Int. Journ. of Fracture 19 (1982) 67-80. 00376-94291821010067-14500.20/0 © 1982 Martinus Nijhoff Publishers, The Hague. Printed in The Netherlands 67 Model for fracture toughness alteration due to cyclic loading I. ROMAN* and KANJI ONO Materials Department, School o[ Engineering and Applied Science, University o[ Cali[ornia, Los Angeles CA 90024, USA (Received July 1, 1980) (The authors wish to dedicate this paper in memory of the late Professor Alan S. Tetelman.) ABSTRACT A new model that is capable of predicting and explaining the effect of cyclic loading on the apparent fracture toughness of materials was developed. The model combines macroscopic fracture criteria with the assumption that transient flow properties of material in the cyclic plastic zone can be simulated by those of macroscopic low cycle fatigue specimens, tested in reversed strain control. Little or no changes in the cleavage fracture toughness due to cyclic loading is predicted or observed for materials that cycle strain harden (e.g., rail steel) and in the fracture toughness of other materials that cycle strain harden (e.g., the commercial 2000 series A1-Cu alloys) and fracture by rupture. However, an increase in the fracture toughness is predicted and observed for materials that cycle strain soften (e.g., ICr-Mo-V and 18 Ni 300 maraging steels), irrespective of fracture mode (cleavage or rupture). The changes in the fracture toughness are predicted and observed to increase with both the number of cycles of applied load and the reversed plastic strain range (or stress intensity range for precracked specimens). I. Introduction Experimental evidence in the literature, e.g. [1-5], has clearly demonstrated that cyclic loading of certain materials affects, to different extents, their apparent fracture toughness. Effects of cyclic loading on the apparent fracture toughness are manifested in two types of observations. First, it has been noticed that for several materials, e.g. 1Cr-Mo-V rotor steel [1], low alloy steels [2] and 7080--T7 aluminum [5], stable fatigue crack growth continued to occur even when the maximum stress intensity in fatigue (Kmax)f exceeded K~c. Unstable crack propagation ensued when (Kmax)l ~ KIC. The other observation, that is consistent with the former one, is the elevation in the apparent fracture toughness, K o, measured for specimens of different materials, e.g. 1Cr-Mo-V rotor steel [1], 7075-T65, aluminum and 18 Ni 300 grade maraging steel [4], that were precracked in constant load control at different load range levels. In contrast, other studies, e.g. [5-7], have shown that similar cyclic loading histories produced no measurable change of the fracture toughness of different materials. Such materials as a variety of aluminum alloys [5], rail steel [6] and a high strength Ni-Cr-Mo-V steel [7] belong to this second group. The available explanations [2, 3, 6] for the phenomena are not satisfactory since each one is partial and refers only to a specific material. Inadequacies of these models have been discussed elsewhere [1]. Since a systematic treatment of the observations is not available, a new model has been developed that is capable of predicting and * Present Address: Materials Science Division, Graduate School of Applied Science and Technology, The Hebrew University, Jerusalem, Israel.