Low-Temperature Reducibility of MxCe1âxO2 (M = Zr, Hf) under

Subscriber access provided by UNIV OF NEBRASKA - LINCOLN

Article x

1-x

2

Low Temperature Reducibility of MCe O (M = Zr, Hf) under Hydrogen Atmosphere Alexander Bonk, Arndt Remhof, Annika C Maier, Matthias Trottmann, Meike V. F. Schlupp, Corsin Battaglia, and Ulrich F. Vogt J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.5b10796 • Publication Date (Web): 25 Nov 2015 Downloaded from http://pubs.acs.org on November 26, 2015

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 free 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 accessible to all readers and 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.

The Journal of Physical Chemistry C 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 37

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

The Journal of Physical Chemistry

Low Temperature Reducibility of MxCe1-xO2 (M = Zr, Hf) under Hydrogen Atmosphere Alexander Bonk,1,2 Arndt Remhof,1 Annika C. Maier,1 Matthias Trottmann,1 Meike V.F. Schlupp,1 Corsin Battaglia,1 Ulrich F. Vogt1,2* 1 Empa, Swiss Federal Laboratories for Materials Science and Technology, Laboratory of Materials for Energy Conversion, 8600 Dübendorf, Switzerland 2 Albert-Ludwigs-University Freiburg, Crystallography, Institute for Geo- and Environmental Science, D-79098 Freiburg i.Br., Germany

ACS Paragon Plus Environment

1


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

Page 2 of 37

ABSTRACT RedOx cycles utilize the reversible oxygen release/uptake of cerium oxide in a variety of renewable energy applications such as fuel cells, water-gas-shift reactions and solar thermochemical fuel production. For all applications the degree of reduction/oxidation determines the overall performance. In this study we report on the redox behavior of MxCe1-xO2-δ (M = Zr, Hf; x = 0, 0.15, 0.2) solid solutions monitored by high temperature in-situ x ray diffraction. During reduction in H2 at 600°C and the successive formation of Ce3+ and oxygen vacancies, the lattice of ceria expands up to 0.3%. The lattice expansion of Hf doped ceria samples is four times larger than in Zr doped or undoped ceria, indicating drastically higher extents of oxygen vacancy formation. The same trends are validated using temperatureprogrammed reduction measurements. Complete reoxidation of the MxCe1-xO2-δ solid solutions in air at 600°C is reflecting the reversibility of the redox process. Scanning electron microscopy and x-ray diffraction analysis before and after redox cycling indicate phase and microstructural stability of all compositions during reduction.

KEYWORDS: doped ceria, chemical expansion, synthetic fuel, redox cycles, low temperature reduction


2

Page 3 of 37

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


INTRODUCTION Nonstoichiometric oxides based on cerium are of growing interest in many fields of application such as fuel cells,1-2 water gas shift reactions,3-4 and thermochemical fuel production5-9. These techniques utilize the capability of ceria to release oxygen from its lattice according to Eq. (1), to form nonstoichiometric CeO2-δ. , + 2

(1)

Upon reduction Ce3+ cations and oxygen vacancies are formed as expressed by the same reaction written in Kröger-Vink notation in Eq.(2). , 1 2 + 2 + + ⦁⦁ 2

(2)

The formed oxygen vacancies are utilized for oxygen ion conduction (fuel cells), CO oxidation (three way catalysis and water-gas-shift), or the chemical reduction of H2O and CO2 (thermochemical cycles) to produce syngas (CO / H2). A high concentration of oxygen vacancies, equal to a high extent of Ce4+ Ce3+ reduction, is desired in all processes, as it relates to a higher performance, more specifically a higher ionic conductivity and / or higher fuel output. Recent investigations have demonstrated that substituting Ce4+ (r(Ce4+) = 0.97 Å) with smaller isovalent Zr4+ or Hf4+ cations (r(Zr4+) =0.84 Å, r(Hf4+) = 0.83 Å) can increase the achievable nonstoichiometry δ since they compensate for the cumulative strain caused by the formation of large Ce3+ (r(Ce3+)= 1.14 Å) cations.5, 10-11 It was shown that the reduction yield of ceria increases linearly with increasing Zr4+ and Hf4+ concentrations at high temperatures.

12-13

This redox performance can be utilized in both high (1500°C) and low (~600°C) temperature


3


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

Page 4 of 37

applications.14-15 Therefore, it is of interest to monitor and evaluate the chemical expansion, induced by the formation of Ce3+, and the influence of dopants on it. In the early 1990’s Chiang et al. monitored the chemical expansion of ceria between 800°C-900°C and found a linearly increasing chemical expansion with increasing nonstoichiometry δ in CeO2-δ.16 Bishop and coworkers described that the expansion can be decreased by the addition of zirconia due to a stronger displacement of Zr4+ towards neighboring vacancies as compared to Ce4+.17 Successively the material compensates for cumulative strain caused by cation expansion during reduction. Studies on Hf doped ceria are not available so far.

In this work we address the reducibility of MxCe1-xO2 (M = Zr, Hf; x = 0, 0.15, 0.2) at even lower lower temperatures (600°C) in H2 using in-situ X-ray diffraction (XRD) and specifically address the phase development during reduction. Furthermore, the chemical expansion, H2 consumption, microstructural stability and reversibility of the oxygen release reaction are evaluated to analyze the low temperature reduction properties of ceria as a function of dopant type and concentration.


4

Page 5 of 37

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


EXPERIMENTAL ZrxCe1-xO2 and HfxCe1-xO2 solid solutions (x = 0, 0.15, 0.2) were synthesized from a Pechini synthesis.13 Cerium (IV) ammonium nitrate (Alfa Aesar, 99.9 %) and zirconium oxynitrate (Alfa Aesar, 99.9 %) or hafnium chloride (Alfa Aesar, 99.9 %), respectively, were dissolved in deionized water (75 ml). A mixture of anhydrous citric acid (CA) (Sigma-Aldrich, 99.5 %) and 1,4-butanediole (BD) (Sigma-Aldrich, 99 %) was stirred at 100°C until all citric acid was dissolved. The aqueous metal-containing solution was added to the CA-BD mixture. At 150°C an esterification reaction was promoted yielding a honey-like polymer, which was dried at 80°C for 24 h. The polymer was transferred into an alumina crucible and calcined for 10 h at 700°C in a constant air flow. For the creation of the desired porosity the pre-calcined powders are mixed with spherical carbon pore former particles (CPF) (0.4-12 µm particle size, HTW Hochtemperatur-Werkstoffe GmbH) in a 1:1 volume ratio in an agate mortar. Samples were uniaxially pressed (Ø 10mm, 10kN) and sintered for 5 h at 1600°C (Carbolite HTF 17/10). The Zr and Hf doped ceria samples will be referred to as Zr15, Zr20, Hf15 and Hf20 hereafter.

Microstructures of the sintered samples were analyzed with a FEI ESEM XL30 using a secondary electron (SE) detector. In back scattered electron (BSE) - mode the presence of dopant rich phases was ascertained. EDX spectra were acquired on a FEI NanoSEM using an acceleration voltage of 20kV. The spectra were evaluated taking into account the sensitivity factors for different elements (k-factor correction).

Temperature programmed reduction by hydrogen (H2-TPR) was performed on a QuantaChrom ChemBET 3000. Experiments were performed in a 5 vol% H2 in N2 gas mixture (80ml/min).


5


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

Page 6 of 37

Pechini samples after sintering were ground in a pestille mortar and 100 mg of sample of each composition was analyzed between room temperature and 900°C.

Phase development and lattice parameters were determined by X-ray diffraction (XRD) on a Bruker D8 (Parallel Beam) using Cu-Kα (λ = 1.5418 Å) radiation (parallel beam, 1 °/min scan speed, 0.014° increment), in an in situ cell. Briefly, the cell consists of a heatable sample stage covered by an X-ray semi-transparent hemispherical beryllium window, described in more detail in Ref.18. The setup is connected to a gas manifold allowing for a switch between air, vacuum and hydrogen (H2, 99.999%). Measurements are performed in static atmosphere. In the in-situ cell thermal expansion coefficients were determined by measuring XRD scans in static air at 30°C, 200°C, 400°C and 600°C. The lattice parameters were derived from Rietveld refinement (Topas 4.2 software) using the structure of CeO2 19 as structural model. Chemical expansion was recorded under 1.5-2.5 bar H2 pressure (static atmosphere). The in situ cell was evacuated at 600°C (to ~10-3 mbar) and flushed with hydrogen three times prior to the XRD scans. The sample was kept at 600°C for 1h to assure thermal equilibrium. Successively, in situ XRD scans were recorded between 20°-65° 2θ, with a scan speed of 1°/min (increment 0.014°) to monitor phase changes. Before and after redox cycling XRD scans were recorded in the 2θ range 10°110° on a PANalytics X’Pert Pro (using the same parameters as on the Bruker D8). The unit cell parameters, zero corrections, scaling factors, peak shape (Pseudo-Voigt profile functions, including FWHM (W)), isotropic thermal factors and side occupancies were refined in the given order. Ionic radii are taken from Shannon’s work.20

The lattice thermal expansion (αLTE) was derived from linear fits of four isothermal XRD scans (Table 1) between 25°C and 600°C and calculated from Eq. (3).


6

Page 7 of 37

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


= /! ∙ #

(3)

where ∆a is the lattice parameter difference at the temperature difference ∆T.21

Expansion of the cubic ceria lattice after the switch from air to H2 at 600°C is attributed to the reduction of ceria which is reflected by the formation of oxygen vacancies, as well as Ce3+ (r(Ce3+) = 1.14 Å) cations that are larger than the “host” Ce4+ (r(Ce4+) = 0.97 Å) cations. The underlying reaction is shown in Eq. (4) for doped ceria: -. % & + '⁄2 ) → % +,--- & , /

&

+ '⁄2 )

(4)

where M is the tetravalent dopant, x its concentration and y the concentration of formed Ce3+. The gained nonstoichiometric cerium oxide can be regarded as a solid solution of CeO2, CeO1.5 and dopant oxide as described by Mogensen et al.22 As long as the fcc structure is maintained the (theoretical) lattice parameter can be calculated assuming Vegard’s law according to (Eq. (5)).

=

4

√3

3456789: + 46:89: ; ∗ 0.9971

(5)

with 456789: = A4B + 31 − A;D31 − ';4 EF + '4 EEE G

(6)

1 1 46:89: = H1 − 31 − A;'I 4 + 31 − A;' ∗ 4. 4 4

(7)

and with


7


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

Page 8 of 37

where y is the concentration of Ce3+ (or the extent of reduction), a is the lattice parameter, x is the dopant concentration (for tetravalent dopants), r indicates the ionic radii of CeIV, CeIII, dopant cation M, oxygen anion O and oxygen vacancy VO, and 0.9971 is a correction factor.23-24 Solving the sum of Eq.(5-7) for y (the extent of reduction) yields Eq. (8). √3 ∗ 4 ∗ 0.9971 − 31 − A;4 EF − 4 − A4J '= 1 3A − 1;4 EF + 31 − A;4 EEE + 3A − 1; 4 + 31 − A; 1 4. 4 4

(8)

Successively, the changes in lattice parameter during reduction can be derived from Rietveld refinement and the reduction extent can be calculated from Eq. (8). RESULTS & DISCUSSION (i) Thermal expansion Undoped and doped ceria samples have a face centered cubic (fcc) structure (Fm-3m space group) after sintering according to the XRD patterns presented in Fig. 1. The lattice parameters of all samples derived from Rietveld refinement are presented in Table 1. When Zr4+ or Hf4+ are doped into the ceria lattice, the lattice parameter decreases due to the replacement of Ce4+ (0.97 Å) with smaller isovalent Hf4+ (0.83 Å) and Zr4+ (0.84 Å), which is reflected by a shift of the peaks to higher angles (2θ) in the XRD patterns (Fig. 1(a) and (b)). This is in agreement with Vegard’s law and previously reported data on similar materials published by the authors.13 All samples exhibit a white yellowish color after sintering, indicating that they are in stoichiometric state (nmetal:noxygen ≈ 1:2) state.25


8

Page 9 of 37

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


Table 1 Existing phases, and phase content in undoped and doped ceria after reduction in H2 at 600°C, as well as average lattice expansion, and successive average reduction extent ([Ce3+]) during the last five scans in H2.

Lattice Parameter [Å] Temperature CeO2

Hf15

Hf20

Zr15

Zr20

25

5.4056

5.3671

5.3491

5.3727

5.3588

200

5.4177

5.3780

5.3576

5.3823

5.3671

400

5.4314

5.3873

5.3686

5.3894

5.3791

600

5.4405

5.3994

5.3798

5.3924

5.3919

αLTE [10-6 K-1]

11.3

10.3

10.0

9.6

10.8

[°C]


9


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

Page 10 of 37

Figure 1 (a) XRD patterns of MxCe1-xO2 (M = Zr, Hf; 0 ≤ x ≤ 0.2) after sintering for 5 h at 1600°C, (b) zoom into the 220peak.

The calculated lattice parameters of undoped and doped ceria at 25°C, 200°C, 400°C and 600°C are plotted in Fig. 3(a). Undoped ceria not only exhibits the largest lattice parameter at 25°C, but also expands the most in the presented temperature range. Fits through four measured lattice parameters (Fig. S1) show that upon doping with 20 mol-% Zr4+ the thermal expansion coefficient of ceria decreases from 11.3 (all CTE values in [10-6 K-1]) to 10.8. A decrease to 10.0 appears when Ce4+ is replaced by 20 mol-% Hf4+ (see Fig. S1). Similar values are obtained for Zr15 (9.6) and Hf15 (10.3), implying that the CTE of ceria can be decreased by the addition of Zr4+ and Hf4+ cations thereby reducing residual stresses during thermal expansion.

(ii) Reduction I – undoped ceria


10

Page 11 of 37

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


All reduction experiments were performed at 600°C by switching the atmosphere from static air to static hydrogen with an intermediate evacuation step. Fig. 2(a) presents a zoom showing the evolution of the 111-peak of undoped CeO2 at 600°C before and after the switch from air to H2 (P0 = KL = 2.5 bar) over time, plotted as a contour map. In a 2D representation the first (in air) and the last (in H2) scan of the contour map are also plotted in Fig. 2(b).

Figure 2 (a) Contour map of the 111 peak of CeO2 during an isothermal switch from air to H2, plotted over time, (b) XRD pattern of the first and the last scan, (c) lattice parameter of the two emerging fcc-phases indicated by #1 and #2. Dotted lines in (c) are guide to the eye. The scale bar indicates the normalized peak intensity and is valid for all contour maps in this study.


11


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

Page 12 of 37

Immediately after the switch from air to H2 atmosphere a shift of the 111 peak to lower angles is observed. The other peaks (200, 220, 311 and 222) shift accordingly (see Fig. S2). Furthermore, a shoulder at lower angles (2θ) appears next to all peaks (exemplarily highlighted for the 111peak in Fig. 2(b)). This indicates the formation of an additional cubic phase after the switch to H2 atmosphere. By Rietveld refinement of the whole 2θ-range (20°-65°), the new phase was identified as an fcc structure with a larger lattice constant than the main component. Hereafter, these phases are referred to as the main, #1, and the secondary fcc phase, #2. The lattice parameters of both phases derived from Rietveld refinement are plotted in Fig. 2(c) for all scans. Phase contents of ~38% (#1) and ~62% (#2) were suggested from Rietveld refinement during the last 5 scans in H2 (Table 2).

Table 2 Existing phases, and phase content in undoped and doped ceria after reduction in H2 at 600°C, as well as average lattice expansion, and successive average reduction extent ([Ce3+]) during the last five scans in H2.

Material

CeO2-δ

Zr15

Zr20

Hf15

Atmosphere

H2

H2

H2

H2

Phase Content

average ∆a/a0

average [Ce3+]

[wt-%]

[%]

[%]

fcc 1

38

0.16

3

fcc 2

62

0.29

6

fcc 1

70

0.04

Low-Temperature Reducibility of MxCe1âxO2 (M = Zr, Hf) under

Recommend Documents

Low-Temperature Reducibility of MxCe1âxO2 (M = Zr, Hf) under