A High Temperature Electrochemical Energy Storage System Based on Sodium Beta-Alumina Solid Electrolyte (Base)

Abstract

Listen

Translate article

This report summarizes the work done during the period September 1, 2005 and March 31, 2008. Work was conducted in the following areas: (1) Fabrication of sodium beta{double_prime} alumina solid electrolyte (BASE) using a vapor phase process. (2) Mechanistic studies on the conversion of {alpha}-alumina + zirconia into beta{double_prime}-alumina + zirconia by the vapor phase process. (3) Characterization of BASE by X-ray diffraction, SEM, and conductivity measurements. (4) Design, construction and electrochemical testing of a symmetric cell containing BASE as the electrolyte and NaCl + ZnCl{sub 2} as the electrodes. (5) Design, construction, and electrochemical evaluation of Na/BASE/ZnCl{sub 2} electrochemical cells. (6) Stability studies in ZnCl{sub 2}, SnCl{sub 2}, and SnI{sub 4} (7) Design, assembly and testing of planar stacks. (8) Investigation of the effect of porous surface layers on BASE on cell resistance. The conventional process for the fabrication of sodium ion conducting beta{double_prime}-alumina involves calcination of {alpha}-alumina + Na{sub 2}CO{sub 3} + LiNO{sub 3} at 1250 C, followed by sintering powder compacts in sealed containers (platinum or MgO) at {approx}1600 C. The novel vapor phase process involves first sintering a mixture of {alpha}-alumina + yttria-stabilized zirconia (YSZ) into a dense ceramic followed by exposure to soda vapor at {approx}1450 C to convert {alpha}-alumina into beta{double_prime}-alumina. The vapor phase process leads to a high strength BASE, which is also resistant to moisture attack, unlike BASE made by the conventional process. The PI is the lead inventor of the process. Discs and tubes of BASE were fabricated in the present work. In the conventional process, sintering of BASE is accomplished by a transient liquid phase mechanism wherein the liquid phase contains NaAlO{sub 2}. Some NaAlO{sub 2} continues to remain at grain boundaries; and is the root cause of its water sensitivity. In the vapor phase process, NaAlO{sub 2} is never formed. Conversion occurs by a coupled transport of Na{sup +} through BASE formed and of O{sup 2-} through YSZ to the reaction front. Transport to the reaction front is described in terms of a chemical diffusion coefficient of Na{sub 2}O. The conversion kinetics as a function of microstructure is under investigation. The mechanism of conversion is described in this report. A number of discs and tubes of BASE have been fabricated by the vapor phase process. The material was investigated by X-ray diffraction (XRD), optical microscopy and scanning electron microscopy (SEM), before and after conversion. Conductivity (which is almost exclusively due to sodium ion transport at the temperatures of interest) was measured. Conductivity was measured using sodium-sodium tests as well as by impedance spectroscopy. Various types of both planar and tubular electrochemical cells were assembled and tested. In some cases the objective was to determine if there was any interaction between the salt and BASE. The interaction of interest was mainly ion exchange (possible replacement of sodium ion by the salt cation). It was noted that Zn{sup 2+} did not replace Na+ over the conditions of interest. For this reason much of the work was conducted with ZnCl{sub 2} as the cathode salt. In the case of Sn-based, Sn{sup 2+} did ion exchange, but Sn{sup 4+} did not. This suggests that Sn{sup 4+} salts are viable candidates. These results and implications are discussed in the report. Cells made with Na as the anode and ZnCl{sub 2} as the cathode were successfully charged/discharged numerous times. The key advantages of the batteries under investigation here over the Na-S batteries are: (1) Steel wool can be used in the cathode compartment unlike Na-S batteries which require expensive graphite. (2) Planar cells can be constructed in addition to tubular, allowing for greater design flexibility and integration with other devices such as planar SOFC. (3) Comparable or higher open circuit voltage (OCV) than the Na-S battery. (4) Wider operating temperature range and higher temperature operation than the Na-S battery. (5) If a cell fails, it fails in the short circuit mode unlike Na-S batteries. Also, cells were successfully subjected to several freeze-thaw cycles. Finally, the feasibility of assembling a planar stack was explored. A two cell stack was assembled and tested. A five cell stack was assembled.