No description available.

A series of model steels containing copper levels up to 0.48wt%, nickel up to 0.22wt% and silicon levels of 0.52wt% were oxidised in air at 1050 and 1150??C, and in a CO2-N2 mixture at 1250??C for times of up to 3 hours. The scaling kinetics were measured and the behaviour of copper-rich phase formation at the scale/metal interface was investigated. When oxidised at 1050/1150??C, significant quantities of copper-rich phase were observed for most model steels. The relatively high oxidation rate under these conditions led to the rapid development of a copper-rich layer with little copper diffusing into the metal. However, when oxidised at 1250??C, the copper-rich phase did not form for a significant amount of time; and for some model steels, not at all. This was attributed to the considerably lower oxidation rate and the fact that more copper was found to have diffused into the metal. Alloying additions of nickel and silicon were found to be beneficial in reducing the amount of copper-rich phase measured at the scale/metal interface under the conditions investigated at 1150??C and 1250??C. This occurred because nickel and silicon addition promoted the occlusion of copper-rich phase into the scale. Copper enrichment during oxidation was modelled using a numerical description of the diffusion processes involved. Predictions of the time for commencement of copper-rich phase formation at 1250??C were in close agreement with observation. Agreement between predicted and observed copper-rich layer thickness was less successful under conditions where occlusion was significant, and the measured thickness varied non-uniformly with time. The cracking susceptibility of the model steels was examined using a hot compression test. Oxidation was performed in air at 1050, 1150 and 1250??C and most specimens were compressed at 1050??C. The amount of cracking was found to increase with the amount of copper-rich phase precipitated at the scale/metal interface during oxidation. In general, nickel addition reduced the amount of cracking at all temperatures; and under some conditions prevented cracking altogether. Silicon reduced or completely suppressed cracking when the subscale formed was liquid. The beneficial effects of nickel and silicon addition were attributed to their effect of promoting copper occlusion.

A series of model steels containing copper levels up to 0.48wt%, nickel up to 0.22wt% and silicon levels of 0.52wt% were oxidised in air at 1050 and 1150??C, and in a CO2-N2 mixture at 1250??C for times of up to 3 hours. The scaling kinetics were measured and the behaviour of copper-rich phase formation at the scale/metal interface was investigated. When oxidised at 1050/1150??C, significant quantities of copper-rich phase were observed for most model steels. The relatively high oxidation rate under these conditions led to the rapid development of a copper-rich layer with little copper diffusing into the metal. However, when oxidised at 1250??C, the copper-rich phase did not form for a significant amount of time; and for some model steels, not at all. This was attributed to the considerably lower oxidation rate and the fact that more copper was found to have diffused into the metal. Alloying additions of nickel and silicon were found to be beneficial in reducing the amount of copper-rich phase measured at the scale/metal interface under the conditions investigated at 1150??C and 1250??C. This occurred because nickel and silicon addition promoted the occlusion of copper-rich phase into the scale. Copper enrichment during oxidation was modelled using a numerical description of the diffusion processes involved. Predictions of the time for commencement of copper-rich phase formation at 1250??C were in close agreement with observation. Agreement between predicted and observed copper-rich layer thickness was less successful under conditions where occlusion was significant, and the measured thickness varied non-uniformly with time. The cracking susceptibility of the model steels was examined using a hot compression test. Oxidation was performed in air at 1050, 1150 and 1250??C and most specimens were compressed at 1050??C. The amount of cracking was found to increase with the amount of copper-rich phase precipitated at the scale/metal interface during oxidation. In general, nickel addition reduced the amount of cracking at all temperatures; and under some conditions prevented cracking altogether. Silicon reduced or completely suppressed cracking when the subscale formed was liquid. The beneficial effects of nickel and silicon addition were attributed to their effect of promoting copper occlusion.

The effects of hydrogen on the fracture behavior of manual SMA welds in carbon steel plate was investigated utilizing modified ½T compact tension specimens. Tests performed on these specimens in the presence of hydrogen were compared to similar tests in helium. These tests showed that hydrogen lowers J_C in both the heat affected zone and the base metal. In 350 psi helium, the experimental value of J_C in the heat affected zone (2826 in.-lbs./in.²) was greater than that obtained in the base metal (1650 in.-lbs./in.²). The tests conducted in 350 psi hydrogen resulted in a reduction in J_C for both the heat affected zone (1425 in.-lbs./in.²) and the base metal (59 in.-lbs./in.²). Furthermore, when compared to specimens tested in helium, it was determined that the material tearing modulus for specimens tested in hydrogen was significantly reduced. Slow stable crack growth occurred in all helium tests and in tests performed on the heat affected zone in hydrogen. However, unstable crack growth (fast fracture) was obtained for base metal tests in hydrogen. Fractographic studies revealed that the mechanism for all slow stable crack growth was microvoid coalescence; whereas, the surface of base metal specimens tested in hydrogen showed that fast fracture occurred by cleavage. Optical microscopy revealed that the fracture path for all base metal tests remained in the base metal, but that the fracture path for all heat affected zone tests moved towards the base. These observations, in conjunction with microhardness readings and quantitative metallography, were used to develop explanations for the observed behavior. These explanations include the combined effects of hydrogen, weld defects, residual stresses, grain size, and test variables such as temperature and specimen size and geometry. / M. S.