The goal of this work was to develop engineered matrix SiC/SiC ceramic composites with crack blunting and self-healing capabilities for 1588 to 1755 K applications. The work optimized the temperature and time conditions for melt-infiltrating SiC/SiC preforms with chromium silicide alloys, and established that these alloys do not react with the coatings on the SiC fibers. Traditional ways of fabricating SiC fiber-based ceramic matrix composites (CMCs) use silicon to melt-infiltrate the CMC preforms, where the Si is often converted to SiC by reaction with carbon. The traditional SiC matrices have poor high-temperature creep properties due to the presence of residual silicon. They also have low fracture toughness and a low matrix cracking stress.

Cross-section of a Tyranno SA3 preform.

The engineered SiC matrix is being developed as an alternative to the traditional SiC matrix for CMCs. The presence of ductile silicide particles in the matrix would blunt a crack, while the presence of self-healing additives leads to the healing of the crack at the operating temperature. There is a large mismatch in the thermal expansions between the silicides and SiC. Thus, directly melt-infiltrating a silicide alloy into a SiC/SiC preform leads to cracking of silicide. The engineered matrix concept involves replacing some of the SiC in the traditional SiC-based matrix with silicide particles, which will be ductile at high temperature, and compensating for the thermal expansion mismatch with SiC by adding suitable amounts of silicon nitride particles. Additionally, self-healing additives are added to heal the cracks.

The composite fabrication consists of three steps. In Step 1, a high-temperature engineered matrix consisting of SiC, Si3N4, CrMoSi, and self-healing additives is formulated and optimized for a suitable combination of properties. In Step 2, the high-temperature engineered SiC-based matrix containing crack self-healing additives and a high-melting silicide, such as Cr-Mo-Si alloy powders, is introduced into a 2D or 3D woven preform as a slurry. Alternatively, traditional tape casting methods can be used for fabricating tapes, which are then laid up in 0/90-degree configuration several layers thick and hot-pressed into preforms. Thus, the composites can be 1D, 2D, or 3D. Step 3 consists of melt-infiltrating this slurry infiltrated preform or tape cast composite with a Cr-Si alloy, such as CrSi2 and Cr-25%Si alloys. After this melt infiltration, the composite is heat treated to homogenize the microstructure. The present technology disclosure relates to Step 3.

Silicide melt infiltration studies were conducted mainly on Sylramic or Tyranno SA3 0/90 preforms eight-ply thick. The procedures are equally applicable for melt-infiltrating other commercially available SiC preforms. Melt infiltration of the preforms with CrSi2 was done between 1765 and 1900 K for infiltration time up to 2 hours. No engineered matrix slurry was used in these studies. The cross-section of a preform was successfully melt-infiltrated with CrSi2 to fill about 25 vol. % of the porosity with about 2.7 vol. % unfilled voids.

At the interface of the CrSi2 matrix with the CVI SiC coating on the BN-coated SiC fibers, the energy dispersion spectra (EDS) from an area close to the interface confirms that there was no chemical reaction between the CrSi2 and CVI SiC. This was confirmed for all melt infiltration temperature and time conditions. The CrSi2 does not react during melt infiltration of the Tyranno SA3 preforms. One Sylramic preform specimen showed evidence of reaction. The carbon and silicon Raman spectra from six different spots in the CrSi2 matrix showed no detectable free carbon and silicon in the CrSi2 matrix. The melt infiltration of the preforms with the Cr-25%Si alloy was done between 1780 and 1960 K for melt infiltration times between 0.25 and 2 hours. The volume fraction of voids that was infiltrated varied between 3 and 10%. There appeared to be an apparent reaction with the CVI SiC and the BN coatings near the surface where the molten metal entered the preforms, whereas no reaction was observed at a distance from the surface of the coupons. On closer observation, it was revealed that the Cr-25%Si molten metal filled the space where BN coating around the fibers was absent and the CVI SiC and fibers had cracked. Careful transmission electron microscopy confirmed that the silicide did not react with either the BN coatings or the CVI SiC. However, if the silicide enters an uncoated cracked SiC fiber, it is likely to react with carbon-rich amorphous SiC present at the grain boundaries of the crystalline SiC grains. The presence of a BN coating protected the fibers from chemical attack.

This work was done by Sai Raj of Glenn Research Center. NASA is seeking partners to further develop this technology through joint cooperative research and development. For more information about this technology and to explore opportunities, please contact This email address is being protected from spambots. You need JavaScript enabled to view it.. LEW-18964-2