Gamma Emission Tomography for Nuclear Fuel Assessment

Gamma Emission Tomography (GET), and in particular its passive variant (PGET), has emerged as a powerful non-destructive assay for characterising the internal distribution of radioactive nuclides in spent nuclear fuel assemblies. By rotating highly collimated gamma detectors around a fuel bundle and recording photon counts at multiple angles, a two-dimensional cross-sectional map of activity and attenuation can be reconstructed. This enables rod-level detection of missing, substituted or partially degraded fuel pins, supporting nuclear safeguards, repository readiness and post-irradiation examination. GET combines principles of single-photon emission computed tomography with bespoke inverse-problem algorithms to account for high self-attenuation, variable fuel geometry and intense radiation fields. Its deployment in water-cooled pools, dry environments and reactor test facilities has demonstrated versatility across fuel types, burnups and cooling times. Ongoing developments target improved spatial resolution, reduced computational cost and enhanced image fidelity, ensuring that GET continues to underpin verification of peaceful use and integrity of nuclear materials worldwide.

Research from Nature Portfolio

First, comparative measurements using mock-up Co-60 rods in both air and water have validated the robustness of PGET instrumentation outside traditional pool environments. Highly collimated CdZnTe detectors captured 360° emission data, confirming the method’s ability to distinguish rod arrangements regardless of medium and paving the way for full-scale spent fuel trials. Second, advances in image reconstruction and automated pin identification have yielded substantial gains in resolution and throughput. A regularised inverse-problem approach outperformed filtered back projection, halving mean-square error and doubling structural similarity, while a convolutional neural network accurately classified bundle types and estimated rod activities to within 5 % uncertainty. Third, a physics-aware reduced-order modelling strategy has slashed the computational burden of Monte Carlo simulation for angular views. By combining limited-view sampling with a real-time forward model, full-angle reconstructions can now be obtained at a fraction of the cost, enabling rapid generation of synthetic datasets essential for validation and instrument design.

Gamma Emission Tomography for Nuclear Fuel Assessment publication trend

The graph below shows the total number of articles in gamma emission tomography for nuclear fuel assessment across all publications each year (not limited to Nature Index journals).

Technical terms

Passive Gamma Emission Tomography (PGET): A non-invasive imaging method that records gamma-ray emission from fuel assemblies at multiple angles to reconstruct activity and attenuation maps.

Collimator: A device composed of parallel slits or channels that restricts gamma-ray trajectories, improving spatial resolution by blocking off-axis photons.

Self-attenuation: The reduction in detected gamma intensity caused by absorption or scattering within the fuel assembly itself.

Sinogram: A two-dimensional representation of projection data collected at different angles, serving as the raw input for tomographic reconstruction algorithms.

References

1. In-air and in-water performance comparison of Passive Gamma Emission Tomography with activated Co-60 rods. Scientific Reports (2023).
2. Simulation of the response of a segmented high-purity germanium detector for gamma emission tomography of nuclear fuel. Discover Applied Sciences (2020).
3. Enhancing passive gamma emission tomography data with deep learning. Annals of Nuclear Energy (2024).
4. Simultaneous reconstruction of emission and attenuation in passive gamma emission tomography of spent nuclear fuel. Inverse Problems and Imaging (2020).
5. Quantitative imaging and automated fuel pin identification for passive gamma emission tomography. Scientific Reports (2021).
6. Vanquishing the computational cost of passive gamma emission tomography simulations leveraging physics-aware reduced order modeling. Scientific Reports (2023).