In the case of Zr-2.5%Nb pressure tube material, certain tramp elements (Cl, P, and C) have a deleterious effect on the fracture properties. In order to dramatically reduce the amount of these impurities, vacuum arc re-melting is adopted. The effect of the melting practice (double or quadruple) on the fracture properties of this material has been previously studied in detail. However, in these studies, the micro-mechanisms of fracture and the role of the trace impurities on crack initiation and propagation in the sub-critical regime were not investigated in detail. In the present study, the mechanisms operating during the three stages of crack growth, i.e., initiation, propagation and fracture, in the case of low toughness double melted material, are proposed based on a detailed study. In our observation, the tramp elements segregate in the form of fine stringers. In the regions away from the segregation, the material is found to be very ductile with the appearance of ligaments with high local plastic deformation before fracture. The presence of such ligaments on the fracture surface is an indication of fracture occurring in a transition zone (ductile to brittle). In the case of the failure in this transition region, both cleavage and ductile mechanisms can occur in the same specimen. In the transition region, near the upper shelf region, the initiation of the crack occurs by cleavage at a local discontinuity, but the toughness increases rapidly and crack propagation occurs in ductile manner by formation of microvoids. The crack front propagates finding the local discontinuities, leaving behind the unbroken regions with high toughness (ligaments). As the crack propagation continues and the crack face opens, the ligaments left well behind the crack tip rupture after the crack front moves further. The electron micro-beam analysis and X-ray mapping show a build-up of Cl concentration at the stringer sites. In the present analysis, the fracture behavior of double and quadruple melted Zr-2.5%Nb material is compared using a “tearing instability” criterion. Using this approach, an attempt is made to assess the enhanced safety margins, in terms of critical crack length, achieved by modification in the melting practice. First, an estimate from the small specimen J-R data was obtained. However, to obtain a realistic estimate for fracture under burst condition an appropriate scaling factor was applied.
