Revised phase stability diagram of rare earth sesquioxides

Abstract

Below 2000°C rare earth sesquioxides (RESOX) have three crystal structures: hexagonal, cubic and monoclinic, designated as A, C and B respectively [1-3]. Early studies, based on low temperature (LT) synthesis, suggested that RESOX phase stability versus temperature is a function of the metallic ion radii (MIR). La2 O3 Ce2 O3 and Nd2 O3 with the highest MIR are A-type, while for Sm2 O3 , Eu2 O3 and Gd2 O3 with intermediate MIR the structure is C-type at LT and B-type at high temperature (HT) [1-3]. All other RESOX including Y2 O3 and Sc2 O3 were assumed to be cubic (C-type) at all temperatures below 2000°C. The transformation from LT cubic to high temperature (HT) monoclinic structure in Sm2 O3 , Eu2 O3 and Gd2 O3 is unusual and therefore Brauer [4] and Yokogawa et al. [5] suggested that the stable phase is monoclinic at all temperatures below 2000°C. To resolve the controversy, we have demonstrated that slowing down grain growth of Sm2 O3 and Gd2 O3 [9] prevented transition from C to B-types in the expected temperatures (1100 and 1300°C respectively). Hence, we suggest that the surface energy plays an important role in determining the structure of nanomaterials [6,7]. The monoclinic Sm2 O3 , Eu2 O3 and Gd2 O3 is the stable structure at all temperatures below 2000 °C when the grain size is large in the nanoscale. However, for smaller nanocrystals the stable structure is cubic since it has a lower surface energy than the monoclinic phase. In addition, Kimmel et al. [9] suggested that for all RESOX with MIR lower than Gd3+ (except Sc2 O3 ) obtained by HT synthesis [10-17] or under high pressure [18-20] the monoclinic phase is the stable phase also at LT. Figure 1 shows the transition from LT monoclinic to the HT cubic phase according to Sato et al. [17]. Figue 2 shows the suggested RESOX stability diagram as function of temperature versus MIR. In sol-gel production the formation of C-type structures is due to the formation of nano-crystals. Subsequent firing at high temperatures yields the HT cubic phase. Thus. the assumption of a continuous cubic structure at all temperatures is wrong. As seen in Figure 1, in Sc2 O3 the monoclinic to cubic transition is below room temperature, in agreement with the fact that HT synthesis yields cubic Sc2 O3 [17]. (The ion radius of Sc3+ is 0.087 nm [21,22])