Fischer–Tropsch synthesis over Pt/Co/Al2O3 catalyst: Improvement in catalyst stability by activation with diluted CO

https://doi.org/10.1016/j.apcata.2020.117645Get rights and content

Highlights

  • Activation by diluted CO prevented rapid catalyst deactivation.

  • Carbon formed and covered cobalt particles during activation by CO.

  • Carbon prevented sintering of cobalt particles in CO-activated catalysts.

  • Activation by CO leads to better catalyst at extended time-on-stream.

Abstract

Fischer–Tropsch cobalt-based catalyst deactivation is one of the major challenges facing gas-to-liquids processes. Changes in catalyst formulation and pre-treatment methods are among the promising routes to address this challenge. In this study, diluted CO was used to pre-treat a 0.5%Pt/25%Co/Al2O3 catalyst, in comparison to H2, and the resulting effects on catalyst stability, activity and product selectivity were evaluated. The investigation was performed using BET, TEM, TPR, XPS and XRD techniques, combined with catalyst evaluation in a 1 L continuously stirred tank reactor. The results reveal that microporous carbon shells encapsulate Co particles during activation by CO at higher temperatures, resulting in lower activity and higher methane selectivity compared to the H2-reduced catalyst, at early reaction times. However, the CO-activated catalyst displayed superior stability, resulting in better performance at extended time-on-stream compared to a deactivating H2-activated catalyst, which showed an increase in CH4 selectivity and a decline in C5+ hydrocarbon formation rate.

Introduction

Fischer–Tropsch (FT) synthesis is an established process that catalytically converts synthesis gas to liquid fuels. Only cobalt- and iron-based catalysts are currently used for commercial applications [1]. For cobalt-based catalysts, formulations and operating conditions resulting in longer catalyst life are desired in order to improve the process economics. To date, one of the major challenges encountered for these catalysts under realistic FT conditions is deactivation with time-on-stream. For example, Saib et al. [2] have reported data that show close to 50% activity loss for a Pt/Co/Al2O3 catalyst within 50 days on stream during realistic FT conditions (230 °C, 20 bar, H2/CO of 2/1 and (H2 + CO) conversion of 50–70%) in a 100-barrel/day slurry bubble column reactor. They identified cobalt sintering, carbon deposition and cobalt surface reconstruction as the causes of deactivation in the presence of clean synthesis gas [3]. Tsakoumis et al. [4] reviewed the literature on cobalt-based FT catalysts and reported that the proposed deactivation mechanisms are (i) poisoning, (ii) cobalt sintering, (iii) carbon formation and fouling, (iv) cobalt re-oxidation, (v) cobalt carbidization, (vi) metal-support solid state reactions and surface reconstruction, (vii) leaching of active phase and (viii) attrition.

On the other hand, appropriate changes in catalyst formulation or pre-treatment conditions have been reported to inhibit the deactivation of cobalt-based catalysts. For example, Lahtinem and Somorjai [5] found that K inhibited the formation of carbon on a cobalt foil model catalyst during CO hydrogenation. Saeys et al. [6] found that promoting a 20%Co/ɣ-Al2O3 catalyst with 0.5% boron improves the catalyst stability without affecting the activity and C5+ hydrocarbons selectivity. With boron present, density functional theory (DFT) calculations showed that C* at step and clock sites on the Co surface became destabilized. That is, boron prevented the adsorption of C* and thus, prevented nucleation and growth of polymeric carbon [7]. Sasol researchers utilized a novel chemical vapor polymerization of acetone technique to produce a carbon layer on ɣ-alumina that served to anchor cobalt particles, resulting in the formation of homogeneous small cobalt crystallites with large inter-particle distances, which might be more resistant to sintering [8]. Recently, the Davis group used phosphorus to anchor cobalt particles and decrease the sintering rate of Co/silica catalysts [9]. Other studies involving TiO2-supported cobalt catalysts [[10], [11], [12]] have reported improved catalyst stability during FT synthesis after pre-treatment using a CO-containing gas. This study aims at extending these findings to a Pt/Co/Al2O3 catalyst system by investigating the effect of catalyst activation using diluted CO on catalyst stability, activity and product distribution during catalyst testing for the FT reaction.

Section snippets

Catalyst preparation and characterization

The catalyst was prepared by a slurry impregnation method reported in an early study [13] and contained 25 wt.% Co and 0.5 wt.% Pt as promoter, on an SBA-200 ɣ-Al2O3 support. Brunauer–Emmett–Teller (BET), Transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS) and X-ray diffraction (XRD) analyses were performed on H2- and CO-reduced samples after passivation. BET analyses were performed on a Micromeritics 3-Flex system using N2 as the analysis gas while TEM analyses were

Catalyst characterization

XRD data for H2- and CO-activated catalyst samples are summarized in Fig. 1. Activation of the catalyst in H2 mainly led to Co°, whereas using CO mainly led to CoO for the samples reduced at 300 and 350 °C, and to Co° for the sample reduced at 400 °C. The results for TPR analyses, respectively in the presence of H2 (H2-TPR) and CO (CO-TPR), are reported in Fig. 2.

A small peak at ca. 100 °C was observed in the H2–TPR profile before a second peak with a maximum value at ca. 200 °C, followed by

Conclusions

At early reaction times, the catalyst activated using diluted CO (5% CO/He) at 300–400 °C, displayed higher methane selectivity and lower activity compared to the H2-activated catalyst. However, this catalyst displayed superior stability, resulting in a better performance at extended reaction times compared to a deactivating H2-reduced catalyst that showed an increase in CH4 selectivity and a decline in the rate of formation of C5+ hydrocarbons. The positive effect of catalyst activation by CO

CRediT authorship contribution statement

Kalala Jalama: Conceptualization, Investigation, Methodology, Validation, Writing - original draft, Writing - review & editing. Wenping Ma: Investigation, Methodology, Validation, Writing - review & editing. Gary Jacobs: Investigation, Resources, Writing - review & editing. Dennis Sparks: Resources. Dali Qian: Investigation. Burtron H. Davis: Conceptualization, Funding acquisition, Resources, Supervision.

Acknowledgments

Financial supports from the Department of Higher Education (DHET) Research development grant funds (2015/2016), the Academic Development and Support division at the University of Johannesburg and the Center for Applied Energy Research at the University of Kentucky are acknowledged. Shelley D. Hopps, and Michela Martinelli are acknowledged for performing XRD and BET measurements

References (36)

  • A. Saib et al.

    Appl. Catal. A

    (2006)
  • A. Saib et al.

    Catal. Today

    (2010)
  • N. Tsakoumis et al.

    Catal. Today

    (2010)
  • J. Lahtinen et al.

    J. Mol. Catal. A

    (1998)
  • K. Tan et al.

    J. Catal.

    (2011)
  • M. Gnanamani et al.

    Appl. Catal. A

    (2017)
  • K. Jalama et al.

    Catal. Commun.

    (2012)
  • K. Jalama

    Catal. Commun.

    (2016)
  • D. Nabaho et al.

    Catal. Today

    (2016)
  • Z. Wang et al.

    Carbon

    (2003)
  • U. Narkiewicz et al.

    Mater. Sci. Eng.

    (2007)
  • H. Li et al.

    J. Alloys. Compd.

    (2008)
  • D. Moodley et al.

    Appl. Catal. A

    (2009)
  • J. Llorca et al.

    J. Catalysis

    (2002)
  • K. Keyvanloo et al.

    J. Catal.

    (2015)
  • K. Cheng et al.

    J. Catal.

    (2016)
  • H. Karaca et al.

    J. Catal.

    (2011)
  • T.K. Das et al.

    Fuel

    (2003)
  • Cited by (0)

    View full text