Supercollimating photonic crystal scintillators

Sachin Vaidya, Seou Choi, Charles Roques‐Carmes, Marin Soljačić2026 · DOI

TL;DR

This paper introduces a novel class of "supercollimating photonic crystal scintillators" that overcome the fundamental trade-off between X-ray detection efficiency and spatial resolution. By engineering light propagation within the scintillator material itself, these new devices can achieve significantly sharper images at lower X-ray doses, paving the way for improved medical imaging and other applications.

What Problem Does This Paper Address?

X-ray imaging technologies, crucial for medical diagnostics and security, rely on scintillators to convert X-ray photons into visible light that can be detected. To capture more X-rays (increasing efficiency), scintillators need to be thicker. However, thicker scintillators lead to a fundamental problem: the emitted light spreads out due to diffraction, blurring the image and reducing spatial resolution. This creates a difficult trade-off: high efficiency means low resolution, and high resolution means low efficiency, often necessitating higher X-ray doses to compensate for the blur. This is particularly problematic in medical imaging, where reducing patient radiation exposure is a major concern. Previous attempts to improve scintillation focused on surface modifications or enhancing emission yield, but none directly addressed the directional propagation of light within the bulk of the scintillator.

Methods

The core innovation lies in designing a three-dimensional photonic crystal structure that exhibits "supercollimation." This phenomenon, previously explored for other optical applications, involves engineering the material's band structure so that light at specific frequencies propagates with minimal diffraction, even over long distances.

The authors developed a sophisticated multi-scale modeling framework to achieve this:

  • Nanophotonic Band Structure Simulation: They used MIT Photonic Bands (MPB) to design and simulate the band structure of 3D photonic crystals (specifically, a single gyroid structure) made from scintillator materials like Lu₂O₃(Eu) and ZnSe(Te). The goal was to create "flat isofrequency contours" (IFCs) in the band structure, which are the key to supercollimation.
  • Monte Carlo Simulation (Geant4): To simulate the entire X-ray imaging process, they integrated the nanophotonic simulations with Geant4, a standard tool for simulating particle transport. This allowed them to model:
  • X-ray energy deposition within the scintillator.
  • The generation of dipole emitters (representing the light-emitting centers).
  • The angular emission of scintillation light, guided by the engineered photonic crystal band structure.
  • The transport of this light to the detector.
  • Image formation and the calculation of metrics like the Point Spread Function (PSF) and Modulation Transfer Function (MTF).

Datasets

The paper does not explicitly mention training datasets. The evaluation relies on simulated X-ray images generated using the described modeling framework. The "datasets" in this context are the simulated input X-ray sources (a pencil beam and the MIT logo) and the simulated X-ray images of a human kidney vasculature. The material properties for ZnSe(Te) and Lu₂O₃(Eu) were taken from existing literature (Refs. 28 and 29).

Key Insights and Ideas

  • Supercollimation as a Solution: The central idea is that by engineering the internal light transport properties of the scintillator, we can directly combat the diffraction-induced blurring that plagues conventional thick scintillators.
  • Flat Isofrequency Contours (IFCs) are Key: The shape of the IFCs in the photonic crystal's band structure dictates how light propagates. Flat IFCs, particularly those oriented towards the detector, lead to supercollimation and suppressed diffraction.
  • Material Bandwidth Matters: The effectiveness of supercollimation is strongly tied to the emission bandwidth of the scintillator material. Materials with narrower emission spectra that fall within the designed supercollimation bandwidth show more pronounced improvements in resolution and DQE.
  • Order-of-Magnitude Improvements Possible: The simulations demonstrate that these supercollimating photonic crystal scintillators can achieve spatial resolution enhancements of up to an order of magnitude compared to bulk scintillators of the same thickness.
  • Dose Reduction Potential: The improved spatial resolution and higher Detector Quantum Efficiency (DQE) at high spatial frequencies translate directly to the ability to achieve comparable image quality with significantly lower X-ray doses, potentially reducing patient exposure in medical imaging.

Key Results

  • Spatial Resolution Enhancement: Simulated X-ray images show significantly sharper details compared to bulk scintillators of equal thickness.
  • MTF Improvement: The Modulation Transfer Function (MTF) shows substantial gains, particularly at higher spatial frequencies, indicating better preservation of image contrast. For Lu₂O₃(Eu), the spatial resolution at MTF₁₀ was enhanced by approximately 10.7x at MTF₃.
  • DQE Enhancement: The DQE at high spatial frequencies was enhanced by one to two orders of magnitude for both materials.
  • Medical Imaging Simulation: In a simulated X-ray angiography of a human kidney vasculature, the supercollimating photonic crystal scintillator achieved similar image quality (measured by FSIM score) using approximately 10 times less X-ray dose compared to a bulk scintillator.

Limitations and Future Work

  • Material Properties: The simulations rely on specific material properties for ZnSe(Te) and Lu₂O₃(Eu). The performance might vary with different scintillator materials.
  • Fabrication Challenges: While the paper discusses potential fabrication approaches (e.g., direct laser writing, infiltration), the practical realization of these complex 3D nanostructures at scale remains a significant engineering challenge.
  • Emission Bandwidth: The effectiveness is limited by the emission bandwidth of the scintillator material. Materials with very broad emission spectra might not benefit as much.
  • Future Work: The authors suggest that their framework can be generalized to other photonic crystal scintillators and that nanophotonic dispersion engineering offers a broad route to simultaneously improve efficiency and resolution in various scintillation-based systems, including high-energy particle detectors. They also highlight the potential for tailoring responses to specific X-ray energies and imaging modalities.

Summary updated

BibTeX
@misc{vaidya2026supercollimating,
  author = {Sachin Vaidya and Seou Choi and Charles Roques‐Carmes and Marin Soljačić},
  title = {{Supercollimating photonic crystal scintillators}},
  year = {2026},
  doi = {10.1117/12.3079431},
}
Figure