In~omation Springheid Va 22151 NAVAL RESEARCH LABORATORY Waui, hd. d I1 This document has been approved for public release and sale; its distribution is unlimited.

The optical caustic technique offers unique opportunities for experimental measurement of fracture parameters under conditions where conventional instrumentation either can not be used or is not available. The apparent simplicity of the method has attracted the interest of metallurgists and materials scientists. However, there are subtle problems with the method which complicate the interpretation of results. Some of the problems have not been fully explored and others need clarification before the method can be routinely applied. Of particular concern is the fact that existing theories for caustic behavior apply only in special cases which are rarely, if ever, attained in test specimens of practical dimensions for fracture testing. Nonetheless, the potential promise of the method justifies a systematic study of the influence of various parameters, both physical and optical, on the observed behavior of the caustic. In previous work, the authors demonstrated that there is a correlation between the measured caustic diameter and the value of the J-integral parameter obtained from mechanical measurements for planar 1T compact tension specimens in various steels, subject to some restrictions. As might be expected, this correlation did not agree with existing theoretical relations since the theories are all based on plane stress assumptions. In the present paper, these correlations are studied further and the influence of imaging distance and specimen thickness examined in detail. In addition, the characterization of caustic behavior in standard bend specimens is investigated. For these studies, a new four-lens caustic camera was used which permits the simultaneous determination of four caustics with varying image distances at each increment of applied load. With the aid of this camera, sufficient data were obtained to determine empirical relations for the calculation of the J-integral values from optical measurement of the caustic.

A systematic investigation was undertaken to examine the influence of experimental variables on the measurement of stress corrosion cracking (SCC) properties of two high-strength steels using fracture mechanics test procedures. The properties of interest were the SCC threshold stress-intensity (KISCC) and time-based crack growth rate (da/dt). Two quench-and-tempered alloy steels in the form of rolled plate material were tested. The steel had yield strengths of 965 MPa (140 ksi) and 1207 MPa (175 ksi). The primary experimental variables investigated included: (1) specimen type (bolt-loaded wedge-opening loaded (WOL) and cantilever beam); (2) bulk solution chemistry (3.5% NaCl in distilled water and natural seawater); (3) dissolved oxygen content of the bulk solution; (4) electrode potential (open circuit and -1.0 V versus Ag/AgCl); (5) method of applying cathodic protection (potentiostat and zinc couple; (6) precracking stress-intensity level; (7) initial applied stress-intensity level; and (8) test duration. The effects of these experimental variables on KISCC and da/dt measurement are reported and the implications of these findings on test method development are discussed.

No abstract

Because the optical caustic technique offers unique opportunities for experimental measurement of fracture toughness under conditions where ordinary instrumentation cannot be used, an effort to correlate the size of the caustic with fracture parameters has been undertaken. The materials of interest are a 4340 steel in a range of thicknesses designed to allow plane-strain, plane-stress, and mixed-mode constraint conditions; an intermediate-strength 5% nickel steel with a high degree of fracture resistance; and a lower-strength, ductile high-strength low-alloy (HSLA) steel, A710. Using compact tension specimens and standard K and J-integral testing procedures, the caustics were photographed as fracture tests were conducted. Experiments showed that there is a clear correlation between the J-integral and the caustic diameter (D) for all three steels. The values did not conform to expected relationships predictable from existing theory for several reasons. In the 4340 steel, the correlation for the three thickest samples was independent of thickness, while the thinnest (plane-stress) sample was distinctly different when the generating circle was contained inside the plastic zone. Under these conditions, the J-D correlation was distinctly different for the other two steels as well. When the generating circle was maintained outside the plastic zone, a single relationship was obtained for all three steels. This behavior is attributed to differences between the steels in plastic flow behavior and slip systems. As a result, the near-field behavior is strongly influenced by factors that have little or no effect on the far-field behavior.