Estimating elastic and thermal contributions to lattice strains from operando X-ray diffraction measurements using fast simulations

Abstract

Lattice strains obtained from operando synchrotron X-ray diffraction measurements during metal additive manufacturing are being increasingly used to estimate temperature evolution during the process. At the minimum, these transient lattice strains have contributions from thermal and elastic strains. Temperature estimates from lattice strains have thus far been extracted assuming that elastic strains are negligible in comparison to thermal strains at high temperatures when the heat source is close to the probed region. However, such an assumption may not only lead to inaccuracies in estimating temperature but also fail to correctly estimate the non-negligible stress evolution occurring at moderate to low temperatures as the heat source moves away. Numerical simulations can be used to predict lattice strains but these predictions are necessarily different from experimental measures.

This work proposes an experimentally corrected numerical approach to improve simulation predictions. It involves first using a recently developed fast numerical thermomechanics model to predict lattice strains. Then, the predicted thermal and elastic strains are corrected using a minimization procedure under the strict constraint that the predicted lattice strains are strictly equal to the measured ones, thus improving the original estimates. This strategy is demonstrated for operando synchrotron X-ray diffraction measurements during directed energy deposition of a thin wall made from 316L stainless steel, which exhibits negligible solid-state phase transformations. Following validation, the corrected thermal and elastic strains are used to estimate temperature and stress evolution and study the difference in temperature and heating/cooling rate prediction caused by neglecting elastic strains.