Year: 2025
Type: article
Abstract: Most semiconductor image sensors capture visible light colors by grouping four pixels together, coating three of them with red, green, and blue (RGB) filters. Each pixel provides intensity information for red, green, and blue, representing the combined pixel size. Without color filters, the same sensor can function as a monochrome sensor with four times the number of pixels as a color-filtered sensor. Recently, a research group led by Dr. Charles Roques-Carmes (Massachusetts Institute of Technology, USA) successfully adapted an RGB color-filtered sensor for x-ray color imaging for the first time. As is well known, image sensors without color filters are widely used for x-rays. X-rays can be converted into visible light when combined with a scintillator crystal. This process enables the creation of monochrome (noncolor) x-ray images. Another popular method is counting the number of electron–hole pairs (charges) for each photon event in a pixel, similar to an ordinary semiconductor x-ray detector. This method produces energy-dispersive, or color, x-ray images, provided that the effects of charge-sharing events are properly corrected. For more details, see K. Sakurai and W. Zhao's X-ray Color Imaging: Static and Dynamic X-Ray Fluorescence to See Chemical Elements (Institute of Physics, 2024), https://doi.org/10.1088/978-0-7503-3215-6. However, a new x-ray imaging method that is neither of the above has now emerged. It combines an RGB color-filtered semiconductor image sensor with three scintillators that emit red, green, and blue colors in response to x-ray irradiation. Three transparent scintillator plates, each 0.1 mm thick, are used: ZnSe:Te (red), Gadox:Tb (green), and NaI:Tl (blue). These plates are stacked and used in a transmissive configuration. X-ray photons of various energies (not monochromatic) are incident on the three scintillator layers, pass through them, and emit light with each color. A detailed Monte Carlo simulation has clarified the attenuation and visible light emission processes in each scintillator layer, as well as the divergence and attenuation of the emitted light. It has been demonstrated that iodine- and gadolinium-based contrast agents could be effectively distinguished from bone, muscle, and soft tissue. For more information, see the paper, Seokhwan Min, Seou Choi, Simo Pajovic, Sachin Vaidya, Nicholas Rivera, Shanhui Fan, Marin Soljačić, and Charles Roques-Carmes, “End-to-End Design of Multicolor Scintillators for Enhanced Energy Resolution in X-Ray Imaging,” Light: Science & Applications 14 (2025): 158, https://www.nature.com/articles/s41377-025-01836-8. A research group led by Professor David W. Messinger of the Rochester Institute of Technology in New York recently developed a method for correlating multi-element x-ray fluorescence imaging with color images obtained using visible light, and in some cases, infrared light. This method is mainly intended for application to works of art. Multimodal image analysis, which involves registering and uniformly handling images obtained from x-ray imaging and other techniques, has become indispensable. X-ray fluorescence imaging can apparently provide information on the chemical composition, that is, the ratio of each element's content, at each point. However, in real applications, such as with artwork, this information is only significant when the correspondence between the points in the x-ray image and the corresponding points in the actual object is confirmed to be accurate. Currently, the alignment procedure between x-ray and visible light (plus infrared) images is often done manually. The research team proposes adopting the well-established mutual information maximization (MIM) technique used in other imaging fields. First, the high-resolution color images obtained using visible light (and infrared light, if applicable) are converted into grayscale images. Next, the structural similarity index (SSI) is calculated between the grayscale images and the images of each element obtained by x-ray fluorescence. The x-ray fluorescence element image with the highest SSI value is then selected as the candidate image. The same correspondence relationship is employed for other element images. Applying image quantization before evaluating the SSI has been found to effectively suppress noise and improve spatial features, which are often problematic in real measurements. The proposed method enables the analysis of x-ray images using a multimodal correspondence method without human intervention, thereby eliminating arbitrariness in manual operation. For more information, see the paper, Morteza Maali Amiri and David W. Messinger, “A New Method for XRF and RGB Image Registration,” npj Heritage Science 13 (2025): 98, https://www.nature.com/articles/s40494-025-01603-3. X-ray ptychography uses coherent x-ray diffraction imaging and does not require focusing optics, such as lenses. Previously, there were two issues with x-ray ptychography: (i) a narrow field of view, which limited its use to small samples, and (ii) loss of spatial resolution due to multiple scattering, which limited its application to thin samples. The former issue has been resolved with the introduction of multi-beam techniques; now, larger samples can be analyzed. The latter has been resolved with multi-slice techniques. However, spatial resolution remains a limitation when dealing with large, three-dimensional samples. Recently, a research group led by Professor Ulrich Vogt at the Royal Institute of Technology in Sweden successfully applied multi-beam and multi-slice techniques simultaneously at the MAX IV and PETRA III synchrotron radiation facilities in Sweden and Germany, respectively. This development paves the way for high-resolution x-ray microscopy imaging of large, three-dimensional samples. In this study, the researchers used a sample consisting of two 200-nm-thick silicon nitride (SiN) membranes with Kapton films of different thicknesses sandwiched between them as spacers. One of the two SiN membranes was coated with FZP, a material commonly used as a focusing optical device, and the other membrane was randomly dispersed with 200-nm-diameter Au nanoparticles. Four types of multi-slice samples were created by varying the thickness of the spacers. The detectors used were Eiger2 X 4M at MAX IV and Eiger X 4M at PETRA III. The research group conducted experiments at the two facilities and confirmed reproducibility, thereby demonstrating the method's effectiveness. For more information, see the paper, Mattias Åstrand, Ulrich Vogt, Runqing Yang, Pablo Villanueva Perez, Tang Li, Mikhail Lyubomirskiy, and Maik Kahnt, “Multi-Beam Multi-Slice X-Ray Ptychography,” Scientific Reports 15 (2025): 9273, https://www.nature.com/articles/s41598-025-93757-0. Perovskite x-ray detectors are one of the most rapidly advancing topics in recent years. Recently, a research group led by Professor Meng Gang (Hefei Institute of Physical Science, Chinese Academy of Sciences, China) developed ultra-low noise CsPbBr₃ perovskite x-ray image detectors. Two types of detectors were investigated: (i) a single-crystal CsPbBr₃ detector cooled with liquid nitrogen and (ii) a polycrystalline CsPbBr₃ detector using a specifically designed wafer. Liquid nitrogen cooling effectively improved the resistivity of the single-crystal detector by two orders of magnitude. This improvement achieved ultra-low noise and enabled the detection of extremely weak radiation signals as low as 0.054 nGy/s. In the polycrystalline case, the innovation lies in the wafer manufacturing method. In this study, CsPbBr₃ wafers were manufactured using a hot-press method to suppress ion migration. Adding 2-bromonaphthalene enabled the effective passivation of interface defects, which suppressed ion migration and improved charge transport properties. Consequently, the activation energy for ion migration increased to 0.56 eV, which significantly suppresses dark current drift under high electric fields. Maintaining image contrast while achieving ultra-low noise levels enabled the detection of extremely weak radiation signals, such as 9.41 nGy/s. Although single-crystal CsPbBr₃ detectors cooled with liquid nitrogen are approximately 200 times more effective at reducing noise, low-noise polycrystalline CsPbBr₃ detectors that can be manufactured cheaply and in large quantities at room temperature are still significant. For more information, see the papers, Xiao Zhao, Shimao Wang, Yanan Song, Toru Aoki, Volodymyr Gnatyuk, Libing You, Zanhong Deng, Ruhua Tao, Xiaodong Fang, and Gang Meng, “Freezing Non-Radiative Recombination in High-Performance CsPbBr3 Single Crystal X-ray detector,” Applied Physics Letters 125 (2024): 083507, https://doi.org/10.1063/5.0224223 and Xiao Zhao, Ziqing Li, Shimao Wang, Yanan Song, Enliu Hong, Tingting Yan, Gang Meng, and Xiaosheng Fang, “2-Bromonaphthalene-Induced Defect Passivation to Suppress Ion Migration in CsPbBr₃ Wafer for X-Ray Detector with Bias-Resistant Stability,” Advanced Function Materials, Early View March 17 (2025), https://doi.org/10.1002/adfm.202500039. The author declares no conflicts of interest.
Source: X-Ray Spectrometry
Author: Kenji Sakurai
Language: English
Subfield: Radiation
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4412938599