Texture-based residual stress analysis of laser powder bed fused Inconel 718 parts.

Although layer-based additive manufacturing methods such as laser powder bed fusion (PBF-LB) offer an immense geometrical freedom in design, they are typically subject to a build-up of internal stress (i.e. thermal stress) during manufacturing. As a consequence, significant residual stress (RS) is retained in the final part as a footprint of these internal stresses. Furthermore, localized melting and solidification inherently induce columnar-type grain growth accompanied by crystallographic texture. Although diffraction-based methods are commonly used to determine the RS distribution in PBF-LB parts, such features pose metrological challenges in their application. In theory, preferred grain orientation invalidates the hypothesis of isotropic material behavior underlying the common methods to determine RS. In this work, more refined methods are employed to determine RS in PBF-LB/M/IN718 prisms, based on crystallographic texture data. In fact, the employment of direction-dependent elastic constants (i.e. stress factors) for the calculation of RS results in insignificant differences from conventional approaches based on the hypothesis of isotropic mechanical properties. It can be concluded that this result is directly linked to the fact that the {311} lattice planes typically used for RS analysis in nickel-based alloys have high multiplicity and less strong texture intensities compared with other lattice planes. It is also found that the length of the laser scan vectors determines the surface RS distribution in prisms prior to their removal from the baseplate. On removal from the baseplate the surface RS considerably relaxes and/or redistributes; a combination of the geometry and the scanning strategy dictates the sub-surface RS distribution.

Reconstruction of incomplete X-ray diffraction pole figures of oligocrystalline materials using deep learning.

X-ray diffraction crystallography allows non-destructive examination of crystal structures. Furthermore, it has low requirements regarding surface preparation, especially compared to electron backscatter diffraction. However, up to now, X-ray diffraction has been highly time-consuming in standard laboratory conditions since intensities on multiple lattice planes have to be recorded by rotating and tilting. Furthermore, examining oligocrystalline materials is challenging due to the limited number of diffraction spots. Moreover, commonly used evaluation methods for crystallographic orientation analysis need multiple lattice planes for a reliable pole figure reconstruction. In this article, we propose a deep-learning-based method for oligocrystalline specimens, i.e., specimens with up to three grains of arbitrary crystal orientations. Our approach allows faster experimentation due to accurate reconstructions of pole figure regions, which we did not probe experimentally. In contrast to other methods, the pole figure is reconstructed based on only a single incomplete pole figure. To speed up the development of our proposed method and for usage in other machine learning algorithms, we introduce a GPU-based simulation for data generation. Furthermore, we present a pole widths standardization technique using a custom deep learning architecture that makes algorithms more robust against influences from the experiment setup and material.

Evaluation of extremely steep residual stress gradients based on a combined approach using laboratory-scale equipment.

Surface treatments characterized by rapid heating and cooling (e.g. laser hardening) can induce very steep residual stress gradients in the direct vicinity of the area being treated. These gradients cannot be characterized with sufficient accuracy by means of the classical sin2Ψ approach applying angle-dispersive X-ray diffraction. This can be mainly attributed to limitations of the material removal method. In order to resolve residual stress gradients in these regions without affecting the residual stress equilibrium, another angle-dispersive approach, i.e. the universal plot method, can be used. A novel combination of the two approaches (sin2Ψ and universal plot) is introduced in the present work. Prevailing limits with respect to profiles as a function of depth can be overcome and, thus, high-resolution surface layer characterization is enabled. The data obtained are discussed comprehensively in comparison with results elaborated by energy-dispersive X-ray diffraction measurements.