The J-integral is widely accepted as a measure of elastic-plastic fracture toughness of engineering alloys. Specimen size and geometry dependence were first noted in fracture toughness measurements using the ASTM E 813 calculation of the deformation J-integral (J_d) by McCabe and Landes [1] in 1983. The modified J-integral (J_m) was introduced by Ernst [2] to attempt to minimize or eliminate size and geometry dependence. Since J_m was introduced, questions have arisen regarding the proper parameter to describe the response of a flawed body to loading. The objective of this research task is to investigate the crack growth and specimen J capacity limitations of J_d and to verify the accuracy and specimen independence of the current J_m formulation. This research is expected to affect decisions concerning the use of small laboratory specimens to predict elastic-plastic crack growth resistance in engineering structures.

The J-R curve tests have been conducted on ½T, 1T, and 2T compact specimens of materials having critical fracture toughness values ranging from J_Ic = 140 to 455 kJ/m². These materials include HSLA-80 steel, A106 steel, 3-Ni steel, and two A533B steels. These tests were conducted in accordance with ASTM E 1152 except that specimen loading was continued until large crack extensions were present, in many cases exceeding 50% of the initial uncracked ligament (b). All specimen data were then analyzed using the equations of ASTM E 1152 for the standard deformation J-integral (J_d) resistance curve. Additional analysis was conducted as well in terms of the modified J-integral (J_m) using equations proposed by Ernst and Landes [3]. This latter quantity was possibly useful to larger crack extensions. The test procedure was directed towards finding the useful limits of the two J quantities for the materials listed above or for materials of equivalent strength and toughness.