Subscriber access provided by UNIV OF LOUISIANA
Energy, Environmental, and Catalysis Applications
Reversible exsolution of dopant improves the performance of Ca2Fe2O5 for chemical looping hydrogen production Davood Hosseini, Felix Donat, Paula M. Abdala, Sung Min Kim, Agnieszka M. Kierzkowska, and Christoph R. Müller ACS Appl. Mater. Interfaces, Just Accepted Manuscript • Publication Date (Web): 30 Apr 2019 Downloaded from http://pubs.acs.org on April 30, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
Reversible exsolution of dopant improves the performance of Ca2Fe2O5 for chemical looping hydrogen production Davood Hosseini, Felix Donat, Paula M. Abdala, Sung Min Kim, Agnieszka M. Kierzkowska, and Christoph R. Müller* Laboratory of Energy Science and Engineering, Department of Mechanical and Process Engineering, ETH Zürich, Leonhardstrasse 21, 8092 Zürich, Switzerland *Corresponding author
[email protected] (Prof. Christoph Müller)
Abstract. Hydrogen (H2) is a clean energy carrier and a major industrial feedstock, e.g. to produce ammonia and methanol. High-purity H2 can be produced efficiently from methane (CH4) using chemical looping-based approaches. In this work, we report on the development of a calcium-iron-based oxygen carrier (Ca2Fe2O5) doped with Ni or Cu and investigate its redox performance for H2 production when CH4 is used as the fuel. The experimental results suggest that the rapid formation of metallic Ni or Cu through exsolution promotes the reducibility of Ca2Fe2O5 with CH4. It was found that the reversible exsolution of Ni or Cu nanoparticles and their re-incorporation in the Ca2Fe2O5 structure is key to avoid particle sintering and deactivation. Having the potential of converting a larger fraction of steam to H2 than pure iron oxide in addition to its higher reactivity with CH4, the doped calcium-iron-based oxygen carrier is a promising material for chemical looping H2 production.
Keywords: chemical looping, hydrogen, oxygen carrier, calcium iron oxide, exsolution
Introduction Iron was used commercially in the early 20th century to produce hydrogen (H2) in the steam-iron process; 1 herein this process, iron is periodically oxidized and reduced using steam and a gaseous fuel, respectively. However, the original steam-iron process was replaced later by the economically more attractive steam reforming of methane (SMR) that is currently the dominant process for producing H2. 2 To produce H of high purity in the SMR process, additional energy-intensive downstream processing 2 steps are required. 3, 4 In contrast, chemical looping-based water splitting or H2 production, a cyclic variant of the steam-iron process, offers the potential to produce pure H2 from small to large scales. The chemical looping-based concept offers the following advantages: (i) Expensive gas separation and purification of the H2 produced is avoided, (ii) the CO2 generated is separated inherently, and (iii) the heat balance for the entire process is potentially neutral. The chemical looping-based production of H2 (CLH) is a cyclic process that involves three principal steps. First, gaseous fuel (e.g. synthesis gas or CH4) reduces iron oxide (Fe2O3), the oxygen carrier. Secondly, in the actual water-splitting step the reduced iron oxide (FeO/Fe) is re-oxidized by steam to produce H2 of high purity after unconverted steam is condensed. For chemically unmodified Fe2O3, the extent of the re-oxidation with steam is limited by thermodynamics, and to fully re-oxidize magnetite (Fe3O4) to its original state (Fe2O3) an additional oxidation step with air is required. Since the oxidation with air is always exothermic, a significant amount of heat is generated that can be utilized within the process.
1 ACS Paragon Plus Environment
ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Over the years, a vast number of Fe2O3-based oxygen carriers have been studied, both synthetic ones and natural ores. 5-9 In most works, the aim was to improve the cyclic performance by adding a metal oxide as a support (e.g. MgAl2O4 10, TiO2 11 or ZrO2 12). In addition, attempts have been also made to increase the reactivity of the material by adding dopants/promoters (e.g. Cu to enhance CH4 conversion 13, 14, Mn to increase purity and yield of H 15, La to increase the iron oxide conversion using CH 16, or 2 4 Ba to modify the electronic properties of iron oxide 17). Note that the costs for the replacement of deactivated iron oxide were one of the main reasons why the original steam-iron process was supplanted by SMR. While recyclability and reactivity of the oxygen carriers are key requirements for the CLH process, the conversion of steam to H2 is similarly important when investigating the redox behavior of the oxygen carrier. With a lower steam to H2 conversion, a larger quantity of expensive steam is required for a certain amount of H2 to be produced, exerting an energy penalty on the process. 18 For unsupported and unmodified iron oxide the maximal steam to H2 conversion at 850 °C is 62% and 21% for metallic iron and wüstite (FeO), respectively. One approach to increase the equilibrium conversion of steam is to modify Fe2O3 by mixing it with other metals to form a ternary oxide phase. For example, the perovskite La0.8Sr0.2FeO3-δ (LSF) has been shown to be capable of converting 77.2% of steam into H2 at 930 °C in a layered reverse-flow reactor. 9 Of course, a higher steam to H2 conversion implies a lowered efficiency for the reduction step of the oxygen carrier with a fuel. It thus depends on the process design, the integration of heat and the recirculation of gas streams what steam to H2 conversion is most desirable from an economic point of view; the detailed analysis thereof is, however, beyond the scope of this paper. Recently, the Ca-Fe-O system has been explored as an oxygen carrier for CLH and it has been reported that it can provide a higher steam conversion than Fe in the thermodynamic limit. 17, 19-22 For example, for pure calcium ferrite (Ca2Fe2O5) the conversion of steam was 75% at 850°C, whereas a typical supported Fe-based oxygen carrier (60 wt.% Fe2O3-ZrO2) gave a conversion of 62%, in line with thermodynamics. 19 Despite having a high steam conversion, Ca2Fe2O5 has shown a poor reactivity with CH4, which may limit its practical applicability in CLH. 21 On the other hand, several studies have reported the beneficial influence of adding Cu to Fe2O3 to improve its reducibility with CH4. 13, 14, 23, 24 It has been reported 14 that that the reactivity of Cu-doped (1 mol%) Fe2O3 with CH4 is increased substantially while the oxygen carrying capacity remains unaffected. Based on DFT calculations, the authors proposed that Cu facilitated oxygen vacancy formation, which would promote the diffusion of lattice oxygen and lower the activation energy for the partial oxidation of methane. In addition, also Ni has been assessed as a promoter for Fe2O3 due to its high reactivity and near complete fuel conversion when using CH4 as feed. 25-28 Based on the above findings, the objective of this work is to investigate in detail the effect of doping (viz. Cu and Ni) on the cyclic performance of Ca2Fe2O5 for CLH, with the aim to increase its reactivity with CH4 and its redox stability. The experimental results obtained indicate that both dopants promote the reactivity of the calcium ferrite with CH4 through the exsolution of Ni and Cu nanoparticles. However, the Ni-doped material cannot be fully regenerated during oxidation, leading to its deactivation with number of redox cycles. On the other hand, we observed that in Cu-doped calcium ferrite Cu is reincorporated in the material during re-oxidation with air, providing an effective means for the redistribution of the Cu nanoparticles counteracting sintering-induced deactivation. Such a redistribution of the dopant is key to maintain the high reactivity of the material with CH4, which results in a higher extent of reduction, leading in turn to higher H2 yields in the oxidation step.
Results and Discussion 1. Composition and structural characterization of the oxygen carriers
2 ACS Paragon Plus Environment
Page 2 of 17
Page 3 of 17 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
Three materials were investigated in this work: (i) Ca2Fe2O5, (ii) Cu-doped (5 wt.%) Ca2Fe2O5, and (iii) Ni-doped (5 wt.%) Ca2Fe2O5. The following nomenclature is used throughout this paper: (i) CaFe, (ii) CaFeCu and (iii) CaFeNi. The actual amount of the Cu and Ni dopants in the as-prepared oxygen carriers were determined by ICP-OES analysis and gave 5.8 wt.% and 4.0 wt.%, respectively (on an oxide basis). X-ray diffraction (XRD) patterns measured for the as-synthesized oxygen carriers CaFe, CaFeCu and CaFeNi are shown in Figure 1(a). The XRD results suggest that CaFe was composed of a pure Brownmillerite Ca2Fe2O5 phase. Other calcium ferrite phases such as CaFe2O4 were not observed. Rietveld analysis of the XRD data (Figure S1(a)) further confirms that CaFe crystallized in a single phase with a brownmillerite-type structure with unit-cell parameters of a = 5.421(1) Å, b = 14.780(4) Å and c = 5.589(2) Å, which are in agreement with those reported elsewhere. 30 Upon doping with Cu or Ni, the brownmillerite-type structure remained unchanged (Figure 1(b)), however, the unit-cell parameters changed to a = 5.405(2) Å, b = 14.846(5) Å and c = 5.578(2) Å for Cu, and a = 5.398(2) Å, b = 14.842(5) Å and c = 5.572(2) Å for Ni. This indicates that Cu or Ni was effectively included in the structure. The influence of Cu- or Ni-doping on the surface morphology of the CaFe oxygen carrier was probed using scanning electron microscopy (SEM). The electron micrographs are shown in Figure S1(b) and indicate that the surface morphology and size of the grains were not influenced by doping. All samples possessed a very low surface area (< 1 m2 g-1). To gain more information on the structure of the oxygen carriers, high-resolution transmission electron microscopy (HR-TEM) was conducted. The obtained images of the lattice fringes and the associated fast Fourier transform (FFT) of CaFe, Figure 1(c) show a lattice spacing of 0.271 nm, i.e. the (022) plane of Ca2Fe2O5. The values of the lattice spacing of CaFeCu and CaFeNi were determined as 0.278 nm and 0.282 nm, respectively. This observation is in line with the shift (towards lower angles) in the position of the (022) peak by XRD (Figure 1(b)) and evidences the inclusion of Cu or Ni into the Ca2Fe2O5 crystal structure (due to the larger ionic radii of the dopants, Fe3+
