TL;DR: The state-of-the-art environmental barrier coatings (EBCs) for Si-based ceramics consist of three layers: a silicon bond coat, an intermediate mullite (3Al2O3-2SiO2), and a BSAS top coat as discussed by the authors.
Abstract: Current state-of-the-art environmental barrier coatings (EBCs) for Si-based ceramics consist of three layers: a silicon bond coat, an intermediate mullite (3Al2O3-2SiO2) or mullite + BSAS (1-xBaO-xSrO-Al2O3-2SiO2) layer, and a BSAS top coat. Areas of concern for long-term durability are environmental durability, chemical compatibility, silica volatility, phase stability, and thermal conductivity. Variants of this family of EBCs were applied to monolithic SiC and melt infiltrated SiC/SiC composites. Reaction between BSAS and silica results in low melting (approx. 1300 C) glasses at T > 1400 C, which can cause the spallation of the EBC. At temperatures greater than 1400 C, the BSAS top coat also degrades by formation of a porous structure, and it suffers significant recession via silica volatilization in water vapor-containing atmospheres. All of these degradation mechanisms can be EBC life-limiting factors. BSAS undergoes a very sluggish phase transformation (hexagonal celsian to monoclinic celsian), the implications of which are not fully understood at this point. There was evidence of rapid sintering at temperatures as low as 1300 C, as inferred from the sharp increase in thermal conductivity.
TL;DR: In this paper, the activation energy for glass crystallization was calculated to be 259 kJ/mol using a kinetic model using a differential thermal analysis (DTA) from variation of DTA peak maximum temperature with heating rate, and the formation of needle-shaped BaSiO 3 and hexacelsian (BaAl 2 Si 2 O 8 ) were the primary crystalline phases.
TL;DR: In this article, the authors showed that the hexacelsian-celsian transformation does not involve Si-0 and Al-O bond openings, since this would be expected to involve an activation energy of at least 60 kcal mole−1.
Abstract: The hexacelsian-celsian transformation from grains of 1/4 in. (0.635 cm) size is slow and erratic. This is because the rate of heterogeneous nucleation in such grains is low and is influenced by contamination present in the furnace. When the grain size is reduced to a −200 mesh powder, heterogeneous nucleation becomes a dominant factor and the transformation is accelerated. The transformation has three stages: the first is controlled by the rate of crystal growth, the second is controlled by the rates of nucleation and crystal growth, and the third is controlled by the rate of nucleation. The activation energy of crystal growth in the first stage of transformation is 20.1 kcal mole−1 ±20%. This suggests that the hexacelsian-celsian transformation does not involve Si-0 and Al-O bond openings, since this would be expected to involve an activation energy of at least 60 kcal mole−1.
TL;DR: In this paper, the activation energies of two refractory glass compositions in the barium aluminosilicate system were studied by differential thermal analysis (DTA), X-ray diffraction (XRD), and scanning electron microscopy (SEM).
Abstract: Barium aluminosilicate glasses are being investigated as matrix materials in high-temperature ceramic composites for structural applications. Kinetics of crystallization of two refractory glass compositions in the barium aluminosilicate system were studied by differential thermal analysis (DTA), X-ray diffraction (XRD), and scanning electron microscopy (SEM). From variable heating rate DTA, the crystallization activation energies for glass compositions (wt percent) 10BaO-38Al2O3-51SiO2-1MoO3 (glass A) and 39BaO-25Al2O3-35SiO2-1MoO3 (glass B) were determined to be 553 and 558 kJ/mol, respectively. On thermal treatment, the crystalline phases in glasses A and B were identified as mullite (3Al2O3-2SiO2) and hexacelsian (BaO-Al2O3-2SiO2), respectively. Hexacelsian is a high-temperature polymorph which is metastable below 1590 C. It undergoes structural transformation into the orthorhombic form at approximately 300 C accompanied by a large volume change which is undesirable for structural applications. A process needs to be developed where stable monoclinic celsian, rather than hexacelsian, precipitates out as the crystal phase in glass B.