VALIDATION OF A BRITTLE INTERGRANULAR FRA CTURE CRITERION BY BIAXIAL TESTS

Citation
E. Kantidis et al., VALIDATION OF A BRITTLE INTERGRANULAR FRA CTURE CRITERION BY BIAXIAL TESTS, Revue de métallurgie, 92(9), 1995, pp. 1043-1051
Citations number
23
Categorie Soggetti
Metallurgy & Metallurigical Engineering
Journal title
Revue de métallurgie
ISSN journal
00351563 → ACNP
Volume
92
Issue
9
Year of publication
1995
Pages
1043 - 1051
Database
ISI
SICI code
Abstract
The existence of ghost lines in the low alloy steels implies a great s catter in the fracture toughness properties (1-6). These weak zones ar e characterized by the segregations of impurities like P and S which c an lead to intergranular fracture. This mode of fracture arise at the prior austenitic grain boundaries which are the main localization for the P segregation. The influence of loading conditions on intergranula r brittle fracture is investigated in a Mn Ni Mo steel used for fabric ation of pressure vessel. This steel was submitted to a heat treatment simulating temper embrittlement. Monotonic biaxial tests within the t ension to torsion range have been performed at -196 degrees C on thin tubular specimens with a wall thickness of I mm and a gage length of 2 0 mm. This kind of mechanical tests is very interesting since ii allow s to show off the existence in the sigma-tau plane of two domains of f racture. Indeed, although the yield surface is very well described by the Von Mises, from the fracture point of view, in the sigma-tau plane the upper part (near torsion) referred as '' P '' field has to be dis tinguished from the lower one (near tension) referred as '' A '' field . Within the '' P '' field, rupture happened after a large among of pl asticity and is due to the unstable growth of defects. In this part, t he maximum principle stress criterion allows to predict the intergranu lar brittle fracture. Within the '' A '' field, fracture happened imme diately after macroscopic plastic yielding. In this case, plasticity b y initiating microcracks is the main factor for controlling the interg ranular brittle fracture for which a criterion based on critical plast ic strain is proposed. These tests underline the importance of the nec essity of plastic deformation to initiate brittle fracture. Indeed, in the '' A '' domain, the maximum principle stress criterion is lower t han the yield stress but fracture could not happen because no defect e xists and as soon as plasticity has created defects, fracture occurs a s unstability conditions are reached. On the contrary, in the '' P '' domain, defects are created by plasticity and induce fracture when pro pagation threshold is reached. Consequently, on the overall sigma-tau plane, the rupture may only occur if the conditions for initiation (pl astic strain criterion) and propagation (principle maximum stress crit erion) are achieved. The condition which actually controls the fractur e is that the second one to be fulfilled. Furthermore, in the '' P '' domain, fracture does not depend on the loading path because tests con sisting, in a first step, of a pure tension (or resp. a torsion) up to a stress maintained constant to the end followed by a pure torsion (o r resp. a tension) lead to the same value of maximum principle stress to failure. A predeformation of torsional prestrain of 18 % at room te mperature on specimens tested subsequently at -196 degrees C produces an increase in the resistance to intergranular brittle fracture. This effect is likely due to the homogenisation of plastic deformation whic h reduces concentration at grain boundaries.Tension-torsion experiment s can be used to test the directionality aspect of fracture associated with the critical stress concept. Measured angles on failed specimens show that fracture occurs along a facet with a normal aligned with th e direction of the maximum principal stress.