Fig. 16 — Vickers microhardness across the different weld joints. Fig. 17 — Impact toughness values of weld metals. straint provided by the relatively harder weld metal and HAZ during deformation under tensile loading. In the notched tensile test, however, YS, notch tensile strength, and percentage elongation of welded notched samples differ considerably although all the samples broke in the base metal (Table 8). The stress is concentrated in the weld metal due to notching effect and experienced strain hardening with the initial increase in tensile load, leading to increasing the strength. However, with further increasing the tensile load, the strain hardening effect becomes saturated and this facilitated the transmission of load to the adjacent base metal. As the base metal is incapable of bearing the load, failure takes place at the base metal. Toughness Charpy impact toughness values of all the welded joints are illustrated in Fig. 17. The impact toughness of the base metal is 29.41 J and impact toughness values of welds J1, J2, J3, and J4 are 32.73, 39.6, 36.21, and 34.33 J, respectively. This clearly indicates there is an increase in weld metal toughness values compared to base metal irrespective of shielding gas conditions. However, among the four shielding gas conditions, the weld metal prepared with 5% CO2 (J2) exhibited the highest impact toughness followed by J3, J4, and J1. Nevertheless, variations in toughness values among J1, J3, and J4 are trivial. The weld metal toughness is undoubtedly dependent upon the several factors such as amount of metastable austenite, martensite transformation, carbide precipitation, and grain size. In general, improved toughness of weld metal is due to the presence of higher martensite content together with equivalent grain structure compare to the base metal — Fig. 17. Martensite colonies arrest secondary cleavages and increase total energy absorbed during fracture (Ref. 31). The grain growth of delta ferrite at high temperatures is also restricted by a higher fraction of austenite on the grain boundaries (which ultimately transform to martensite on cooling and under deformation). However, the impact toughness values of the weld metal show relatively complex behavior involving several dependent factors. In the present study, weld metal J1 with pure Ar shielding has a higher amount of stable austenite together with very less carbide precipitation along the grain boundary (Fig. 13A), resulting in relatively lower toughness. This phenomenon only indicates the dominant role played by the grain size and the lath martensite. Weld metal J1 has a comparatively coarse grain structure (Fig. 10) with a lower amount of lath martensite (Fig. 6), which ultimately resulted in a comparatively lower toughness by decreasing the absorbed energy during fracture. In welds J2 and J3, higher martensite formation (Fig. 6C) with fine grain structure (Fig. 10) eventually nullify other factors (i.e., metastable g-phase and carbide precipitation) by hindering secondary cleavages during fracture and thus increases toughness. Interestingly, weld J4, even with higher carbide precipitation (Figs. 11 and 12), shows comparable toughness value with other welds. This is probably due to the fact that higher carbon content in weld J4 (Fig. 5) increases the amount of stable austenite by expanding the g loop and enhances the toughness to some extent by taking priority over the precipitation. In addition, apart from microstructural constituents, inclusion content of the weld metal also manipulates the toughness values. In general, it is accepted that higher inclusion content in weld metal can drastically decrease the toughness values (Ref. 32). Again, inclusion content increases with the increase in oxygen potential (OP) of shielding gas mixtures. Accordingly, pure Ar shielding has zero OP and as the CO2 content increases, OP increases simultaneously. Higher OP obviously increases the inclusion content in the weld metal, and it should decrease the toughness. However, based on the observed toughness, it can be assumed that the inclusion formed in the weld metal is very fine, which was reported to have almost negligible effect on the toughness of the weld metal (Ref. 33). Additionally, SEM fractographs (Fig. 18) show the size and distribution of dimples on the surface of broken Charpy impact specimens. By comparing the fractographs in Fig. 18B and C, it is observed that the 5% and 10% CO2 welds have mainly ductile rupture with very few cleavages, which clearly indicate the enhancement in toughness. Conversely, other welds have extensive distribution of cleavage facets with ductile rupture on the fracture surfaces, which undoubt- WELDING RESEARCH 112-s WELDING JOURNAL / APRIL 2015, VOL. 94
Welding Journal | April 2015
To see the actual publication please follow the link above