Highly Resolved and Robust Dynamic X‐Ray Imaging Using Perovskite Glass‐Ceramic Scintillator with Reduced Light Scattering

Abstract All‐inorganic perovskite quantum dots (QDs) CsPbX3 (X = Cl, Br, and I) have recently emerged as a new promising class of X‐ray scintillators. However, the instability of perovskite QDs and the strong optical scattering of the thick opaque QD scintillator film imped it to realize high‐quality and robust X‐ray image. Herein, the europium (Eu) doped CsPbBr3 QDs are in situ grown inside transparent amorphous matrix to form glass‐ceramic (GC) scintillator with glass phase serving as both matrix and encapsulation for the perovskite QD scintillators. The small amount of Eu dopant optimizes the crystallization of CsPbBr3 QDs and makes their distribution more uniform in the glass matrix, which can significantly reduce the light scattering and also enhance the photoluminescence emission of CsPbBr3 QDs. As a result, a remarkably high spatial resolution of 15.0 lp mm−1 is realized thanks to the reduced light scattering, which is so far a record resolution for perovskite scintillator based X‐ray imaging, and the scintillation stability is also significantly improved compared to the bare perovskite QD scintillators. Those results provide an effective platform particularly for the emerging perovskite nanocrystal scintillators to reduce light scattering and improve radiation hardness.


Supporting Information
Highly resolved and robust dynamic X-ray imaging using perovskite glass-ceramic scintillator with reduced light scattering

Materials and Methods
Sample fabrication. CsPbBr 3 :xEu QDs glass-ceramic (GC) samples were synthesized by a melting-quenching route and subsequent crystallization (heat-treatment). The GC matrix were designed with molar compositions of 35B 2 O 3 -35SiO 2 -12ZnO, and the added perovskiterelated components were 9Cs 2 CO 3 -6PbBr 2 -3NaBr-(6-x)Eu 2 O 3 (in mol%). The raw materials of B 2 O 3 (99%), SiO 2 (99.99%), ZnO(99.9%), Cs 2 CO 3 (99%), PbBr 2 (99.9%), NaBr(99.99%) and high purity Eu 2 O 3 (99.99%) were well mixed and ground into powders with an agate mortar and pestle. Then the well-ground stoichiometric compounds were put into an alumina crucible and melted at 1200 °C for 20 min in air atmosphere. After that, the melt was poured onto a 420°C pre-heated stainless steel plate and then pressed by another brass plate to form precursor glass. In addition, the glasses are annealed in a muffle furnace at 420°C for 3 hours to release the thermal stress, precursor glasses (PG) were formed. Subsequently, PG were heat-treated for 17 h at 500 °C to form transparent GCs. The Eu doped CsPbBr 3 GC products were optical polished or ground into powders for further characterization and usage.
Material characterization. The phase samples were identified via XRD measurement (D8ADVANCE/Germany Bruker X-ray diffractometer), with Cu-Kα radiation (λ=0.15405 nm) in the 2θ range from 10° to 60°. The microstructures of the GCs were analyzed by transmission electron microscopy (TEM) and high-resolution field transmission electron microscopy (HRTEM) using U.S. FEI TecnaiG2 F20 operating at 200 kV. The photo-luminescence (PL) spectra were measured with a HITACHI F-7000 fluorescence spectrophotometer using a 150 W Xe lamp as the excitation source. The afterglow and PL decay were detected using a home-setup microfluorescence system. The excitation light (515nm) was generated by femtosecond laser, (Light Conversion Pharos, 1030 nm, <300 fs, 1 MHz). TRPL decay kinetics were collected using a TCSPC module (PicoHarp 300) and a SPAD detector (IDQ, id100). The transmittance spectra and absorption spectra were recorded in the wavelength range from 200 to 800nm using a Model HITACHI U-4100 type spectrophotometer (Hitachi, Tokyo, Japan).
The X-ray attenuation efficiency (AE) could be calculated using the following formula:  fig. 2d. The X-ray absorption coefficient used for light yield calculation is at 22keV, which is the main X-ray output of our X-ray tube.
Light yield is defined as the ratio of photon numbers emitted from the luminescent sites to the total absorbed X-ray energy, it represents an internal X-ray conversion efficiency. In view of this definition, the emission photon counts of scintillators should be normalized to same X-ray attenuation (100%) as the following formula: where AE% is the X-ray attenuation efficiency (%) of scintillators with certain thickness as described in the last section. The light yield of CsPbBr 3 :xEu QDs glass-ceramic (GC) (LY perovskite ) can be calculated by the following formula: X-ray imaging. The X-ray source was Mini-X X-ray tube (target material: Ag) supplied by Amptek Inc and operated at 50 KV, exhibiting the X-ray output spectrum with both intense peak and average photon energy at about 22 KeV. The current of tube was tuned between 5 μA and 79 μA to modulate the X-ray dose rates. The objects for imaging and GC scintillators were placed perpendicular to the incident X-ray, and the scintillators were fixed just behind the objects. In order to extinguish the negative influence induced by direct radiation from Xray source to camera, a reflector was used to deflect the optical path by 90 degrees. Lastly, a CMOS camera (Photometrics 95B) was used to collect X-ray images. This camera consists of 1200×1200 pixels and the area of each pixel is 11μm × 11 μm.
MTF measurements. MTF determines the spatial resolution of imaging system and represents the ability to transfer input signal modulation of spatial frequency relative to its output. An MTF value of 1 indicates the perfect detection of a given spatial frequency. Using slanted-edge method to calculate MTF, we took the X-ray images of a sharp edge from a piece of aluminum (thickness: ~1mm), and the X-ray dose rate was 47.2 μGyair/s. Next the edge spread function (ESF) was derived by the edge profile, from which we could deduce the line spread function (LSF) by calculating derivative. Finally, the Fourier transform of the LSF defines the MTF, meaning the MTF curves could be calculated by the following formula: Where the ν is spatial frequency, x is the position of pixels. Due to using different optical system, the position of pixels (x) are defined by following formula: Where the N is the ordinal number of pixels in X-ray edge image, d is the pixel size (11 μm) and β is optical magnification.