1972 Joumal of the American Ceramic Society-Boakye et al. Vol 87. No. 10 t动m SNexte 720 graphitic carbon Nextel 20 on NexteJTM1720 Fig. 7. TEM images of ZNS-C-derived coatings on Nextel 720 Coatings were applied at 1300.C in argon Coatings were an intimate mechanical mixture of 5 nm t-ZrO2, SiOz, and carbon. In a few places, an overlayer of 20 nm t-ZrO2 particles was present. In some places, the carbon was graphitic. Vanadium from the flux was only occasionally detected at locally high concentrations in particles, which were inferred to be vanadium carbide. Coatings deposited at 1000C on Nextel 720 and heat-treated Grain growth was not measurable in the ZNS-C coated fibers for I h in argon at 1200C showed spatially nonuniform coarsen- after I or 100 h at 1000C in air(Fig. 10). Lognormal Al,O3 ing of the I-ZrO2 particles to >10 nm, with no trace of ZrSio4 grain-size distributions with an inverse logaverage of 63 nm(long formation(Fig 8). In most coatings, the Sio2-ZrO2 dispersion was dimension) was found in both cases, which was identical to that in uniform. However, in isolated locations on some coatings, the as-received fiber. ,Slight grain growth to 89 nm was detected in dispersion was nonuniform on a local scale, with 5 nm 1-ZrO ZP-C-coated fiber after I h at 1000C in air(Table m). Significant articles dispersed over 30-60 nm Sio, spheres after I h at rain growth was evident in the ZNS-C-coated Nextel 720 after 1000C in argon, and 20 nm t-ZrO2 particles dispersed over 100 h at 1200oC, with noticeable faceting of the Al,O3 grains(Fig 100-300 nm SiO, spheres after I h at 1000 C in air(Fig 9). There 11). A lognormal Al O3 grain-size distribution with an inverse was no obvious cause of the sporadic nonuniformity in the logaverage of 112 nm was found. It is tempting to attribute this di grain growth to the V2O3 flux, but the effect of this flux on grain Although heat treatments of vanadium-doped powders formed growth in the Al2O3, Al2O3-mullite, or mullite systems has, to our ZrSio, at 900%C for I h in air heat treatments of 100 h at 1000 knowledge, not been studied. and 1200C in air did not form ZrSio, in ZNS-C-derived fiber (B) Fiber Strength: The as-received Nextel 720 tensile coatings On Nextel 720 fibers, dispersions of 20 nm t-ZrO2 grai strength of 2 GPa was retained after it was coated with the in dense, amorphous SiO, formed after I and 100 h at 1000.C ZNS-C precursor at 1000C in argon, but there was a slight (Fig. 10). ith log mean(nm) of 1.28+ 0.25 and 1.33 0.22 These I-ZrO2 grains had a lognormal particle-siz strength decrease after it was coated at 1200C(Fig. 13 and Table distribution I). Coated-fiber strength was independent of precursor concentra- after I and 100 h, respectively. Little or no coarsening occurred at tion. Strength was not degraded after heat treatment in argon at 1000%-1200C for 1 h. Coated-fiber strength also was retained amorphous SiO, after 100 h at 1200C(Fig. 11). The metastability after 1 h in air at 600 C and I h in argon at 1000C, but, after I h of the I-ZrO, phase for small particle sizes was frequently in air at 1000C, the fiber strength was severely degraded; the fiber observed and has been extensively discussed elsewhere. 75 Vana- had only -10% of the as-received strength. ZP-C-coated Nextel um was not detected in the coating using EDS. Lack of ZrSio4 720 was degraded only slightly after it was coated at 1000oC in formation was attributed to loss of V,Os flux, perhaps to the fibers argon, but was not degraded nearly as much as ZNS-C coated However, vanadium was not detected along fiber grain boundaries Nextel 720 after 1 h in air at 1000 C. However, ZN-C-coated or triple junctions using EDS with analytical TEM. The 2.5 wt% Nextel 720 strength was degraded by almost 50% after it was V2Os flux in a 100 nm thick coating would have been diluted by coated in argon at 1000 C, but it underwent only little further a factor of 30 if it had partitioned equally with the fiber, and it degradation after heat treatment in air at 1000C for I h would not have been detectable unless it concentrated as precipi- Hi-Nicalon fibers coated with the ZNS-C precursor and heat tates. ZP-C-derived coatings(no SiO,) formed porous 15-20 nm treated in air at 1000oC for I h were so weak that they could no I-ZrO, after I h at 1000C in air(Fig. 12) be tested for tow-bundle strength. This contrasted with resultsCoatings deposited at 1000°C on Nextel 720 and heat-treated for 1 h in argon at 1200°C showed spatially nonuniform coarsen￾ing of the t-ZrO2 particles to 10 nm, with no trace of ZrSiO4 formation (Fig. 8). In most coatings, the SiO2–ZrO2 dispersion was uniform. However, in isolated locations on some coatings, the dispersion was nonuniform on a local scale, with 5 nm t-ZrO2 particles dispersed over 30 – 60 nm SiO2 spheres after 1 h at 1000°C in argon, and 20 nm t-ZrO2 particles dispersed over 100 –300 nm SiO2 spheres after 1 h at 1000°C in air (Fig. 9). There was no obvious cause of the sporadic nonuniformity in the dispersion. Although heat treatments of vanadium-doped powders formed ZrSiO4 at 900°C for 1 h in air, heat treatments of 100 h at 1000° and 1200°C in air did not form ZrSiO4 in ZNS-C-derived fiber coatings. On Nextel 720 fibers, dispersions of 20 nm t-ZrO2 grains in dense, amorphous SiO2 formed after 1 and 100 h at 1000°C (Fig. 10). These t-ZrO2 grains had a lognormal particle-size distribution with log mean(nm) of 1.28 0.25 and 1.33 0.22 after 1 and 100 h, respectively. Little or no coarsening occurred at 1000°C. Dispersions of
200 nm m-ZrO2 grains formed in dense, amorphous SiO2 after 100 h at 1200°C (Fig. 11). The metastability of the t-ZrO2 phase for small particle sizes was frequently observed and has been extensively discussed elsewhere.75 Vana￾dium was not detected in the coating using EDS. Lack of ZrSiO4 formation was attributed to loss of V2O5 flux, perhaps to the fibers. However, vanadium was not detected along fiber grain boundaries or triple junctions using EDS with analytical TEM. The 2.5 wt% V2O5 flux in a 100 nm thick coating would have been diluted by a factor of 30 if it had partitioned equally with the fiber, and it would not have been detectable unless it concentrated as precipi￾tates. ZP-C-derived coatings (no SiO2) formed porous 15–20 nm t-ZrO2 after 1 h at 1000°C in air (Fig. 12). Grain growth was not measurable in the ZNS-C coated fibers after 1 or 100 h at 1000°C in air (Fig. 10). Lognormal Al2O3 grain-size distributions with an inverse logaverage of 63 nm (long dimension) was found in both cases, which was identical to that in as-received fiber.33,53 Slight grain growth to 89 nm was detected in ZP-C-coated fiber after 1 h at 1000°C in air (Table II). Significant grain growth was evident in the ZNS-C-coated Nextel 720 after 100 h at 1200°C, with noticeable faceting of the Al2O3 grains (Fig. 11). A lognormal Al2O3 grain-size distribution with an inverse logaverage of 112 nm was found. It is tempting to attribute this grain growth to the V2O3 flux, but the effect of this flux on grain growth in the Al2O3, Al2O3–mullite, or mullite systems has, to our knowledge, not been studied. (B) Fiber Strength: The as-received Nextel 720 tensile strength of
2 GPa was retained after it was coated with the ZNS-C precursor at 1000°C in argon, but there was a slight strength decrease after it was coated at 1200°C (Fig. 13 and Table I). Coated-fiber strength was independent of precursor concentra￾tion. Strength was not degraded after heat treatment in argon at 1000°–1200°C for 1 h. Coated-fiber strength also was retained after 1 h in air at 600°C and 1 h in argon at 1000°C, but, after 1 h in air at 1000°C, the fiber strength was severely degraded; the fiber had only
10% of the as-received strength. ZP-C-coated Nextel 720 was degraded only slightly after it was coated at 1000°C in argon, but was not degraded nearly as much as ZNS-C coated Nextel 720 after 1 h in air at 1000°C. However, ZN-C-coated Nextel 720 strength was degraded by almost 50% after it was coated in argon at 1000°C, but it underwent only little further degradation after heat treatment in air at 1000°C for 1 h. Hi-Nicalon fibers coated with the ZNS-C precursor and heat￾treated in air at 1000°C for 1 h were so weak that they could not be tested for tow-bundle strength. This contrasted with results Fig. 7. TEM images of ZNS-C-derived coatings on NextelTM 720. Coatings were applied at 1300°C in argon. Coatings were an intimate mechanical mixture of 5 nm t-ZrO2, SiO2, and carbon. In a few places, an overlayer of 20 nm t-ZrO2 particles was present. In some places, the carbon was graphitic. Vanadium from the flux was only occasionally detected at locally high concentrations in particles, which were inferred to be vanadium carbide. 1972 Journal of the American Ceramic Society—Boakye et al. Vol. 87, No. 10