Super duplex stainless steels Part 1 Heat affected zone microstructures

Abstract

The development of heat affected zone microstructures in super duplex stainless steels (nominal composition Fe–25·0Cr–3·5Mo–7·5Ni–O·5Cu–0·25N–0·5W wt-%) has been examined on multipass welds, bead-on-plate welds, and simulated weld samples. Three-dimensional heat flow equations, for different plate thicknesses, have been applied to the grain growth kinetics of δ ferrite. The grain growth constants have been rationalised using an equation of the Arrhenius type. The level of agreement between the measured δ ferrite grain size and the predicted values is found to be satisfactory. The effect of heat input on the δ ferrite grain growth has been examined in weld metals and the results indicate that the large δ ferrite grain structure can be eliminated using low levels of heat input. The δ + γ → #x03B4; transformation kinetics and the degree of ferritisation have been studied as functions of the welding variables. The results are discussed in an attempt to quantify a minimum plate thickness and optimum heat input to prevent a high level of ferritisation. Particular emphasis is given to the effect of reheating on the formation of intermetallics, nitrides, and carbides. During the first thermal cycle, the heat input can be sufficient to decompose some of the austenite to a metastable ferritic microstructure. Competition between the reformation of austenite and the nucleation of nitrides and carbides takes place in this supersaturated structure on reheating by subsequent passes. The relative positions of the C curves for formation of austenite and nitrides, and the peak temperature of the second cycle, determine which phases form. The depletion of chromium adjacent to the nitride/δ ferrite interface is examined using a finite differences model and the implications of the results for the localised corrosion resistance are discussed. The compositions of phases have been measured and the experimental values are compared with a thermodynamic analysis which predicts the equilibrium compositions of both the δ and γ phases as a function of temperature and chemical composition. Deviation from equilibrium partitioning is observed owing to non-equilibrium cooling conditions, which result in the formation of configurationally frozen structures. Extensive partitioning is observed for high heat input welds but, conversely, low heat input eliminates the partitioning in the heat affected zone. The results are summarised in an attempt to rationalise the relationship between chemical composition, welding conditions, and heat affected zone microstructures.MST/1598