Comparative study of Fischer–Tropsch synthesis with H2/CO and H2/CO2 syngas using Fe- and Co-based catalysts

https://doi.org/10.1016/S0926-860X(99)00173-8Get rights and content

Abstract

Hydrogenation of CO, CO2 and their mixtures has been comparatively studied with a Co–MnO–Aerosil–Pt and a Fe–Al2O3–Cu–K catalyst at the University of Karlsruhe.

With iron catalysts as promising for CO2 hydrogenation, their composition was varied: (1) several supports (SiO2, TiO2, Al2O3), (2) alkali promotion (Li, Na, K, Rb), (3) usage of Zeolite Y as catalyst component. The catalysts were characterised by adsorption methods, XRD, TPR and temperature programmed decarburisation after a H2/CO2 treatment (Korea Research Institute of Chemical Technology).

Iron and cobalt catalysts behaved differently in CO2 hydrogenation. With the alkalised iron catalyst the same hydrocarbon product composition was obtained from a H2/CO2 and from a H2/CO synthesis gas in spite of the CO partial pressure remaining low, specifically due to water gas shift equilibrium constraints. With the cobalt catalyst at increasing CO2 and respectively decreasing CO content of the syngas, the product composition shifted from a Fischer–Tropsch type (mainly higher hydrocarbons) to almost exclusively methane. These basically different catalyst behaviours are explained by different modes of formation of the kinetic regime of FT synthesis—selective inhibition of methane formation and the selective inhibition of product desorption as a prerequisite for chain growth—in the case of iron through irreversible carbiding and alkali surface coverage and in case of cobalt through strong reversible CO adsorption.

Investigation of the various modified iron catalysts showed alumina to be the best support for CO2 hydrogenation and potassium to act as a powerful promotor.

With the Fe–Y–zeolite–alkali catalysts, a decrease of methane selectivity was observed in the order Li < Na < K < Rb being applied as promotors.

Introduction

Chemical CO2 fixation has received much attention since anthropogenic green house gases—and among these particularly the CO2—appear to warm up the atmosphere [1]. Among conversion processes, those for liquid automotive fuels are generally much larger in scale than processes producing chemicals. This is substantialised by Fig. 1 in more detail, showing the annual world mass flow of major fuels and petrochemicals in terms of the associated carbon mass flow. Liquid fuels production through FT CO2 hydrogenation thus could act as a substantial CO2 sink, provided hydrogen could be produced on a no-CO2-release-basis (hydroelectric, solar or nuclear energy). FT hydrocarbon products would benefit from (1) present infrastructures, (2) high specific value, high cleanliness (no sulphur, no aromatics, high quality city diesel) and suitability of organic bulk chemicals themselves or respective feedstocks. CO2 would be available from fossil fuel fired power plants. Thermodynamic considerations indicate no equilibrium constraints for FT CO2 hydrogenation (, ).CO2+3H2CH2+2H2RH500K0=−125kJ/molΔRG500K0=−20kJ/molCO2+2H2CH2+H2RH500K0=−165kJ/molΔRG500K0=−40kJ/molwhere values of ΔRH0 and ΔRG0 per CH2 have been calculated as 1/6 of those per n-C6H14

In FT processes not CO2 consumption but CO2 formation can play an important role, specifically when using CO rich syngases (H2/CO∼1) from high temperature coal gasification [6] (Eq. (3)).2CO+H2CH2+CO2

This overall stoichiometry is achieved by the water gas shift activity of alkalised iron catalysts which allows also the reaction of H2O, produced in the FT conversion, with the CO of the feed gas Eq. (4).H2O+COCO2+H2

H2O and CO2 in the reacting mixture can affect the catalysts themselves and of course the reaction kinetics and this differently for iron and cobalt catalysts. With iron catalysts H2O acts strongly inhibiting [7], [8], thereby reducing the possibility of attaining high syngas conversion per pass. The simultaneously produced CO2 is only weakly inhibiting [9]. Both, H2O and CO2 are oxidising compounds in the reacting mixture and may cause oxidation and structural changes of the iron catalyst [10], [11], [12]. Lowering the CO partial pressure of the syngas would decrease its carbiding potential [14].

With cobalt catalysts partial pressures of H2O can act beneficially [7], [15] as reducing methane selectivity substantially. Water has no inhibiting influence with cobalt catalysts and even promoting effects have been reported. This allows for high degrees of conversion with cobalt catalysts as no product inhibition is present. However, under certain circumstances water might oxidise the cobalt unfavourably.

With cobalt catalysts CO2 is neither formed nor produced during FT synthesis and also neither strongly adsorbed nor hydrogenated, thus playing the role of a diluting gas. Only if the cobalt catalyst is mixed with a water gas shift catalyst, also the WGS-reaction proceeds during the FT conversion and such hybrid catalysts might be potential candidates for FT CO2 hydrogenation, provided, no unfavourable features of the system prevail. This problem will be addressed below.

FT CO2 hydrogenation with iron catalysts has been addressed in a number of investigations. Promoting of the iron with e.g. Cr, Mn, Mo, Zn was studied [16], [17], [18]. The most beneficial promoting was observed with potassium, applied in high concentration (up to 0.5 mol K/mol of Fe) [19], [20].

Characterisation studies—in situ and with spent catalysts—indicated the essential formation of iron carbide phases [13], [14]. Mössbauer studies on iron catalysts used for CO2 hydrogenation revealed χ-Fe5C2, Θ-Fe3C and Fe3O4 phases to occur [17]. Carbide formation was favoured by K-promotion of the catalyst. By in situ dynamic XRD experiments [21] a rapid transformation of Fe2O3 to Fe3O4 was observed, followed by slow transformation of Fe3O4 to χ-Fe5C2. Starting with metallic iron the formation of ε′-Fe2,2C and χ-Fe5C2 was found. With supported iron catalysts, the carbide phase mainly consisted of χ-Fe5C2 [22]. A strong interrelation between iron carbide formation and catalyst FT activity for CO2 hydrogenation has been reported by Fiato et al. [23]. About the mechanism of FT CO2 hydrogenation the general view is a first reverse WGS reactionCO2+H2CO+H2Oto produce CO which is subsequently consumed in the FT conversionCO+2H2CH2+H2OHowever, the additional reaction of a direct hydrogenation of CO2 has also been proposed [23].

Section snippets

Experimental

The comparative FT synthesis experiments with a cobalt and an iron catalyst at varied CO and CO2 content of the syngas were carried out in a fixed-bed reactor with the finely powdered catalysts (dP < 0.1 mm) covering larger fused silica particles (dP = 0.25–0.4 mm) as an adhering layer, the weight ratio of catalyst to fused silica particles being 1 : 10.

Thus intraparticle mass transfer influences even in catalyst pores filled with liquid products were excluded due to the small particle size. On the

CO and CO2 hydrogenation on the Co/MnO/SiO2/Pt catalysts.

It is known for FT synthesis with Co catalysts that CO2 is not being formed [15], [29]. Co catalysts exhibit no activity for the CO shift reaction, correspondingly, hybrid catalysts containing a Co shift component would be required to convert CO2 into CO which subsequently reacts to hydrocarbons on the FT catalyst, provided no direct FT synthesis is possible with CO2. Because of its activity for the CO shift reaction, MnO was included into the catalyst.

At a constant total synthesis pressure of 1

Conclusions

With cobalt and iron catalysts CO and CO2 play different roles in Fischer–Tropsch synthesis. With cobalt catalysts CO2 acts merely as a diluent. Thus replacing CO by CO2 in the synthesis gas means only reducing its CO content. This leads specifically to a change of product composition from a FT type (mainly hydrocarbons C2+) to a methanation type (only methane obtained as product). This change in product composition is explained through the principle of selective inhibition as dominating the FT

References (41)

  • D.B. Bukur et al.

    J. Catal.

    (1995)
  • H. Schulz et al.

    Appl. Catal. A

    (1999)
  • H. Schulz et al.

    Stud. Surf. Sci. Catal.

    (1997)
  • M.D. Lee et al.

    Appl. Catal. A

    (1991)
  • K.W. Jun et al.

    Stud. Surf. Sci. Catal.

    (1998)
  • H. Jung et al.

    J. Catal.

    (1992)
  • G.B. Raupp et al.

    J. Catal.

    (1979)
  • R.A. Fiato et al.

    Stud. Surf. Sci. Catal.

    (1998)
  • H.P. Bonzel et al.

    Surf. Sci.

    (1982)
  • G. Ertl et al.

    Chem. Phys. Lett.

    (1979)
  • I.R. Leith et al.

    Appl. Catal.

    (1988)
  • T. Inui, M. Anpo, K. Izui, S. Yanagida, T. Yamaguchi (Eds.), Advances in chemical conversions for mitigating carbon...
  • B.V. Bora, T.L. Marker, Abstracts of the 15th World Petroleum Congress, Topic 4 (P3)...
  • K. Weissermel, H.J. Arpe, Industrial Organic Chemistry, VCH, Weinheim...
  • OECD Environmental data...
  • Deutscher Mineralölwirtschaftsverband, Hamburg 1998, personal...
  • H. Kölbel et al.

    Erdöl und Kohle

    (1949)
  • H. Schulz et al.

    Stud. Surf. Sci. Catal.

    (1994)
  • W. Zimmermann et al.

    Can. J. of Chem. Eng.

    (1990)
  • R.B. Anderson, The Fischer–Tropsch Synthesis, Academic Press, London,...
  • Cited by (0)

    View full text