Elsevier

Journal of Catalysis

Volume 237, Issue 1, 1 January 2006, Pages 152-161
Journal of Catalysis

Investigation of promoter effects of manganese oxide on carbon nanofiber-supported cobalt catalysts for Fischer–Tropsch synthesis

https://doi.org/10.1016/j.jcat.2005.10.031Get rights and content

Abstract

The effects of the addition of MnO were studied on a carbon nanofiber-supported cobalt catalyst. The starting sample, cobalt loading 9.5 wt% and 8% cobalt dispersion, was promoted by impregnation with small amounts of MnO (0.03, 0.1, 0.3, 0.6, and 1.1 wt%). XPS and STEM-EELS showed MnO to be associated with Co in both dried and reduced catalyst. In the drying step, MnO was deposited on the passivated cobalt particles, because of the tendency of both metals to form stable mixed compounds. After reduction, the MnO remained close to the cobalt particles, because the support material lacked sites with significant interaction with MnO. The promoter suppressed both the hydrogen chemisorption uptake and the cobalt reducibility even with the lowest MnO loading. At 1 bar, large improvements in the selectivity toward C5+ products (from 31 to 45 wt%) were found with MnO loadings of 0.3 wt% and higher. At 20 bar, the addition of only 0.03 wt% MnO improved the C5+ selectivity from 74 to 78 wt%, but larger amounts decreased the selectivity to 52 wt% at 1.1 wt% MnO. The surface-specific activity (TOF) first increased with MnO loading from 26 to 60×10−3s−1 for 0.3 wt% MnO, whereas it decreased, probably as a consequence of excessive Co surface coverage at MnO loadings >0.3wt%. Product analysis (paraffin:olefin ratio) indicates that a major role of MnO involves moderation of hydrogenation reactions.

Introduction

In the Fischer–Tropsch (FT) reaction synthesis gas (CO/H2) is catalytically converted into hydrocarbons via surface polymerisation. Using synthesis gas produced from natural gas, coal, or biomass, transportation fuels can be produced from feedstocks other than crude oil. The quality of the products formed in combination with a non-crude oil feedstock support the FT process to play a crucial role in the worldwide energy supply in the coming decades.

Supported cobalt catalysts are well known for their activity and selectivity in the FT reaction [1]. The catalysts are often promoted with small amounts of noble metals to decrease the reduction temperature and increase the activity [2], [3], [4]. To achieve better selectivities toward long-chain products, special metal oxides can be added [4], [5], [6], [7]. In this paper we investigate the influence of manganese oxide on cobalt supported on carbon nanofiber (CNF) catalysts. MnO is reported to be a promoter for cobalt-based FT catalysts in both academic and patent literature. Originally most research was devoted to cobalt on MnO2 supports and to systems with mixed oxides of cobalt and manganese [8], [9], [10], [11], [12]. All of these systems have relatively high manganese loading. MnO promoter effects on cobalt catalysts supported on oxidic carriers were investigated recently [13], [14], [15], [16], [17], [18], [19], [20]. The promoting effect of MnO is suggested to originate from a lower degree of cobalt reduction [12], [13], [20]. In all cases, a metal oxide was also used as support material, also decreasing cobalt reducibility and thus complicating the analysis. Therefore, we decided to reduce support effects by using an inert carrier, CNF. CNF is a novel graphitic support material with promising applications, including use as a support for FT catalysts [21], [22]. Recently, we showed, using X-ray photoelectron spectroscopy (XPS) and STEM-EELS, that in Co/CNF catalysts promoted with MnO, the promoter is present only in the vicinity of the cobalt particles and not elsewhere on the support [23].

This paper provides a comprehensive characterization and systematic investigation at low and high pressure of the catalytic effects of manganese loading on Co/CNF catalysts. Catalysts with Co/Mn molar ratios varying from 11 to 431 were prepared and were characterized by acid–base titration, H2 chemisorption, scanning transmission electron microscopy–electron energy loss spectroscopy (STEM-EELS), transmission electron microscopy (TEM), temperature-programmed reduction (TPR), XPS, and X-ray diffraction (XRD), whereas FT catalysis experiments were carried out in fixed-bed reactors at both atmospheric pressure and 20 bar.

Section snippets

Catalyst preparation

CNF of the fishbone type (average diameter, ca. 30 nm) were grown from syngas as described previously [24]. After purification by refluxing in 1 M KOH, adsorption sites were created by refluxing the CNF in concentrated HNO3, as described by Toebes et al. [25]. After washing and drying at 120 °C, CNF with a BET surface area of 160 m2/g and a bulk density of 0.50 g/ml were obtained.

Cobalt was loaded on the activated CNF by incipient wetness impregnation (pore volume, 0.56 ml/g of a solution

Titration

The activated CNF used for loading Co contained 0.15 mmol of acidic groups per gram. MnO was loaded on the Co/CNF after a reduction treatment at 350 °C for 2 h. Unfortunately, the amount of acidic groups on Co/CNF could not be measured directly, because the presence of cobalt interfered. Therefore, a sample of the original oxidised, unloaded CNF batch was reduced in H2 at 350 °C and subsequently titrated. This sample contained only 0.06 mmol/g, which is an upper limit for the acid sites in

Conclusion

In this paper we have shown that CNF provides a suitable support for studying the manganese promotion effect in cobalt-based FT catalysis, because it enables the study of a promoter effect without interference of support effects. Therefore, cobalt properties can be influenced by the addition of MnO loadings as low as 0.03 wt%. XPS and STEM-EELS demonstrated that manganese is closely associated with cobalt both in the catalyst precursor and in the final catalyst. We found that manganese retards

Acknowledgements

The authors thank J.W. Geus and H. Meeldijk for the TEM studies, V. Koot for the TPR and H2 chemisorption measurements, and A.J.M. Mens for the XPS measurements. Financial support was provided by Shell Global Solutions.

References (39)

  • D. Schanke et al.

    J. Catal.

    (1995)
  • A. Kogelbauer et al.

    J. Catal.

    (1996)
  • G. Jacobs et al.

    Appl. Catal. A

    (2002)
  • A. Guerrero-Ruiz et al.

    Appl. Catal. A

    (1994)
  • G.J. Haddad et al.

    J. Catal.

    (1996)
  • G.R. Moradi et al.

    Catal. Commun.

    (2003)
  • M.J. Keyser et al.

    Appl. Catal. A

    (1998)
  • Y.-Q. Zhang et al.

    J. Nat. Gas Chem.

    (1997)
  • M. Jiang et al.

    Appl. Catal. A

    (2001)
  • F. Morales et al.

    J. Catal.

    (2005)
  • G.L. Bezemer et al.

    Stud. Surf. Sci. Catal.

    (2004)
  • M.L. Toebes et al.

    Catal. Today

    (2002)
  • M.L. Toebes et al.

    Carbon

    (2004)
  • H.P.C.E. Kuipers

    Solid State Ionics

    (1985)
  • F. Winter et al.

    Carbon

    (2005)
  • R.C. Reuel et al.

    J. Catal.

    (1984)
  • M. Oku et al.

    J. Electron. Spectrosc. Relat. Phenom.

    (1976)
  • M.A. Langell et al.

    Appl. Surf. Sci.

    (2000)
  • C.J. Bertole et al.

    J. Catal.

    (2004)
  • Cited by (0)

    View full text