Melt infiltration studies of 2D Tyranno SA3 @ ceramic matrix composite preforms with CrSi 2 intermetallic alloy

• Published in: Journal of the American Ceramic Society 2021-07, Vol.104 (7), p.2966-2980
• Main Author: Raj, Sai
• Format: Article
• Language: English
• ISSN: 0002-7820; 1551-2916
• DOI: 10.1111/jace.17722

This paper reports the results of melt infiltrating 2D Tyranno SA3 @ fiber woven preforms with molten CrSi 2 between 1768 and 1896 K under a vacuum of 1.3 x 10 −4  Pa (10 −6  torr). The infiltration times varied between 1800 and 7200 s, which did not have any significant effect on the volume fraction of voids filled with the metal. Optical and scanning electron microscopy, backscattered electron imaging, energy dispersion spectroscopy, and Raman spectroscopy were used to characterize the infiltrated preforms. A plot of the volume fraction of open voids infiltrated against the absolute melt infiltration temperature showed a sharp peak at 1773 K with almost complete infiltration of the voids at this temperature. The extent of silicide infiltration of the preforms dropped steeply above 1773 K with an increasing melt infiltration temperature irrespective of the amount of infiltration time. Above 1805 K, the volume fraction of voids infiltrated with the melt was nearly 0%. It is demonstrated that CrSi 2 did not show evidence of a reaction with the SiC fibers. The possibility that a resisting force due to contact angle hysteresis (CAH) may have influenced the diminishing amount of voids filled with increasing temperature above 1773 K was examined. The resistance force was estimated to be extremely small to be consequential. Another possibility that the CrSi 2 may have decomposed into Cr(g) and Si(g) in the vacuum melt infiltration furnace appeared to be more plausible based on thermodynamic analyses. An empirical equation is proposed to calculate the amount of remaining charge left to infiltrate the preforms at the infiltration temperature. It is shown that an initial charge of 10 g would rationalize the present observations. While the decomposition of the CrSi 2 appeared to mostly explain the present results, some discrepancies were observed, which were inconsistent with the decomposition model.