S Nogami et aL Fusion Engineering and Design 83(2008)1490-1494 93 Si-O Si-C Si-O2 Si-C D=400 D=0 105100 Binding Energy [e\ Fig. 6. The Si-2p XPS spectra for the(a) sic-matrix and (b) sic-fiber of the Pyc-sicsiC composite after oxidation at 1200C as a function of the depth from the oxidized surface(D). peaks with the biding energy of about 107ev for the Sic-matrix 3.3. Oxidation mechanism of Sic/Sic composite and 105 eV for the Sic-fiber were clearly observed near the speci- men surface(D< 150nm). The binding energy of these peaks was Fig 7 shows the schematic illustration of the oxidation behav- larger than that of Si-Oz(103-104ev [6)). Though the reason of ior of the Pyc-Sic/SiC composite at 1000 C and 1200C[4, 9]. For this difference was not clarified in this work, it might be due to an the oxidation of the Pyc-Sic/Sic at 1000C(see(a) and(b) of interface strain between Sioz layer and Sic layer and some impu- Fig. 7). since the formation rate of Sioz layer on the specimen sur- rity adulterated during oxidation. 7,8. In the region deeper than face is relatively low, the recession of Py c interface layer might about 150 nm from the oxidized surface, the peaks with the biding continue and the interface region was not sealed by Sioz. There- energy of about 101-102ev were observed for the Sic-matrix and fore, the tGa spectrum was considered to show the significant SiC-fiber, which might correspond to Si-C with a binding energy of weight loss up to 20h due to the Pyc interface recession and 100-101 ev [6. It was also clearly observed that the region with almost no change after 20 h due to the very small Sioz forma- 50-180 nm from the oxidized surface(bold line in Fig. 6)con- tion sisted of both SiOz peak and Sic peak. The FWHM(full width half For the oxidation of the Pyc-SiC/SiC at 1200C(see(c)and( maximum)of those peaks was almost the same(2.3 for Sio2 and of Fig. 7). since the formation rate of Sioz layer on the specimen 1.7 for SiC). And these values were similar to the FWHM of the surface is relatively high, the interface region might be sealed by region consisting of only Sio2(D< 150 nm)and the region consist- Oz at the early stage of the oxidation test before the Py interface ing of only SiC(D>180 nm). Therefore, this region( D=150-180 nm) disappear due to its recession. Therefore, the TGa spectrum was might consist of mixture of Sio and Sic. considered to show very small weigh (a) (b) 02 SiC-matrix SiC-matrix 000°C Py C-interface SiC-matrix SiC-fiber SiC-fiber PyC SiC-fiber SiC-matrix SiC-matrix 1200°C PyC-interface SiC-fiber (c) (d) Fig. 7. Schematic illustration of oxidation behavior of Pyc-SiC/SiC composite at 1000'C and 1200"C.S. Nogami et al. / Fusion Engineering and Design 83 (2008) 1490–1494 1493 Fig. 6. The Si-2p XPS spectra for the (a) SiC-matrix and (b) SiC-fiber of the PyC-SiC/SiC composite after oxidation at 1200 ◦C as a function of the depth from the oxidized surface (D). peaks with the biding energy of about 107 eV for the SiC-matrix and 105 eV for the SiC-fiber were clearly observed near the speci￾men surface (D < 150 nm). The binding energy of these peaks was larger than that of Si–O2 (103–104 eV [6]). Though the reason of this difference was not clarified in this work, it might be due to an interface strain between SiO2 layer and SiC layer and some impu￾rity adulterated during oxidation. [7,8]. In the region deeper than about 150 nm from the oxidized surface, the peaks with the biding energy of about 101–102 eV were observed for the SiC-matrix and SiC-fiber, which might correspond to Si–C with a binding energy of 100–101 eV [6]. It was also clearly observed that the region with 150–180 nm from the oxidized surface (bold line in Fig. 6) con￾sisted of both SiO2 peak and SiC peak. The FWHM (full width half maximum) of those peaks was almost the same (2.3 for SiO2 and 1.7 for SiC). And these values were similar to the FWHM of the region consisting of only SiO2 (D < 150 nm) and the region consist￾ing of only SiC (D > 180 nm). Therefore, this region (D = 150–180 nm) might consist of mixture of SiO2 and SiC. 3.3. Oxidation mechanism of SiC/SiC composite Fig. 7 shows the schematic illustration of the oxidation behav￾ior of the PyC-SiC/SiC composite at 1000 ◦C and 1200 ◦C [4,9]. For the oxidation of the PyC-SiC/SiC at 1000 ◦C (see (a) and (b) of Fig. 7), since the formation rate of SiO2 layer on the specimen sur￾face is relatively low, the recession of PyC interface layer might continue and the interface region was not sealed by SiO2. There￾fore, the TGA spectrum was considered to show the significant weight loss up to 20 h due to the PyC interface recession and almost no change after 20 h due to the very small SiO2 forma￾tion. For the oxidation of the PyC-SiC/SiC at 1200 ◦C (see (c) and (d) of Fig. 7), since the formation rate of SiO2 layer on the specimen surface is relatively high, the interface region might be sealed by SiO2 at the early stage of the oxidation test before the PyC interface disappear due to its recession. Therefore, the TGA spectrum was considered to show very small weight change. Fig. 7. Schematic illustration of oxidation behavior of PyC-SiC/SiC composite at 1000 ◦C and 1200 ◦C