PSI - Issue 42

Diego F. Mora et al. / Procedia Structural Integrity 42 (2022) 224–235 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

228

5

Table 1. Thermophysical properties (Stumpfrock and Taylor 1999). Property 17 MoV 84 Density [Kg/m 3 ] 7806 Thermal expansion [1/°C] 12.3x10 -6 Thermal conductivity [W/mK] 47.4 Specific heat [J/kg-K] 494

The temperature dependent elastic properties of the material are listed in Table 2. Linear elastic isotropic material is assumed in the FE simulation. This is a conservative assumption but it is known that RPV steel at the end of the lifetime exhibits irradiation induced hardening and loss of ductility (Meslin et al. 2010; Soneda 2014).

Table 2. Temperature dependent Young modulus.

Temperature [°C]

E [MPa] 214000 206900 207600 210100 213200

20 50

100 150

200-700

In order to simulate the highly-irradiated RPV steel at the end of the lifetime, the 17 MoV 8 4 steel was heat treated to increase the DBTT. Thus, the resultant 17 MoV 8 4 mod appears to have a much higher DBTT than the original material and the corresponding Nil-ductility transition reference temperature NDT RT was 253°C after the heat treatment. The fracture toughness of the material can be described according to the ASME model, ( ) ( ) 36.5 22.7exp 0.036 Ic NDT K T RT = + − (1)

The corresponding fracture toughness for arrest is ( ) ( ) 29.45 1.368exp 0.0261 Ia NDT K T RT = + − In Eqs. (1) and (2), T is the temperature of the material. 2.3. FE-model

(2)

The finite element model was developed to determine the temperature distribution in the cylinder and later the stresses. Axisymmetry should be the most convenient approach to perform the simulation in this case since both geometry, loads and initial flaw are axisymmetric. However taking into account that the implemented XFEM in ABAQUS is only available for plane strain (CPE4), plane stress (CPS4) and 3D solid (C3D8), the use of axisymmetric element is not feasible. In order to sort out the problem, cyclic symmetry in order to reduce the FE-model is applied. Thus, the cylinder is reduced to a slice as indicated in Fig. 3(a), which covers one degree of the base circumference. The slice is meshed with 3D solid brick elements (DC3D20 and C3D8 for thermal analysis and mechanical analysis, respectively) as in Fig. 3(b) and refinement is applied in the region near to the crack location. Two refinements consisting of regular mesh were applied (Fig. 3(c)), which will be presented in section 3. Table 3 summarizes history of the fluid temperatures and of the heat transfer coefficients (HTC) applied to the inner surface of the cylinder to simulate the thermal shock produced by the water injection. Fig. 4(a) shows in red the surface on which the heat transfer between the water and the steel occurs. On the other surfaces of the cylinder,