Proton Conducting Properties of Sr1+XLn1-XAlO4-δ (Ln=Pr,Sm) with Layered Perovskite Structure for Solid Oxide Fuel Cells

Abstract

Perovskite material, such as SrCeO3,BaCeO3, SrZrO3 and BaZrO3 are expected for as electrolyte material of proton conductive SOFC due to their high proton conductivity at 500-700 ºC.[1] Proton conductive electrolyte is required not only high conductivity but also high ionic transport number, high proton transport number and high stability. BaCe1-xYxO3-δ(BCY) is reported as a high conductivity material but the chemical stability is not sufficient especially in CO2 atmosphere. Recently, we reported Pr doped Ba2In2O5(PBI)[2] for proton conducting SOFC. However, the SOFC with PBI demonstrated relatively low power density of 4.0mW/cm2 at 700○C. Therefore, the development of electrolyte material is still required for PC-SOFC.  To develop the novel proton conductor, we focused on layered perovskite structures. We have reported the proton conductivity of Sr1+xPr1-xInO4-δ (SPI)[3], which has Ruddlesden-Popper structure (expressed in A’OAnBnO3n, when n=1 called as K2NiF4 structure). Due to introduction of rock salt layer (A’O layer), improving proton transport number was suggested due to the selective inhibition of oxide ion conduction and improvement of stability were also expected. With considering proton transport mechanism which is suggested as shown in Eq.1, the basicity (relating to electronegativity) of constituent element is also an index of proton transport number at high temperature by stabilizing OH-. As a result, the total conductivity of SPI was logσ≒-1.5 in 900 ºC in 8% humidified O2, and power generation was not stably operated at 500 ºC in H2 fuel. Their conductivity, stability and transport number in high temperature is a remaining issue to be improved.  To improve conductivity and stability, we tentatively applied the tolerance factor of simple perovskite structure (Eq. 2) as an index of the structure symmetry of K2NiF4. When tolerance factor is near 1, the structure has good symmetry, so higher stability and better ionic conductivity is expected. For A and A’ site with large ionic radius and B site with small element are required to keep the tolerance factor near 1. Thus, in this work we chose Al for B site which has small ionic radius instead of In. Sr1+xPr1-xAlO4-δ (SPA) has good tolerance factor of 1.01. Moreover, we also focused on ionic valence which has the connection with oxygen deficiency to improve conductivity and proton transport number. Valence change will change amount of oxygen vacancy and which are estimated by Eq. 3-4. Pr ion is Pr3+ or Pr4+ and Sm ion is Sm3+ or Sm2+ .They have almost same ionic radius and electronegativity as shown in Table1. Therefore, we also investigated Sm instead of Pr for A site element to discuss the effect of ionic valence for the electrolyte properties. Sr1+xSm1-xAlO4-δ(SSA) also has good tolerance factor of 1.01. We synthesized SPA and SSA with different A/A’ ratio and observed their power generation characteristics to investigated the confirm the validity to use such index for the development of proton conductive materials. 　All samples were synthesized by solid phase reactions using the powders of Pr6O11, Sm2O3, Sr2CO3  and Al2O3 .After making the disk of SPA and SSA and sintering in 1400 ºC for 10 h, analysis of crystal structure was carried out by XRD measurement. The conductivities were measured by using AC impedance spectroscopy. Ionic transport numbers were evaluated by the configuration of H2 and O2 concentration cells. The power generation of SOFC with SPA and SSA electrolyte are carried out at 500 ~ 900℃. 　The total conductivities of SPA and SSA are lower than that of SPI in dry and humidify O2.On the other hand, SSA and SPA has proton conductivity and high proton transport number especially at 700 ~ 900 ºC. The power generation experiments using SOFC with the disks of SPA and SSA were carried out. [1] N. Kochetova et.al. RSC Adv., 2016, 6, 73222 [2]X.Li, M.Ihara, JOURNAL OF THE ELECTROCHEMICAL SOCIETY, 2015,162,8 [3]X.Li, H.Shimada, M.Ihara, ECS Transactions, 50(27), 3-14 (2013) Figure 1