Molten Silicate Reactivity with Environmental Barrier Coating Materials for Gas Turbine Engine Applications
Open Access
- Author:
- Stokes, Jamesa
- Graduate Program:
- Materials Science and Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- September 12, 2019
- Committee Members:
- Douglas Edward Wolfe, Dissertation Advisor/Co-Advisor
Douglas Edward Wolfe, Committee Chair/Co-Chair
John Richard Hellmann, Jr., Committee Member
John C Mauro, Committee Member
Peter J Heaney, Outside Member
Bryan Harder, Special Member
Valerie Wiesner, Special Member
John C Mauro, Program Head/Chair - Keywords:
- environmental barrier coatings
CMAS
glass
rare earth disilicate
crystallization
EBC
yttrium disilicate
ytterbium disilicate
apatite
melting - Abstract:
- Environmental barrier coatings (EBCs) are employed in gas turbine engines to protect silicon carbide (SiC)-based engine components from deleterious reactions with water vapor in the harsh combustion environment. Yttrium disilicate (YDS) and ytterbium disilicate (YbDS), the current state-of-the-art EBC topcoat materials, degrade when exposed to calcium-magnesium-aluminosilicates (CMAS) caused by the ingestion of dust particles during engine operation. This damage mechanism will be exacerbated as gas turbine engines continue to operate at temperatures much greater than the melting temperature of the ingested particulates. The degradation of EBC materials is largely thermochemically driven, resulting in dissolution of coatings via hot corrosion. After dissolution, various reaction products are precipitated based on the thermodynamic equilibria between the coating and molten deposit. Therefore, novel coating chemistries based on YDS and YbDS were formulated to utilize this thermochemical interaction advantageously to aid in crystallizing molten deposits at high temperature to prevent further degradation. The thermochemical behavior of EBCs was studied primarily through heat treatments of mixtures of EBC material and various CMAS compositions. The products of these interactions were evaluated using X-ray diffraction, scanning electron microscopy, energy dispersive spectroscopy, and electron probe microanalysis. Four quaternary CaO-MgO-Al2O3-SiO2 silicates were investigated, having CaO:SiO2 ratios equal to terrestrial sources of ingested particles relevant to engine operating environments. The silicates exhibited a wide range of melting temperatures from ~1240°C up to ~1500°C, with most of the compositions exhibiting incongruent melting behavior. These CMAS compositions were mixed with YDS and YbDS and heat treated at 1200°C, 1300°C, and 1400°C for 1 hour, 10 hours and 50 hours. Below the full melting temperatures of the CMAS compositions, reactions were kinetically limited due to intrinsic crystallization of CMAS, with YbDS exhibiting little to no interaction with the molten silicates up to 50 hours. Above glass melting temperatures, exposure of the disilicates resulted in full dissolution into the molten glass. Reaction products included reprecipitation of the disilicate phase and a Ca2RE8(SiO4)6O2 apatite-type silicate. However, the formation of the apatite phase was also dependent on cation size and CaO content in the molten deposit. Apatite formation decreased linearly with a decrease in CaO:SiO2 ratio below 0.635 in CMAS, because of the decreased stability of smaller RE cations like Yb3+ or Y3+ in the apatite structure. With high CaO:SiO2 ratio CMAS compositions, YDS formed ~50% more apatite than YbDS, indicating that apatite stability increases with increasing RE cation size. Investigations with additional RE disilicates (RE = Er, Dy, Gd, Nd) with CMAS were carried out at 1400°C. With larger RE cations like Dy3+, Gd3+, and Nd3+, additional stoichiometries of apatite were stabilized (Ca2+yRE8+x(SiO4)6O2+3x/2+y). As the CaO:SiO2 ratio decreased, apatite formation remained consistent for ratios above 0.277, although the apatite precipitates became deficient in CaO and rich in REO1.5 content. Below CaO:SiO2 ratios of 0.277, GdDS reacted with CMAS to form the apatite, and no apatite was observed for DyDS or NdDS. From these studies, it was concluded that GdDS is in an optimum cation size range in which the Gd3+ cation, which is ~2% larger than Dy3+, stabilized apatite at very low CaO:SiO2 ratios. Because Nd3+ is ~5% larger than Gd3+, however, CMAS reactions with NdDS promoted ~20% greater liquid formation in the system due to its larger size. GdDS was incorporated into YDS and YbDS by synthesis of Y2-2xGd2xSi2O7 and Yb2-2xGd2xSi2O7 solutions via solid state reaction method at 1580°C. YDS formed a complete solid solution series with GdDS, with the resulting single phase compounds exhibiting an orthorhombic crystal structure with a very high coefficient of thermal expansion (CTE) (>~8×10-6/K), not recommended for use as EBC materials. In the YbDS-GdDS system, 20 mol% GdDS was soluble in YbDS before the phase transformation occurred, making this material potentially useful as an EBC. Additionally, incorporation of 30 mol% GdDS resulted in a monoclinic single phase compound. Above 30 mol% GdDS, phase segregation was observed between a Yb-rich monoclinic compound and a Gd-rich orthorhombic compound, with results indicating that there is a maximum solubility of ~40 mol% GdDS in YbDS. Baseline YDS and YbDS pellets were tested alongside 5 mol% GdDS, 20 mol% GdDS and 30 mol% GdDS Yb2-2xGd2xSi2O7 samples in diffusion couple experiments with CMAS to determine extent of melt infiltration. Additions of GdDS to YbDS promoted more apatite formation over baseline YbDS and reduced CMAS infiltration on short time scales by ~60%. However, the formed apatite layer was observed to diffuse inwards at longer times at high temperature, in addition to CMAS infiltration occurring along grain boundaries. Thus, these materials were found not suitable for CMAS mitigation in EBCs.