Some innovations in physics come from entirely new technologies, others from fresh theoretical insights. Others still take shape by bringing together existing tools in new ways, working out how to combine them to outperform other solutions. The branch of particle physics that studies weakly interacting particles – such as neutrinos and some types of dark-matter candidates – could use innovative detection approaches: technological challenges in this research area quickly become practical as well as economic, as increases in detector volume and spatial resolution improve the sensitivity to the processes producing the particles of interest. Similarly, demanding targets on instrument capability apply to the calorimeters used in collider experiments.
Three-dimensional (3D) tracking of elementary particles in large-volume, dense materials is required in most particle physics experiments. In a scintillator, this is commonly achieved through fine segmentation of the material into many smaller active units, with each unit emitting light in the visible frequency range when a charged particle passes through it. Typically, the photons produced in every active unit are collected by optical fibres and carried outside of the scintillator to the photomultiplier tubes or silicon photomultipliers used for photon counting.
In the T2K neutrino-oscillation experiment in Japan, for example, one detector boasts about two tons of sensitive volume assembled from approximately two million cubes and 60,000 fibres. Over at CERN and the Paul Scherrer Institute, the LHCb and Mu3e experiments achieve sub-millimetre spatial resolution thanks to millions of thin scintillating optical fibres. With these figures, it’s clear that the scalability of this kind of scintillator material segmentation may turn into a bottleneck when larger volumes become necessary.
Now, a collaboration between ETH Zurich and EPFL invites the research community to change radically the way elementary particles are detected. PhD student Till Dieminger, senior scientist Dr Saúl Alonso-Monsalve, Professor Davide Sgalaberna and colleagues in his group, together with members of the Advanced Quantum Architecture Lab at EPFL in Lausanne led by Professor Edoardo Charbon, proposed and tested the first prototype of a new detector capable of performing ultrafast, 3D and high-resolution imaging of particles in large volumes of unsegmented scintillator material. Their demonstration, along with a comprehensive simulation study, appeared recently in Nature Communications.
Some years ago, photography enthusiasts may remember the buzz around so-called plenoptic or light field cameras: these devices differ from more conventional ones because they record light intensity while also capturing depth information. This is made possible by a micro-lens array (MLA) placed between the standard objective lens and the imaging sensor: each lens in the array acts as a tiny camera that contributes to reconstructing the light field intended as a map of light intensity at a given spatial location and direction. Plenoptic cameras hold great potential for imaging and, if combined with single-photon avalanche diode (SPAD) array sensors, can achieve high-resolution 3D tracking of elementary particles even in photon-starved conditions. Nonetheless, the use of light field cameras for particle tracking remained unexplored until now.
Within the PLATON project funded by the Swiss National Science Foundation, the ETHZ-EPFL team built a first concept demonstrator based on a light field camera featuring an MLA as well as a SPAD array imaging sensor. The SPAD sensor, called SwissSPAD2, was developed by the team at EPFL; the MLA was designed and mounted onto the sensor, thus forming the plenoptic system, by Raytrix GmbH. Crucially, SwissSPAD2 adds gated photon detection to the setup: this means that detection events fall into fixed temporal windows, making it possible to isolate the time intervals where the signal coming from detected photons dominates over spurious counts.
The researchers study the performance of the PLATON demonstrator by characterising its spatial resolution with data collected in the laboratory for light intensities ranging from a few hundred down to five detected photons. The team also tests the prototype on its ability to detect and reconstruct the position of electrons in a plastic scintillator block from a strontium-90 source. In all considered cases, simulations show good agreement with the measurements performed in the laboratory.
The experience gained with the first PLATON demonstrator has already steered the team’s short-term upgrade plans: indeed, the researchers are developing a novel SPAD array sensor that will allow them to achieve higher photodetection efficiency as well as single-photon sub-nanosecond temporal resolution. The latter means that detected photons won’t just be assigned a time window, but an actual time stamp. Additionally, the design of the plenoptic camera has already been optimised to maximise both its field of view and its light collection capability. All these developments are expected to improve PLATON’s spatial resolution, as already suggested by the simulations presented in the paper.
The published simulation results, which test PLATON’s projected performance when detecting neutrinos, are based on the upgraded version of the system that’s currently under development. These simulations also feature new image post-processing methods that take advantage of a neural network (NN) built around a so-called Transformer architecture adapted from those commonly used for large language models; importantly, the proposed NN is shown to efficiently capture the correlations among detected scintillation photons. The presented simulations indicate that a spatial resolution below 1mm is realistic for a PLATON system with an unsegmented volume of (10x10x10)cm3, and that the selection of neutrino interactions with final-state low-momentum protons can be achieved with high purity and efficiency.
When scaling up to a one-cubic-metre unsegmented scintillator detector, the team doesn't run neutrino detection simulations due to limited computational resources but considers a scenario with a simplified point-like photon source. In this case, the outcome of the simulations shows that a spatial resolution of a few millimetres is within reach – a result that's on par with state-of-the-art plastic scintillator detectors. Based on such encouraging findings, the authors expect that further work on the optical design, among others, will unlock sub-millimetre resolution in PLATON-type particle detectors with volumes larger than 1m3.
According to the ETH Zurich researchers, the potential of the new approach discussed in the paper extends beyond particle physics. Indeed, they believe that their plenoptic-camera-based system could provide a performance boost in a variety of imaging applications.
Dieminger, Alonso-Monsalve and Sgalaberna have now filed three separate patents on PLATON for positron emission tomography (PET). The patents cover the scanner concept and the image post-processing methods, including the NN developed by Alonso-Monsalve. Particle physics has a long tradition of impactful technology transfer, from the world wide web to proton therapy: the PLATON project may well turn into yet another success story.
Nature Communications
An ultrafast plenoptic-camera system for high-resolution 3D particle tracking in unsegmented scintillators
21-Mar-2026