Elsevier

Solid State Ionics

Volume 178, Issues 3–4, February 2007, Pages 213-220
Solid State Ionics

Stability and conductivity study of the BaCe0.9−xZrxY0.1O2.95 systems

https://doi.org/10.1016/j.ssi.2006.12.007Get rights and content

Abstract

Solid oxide components such as protonic separation membranes for the hydrogen purification and electrolyte for solid oxide fuel cell require thermo-chemical stability and high conductance. The perovskite BaCe0.9Y0.1O2.95 exhibits good proton conduction at high temperatures, but shows poor thermo-chemical stability. Substituting Zr for Ce in BaCe0.9Y0.1O2.95 improves the thermo-chemical stability but reduces proton conduction. The objective of this work was to study the optimization of protonic conductance and thermo-chemical stability by changing the ratio of Ce to Zr in BaCe0.9−xZrxY0.1O2.95. To elucidate the dopant effect, a coprecipitation and freeze drying method has been developed to produce single phase perovskites of BaCe0.9−xZrxY0.1O2.95 (0  x  0.9). The method has been optimized to yield high purity and homogeneous powders with a particle size of 50–100 nm in diameter. The sintering characteristics were studied in the temperature range of 1400–1650 °C. BaCe0.9Y0.1O2.95, BaCe0.7Zr0.2Y0.1O2.95 and BaCe0.5Zr0.4Y0.1O2.95 can be sintered to high density at 1650 °C. Sintered BaCe0.5Zr0.4Y0.1O2.95 and BaCe0.3Zr0.6Y0.1O2.95 show good chemical stability against water and carbon dioxide. Electric conductivities of sintered samples, which have been measured by impedance spectroscopy, decrease with their Zr content.

Introduction

Since the discovery of proton conduction in sintered oxides by Iwahara et al. in the early 1980s, [1], [2] there is a tremendous interest in research and the application of this type of material. These sintered metal oxides, which are crystalline and demonstrate good proton conduction between 300 °C and 1000 °C, are generally known as high temperature protonic conductor (HTPC), solid state protonic conductor, and high temperature proton conducting ceramics [3], [4], [5], [6], [7]. Similar as the stabilized zirconia solid electrolytes for solid oxide fuel cell, the HTPCs have oxygen ion vacancy in the crystal structure. Upon exposure to water, hydrogen and other hydrogen-containing atmosphere, they become protonic conductor by incorporating proton to their oxygen ion. The proton can hop between oxygen ions in the crystal and therefore the ceramic exhibits proton conduction. The mechanism of the proton conduction was an assumption based on the dependence of conductivity upon atmosphere [2]. It has been observed and verified by various microscopical investigations of the materials [8], [9], [10], [11], [12]. Applications of HTPC membranes include hydrogen and steam sensors, electrolytes for fuel cell, hydrogen purification, hydrogen pump and steam electrolyzer [5], [13], [14]. CaZr0.9In0.1O3−δ has been commercialized as hydrogen sensor for molten aluminum casting process by TYK Corporation of Japan [15], [16].

One important aspect of HTPC is the chemical stability to the environment. BaCeO3 based ceramics show the highest proton conductance, but it is thermodynamically the least stable among all HTPC materials [7], [17], [18], [19], [20], [21], [22]. As its lattice energy (or Gibbs free energy) is higher, the crystal structure is intrinsically less stable. For powder preparation and pellet sintering at high temperature (∼ 1400 °C), it reacts with alumina and zirconia sample holders. It is known that sintered pellets disintegrate and decompose upon contacting boiling water [17], [21]. Additionally, the material shows reactivity with carbon dioxide. The BaCeO3 based material can be stable between 600 and 800 °C, the most important range for intermediate temperature solid oxide fuel cell [19]. Nevertheless, the lack of stability toward water around water boiling temperature remains a concern for many applications [21]. On the other hand, BaZrO3 based ceramics show high stability towards water and carbon dioxide [22], [23]. However, the material exhibits poor protonic conductivity. Solid solutions of BaCeO3–BaZrO3 has been prepared by solid-state reactions and investigated to improve environmental durability and maintain high conductivity [24], [25], [26], [27]. Zirconium substitution enhances environmental durability but decreases conductivity. Moreover, modeling based on quantum mechanics and conductivity measurement based on single crystal concluded it should be a better protonic conductor. The perceived lower conductivity of the sintered ceramics is possibly due to high grain boundary impedance [23]. For HTPCs, high grain boundary resistance is usually caused by impurities introduced during processing. If the materials with better purity can be prepared, the solid solutions of BaCeO3–BaZrO3 deserve further investigation.

Coprecipitation method has been reported to prepare BaCeO3 based HTPC [28], [29]. Finer powders can be prepared at lower temperature. The sintered oxides also exhibited favored properties such as reduced grain boundary activation energy for conductivity. Ammonium oxalate was used as a precipitant, which does not stoichiometrically precipitate zirconium from solution containing zirconium dinitrate oxide based on this investigation. A different coprecipitation method is reported here. It can be used to prepare both of BaCeO3 and BaZrO3 based oxides and their solid solution BaCe1−xZrxO3. Carbonate combined with moderate alkaline solution is used as a precipitant that can stoichiometrically precipitate most of metals except group IA metals. It is somewhat different from the procedure for the high temperature superconductors [30]. Since none of the metal in the system forms soluble complex compound with ammonia, ammonium hydroxide and ammonium carbonate are used a precipitant and for pH adjustment instead of organic bases such as tetramethyl ammonium hydroxide and methylamine [30], [31], [32]. All the by products of the process are volatile upon heating, therefore the precipitates do not have to be carefully washed for a pure product. In addition to the coprecipitation process, the precipitates are freeze dried instead of oven dried to minimize coarsen of the powders. BaCe0.9−xZrxY0.1O3−δ (x = 0, 0.2, 0.4, 0.6, 0.9) have been prepared by the coprecipitation/freeze drying method. Aliovalent doping of Y (or other trivalent elements such as Nd, Yb, etc.) is necessary to introduce oxygen ion vacancy (in the crystal structure) that allows the insertion of water to facilitate proton conduction.

BaCeO3 and BaZrO3 are notoriously difficult materials to densify. Sintering temperatures above 1600 °C are needed to achieve densities ≥ 90% theoretical. Traditionally, solid state methods are used to prepare coarse (1–5 μm) HTPC powders. It is advantageous to process at lower temperatures. Reduction in particle size can reduce sintering temperatures. One purpose of this investigation is to examine the feasibility of sintering the material at a lower temperature. Crystalline fine powders generated from the method have been sintered. Both of the powders and the sintered oxides are tested for chemical stability. Electrical conductivity has been measured by impedance spectroscopy.

Section snippets

Powder preparation

Ba(NO3)2 (99.95%, Alfa Aesar), Ce(NO3)3·6H2O (99.99%, Alfa Aesar), Y(NO3)3·xH2O (99.99%, Alfa Aesar, the x is nominally equal to 6.00 to guarantee at least 99.9% purity), ZrO(NO3)2·xH2O (99.9%, Alfa Aesar, the x is determined by ICP and gravimetric method to be 2.46 for the batch we received) are used for metal sources. The precursors were weighed to yield BaCe0.9−xZrxY0.1O2.95 (x = 0, 0.2, 0.4, 0.6, 0.9). The metal nitrates were dissolved in deionized water. In a separate beaker, (NH4)2CO3

Powder properties and sintering

The precipitates were freeze dried to reduce formation of hard agglomerations during drying or calcination. The freeze dried powder flows freely and it remains so after calcination. As a result, mechanical grinding is not necessary and avoids contamination of grinding media. Fig. 1 shows the results of TG for precursor compositions of BaCe0.9Y0.1O2.95 and BaZr0.9Y0.1O2.95. The TG results indicate the weight loss (i.e. the conversion of barium carbonate in the precursors) is completed by

Conclusions

Crystalline fine powders of BaCe0.9−xZrxY0.1O2.95 (x = 0, 0.2, 0.4, 0.6, 0.9) with excellent stoichiometry have been prepared by a coprecipitation/freeze dry process. Sintering temperatures above 1600 °C are still needed to densify the materials. At 1650 °C, BaCe0.9Y0.1O2.95, BaCe0.7Zr0.2Y0.1O2.95, and BaCe0.5Zr0.4Y0.1O2.95 can be sintered above 95% of theoretical density. BaCe0.3Zr0.6Y0.1O2.95 is only 80% and BaZr0.9Y0.1O2.95 barely densified under the same conditions. The dense pellets show

Acknowledgements

Permission to publish this work from NASA Glenn Research Center is greatly appreciated. Technical assistances of Ralph Garlick, Paula Heimann, Anna Palczer, and Jami Olminsky are gratefully acknowledged. This research was partially funded by the strategic research fund of NASA Glenn Research Center through NASA cooperative agreement NCC3-850.

References (42)

  • H. Iwahara et al.

    Solid State Ionics

    (1981)
  • H. Uchida et al.

    Solid State Ionics

    (1983)
  • T. Norby

    Solid State Ionics

    (1999)
  • H. Iwahara

    Solid State Ionics

    (1995)
  • H. Iwahara

    Solid State Ionics

    (1996)
  • N. Bonanos

    Solid State Ionics

    (2001)
  • T. Matzke et al.

    Solid State Ionics

    (1996)
  • W. Münch et al.

    Solid State Ionics

    (1996)
  • H. Maekawa et al.

    Solid State Ionics

    (1999)
  • W.G. Coors

    J. Power Sources

    (2003)
  • G. Zhang et al.

    Solid State Ionics

    (2003)
  • T. Yajima et al.

    Sens. Actuators, B, Chem.

    (1993)
  • T. Yajima et al.

    Solid State Ionics

    (1995)
  • K.D. Kreuer

    Solid State Ionics

    (1999)
  • S. Wienströer et al.

    Solid State Ionics

    (1997)
  • K.H. Ryu et al.

    Solid State Ionics

    (1999)
  • K. Katahira et al.

    Solid State Ionics

    (2000)
  • K.D. Kreuer et al.

    Solid State Ionics

    (2001)
  • S.D. Flint et al.

    Solid State Ionics

    (1995)
  • S. Gallini et al.

    Solid State Ionics

    (2003)
  • F. Chen et al.

    J. Eur. Ceram. Soc.

    (1998)
  • Cited by (236)

    View all citing articles on Scopus
    View full text