PSI - Issue 52

A.D. Cummings et al. / Procedia Structural Integrity 52 (2024) 762–784 A. Cummings / Structural Integrity Procedia 00 (2023) 000–000

765

4

and manufacturing process of the materials to increase their impact performance at -40 ◦ C. Similarly, an optimisa tion programme was carried out on carbon-manganese-nickel steel ASTM A350 LF5, used for the Magnox M2 and Excellox 6 package designs, Price (1997). The study highlighted the importance of low carbon and sulphur content in forged bodies. More recently, in 2013, NTS conducted a study on the quality aspects of casting a thick, marten sitic, stainless steel, designated GX3CrNi13-4 1.6982. This included understanding a variation of impact properties throughout the structure and Non Destructive Testing (NDT) detection criteria Grainey et al. (2013). Further work on this material, as yet unpublished, has been carried out to characterise the fracture toughness of the package body. Analytical approaches in fracture mechanics for transport packages have been pioneered by the German competent authority, BAM, see for example Zenker et al. (2001, 2002). In part motivated by recycling of decommissioned nuclear plant steels that would otherwise be scrapped, one study considers recycled ductile cast iron storage containers for transport and storage of intermediate level waste. Zenker et al. (2001) developed 2-D solutions for geometries related to such cuboid transport packages, in particular internal radii between the package side wall(s) and base. They compared their results to dynamically calculated stress intensity factor solutions and found that under particular conditions the dynamically calculated elastic-plastic stresses could be used to obtain crack driving forces in quasi static solutions. This approach provided a means to safely justify stresses greater than half the yield stress, by taking advantage of sophisticated Finite Element Analysis (FEA) techniques. Post 2002, to meet the new regulatory requirements, Zenker et al. (2007) carried out a testing and analytical validation programme on cuboidal, low ductility cast iron containers. The challenge was to prepare a safety case for the new material at -20 ◦ C, with a fracture toughness value varying between 22 - 27 MPa.m 0 . 5 . Fracture toughness and subzero tensile testing at elevated strain rates were performed. Uncracked FEA models were compared favourably to strain gauge and accelerometer measurements taken during drop testing. Komann et al. (2013) presented a detailed assessment methodology in preparation for the licensing of German transport packages made from low-ductility cast iron. Applying the validated modelling approach described by Zenker et al. (2007) to obtain dynamic loading and crack driving force, a submodelling approach was described. Dynamically calculated displacements from appropriate cut-boundaries were used to drive cracked body submodels where the crack size was postulated based upon NDT detection capabilities. The submodels were solved either statically or dynamically with Abaqus / Standard or Abaqus / Explicit Abaqus (2023). In the case of Abaqus / Standard appropriate J D (orK JD ) are obtained directly from the solution output. For dynamic cases a validated, in-house code is required to perform the contour integral evaluation about the crack tip to obtain J D (orK JD ). This approach was recently applied to a new transport package design required to have a low service temperature of -40 ◦ C Komann et al. (2019). A more generalised elastic-plastic assessment approach is provided in the EPRI (1981) assessment methodology. The EPRI (1981) approach aimed to provide a similar assessment method to that established for Linear Elastic Fracture Mechanics (LEFM) for Elastic Plastic Fracture Mechanics (EPFM). The original handbook was limited to 2-D cases and during time its maintenance was not upkept. Recently Anderson et al. (2023) have planned an update to the EPRI handbook. The EPRI (1981) method was modified through the use of the reference stress method Ainsworth (1984) which has been implemented in most modern standards BS7910 (2019); ASME FFS-1 / API579-1 (2016); R6 (2001). One of the challenges relevant to dynamic fracture is the influence of the time to fracture on both the determination of stress intensity and the fracture toughness. Meyers (1994) presented arguments, without demonstration, on why very short stress pulses, of the order of microseconds, may increase the dynamic fracture toughness. Conversely BS7910 (2019) decreases fracture toughness to account for dynamic loading. This is in-keeping with research studies on high strength alloyed steels, which indicate dynamic fracture initiation decreases with shorter stress pulses Foster et al. (2008); Xu et al. (2011) - this may not be true for all materials. Zenker et al. (2001); Komann et al. (2013) discuss the influence of the time to fracture and inertial e ff ects on numerically calculated K (or J) values. They suggest that stress pulses over milliseconds for thick walled transport packages are likely to be quasi-static. They also warn that when submodelling, artificial contamination of dynamically calculated values is possible with poor cut boundary choices or insu ffi cient temporal resolution of cut boundary displacements.