LAUSR

Abstract Experimental techniques for measuring the crack length and crack closure behavior of single edge crack tension [SE(T)] specimens with fixed ends were developed with compliance method employing a high-resolution laser displacement meter together with data processing for noise elimination. Then, an automatic force-shedding fatigue crack propagation test system was developed. Using this system, we measured the crack length and crack closure behavior of thin films with a thickness of 5 or 10 µm and width of 10 mm, with which the accuracy of crack length measurement was better than 0.46 mm. The crack closure behavior could also be identified by the compliance method. The relationship between the crack propagation rate da/dN and the stress intensity factor range ΔK for nanocrystalline (NC) and ultrafine-grained (UFG) thin film was independent of the force ratio, and the relationship for the 10 µm-thick film was almost identical to that for the 5 µm thick film. No crack closure was observed in the UFG and NC thin films, whereas it occurred in the medium-size-grained (MG) thin film. The da/dN–ΔKeff relationships for thin films were independent of the force ratio and film thickness. In the Paris law region, da/dN for the MG film was higher than that for an MG plate at comparable ΔKeff, showing that crack closure could not account for the thickness effect. However, no difference in ΔKeff,th was observed for both the MG film and the MG plate. In the Paris law region (Region B), no effect of grain size on the da/dN–ΔK relationship was observed for the 10-µm-thick film; however, the crack propagation rate was higher for a film with smaller grain size and thickness from 100 to 180 µm. For NC and UFG films, the crack propagation rate of the 100-µm-thick film was higher than that of the 10-µm-thick film, but it was almost identical for MG films; however, the rate for a thin film was higher than that for a bulk plate. The threshold stress intensity factor range was always lower for a film with smaller grain size.