Nuclear fusion energy has strong potential as an alternative clean source of energy, which can help resolve our dependence on fossil fuels, reducing our carbon emissions and providing a major contribution to hitting net zero targets. Assembling and maintaining components requires remote joint techniques, such as laser welding, that is extensively used to overcome complicated assembly and maintenance difficulties. Previous studies have demonstrated the feasibility of using remote laser tools to weld components. However, the joining process can cause substantial residual stress, leading to premature failure, such as cracking or ‘crack initiation’. Therefore, the investigation of residual stress is critical to maintaining the structural integrity of components and predicting where crack initiation could form.
Non-destructive evaluation techniques are the answer to achieving a better understanding of residual stress. While two-dimensional (2D) residual strain can be determined by using synchrotron-based X-rays. 3D residual strain is accessible to view by using neutron diffraction due to the high penetration of neutron particles. However, investigation of subtle changes is challenging due to the millimetre-level resolution and slow scanning speed limits the capacity of analysing across a large area.
Detailed residual strain maps for such an area can be obtained by using neutron Bragg edge imaging (NBEI). Originating around the year 2000, Bragg-edge imaging offers the possibility for non-destructive visualization of microstructural characteristics of the sample with high resolution, such as texture variations and lattice strains. Recently developed 3D tomographic reconstruction techniques can provide valuable insights into the bulk of joints via a 3D model, which is reconstructed using multiple angular projections. Combining the tomographic reconstruction technique with NBEI allows the inspection of microstructures and strain fields for a large area simultaneously in a non-destructive way. This can reveal the differences undetected by standard diffraction and imaging techniques. Further reduction of the angular projections is desirable to develop a cost-effective strain tomography method. Up until now, only a few studies have examined the tomographic strain reconstruction on a large scale.
A team of researchers from ISIS Neutron and Muon Source, University of Surrey, the UK Atomic Energy Authority and the National Physical Laboratory have combined efforts to work on a residual strain tomographic reconstruction of a laser-welded Eurofer97. Reduced-activation ferritic/martensitic (RAFM) steels are widely used as structural materials in a wide range of industry components (from example to example). Eurofer97, one of the RAFM steels, uses lower activation elements and is used to build fusion reactors. The team’s previous study, led by Dr Sui (Revealing the residual stress distribution in laser welded Eurofer97 steel by neutron diffraction and Bragg edge imaging - ScienceDirect) was one of the most discussed papers published in association with ISIS in 2022 with a high Almetric score, which also investigated Eurofer97 steel and revealed the residual stress distribution.
In this study, the bulk microstructure and residual strain distribution in a laser-welded Eurofer97 joint was measured by reconstructing a small number of NBEI projections. The 3D volumetric reconstruction was achieved using the filtered back projection (FBP) technique. This technique is the basis for image reconstruction (converting from the measured data to the image) on modern CT scanners. The reconstructed microstructures via neutron attenuation are consistent with that obtained from reflected light microscopy.
The residual strain reconstruction results discovered that the reconstruction presented a similar strain situation to the conventional neutron diffraction and Bragg edge imaging residual strain measurements, implying a uniform strain distribution in the joint.
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.