially sul�dation of the catalyst gives the sul�de phases which act as support and hence maintains the dispersion of the iron centers.Vacancies created by the loss of H 2 O and sulfur also increase the porosity of the sul�de salt of catalysts.Surface area of the sulfur-treated and untreated catalysts was determined using N 2 absorption desorption.e results show signi�cant e�ects of sulfur treatment on porosity and  speci�c surface area of catalyst (Table 3).e results show that speci�c surface area increases with increasing the H 2 S volume fraction until to 6 v%.e data could imply that the S increases the dispersion of the Fe and Mn which might be a reason for the better catalytic performance of the above catalyst [27,28,31].It can be seen from Table 3 that sulfur treatment in excess of 6 v% resulted in a decrease in the speci�c surface area and CO conversion.According to Table 2, it seems that it is the speci�c surface area, pore volume, and pore size distribution are dependent to sulfur treatment.e H 2 S-treated (6 v%) catalyst was subjected to XRD characterization before and aer catalyst performance test, and the corresponding XRD patterns are presented in Figure 6.Before the performance test, the XRD pattern shows the presence of Fe 1− S (hexagonal) phase in addition to  Fe 2 O 3 (rhombohedral), Mn 2 O 3 (cubic), and SiO 2 (hexagonal).ese patterns disappear aer the performance test and instead another pattern that belongs to Fe 3 O 4 (cubic), MnO (cubic), and carbide phases FeC and Fe 2 C appears.e results show that the sul�de phase has been removed during the reaction.In addition, metallic iron is rapidly converted to iron carbide during the reaction which may be subjected to further oxidation into Fe 3 O 4 .It is well known that the iron carbides phases are active for FTS and oxidic species are responsible for production of ole�ns [32][33][34].

Effect of Operational
Conditions.One of the other major factors which have a marked effect on the catalytic performance of a catalyst is the operating conditions.For optimizing of the reaction conditions in this study, the effects of operating conditions such as H 2 /CO feed molar ratios, GHSV, reaction temperatures, and reactor total pressures were examined to investigate the catalyst stability and its performance for the light ole�ns production.

3.3.1.
Effect of H 2 /CO Molar Feed Ratio.e in�uence of the H 2 /CO molar feed ratio on the steady state catalytic performance of the catalyst treated with 6 v% of H 2 S was investigated for the FTS at 250 ∘ C, GHSV = 1000 h − , and atmospheric pressure.e CO conversion and light ole�n products selectivity percent are shown in Table 4. e results showed that with variation in H 2 /CO molar feed ratios from 1/1 to 3/1, different selectivity with respect to C 2 -C 4 light ole�ns was obtained.Among them, for H 2 /Com = 3/2 (GHSV = 1000 h − ), the total selectivity of C 2 -C 4 light ole�ns was the highest while, the CH 4 and CO 2 selectivity was the least.erefore, the H 2 /CO = 3/2 ratio was chosen as the optimum ratio for conversion of the syngas to C 2 -C 4 ole�ns over the 20 wt.% (Fe, Mn)/SiO 2 nanocatalyst treated with 6 v% of H 2 S.

Effect of Gas Hourly Space Velocity (GHSV).
To obtain a better understanding of the factors affecting the catalytic performance of 20 wt.% (Fe, Mn)/SiO 2 nanocatalyst treated with 6 v% of H 2 S, a series of experiments were carried out at different GHSV from 800 to 1300 h − under the reaction conditions (H 2 /CO = 3/2,    bar at 250 ∘ C), and the results are presented in the Table 5. e CO conversion increased with increasing space velocity and reached a maximum CO conversion of 72% for space velocity of 1100 h − and then decreased with further increasing of space velocity.At the same time, methane and CO 2 selectivity decreased till space velocity of 1100 h − then increases markedly.
Madon and Taylor [35] studied the effect of space velocity on the ole�ns and para�ns selectivity for Ru catalyst and studied at a range of temperatures between 220-290 ∘ C under the same reaction conditions (   bar, H 2 /CO = 3/2, and GHSV = 1100 h − ), and the results are presented in the Table 6.e results show that for the reaction temperature at 260 ∘ C, the total selectivity of light ole�ns products was the highest.In addition, the CO conversion increases with increasing the operating temperature.In the same way, it has been reported that at low reaction temperatures, the conversion percentage of CO is low and so it causes a low catalytic performance [9].On the other hand, increasing the reaction temperature leads to increasing of methane as an unwanted product.erefore, in this study, 260 ∘ C is considered the optimum operating temperature.ese results indicate that the reaction temperature is a parameter of crucial importance in the catalytic performance of ironmanganese catalyst for hydrogenation of CO.

Effects of Total Pressure.
A series of experiments were carried out to investigate the performance of the 20 wt.% (Fe, Mn)/SiO 2 nanocatalysts treated with 6 v% of H 2 S during variation of total pressure in the range of 1-10 bar, at the optimal reaction conditions of H 2 /CO = 3/2, GHSV = 1100 h − , and 260 ∘ C (Table 7).e results indicate that at the total pressure of 1 bar, the optimal catalyst showed a high selectivity respect to C 2 −C 4 light ole�ns.It is also apparent that the C 5 -C 9 and C  + selectivities increase with increasing the pressure [36].e results also indicate that the CO conversion and the total selectivity with respect to C 2 −C 4 light ole�ns decrease with increasing the pressure.Increase in the selectivity of higher molecular weight hydrocarbons of Fe-Mn catalyst upon increasing the pressure can be explained by the increased concentration of -ole�ns and readsorption and chain initiation of these primary products on catalyst surface which lead to the ultimate desorption of these -ole�ns as larger products.Hence, because of high CO conversion and higher total selectivity with respect to C 2 -C 4 ole�ns at the total pressure of 1 bar, this pressure was chosen as the optimum pressure.

Conclusions
In conclusion, it is found that the activity and selectivity of the catalyst are affected by the level of sulfur adsorbed on the catalyst, and the catalyst treated with 6 v% of H 2 S showed the best catalytic performance for light ole�ns production.e operational conditions such as H 2 /CO molar feed ratio, gas hourly space velocity (GHSV), reaction temperature, and reaction total pressure were very effective and the optimal operating conditions for production of light ole�ns were found to be 260 ∘ C with molar feed ratio of H 2 /CO = 3/2 (GHSV = 1100 h − ) under the total pressure of 1 bar.e optimal nanocatalyst treated with 6 v% of H 2 S was found to be superior to the other catalysts in terms of better C 2 -C 4 selectivity in the FTS products and higher ole�n/para�n ratio (2.6).In addition, methane formation by using this modi�ed catalyst was suppressed, which caused decreasing of methane selectivity from 24.3 to 15.8% at 260 ∘ C with molar feed ratio of H 2 /CO = 3/2 (GHSV = 1100 h − ) under the total pressure of 1 bar.

F 3 :
SEM images of precursor (a) and calcined catalyst containing 20 wt.% (Fe, Mn)/SiO 2 .A possible reaction proposed by Van der Kraan et al.[27] may be accounted for the advantages of sulfur treatment of iron-based FT catalysts (3):
[28]Catalyst Preparation.Fe(NO 3 ) 3 ⋅9H 2 O and Mn(NO 3 ) 2 ⋅4H 2 O (Fe/Mn molar ratio is 3/1[28]), tetraethyl orthosilane were dissolved separately in ethanol at 60 ∘ C and mixed together.An ethanolic solution of oxalic acid (H 2 C 2 O 2 S Treatment of the Catalyst.As Figure1shows, the system composed of two parts, H 2 S generator and sulfur treatment reactor.eH 2 S was produced on addition of H 2 SO 4 (from dropping funnel �) to a �ask containing Na 2 S⋅H 2 O (�ask 5) and then was mixed with N 2 and conducted to a stainless steel reactor.ecarriergascontainingdesired volume fraction of H −1 , and    bar at 250 ∘ C. b H 2 S volume fractions =  H 2 S /( H 2 S +  N 2 ).3.2.1.Effect of H 2 S VolumeFraction.ecatalystcontaining 20 wt.% (Fe, Mn)/SiO 2 was treated with different H 2 S volume fraction at 200 ∘ C and 1 bar for 5 h and then the treated catalysts were sub�ected to FTS production of light ole�ns under same reaction conditions (H 2 /CO = 2/1, GHSV = 1000 h − ,    bar at 250 ∘ C). e results show the highest total selectivity respect to C 2 -C 4 light ole�ns products (C 2 -C 4 ole�ns/C 2 -C 4 paraffin = 2.26) as well as the least CH 4 and CO 2 selectivities were achieved using 6/100 H 2 S volume fraction (Table2).Higher volume fractions led to less total CO conversion which is a drawback in the industrially point of view.us, the effects of sulfur strongly depend on the S loading, possibly because different catalyst functions are affected by sulfur.T 3: N 2 adsorption-desorption measurements of iron-manganese catalyst treated with H 2 S. H 2 S volume fractions a Speci�c surface area (m 2 g −1 ) 2 S ( H 2 S ( H 2 S +  N 2 )) was passed over the meshed catalyst (1.0 g) held in the middle of a �xed bed stainless steel reactor (  2 ∘ C and    bar for 5 h).Before H 2 S treatment of the catalyst, the system should be exposed to the stream of pure N 2 for 30 min to eliminate the oxygen.T 1: Effect of loading of (Fe-Mn) on the catalytic performance of catalysts.Reaction conditions: H 2 /CO = 2/1, GHSV = 1000 h −1 , and    bar at 250 ∘ C. T 2: Effect of different H 2 S volume fractions on the catalytic performance of catalyst a .a H 2 S volume fractions =  H 2 S /( H 2 S +  N 2 ).
T 4: Effect of different H 2 /CO feed ratio on the catalytic performance of catalyst.
T 5: Effect of different GHSV on the catalytic performance of catalyst a .Reaction conditions: H 2 /CO = 3/2,    bar at 250 ∘ C. T 6: Effect of different reaction temperature on the catalytic performance of catalyst.According to the results in Table5, at the ranges of 800-1100 h − , signi�cant increasing on light ole�ns selectivity was observed.It is apparent that in GHSV = 1100 h − the selectivity for C 2 -C 4 light ole�ns was increased.erefore, in this study, GHSV = 1100 h − is considered to be better GHSV at 250 ∘ C, because in this GHSV a high CO conversion and total selectivity of light ole�ns products and low CH 4 and CO 2 selectivity were observed.eseresultsindicatethat the GHSV is a parameter of crucial importance on the catalytic performance of iron-manganese catalysts for hydrogenation of CO.3.3.3.Effect of ReactionTemperature. e effect of reaction temperature on the catalytic performance of the 20 wt.% (Fe, Mn)/SiO 2 nanocatalyst treated with 6 v% of H 2 S was T 7: Effect of different total reaction pressure on the catalytic performance of catalyst.Reaction conditions: H 2 /CO = 3/2, GHSV = 1100 h −1 , and   2 ∘ C.