Authors: Dr Bin Zhu and Dr Tan Sui
The Challenge: Residual Stress in Fusion Reactor Joints
The development of in-vessel components for nuclear fusion reactors requires precise mechanical characterization of their joints, such as laser-welded Eurofer97 and P91, as well as dissimilar Cu-W and Eurofer97-W joints. Despite advanced joining techniques, residual stresses induced during welding can significantly impact material performance, potentially leading to premature failure. Accurately assessing residual stress distribution within the narrow fusion zone (FZ) and heat-affected zone (HAZ) remains a significant challenge.
University of Surrey, led by Dr Sui has pioneered residual stress evaluation for structural integrity research for nuclear fusion materials, particularly in metallic joints for in-vessel fusion components. In collaboration with the UK Atomic Energy Authority (UKAEA) and National Physical Laboratory (NPL), Surrey has developed innovative experimental and modelling approaches to understand how joining-induced residual stresses degrade mechanical properties and lead to premature failure.
Dr Zhu is a key contributor in establishing new experimental capabilities for residual stress evaluation in multiple fusion joints, using neutron diffraction and imaging at STFC’s ISIS Neutron and Muon Source and plasma focused ion beam – digital image correlation (PFIB-DIC) at University of Surrey.
Methodology: High-Resolution Residual Stress Measurement
To evaluate residual stresses in laser-welded joints, The Xstress DR45 system was used to perform residual stress linear mapping over both P91 and Eurofer97 joints, with measurements taken at precise locations, as illustrated in Fig. 1. The line-scan measurements (green circles) were conducted using a 0.3 mm collimator with a spacing of 0.1 mm, and a 2θ angle of 156.4° was selected for measuring the {211} peak. Calibration was carefully carried out using a powder sample, and the device’s performance was verified against a reference sample. Additionally, the same samples were analyzed using plasma-focused ion beam digital image correlation (PFIB-DIC) [1] and neutron diffraction techniques [2,3]. Detailed methodologies for these complementary approaches can be found in previously published research.

Fig. 1: Experimental setup for evaluating the residual stress distribution using the Xstress DR45.
Results from Xstress DR45 and Compared with PFIB-DIC and Neutron Diffraction
The measurement time with the very small collimator was 2 minutes per point. The Xstress DR45 with the XY-table enables high spatial resolution evaluation of the residual stress distribution. The Xstress DR45 achieved the evaluation for the residual stress distribution, where the peak tensile stress of around 350 MPa was observed near the FZ/HAZ interface. The residual stress distribution trends obtained with DR45 closely align with those from PFIB-DIC. Whilst the PFIB-DIC technique offers the highest resolution at approximately 0.03 mm for each measurement point, Xstress DR45 demonstrates an excellent balance between resolution and efficiency. Although the DR45 does not match the ultra-high resolution of PFIB-DIC, it successfully captures the overall residual stress distribution trends within these narrow regions, demonstrating its capability to measure stress in highly localized zones.
![Fig. 2 DR45-XRD residual stress results for Eurofer97, with correlation to PFIB-DIC [1] and neutron diffraction [2] at 0 degree direction.](https://www.stresstech.com/wp-content/uploads/2025/04/Fig-_2_DR45_residual_stress_results_comparison_NE_PFIB-DIC.png)
Fig. 2 DR45-XRD residual stress results for Eurofer97, with correlation to PFIB-DIC [1] and neutron diffraction [2] at 0 degree direction.
When compared to neutron diffraction, Xstress DR45 provides a more detailed residual stress distribution within the FZ and HAZ regions. The limitation of neutron diffraction, due to its lower spatial resolution of approximately 1.4 mm [2], often results in stress values being averaged out within different regions of the laser-welded joints. In contrast, DR45 achieves significantly greater resolution, allowing for the identification of localised residual stress variations that would otherwise be missed. Furthermore, the trend of full-width half maximum (FWHM) distribution measured using Xstress DR45 was cross-validated through nanoindentation experiments, confirming its proportional relationship to micro-hardness. For the P91 joints, DR45 results were found to be consistent with PFIB-DIC measurements in both scanning directions [4], reinforcing its reliability in capturing stress variations across different welding configurations.
![Fig. 3: DR45-XRD residual stress results for P91, with correlation to PFIB-DIC [1] and neutron diffraction [2], at (a) 0-degree direction and (b) 90-degree direction.](https://www.stresstech.com/wp-content/uploads/2025/04/Fig_3_DR45_RS_results_P91_correlation_PFIB-DIC_NE.png)
Fig. 3: DR45-XRD residual stress results for P91, with correlation to PFIB-DIC [1] and neutron diffraction [2], at (a) 0-degree direction and (b) 90-degree direction.
Conclusion: Xstress DR45 as an Efficient Residual Stress Analysis Tool
The Xstress DR45 XRD system offers a fast and reliable method for surface residual stress detection, making it a valuable tool for initial assessments before applying more complex and time-consuming techniques such as PFIB-DIC. For cases that do not demand such extreme precision, Xstress DR45 serves as a practical and efficient alternative, particularly for analyzing narrow laser-welded joints. Its ability to capture high-resolution residual stress distributions in a short timeframe makes it a compelling option for the evaluation and optimization of fusion reactor components.
Reference
[1] B. Zhu, Y. Wang, J. Dluhoš, A.J. London, M. Gorley, M.J. Whiting, T. Sui, Sci Adv 8 (2022).
[2] B. Zhu, N. Leung, W. Kockelmann, S. Kabra, A.J. London, M. Gorley, M.J. Whiting, Y. Wang, T. Sui, J Mater Sci Technol 114 (2022) 249–260.
[3] B. Zhu, N. Leung, Y. Wang, H. Zhang, J. Dluhoš, T. Pirling, M. Gorley, M.J. Whiting, T. Sui, Materials Science and Engineering: A 877 (2023) 145147.
[4] B. Zhu, O. Mohamed, A. Koko, H. Zhang, J. Dluhoš, Y. Wang, M. Gorley, M.J. Whiting, T. Sui, Journal of Materials Research and Technology 35 (2025) 6341–6347.
