In high-carbon steels, the austenite-to-martensite transformation under applied stress initiates at a critical value, contrary to low-carbon steels where the transformation occurs at the early stages of deformation. There are few models that account for the delayed transformation, because previous models were mostly developed from low-carbon steels that exhibit martensite transformation behaviour that is different from high-carbon steels. Also, few studies have been done on methods of tailoring chemical composition and processing to design steels that possess optimal austenite stability, but do not require too many expensive austenite stabilising elements such as nickel.

This research aims to develop equations for modelling the critical stress and progress of deformation-induced martensite transformation in high-carbon steels. Another aim is to design a new high-manganese steel that is comparable to the high-nickel AISI 3310 carburising steel in terms of austenite stability and mechanical properties.

The research investigated carburising-grade steels used in bearing applications. Various thermomechanical tests were conducted to investigate the influence of chemical composition and microstructure on retained austenite stability. Dilatometry experiments showed that the martensite-start temperature decreased with increasing grain size when the steels were austenitised below the A_cm temperature, but increased with grain size above the A_cm temperature. These observations were linked to the amount of carbon dissolved in austenite.

Analytical models that describe the critical stress and progress of martensite transformation in high-carbon steels were developed. The critical stress model is a function of chemical composition, deformation temperature, and initial retained austenite fraction. Higher austenite stabiliser concentrations or deformation temperatures result in smaller magnitudes of the chemical driving force, leading to a higher critical stress predicted and represent higher austenite stability. A lower initial austenite fraction represents fewer martensitic nucleation sites, resulting in deformation-induced martensite transformation that occurs at a higher critical stress. The transformed amounts of austenite under applied stress depends on the critical stress for martensite transformation. The predicted results were in good agreement with experimental data.

Model applications were demonstrated through examples that involve the determination of alloying combinations and service conditions that are suitable for desired range of critical stresses and austenite fractions predicted. The examples highlight the utility of these models as tools for alloy design.

A new high-manganese carburised steel was developed based on thermodynamic modelling and literature-informed design criteria. Samples of the new steel and AISI 3310 steel were carburised and subjected to martensitic quench-and-temper heat treatments. The mechanical properties and austenite mechanical stability of the new steel were found to be not on par with the 3310 steel. The reasons include a lower bulk carbon content, and the presence of massive carbides and oxide inclusions in the new steel. Practical solutions to improve the low-pressure carburisation process of the new steel are suggested in Section 7.5.2.

The possibility of designing new carburising steels with lower costs by tailoring composition and processing routes based on enhancing austenite stability was explored and demonstrated. Process improvement aspects identified from the carburisation and heat treatment of the new steel can be used to inform the processing of such steels in the future.