ICF13B

-8- defect. They both promote the propagation of surface crack dramatically. Interestingly, as the interface defect length enlarges, the transformation of the SERR curves from oscillation (e.g. 0.05 L = ) to monotone (e.g. 0.15 L = ) and then back to oscillation (e.g. 0.25 L = ), which, as previous discussed, is induced by the extra deviating force, can also been found in Fig. 4. As discussed above, it is concluded that the interface defect would promote the surface cracking behavior by increasing the SERR around the crack tip. Herein, it is generally accepted that the maximum increment of SERR induced by the interface defect is critical to estimate emanation of surface crack. Fig. 5 shows the maximum normalized SERR increment as a function of interfacial defect length and location. Two most important characteristics of interface defect, the length and the location, have been taken into consideration to investigate the their combined effect on surface crack. As shown in Fig. 5, the maximum normalized SERR increment for a remote interface defect ( 0.15 c = ) is near to zero for a relatively small critical defect length (e.g. when L is less than 0.15), which means the remote and small defect has less potential to affect the surface cracking behavior. However, when interface defect is closer to the surface crack(e.g. 0.05 c = ), the critical defect length below which the defect can be ignored correspondingly decreases (e.g. L is less than 0.05). The most extreme situation is that the interface defect locates under the surface crack, in which the small defect cannot be ignored (as the black line in Fig. 5). In general, the maximum normalized SERR increment for a central interface defect is higher than the counterpart for the offsetting defect excepted the appearance of the extra deviating force for the surface crack, as discussed earlier, induced by asymmetric loading and restraint. The oscillation of the maximum normalized SERR increment curve stands for the extra deviating force. Figure 5. The maximum normalized SERR increment as a function of interfacial defect length and location. 4. Conclusions The effects of interface defect on the surface fracture behavior of thermal barrier coating system (TBCs) are investigated in this work. The results show that the two important characteristic parameters, defect location and defect length, both have a significant influence on the driving force of surface crack. It is concluded a surface crack initiated above an interface defect propagates more easily since the enhancement of surface crack driving force is largest for the defect directly beneath the surface crack. Similarly, the surface crack propagates easier as the defect length increases. In present work, the critical value of the defect offset (below which the effect of an interface defect can be ignored) has been obtained and so does the critical value of defect length. By understanding the effect of the interface detect on the surface crack and using the critical value of the defect offset and defect length, the surface cracking behavior can be partially controlled. It is beneficial for

RkJQdWJsaXNoZXIy MjM0NDE=