No abstract

A primary failure mode for environmental barrier coatings (EBCs) on SiC ceramic matrix composites (CMCs) is the oxidation of the intermediate Si‐bond coating, where the formation of SiO2 at the bond coating–EBC interface results in debonding and spallation. This work compares the microstructure evolution and steam oxidation kinetics of the Si‐bond coating beneath yttrium/ytterbium disilicate ((Y/Yb)DS) and ytterbium disilicate/monosilicate (YbDS/YbMS) EBCs to better understand the impact of EBC composition on oxidation kinetics. After 500 1‐h cycles at 1350°C, (Y/Yb)DS displayed a decreasing concentration of the monosilicate minor phase and increasing concentration of porosity as furnace cycling time increased, whereas the YbDS/YbMS EBC displayed negligible microstructural evolution. For both EBC systems, thermally grown oxide growth rates in steam were found to increase by approximately an order magnitude compared to dry air oxidation. The (Y/Yb)DS EBC displayed a reduced steam oxidation rate compared to YbDS/YbMS.

Environmental barrier coatings (EBCs) are a commercially proven means of protecting SiC-based materials in gas turbine environments. However, there are little specific data in the literature on the impact of coatings like Yb 2 Si 2 O 7 on preventing accelerated SiO 2 growth in the presence of H 2 O. Quantification of reduced rates are necessary for evaluating and comparing EBC effectiveness and incorporation of silica growth rates into future EBC lifetime models. In this study, baseline kinetics of silica formation on bare Si and chemically vapor deposited (CVD) SiC in the 1250-1425℃ range were obtained via 100 h isothermal exposures in dry air and steam environments utilizing a SiC reaction tube to mitigate specimen volatility. An Arrhenius plot of the resulting rates was constructed, representing baseline minimum and maximum rates for Si and SiC oxidation at ambient pressure. Various EBC systems on CVD SiC substrates including air plasma sprayed (APS) EBCs with and without a Si bond coating and with surface roughening to enhance Yb 2 Si 2 O 7 adhesion were subjected to 1-h furnace cycle testing in air with 90vol%H 2 O at 1250-1350℃ for up to 500 cycles. After exposure, silica formation rates were measured and compared to the baseline rates to assess EBC effectiveness, where EBC effectiveness is gauged as the propensity to reduce underlying rates of silica formation. With a Si bond coating, a ~180 µm Yb 2 Si 2 O 7 (YbDS) top coating reduced rates over the entire 1250°-1350℃ range. Without a Si bond coating, ~60 µm (YbDS) coatings deposited directly onto CVD SiC exhibited poor adhesion, and had to be deposited onto substrates with enhanced roughness at 1350℃. While exhibiting good adhesion at 1350℃, overall the single layer YbDS coating exhibited a decreasing effectiveness from 1250° to 1350℃.

High‐temperature, high‐velocity water vapor (steam‐jet) exposures were conducted on Y2O3, Y2SiO5, Y2Si2O7, and SiO2 for 60 hours at 1400°C. Volatility of Y2O3 was not observed. Phase‐pure Y2SiO5 exhibited SiO2 loss forming Y2O3 and porosity. A mixed porous and dense Y2SiO5 layer formed on the surface of Y2Si2O7 due to SiO2 depletion. The mechanisms and kinetics of the reaction between SiO2 and H2O(g) to form Si(OH)4(g) from Y2SiO5, Y2Si2O7, and SiO2 are discussed.

Silicon carbide Schottky diodes with thick i-regions are reported. Compared with previously reported p-i-n photodiodes, a shift of the absorption peak from 270 nm to 350 nm was observed. The responsivity curves of the Schottky diode are modeled and compared with the experimental data.

No abstract

public-access-plan).

This milestone was originally envisioned for completion in FY21 and was delayed due to the COVID-19 response and difficulties in fabricating/obtaining fluoride salts. A monometallic type 316H stainless steel thermal convection loop (TCL) was operated for 1000 h with flowing LiF-BeF2 (i.e. FLiBe) salt and a peak temperature of 650°C. In general, the attack was minimal. However, classic mass transfer was not clearly observed with mass losses in both the hot and cold legs. There was an issue with cleaning the loop with water and specimens may have oxidized, which was not an issue with the previous FLiNaK TCL.