Materials Research Activities

Beta-alumina

Beta-alumina

b-alumina is an ordinary ternary oxide first identified as an allotropic form of alumina in 1916 in the Bayer process of preparation of alumina . It's an inexpensive material largely available in its form 11 Al2O3 - x Na2O, with x between 1.25 and 1.4 (later on, chemists have synthesized the material with x over the whole range 1.0 - 1.6) .The main characteristics of b-alumina's structure were determined by means of X-ray diffraction in the 1930s [Bragg W.L., Gottfried C., West J. Z. Kristallographie 77, (1931)  255; Beevers C.A., Ross  M.A S, Z. Kristallographie, 97,  (1937)59]. The salient feature of this structure is its two-dimensional character. The crystal is made of parallel thin layers of dense alumina separated from each other by scarcely occupied planes where all sodium ions are confined. 

Up to 1967, b-alumina was of scientific interest only for a few cystallogrophers and  ceramists. As it has very good refractory properties, it was used for the construction of industrial furnaces.The identification and characterization of a new form of b-alumina, named b''-alumina, in a laboratory of solid state chemistry (- Jeanine Théry, Daniel Briançon,  "Sur les propriétés d'un nouvel aluminate de sodium NaAl5O8", Compt.Rend.Acad.Sc. 254 (1962) 2782-2784;  "Structure et propriétés des aluminates de sodium" Rev. Hautes Tempér. et Réfract. 1 (1964) 221-227) were hardly noticed. Suddenly in the 1970s, b-alumina became a star material, intensely investigated by hundreds of chemists and physicists around the world. This modest material inspired the creation of a new scientific sub-discipline named “superionic conductors” in 1972 by W.L. Roth and later renamed “Solid State Ionics”.

 

The new area of research emerged from industrial research in the Ford Motor Company at Dearborn, Michigan where three researchers Yung-Fang YuYao, J. T. Kummer and Neill Weber “reinvented” b-alumina. Two articles revealed that b-alumina's structural peculiarities allowed a high ionic conductivity and that consequently b-alumina could be used in some applications as a substitute to traditional liquid electrolyte. In addition to a high ionic conductivity, b-alumina presented other attractive properties such as small electronic transfer, a mandatory characteristic for a substance to be used as an electrolyte, as well as chemical passivity. Moreover, it could be easily shaped (Yao Y.F.Y, and Kummer, J.T. “Ion Exchange properties of and rates of ionic diffusion in Beta-Alumina” J. Inorg Nucl. Chem29 (1967) 2453.-2475.Weber N., Kummer J.T. ,“Sodium-Sulfur Secondary batteries“, Proceedings of the 21st Annual. Power Sources Conference . 21 (1967) 37-39). These two publications sparkled a booming activity of research all around the world, in the early 1970s when the oil crisis raised the concern with energy.

Two distinctive features characterize the science of b-alumina.

First, there was an immediate shift from the industrial impetus to theoretical issues. Between 1967 and 1970 most of the research on b-alumina was devoted toward a sodium-sulfur cell for load-leveling and electrical vehicles. Then b-alumina, together with another famous solid electrolyte, silver iodide (AgI), known for a much longer time, but having less interesting perspectives of applications, became a model-material for developing methods of investigation and understanding the physical laws of high ionic conductivity. In the series of 8 conferences on “Fast Ion Transport in Solids” starting in Belgirate 1972 and ending in Belgirate in 1992, one can see that in the theoretical section most of the papers dealt with b-alumina in the 1970s. Practical problems lead to fundamental research. It soon became clear that two characteristics of the material, namely non-stoichiometry and structural disorder, play predominant roles in the explanation of its exceptional ion transport properties. Non-stoichiometry is characterized by the fact that the material composition is not set to a unique value, as can be seen from the wide range of values that the x parameter can take in the formula 11 Al2O3 – x Na2O. In 1976, William L. Roth proposed a new structural model that accounted for non-stoichiometry. Basically, he showed that many cristallographic sites are available for sodium ions, in contrast with former descriptions which had identified only one possible site for each sodium ion. In other words, each sodium "has the choice" between various sites, and, when located on a given site, it "sees" several empty sites in its neighborhood. This was an important step, but it remained to understand why jumping to an empty site requires such a small amount of thermal energy that it can occur at room temperature. Two years later, Colomban and coworkers, the first chemists to accomplish the synthesis of the stoichiometric compound (i.e. with x equal to 1), showed that this material has poor ionic transport properties. This crucial experimental result gave strong support to the idea of cooperative jumps involving two or more sodium ions simultaneously, which was instigated by Whittingham and Huggins in 1971 and developed by Wang and co-workers  in 1975. Put simply, non-stoichiometry allows for the formation of sodium ion pairs or small clusters that move as a whole. The energy per ion required for an elementary process is thus smaller than if each individual ions had to move for its own. The key principles of fast ion transport in b-alumina were thus settled by the end of the 70s, but a lot of work remained to be done  in order to get a comprehensive view of the detailed structural and physical properties. In the early 1980s when the journal Solid State Ionics was created there were as many research activities conducted in industrial laboratories as in academic settings.

The second distinctive feature of b-alumina research was the collaboration between physicists and chemists. Physicists were crucial for the in-depth characterization of the b-alumina structure and for the understanding of its dynamics. The problems that solid state physicists had to tackle with were new, because they were trained to study the dynamics of electrons more than those of ions. The role played by b-alumina and other superionic conductors in the development of new experimental methods and techniques is very important for the history of solid state science. Most of the nuclear magnetic resonance (NMR) and relaxation techniques, of the methods of X-ray and neutron diffraction and diffusion, and of the instruments able to measure the electrical response of a solid in a very wide frequency range from 10-6 to 1015 Hz have been designed for research on superionic conductors. They are now widely used in various sectors of materials science. Chemists were crucial to grow the pure and large single crystals of b-alumina needed for the close investigation of their structure. Two chemists were recognized as experts in this art : Gregory C. Farrington from the General Electric Laboratory (then at Penn University) and Philippe Colomban in the R. Collongues's Solid State Chemistry Laboratory in Paris. By contrast to a number of research fields in materials science, the community of b-alumina researchers presented a good balance between U.S American and West European laboratories, while Japan played a minor role in this domain. Chemists from many countries searched inspiration in the peculiar structure of b-alumina to invent new materials with better performance. Industrial chemists were in charge of sintering and shaping by injection molding or other techniques.

After a peak of publications in the mid 1980s the interest in b-alumina declined for several reasons. A number of industrial companies reduced their research investments while the concern for energy storage no longer appeared as a priority. Indeed, other applications were suggested for b-alumina such as in non-linear optics. But, the material did not bring really new insights in this field. Moreover, from a theoretical point of view, the intense exploratory activity of the previous decade came to an end and only minor advances could be expected. Physicists retrieved from the field while chemists went hunting for other families of ionic solid conductors. Materials from the NASICON family raised a lot of attention in the late 1980s, but they proved less rewarding as b-aluminas.

Finally, b-alumina may not have the great commercial future that was expected in the 1970s. It is a model material in itself but its association with electrode materials in a system raised many difficulties. In the mid-seventies sodium-sulfur and lithium iron sulphide batteries were enjoying world-wide attention. In spite of interfacial problems between the electrolyte and the sulphur electrode, the system has been continuously improved over the past decades. Pilot systems have been developed and tested by Chloride Silent Power Ltd in a joint venture with RWE in England in order to equip electrical vehicles and by ABB Hochenergiebatterie GmbH in Germany (I.W. Jones, S.N. Heavens and F.M. Stackpool “Current trends in sodium-sulphur battery development at CSPL” and W. Fischer “Status and prospects of High Energy Na-S Batteries at ABB in Beta-aluminas and Beta-Batteries (Key Engineering materials vols 59-60 (1991) pp. 305-314 and 315-326). A promising alternative is the sodium-chloride battery, the Zebra battery developed by AEG Anglo-American batteries (Johan Coetzel and Jim Sudworth “Out of Africa, The Story of the Zebra battery”, reference missing). Whatever its commercial applications, b-alumina proved to be a main pillar in the construction of materials science.

This page was written by Hervé Arribart and Bernadette Bensaude-Vincent and last updated on 16-Feb-2001 by Arne Hessenbruch.