Measurement and prediction of phase transformation kinetics in a nuclear steel during rapid thermal cycles

G. Obasi, E. J. Pickering, A. N. Vasileiou, Y. L. Sun, D. Rathod, M. Preuss, J. A. Francis, M. C. Smith

Research output: Contribution to journalArticleResearchpeer-review

11 Citations (Scopus)

Abstract

Accurate prediction of the residual stress distributions in steel welds can only be achieved if consideration is given to solid-state phase transformation behavior. In this work, we assess the ability of a model for reaction kinetics to predict the phase transformations, and corresponding evolution of volumetric strain, in a nuclear pressure vessel steel when subjected to rapid weld-like thermal cycles. The cases under consideration involved the rapid heating of SA508 steel to a temperature of either 900 °C or 1200 °C for a period of 10 seconds, and subsequent cooling of the material to room temperature at rates between 0.1 and 100 °C s −1 . Predictions for the microconstituent proportions and transformation temperatures for each thermal cycle are compared to those measured through a combination of dilatometry, optical and electron microscopy, and synchrotron X-ray diffraction. In general, there was good agreement between measured and predicted transformation start temperatures and microconstituent fractions for cooling rates relevant to welding (≥ 10 °C s −1 ). Even in the cases in which discrepancies were found for start temperatures, examination of the corresponding dilatation curves showed a good match between predicted and experimental transformation strain evolution. This is a very positive result in terms of residual stress prediction in welds. At slower cooling rates, significant discrepancies arose owing to the model’s incapacity to predict Widmanstätten ferrite or retained austenite, and its failure to account for the effects of carbon redistribution during transformations involving diffusion. Although not relevant to welding, improvements to the model to rectify these issues would be beneficial in terms of its wider predictive capabilities.

Original languageEnglish
Pages (from-to)1715-1731
Number of pages17
JournalMetallurgical and Materials Transactions A: Physical Metallurgy and Materials Science
Volume50
Issue number4
DOIs
Publication statusPublished - 15 Apr 2019
Externally publishedYes

Cite this