Elsevier

Catalysis Today

Volume 215, 15 October 2013, Pages 95-102
Catalysis Today

Effect of precursor on the catalytic performance of supported iron catalysts for the Fischer–Tropsch synthesis of lower olefins

https://doi.org/10.1016/j.cattod.2013.03.018Get rights and content

Highlights

  • A more uniform distribution of iron particles was achieved using AIC as precursor.

  • High coke levels were observed for catalysts with highly aggregated iron particles.

  • Large iron particles fragmented due to carbon deposition during the FTO.

Abstract

Lower olefins are traditionally produced from cracking of naphtha and other crude oil fractions. The Fischer–Tropsch-to-Olefins process (FTO) enables the direct synthesis of lower olefins from synthesis gas (CO + H2) derived from alternative feedstocks such as natural gas, coal or biomass. A catalyst suitable for this process must comply with different requirements: high selectivity for C2single bondC4 olefins, low methane selectivity, high catalytic activity and excellent mechanical and chemical stability under demanding reaction conditions (high temperatures and low H2/CO ratios). These features have been reported for a catalyst consisting of iron-containing nanoparticles promoted with sodium and sulfur dispersed on a weakly interactive support. In this study, Na plus S promoted α-alumina supported catalysts with loadings of 1–20 wt% Fe have been prepared using different iron precursor salts to investigate their effects on catalytic performance. The catalysts prepared from iron nitrate or ammonium iron citrate both consisted of iron nanoparticles of 15-20 nm and displayed high selectivity to lower olefins (>50% C) in combination with low methane selectivity (<20% C) when tested under industrially relevant conditions (340 °C, 20 bar and H2/CO = 1 v/v). The catalyst synthesized with ammonium iron citrate exhibited a higher catalytic activity and a much lower rate of carbon lay-down (factor 4–6) compared to the sample prepared with iron nitrate. Tentatively, the differences in catalytic performance are attributed to a more uniform distribution of the iron particles observed when ammonium iron citrate was used as precursor. These results suggest that the extent of aggregation of iron (carbide) nanoparticles affects their catalytic performance.

Introduction

Lower olefins are used for the production of plastics, solvents, lubricants, pharmaceuticals, cosmetics and many other products. They are traditionally produced from oil refining processes, steam cracking of naphtha and by dehydrogenation of ethane and propane from natural gas. The need for alternative feedstocks and processes to produce these major building blocks from carbon sources other than oil has pushed research in this area during the last 40 years [1], [2].

An interesting alternative is to produce lower olefins from synthesis gas, a mixture of CO and H2 derived from natural gas, coal or biomass. One of the processes that allow the conversion of synthesis gas into lower olefins without intermediate steps is the so called Fischer–Tropsch-to-Olefins (FTO) process [3], [4], [5]. Recently, selective, active and stable iron catalysts for this process have been reported [6], [7], [8], [9]. These catalysts consist of iron-containing nanoparticles promoted with sodium and sulfur dispersed on a carrier material with low interaction toward iron. It has been shown that the nature of the support [6], the size of iron carbide particles [8] and the amount and nature of promoters [9] play a crucial role in determining the performance for the selective production of lower olefins from synthesis gas.

Another aspect that may have a major impact on catalytic activity and product selectivity is the spatial distribution of the active carbide particles on the support material. Nanoparticle growth is one of the main deactivation pathways for many commercial catalysts [10], [11], [12], [13], [14], [15], [16], [17]. This phenomenon has been studied extensively for processes involving the conversion of synthesis gas such as the production of methanol [18], [19], [20], [21] and the Fischer–Tropsch synthesis of transportation fuels over cobalt catalysts [22], [23], [24]. Most recently our group has reported a tremendous impact of the distribution of copper nanoparticles on their stability for methanol synthesis [18].

Strategies to mitigate active phase particle growth comprise alloying with a higher melting point metal to reduce their mobility [25], [26], increasing the metal-support interaction by using specific carriers [27], [28] and the development of improved catalyst synthesis methods that allow a homogeneous distribution of the active particles on the support material [18]. The modification of the active phase by adding a metal with a higher melting point or the use of a carrier material with a stronger interaction with the active phase might have a negative impact on catalytic activity and product selectivity [6]. For this reason, modified catalyst preparation techniques that maximize the inter-particle spacing without affecting its activity and selectivity are preferred. Improvements on the distribution of metal nanoparticles on a support have been reported using controlled thermal decomposition of metal nitrates [18], [29], [30], [31], confinement of metal nanoparticles [32], [33], [34], and using chelated metal complexes as precursors during the impregnation of the support [35], [36].

The Fischer–Tropsch synthesis carried out in the presence of an iron catalyst has been mainly directed for the production of gasoline and light hydrocarbons. These catalysts are in general unsupported (bulk) materials and they have been investigated extensively. Limited studies have been performed on supported iron catalysts for the selective production of lower olefins from synthesis gas. The supported iron catalysts reported in literature were mainly prepared using high surface area carrier materials such as γ-Al2O3, SiO2 or activated carbon. Generally, iron nitrate was the metal precursor of choice because of its high water solubility and low cost. The distribution of iron-containing nanoparticles on catalysts prepared using iron nitrate and high surface area supports may be quite uniform. Nevertheless, strong metal-support interactions hinder the formation of the active carbide phase thus resulting in a low catalytic activity [6]. The decrease in activity due to a strong interaction between the support and the iron-containing particles can be overcome by using an inert support [6], [7]. The main disadvantage of using a carrier with low interaction toward the iron phase is that the metal (carbide) particles tend to aggregate during reaction resulting in deactivation by loss of active surface area. For this reason, it is necessary to use an iron precursor that will allow obtaining a homogeneous distribution of nanoparticles on the inert support. To this end, we have chosen to use chelated metal complexes.

In the present work we have prepared series (1–20 wt% Fe) of unpromoted α-alumina supported iron catalysts using two different iron precursors: an inorganic salt (iron nitrate) and a chelated metal complex (ammonium iron citrate). An additional series of sulfur and sodium promoted catalysts using both precursors were prepared for comparison. The catalysts have been tested for FTO to investigate the influence of the iron precursor on catalytic performance.

Section snippets

Preparation of unpromoted Fe/α-Al2O3 catalysts with iron nitrate

Five supported catalysts with different nominal iron loadings (1 wt%, 2 wt%, 5 wt%, 10 wt% and 20 wt% Fe) were prepared using the incipient wetness impregnation technique. The series of catalysts XNFe, where X corresponds to the iron loading, were prepared using aqueous solutions of Fe(NO3)3·9H2O (Acros, ACS reagent ≥98%). The amounts of iron salt and demineralized water used to prepare the solutions are specified in the supplementary data, Table S1. For the preparation, 5 g of α-Al2O3 (10 m2 g−1; pore

Results and discussion

The samples with 5 wt% iron loading were analyzed with ICP–optical emission spectroscopy to determine the composition of the calcined catalysts. The results of this analysis are summarized in Table S3 (supplementary data). The promoted catalysts 5NFeP and 5AFeP showed similar sulfur and sodium contents (∼0.2 wt% Na and 0.03 wt% S).

The XRD analysis of 5 wt% Fe and 10 wt% Fe catalysts demonstrated that samples prepared with different iron precursors and with the same iron loading had similar

Conclusions

The influence of the iron precursor at loadings of 1–20 wt% Fe on the catalytic performance of Fe-based FTO catalysts was studied. The catalysts prepared with ammonium iron citrate exhibited a more uniform distribution of iron (oxide) particles on α-Al2O3 compared with the samples prepared with iron nitrate based on analysis with TEM of the calcined and the reduced catalysts (before catalysis). The unpromoted and promoted samples with 5 wt% Fe showed similar average iron (oxide) crystallite sizes

Acknowledgements

Financial support was provided by ACTS-ASPECT (NWO). We thank G. Bonte and A. Chojecki for performing the catalytic tests at 20 bar performed at Dow Benelux. The contribution Prof. E.J.M. Hensen in the discussion of the results of in situ Mössbauer spectroscopy experiments is gratefully acknowledged. We would like to thank C. van de Spek and J.D. Meeldijk for the TEM images of fresh and spent catalysts and T. Zalm for ICP analysis.

References (42)

  • C.H. Bartholomew

    Applied Catalysis A

    (2001)
  • D.L. Trimm

    Applied Catalysis A

    (2001)
  • Y. Nagai et al.

    Catalysis Today

    (2011)
  • R.J. Liu et al.

    Applied Catalysis A

    (2005)
  • K.O. Christensen et al.

    Applied Catalysis A

    (2006)
  • A.K. Datye et al.

    Catalysis Today

    (2006)
  • G. Jacobs et al.

    Applied Catalysis A

    (2002)
  • A.M. Saib et al.

    Catalysis Today

    (2010)
  • Ø. Borg et al.

    Journal of Catalysis

    (2008)
  • C.S. Chen et al.

    Applied Catalysis A

    (2004)
  • R. Riva et al.

    Applied Catalysis A

    (2000)
  • A.J. van Dillen et al.

    Journal of Catalysis

    (2003)
  • D. Chen et al.

    In situ catalyst characterization by the oscillating microbalance catalytic reactor (TEOM)

  • Z. Liu et al.

    Fuel Processing Technology

    (2000)
  • T. Liu et al.

    Nanostructured Materials

    (1999)
  • D.B. Bukur et al.

    Applied Catalysis A

    (1995)
  • D.S. Kalakkad et al.

    Applied Catalysis A

    (1995)
  • C. Wang et al.

    Journal of Natural Gas Chemistry

    (2003)
  • M. Janardanarao

    Industrial and Engineering Chemistry Research

    (1990)
  • E. Schwab et al.

    Oil Gas European Magazine

    (2010)
  • B. Sun et al.

    ChemCatChem

    (2012)
  • Cited by (0)

    View full text