The use of strontium ferrite perovskite as an oxygen carrier in the chemical looping epoxidation of ethylene

https://doi.org/10.1016/j.apcatb.2020.119821Get rights and content

Highlights

  • Ag based catalysts were investigated for ethylene epoxidation via chemical looping.

  • The effect of structural modifications of the oxygen carrier was explored.

  • Two perovskite phases were used: cubic SrFeO3, and Ruddlesden-Popper (RP) Sr3Fe2O7.

  • Intermediate perovskite ratios led to maximum yields of ethylene oxide.

Abstract

In the epoxidation of ethylene to ethylene oxide using chemical looping, Ag is supported on strontium ferrite perovskite, where Ag acts as the catalyst while the perovskite acts as an oxygen carrier. This study explores how various oxygen carriers based on strontium ferrite affect chemical looping epoxidation. The structure of the oxygen carrier was varied by incorporating different ratios of two perovskite phases: cubic SrFeO3, and a layered Ruddlesden-Popper (RP) phase: Sr3Fe2O7. Maximum yield of ethylene oxide was obtained for the sample with 1:1 SrFeO3:RP ratio, leading to an increase in the yield of ethylene oxide around 4 times, compared to only SrFeO3 (i.e. 0.4–1.6% ethylene oxide yield). These results confirm a possibility of designing the oxygen carriers used in chemical looping epoxidation towards optimal performance. Such a design approach can be expanded to other chemical looping processes to tune the performance of oxygen carrier for the reaction in question.

Introduction

Chemical looping (CL) depends on conveying lattice oxygen, instead of gaseous oxygen, to a reaction. A metal oxide, termed an oxygen carrier, donates its lattice oxygen to a gaseous reactant, and consequently is reduced to a lower oxidation state of the metal. The reduced oxygen carrier is then withdrawn from the site of this reaction, reoxidised with air, and then recycled to repeat the first step of donating oxygen to the reaction. Such a scheme enables the physical, or temporal, separation between air and the gaseous reactant. In the context of selective oxidation of organic chemicals, the chemical looping approach would eliminate the need for co-feeding gaseous oxygen with the organic substrates. As a result, only the organic component would be fed to a reactor containing the oxygen carrier coupled with a catalyst for the reaction. This represents an important improvement in safety because, in the CL route, there is a minimal risk of creating explosive mixtures (organic components mixed with air). Such improved safety means that dilution of feed gases is no longer necessary, leading to lower separation costs and possible intensification of the process. Additionally, by changing the nature of the oxygen species contributing to the reaction, from adsorbed gaseous oxygen to lattice oxygen, the CL approach might help in increasing the selectivity of multi-reaction processes towards the desired products.

CL has been explored for the production of a number of chemicals, including syngas [1], ethylene [2,3] and ethylene oxide [4]. For these processes, a suitable oxygen carrier must possess several characteristics, such as: (i) sufficient oxygen capacity, (ii) favourable thermodynamics, (iii) the capability to be reoxidised rapidly, (iv) resistance to attrition, melting and agglomeration, and (v) reasonable cost [5]. An in-silico screening of suitable materials for CL has revealed that strontium ferrite is a promising oxygen carrier owing to its high stoichiometric oxygen capacity, resistance to carbonation and the ability to withstand many redox cycles [6,7]. Capitalising on these favourable characteristics, Chan et al. (2018) explored the feasibility of supporting Ag on SrFeO3-δ and using the resultant particle as a catalyst for the epoxidation of ethylene in a CL arrangement. The rationale behind using Ag for such catalysts is that the commercial production of ethylene oxides from ethylene already depends on Ag-based heterogenous catalysts, where Ag is the active metal. The chemical looping catalyst was active for epoxidation, resulting in 4% conversion of ethylene and 25% selectivity towards ethylene oxide (EO). Such activity is significantly lower than the activity exhibited by industrial catalysts in direct epoxidation with O2(g) (e.g. ∼85% EO selectivity and 10% ethylene conversion) [8]. Improving the competitiveness of the CL approach will require higher conversions to EO. Clearly, the oxygen carrier has an important influence over this, being the only source of oxygen for the reaction. Indeed, we showed in a recent study that modifying SrFeO3 with ceria led to catalysts with 60% EO selectivity at 10% ethylene conversion, representing a significant improvement in performance compared to using SrFeO3 only [9].

In the present study we have investigated a mixture of the two perovskite-type phases, SrFeO3-δ (0 ≤ δ ≤ 0.5) and Sr3Fe2O7-γ (0 ≤ γ ≤ 1) as oxygen carriers. The two phases differ with respect to the crystal structure, chemical potential and solid-state oxygen transport [10]. Depending on oxygen vacancy ordering, the crystal structure of SrFeO3-δ can be cubic (δ = 0), tetragonal (δ = 0.125), orthorhombic (δ = 0.25) or brownmillerite (δ = 0.5) [11]. On the other hand, Sr3Fe2O7-γ is a Ruddlesden-Popper oxide with a tetragonal structure [12], that uses different pathways for oxygen-ion transport, which give such oxide excellent oxygen transport properties [13].

So far, most of the studies focusing on strontium ferrites have used one material alone, viz. SrFeO3-δ [4,14] or Sr3Fe2O7-γ [15,16]. In this study, we propose a simple preparation method that results in a mixture of the two perovskite phases. The resulting multi-phase materials are then used as oxygen carriers for chemical looping epoxidation of ethylene. We use the experimental results to discuss the feasibility of designing oxygen carriers of tailored properties for epoxidation.

Section snippets

Catalyst preparation

Here, the word “catalyst” will be used to describe a composite comprising an oxygen carrier and, deposited on its surface, metallic silver. Ruddlesden-Popper (Sr3Fe2O7-γ) and cubic perovskite (SrFeO3-δ) phases will be referred to as ‘RP phase’ and ‘cubic phase’, respectively. Using the technique of Chan et al. (2018), the catalyst was made in two stages: preparation of the support material (oxygen carrier) followed by the impregnation of the support with silver. The support was synthesised in

Phase identification and morphology

Fig. 1 shows the XRD patterns for the prepared oxygen carriers, while Table 1 shows the results for the quantitative phase analysis. As planned, the phase composition varied between all prepared samples. For each material, the evaluated composition contained more Sr3Fe2O7-γ than intended, which is consistent with the observation of Lau et al. (2017), that preparing pure SrFeO3 can yield a mixture of SrFeO3 and Sr3Fe2O7-γ. Due to the detection limit of the XRD, the experimental uncertainty

Characterisation of oxygen carriers

Fig. 3, Fig. 4 show the results of a temperature-programmed reduction and oxidation in air for the prepared materials. The results describe the basic thermochemical properties of these oxygen carriers. Overall, all carriers released similar amounts of oxygen upon heating to 900 °C, with SFO showing marginally higher amounts of oxygen released compared to 50 SFO and 100 SFO. At 900 °C, the amount of oxygen released for all perovskites is linked to their respective oxygen capacities, since the

Conclusions

This study proposes using different phase ratios of strontium ferrite as oxygen carriers for chemical looping epoxidation of ethylene. The different perovskite phase ratios induced a change in oxygen uptake and release patterns, due to the ability of the two mixed perovskites, SrFeO3 and Sr3FeO7, to exhibit multiple stoichiometries and oxygen capacities. This ultimately led to different behaviours in catalytic performances for ethylene epoxidation in chemical looping mode. The results showed

CRediT authorship contribution statement

S. Gabra: Investigation, Formal analysis, Writing - original draft. E.J. Marek: Resources, Validation, Writing - review & editing. S. Poulston: Project administration, Writing - review & editing. G. Williams: Project administration, Writing - review & editing. J.S. Dennis: Supervision, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors are grateful for financial support from the Engineering and Physical Sciences Research Council and >GS2>Johnson<\GS2> Matthey plc. through an Industrial CASE studentship. Samuel Gabra gratefully acknowledges funding from Cambridge Trust.

References (31)

Cited by (10)

  • The investigation of SrCoO<inf>3-δ</inf> perovskite doping cobalt oxide thermochemical energy storage system

    2022, Journal of the Taiwan Institute of Chemical Engineers
    Citation Excerpt :

    Besides, the TCES performance of SrCoO3-δ perovskite also is demonstrated [26]. More than that, perovskite materials are widely used as the oxygen carrier in the chemical looping technology or the materials for the oxygen permeable membrane, such as SrFeO3 and SrCoO3-δ perovskite [27,28]. Therefore, if some certain perovskites which have excellent performance of oxygen transport performance were added into the traditional metal oxide TCES materials system, it may not decrease the reactivity of the TCES material.

  • SrFeO<inf>3</inf> peculiarities and exploitation in decontamination processes and environmentally-friendly energy applications

    2022, Current Research in Green and Sustainable Chemistry
    Citation Excerpt :

    Chemical looping offers a potentially efficient and cost-effective approach for carbonaceous energy conversion, CO2 capture, and chemical production (see syngas), with different configurations and chemical feeds, i.e., methane, ethane, ethylene. In chemical looping, an oxygen carrier spontaneously releases and replenishes lattice oxygen under oxygen partial pressure and/or temperature swings (scheme in Fig. 4c) [6,9,30,31]. Several doped SF sorbents, such as SrxCe1−xFeO3-δ, SrxCa1−xFeO3-δ, SrFeyCu1−yO3-δ, SrFeyCo1−yO3-δ and SrFeyMn1−yO3-δ, have demonstrated suitable thermodynamic properties, excellent redox stability and oxygen-carrying capacity at relatively low temperatures.

View all citing articles on Scopus
View full text