Search

Fusion In-vessel components, assembled and maintained using laser welding, one of the most promising techniques, exhibit complex distributions of residual stress, microstructures, and material properties. These residual stresses can compromise structural integrity and lifespan of critical components. Although using advanced experimental measurements can evaluate the residual stress for individual case, extending the measurements to massive number of components are costly and time-consuming. Traditional machine learning (ML) models struggle to account for the heterogeneity and anisotropy of these stress distributions. Here, we develop a novel ML framework based on the Eurofer97 steel, the structural material for in-vessel components. The ML framework is trained on high-resolution residual stress data derived from recently-developed evaluation techniques. Combining with microstructures, the model enables prediction of heterogenous and anisotropic residual stress distribution. It successfully predicts the compressive residual stress in fusion zone (∼−200 MPa) balanced by tensile residual stress in heat affected zone (∼300 MPa), aligning closely with experimental results with the R-squared value of 0.989 and the mean square error of 10−4. Unlike experiments that take hours, the ML model provides predictions within seconds. It offers valuable insights into residual stress prediction for various joints, enhancing the reliability and lifetime prediction of in-vessel components.

The European DEMOnstration power plant (EU DEMO) project is currently leading the research endeavours within the fusion field with the goal of developing a next generation fusion reactor. Given the challenging nature of the application, it is essential to establish methodologies that can provide convenient and reliable characterisation of the chosen materials within DEMO. In this paper, the recently developed Time-of-Flight Neutron Bragg Edge Imaging (TOF-NBEI) was used on Tungsten (W)/ Copper (Cu) dissimilar joints sample mock-ups of the cooling system design used in the critical divertor component with the goal of mapping the residual stresses across the sample. Residual strain mapping was performed on the W phase with considerable tensile residual strains identified close to the W-Cu interface. The large-grain microstructure of the Cu phase was analysed using the energy-resolved neutron radiographs. The results will be used as a basis for future TOF-NBEI experiments of tungsten monoblocks related to DEMO.

Nuclear fusion is a potential source of electricity which can address the environmental problems posed by fossil fuels. Eurofer97 steel is a primary structural material for breeding blanket and divertor components in fusion Tokamaks. Assembling and maintaining the structural integrity of these in-vessel components requires remote joint techniques, such as laser welding, although it induces immersive residual stress. The interaction of the residual strain and the heterogeneous microstructure degrades the mechanical performance of fusion components. However, an inspection of bulk residual strain distribution is still challenging. This study presents the residual strain distribution in the bulk of the weldment using volumetric tomographic reconstruction. A neutron Bragg edge imaging technique is used to obtain 2D angular projections. The 3D volumetric strain map is reconstructed from 2D residual strain projections using the filtered back projection technique. It is found that the laser welding technique generates a uniform residual strain field in the through-thickness direction. The results also demonstrate the potential of reconstructing volumetric residual strain distribution in bulk materials using fewer projections to reduce data redundancy and acquisition time for the neutron Bragg edge imaging technique.

In this work, novel bioinspired polyurethane (PU) scaffolds were fabricated via freeze casting for PU-based Pancreatic Ductal Adenocarcinoma (PDAC) model. In order to reproduce the tumour micro-environment that facilitates cellular kinetics, the PU scaffolds were surface modified with extracellular matrix (ECM) proteins including collagen and fibronectin (Col and FN). Synchrotron-based small- and wide-angle X-ray scattering (SAXS/WAXS) techniques were applied to probe structural evolution during in situ mechanical testing. Strains at macroscopic, nano-, and lattice scales were obtained to investigate the effects of ECM proteins and pancreatic cell activities to PU scaffolds. Significant mechanical strengthening across length scales of PU scaffolds was observed in specimens surface modified by FN. A model of stiffness modulation via enhanced interlamellar recruitment is proposed to explain the multi-scale strengthening mechanisms. Understanding multi-scale deformation mechanisms of a series of PU scaffolds opens an opportunity in developing a novel pancreatic cancer model for studying cancer evolution and predicting outcomes of drug/treatments.

Many natural materials demonstrate ideal design inspirations for the development of lightweight composite materials with excellent damage tolerance. One notable example is the layered architecture of nacre, which possesses toughness an order of magnitude higher than its constituent parts. Man-made nacre-like ceramic/polymer composites obtained through direct infiltration of polymer in ceramic scaffolds have been shown to produce improved mechanical properties over other composite architectures. Replacing the polymer phase with metal could provide higher damage tolerance but the infiltration of metal into complex ceramic scaffolds is difficult due to the surface tension of molten metal. To address this, bioinspired nacre-like micro-layered (µL) alumina scaffolds with different ceramic fractions from 18 to 85% were infiltrated with aluminium alloy 5083 via pressureless and squeeze casting infiltrations techniques. The scaffolds were created using a bi-directional freeze-casting and one-step densification method. As a result, the µL alumina/aluminium composites displayed significant extrinsic toughening mechanisms with both high strength and toughness. The mechanical performance was highly dependent on the interface, microstructure, and composition. The nacre-like composites with 18% alumina and AlN interface displayed a maximum resistance‐curve toughness up to around 70 MPa.m½ (35 MPa.m½ at the ASTM limit) and a flexural strength around 600 MPa.

In this manuscript, we elucidated, for the first time, the substructural mechanisms present in our recently developed bioinspired polyurethane-based pancreatic tissue models. Different protein coatings of the model, i.e., collagen and fibronectin were examined. More specifically, analysis took place by combined real-time synchrotron X-ray scattering techniques and confocal laser scanning microscopy, to quantify the structural alteration of uncoated-polyurethane (PU) and protein-coated PU as well as the time-resolved structural reorganisation occurring at the micro-, nano- and lattice length scales during in situ micromechanical testing. We demonstrate that a clear increase of stiffness at the lamellar level following the fibronectin-PU modification, which is linked to the changes in the mechanics of the lamellae and interlamellar cohesion. This multi-level analysis of structural-mechanical relations in this polyurethane-based pancreatic cancer tissue model opens an opportunity in designing mechanically robust cost-effective tissue models not only for fundamental research but also for treatment screening.