Development of broad-band high-reflectivity multilayer film for positron emission tomography system

Abstract

The use of non-ideal reflective materials and low-precision manual manufacturing technologies is a fundamental technical obstacle blocking the positron emission tomography (PET) systems from achieving better performances. We propose to address that long-standing obstacle with advanced multilayer dielectric coating technologies. We designed an broad-band multilayer high-reflectivity (HR) film that can be coated directly on the surface of ultra-precision polished lutetiumyttrium oxyorthosilicate (LYSO) scintillators. The film consists of 48 layers of TiO2/SiO2/HfO2 which are deposited on LYSO scintillator crystal alternately using the electron beam evaporation method. The overall thickness of the HR film is about 3μm. The HR film combines 3 quarter wavelength reflective films, with the central wavelengths of 365 nm, 430 nm and 570 nm respectively, to match the emission spectrum of the LYSO scintillator. The optical experimental results show that the HR film achieved an excellent average reflectivity of 99.50% at 8° incident angle for incident lights with wavelengths between 360 to 620 nm. The average reflectivity at 60° incident angle is higher than 90%. The results of the hardness experiments and the adhesive strength experiments show that the HR film has an excellent mechanical strength. The HR coating technology developed in this study is very attractive because it allows to "print" high-performance reflectors on a scintillator directly with high-precision, instead of manually gluing reflective films on the scintillator. Thus, we conclude that the HR film provides a viable solution to the long standing technical bottleneck that limits the development of high-performance detectors for advanced PET imaging.

Keywords: Detector design and construction technologies and materials; Gamma detectors (scintillators, CZT, HPG, HgI etc); Mirror coating; Optics.

Figure 1.

(a) The refractive index of LYSO. Three samples (20 mm × 30 mm × 1 mm in size) were measured. The average of the three measurements are represented by the solid curve. (b) The transmittance of a 4 mm thick LYSO sample.

Figure 2.

The theoretic spectral reflectance of the 48-layer HR film at normal incidence.

Figure 3.

(a) The polished LYSO samples. The large sample is 20mm × 30mm × 1 mm in size. The small one is 5 mm × 6 mm × 30mm in size. (b) The ZYGO image of a polished crystal surface. (c) The AFM image of a polished crystal surface.

Figure 4.

(a) The SEM image of the cross section of the tri-layer film on a Silicon substrate. The AFM surface mapping of (b) the first layer (Ta₂O₅), (c) the second layer (SiO₂) and (d) the third layer (Ta₂O₅) of the tri-layer film.

Figure 5.

(a) Picture of an uncoated sample and two coated samples. (b) Pictures of an ESR file, a coated JGS1 fused Silica, a coated LYSO sample and a coated K9 glass slide. (c) SEM image of the cross-section of the 48-layer dielectric film on a LYSO substrate. (d) The AFM image of the HR film surface.

Figure 6.

(a) The measured reflectance spectrum of the HR film deposited on LYSO samples. The incident angles are 0°, 8°, 15°, 45° and 60° respectively. (b) Comparison of the simulated and the measured reflectance spectrums at the 0° incident angle.

Figure 7.

(a) The surface image of indentation with maximum indentation depth set at 215 nm. (b) The profile of indentation at the position marked by the blue line in (a). (c) The nanohardness and (d) the elastic modulus measured at different depths of indentation.

Figure 8.

The results of the interfacial adhesion strength experiments. X-axis and Y-axis represent the loading force on the tip and the friction while the tip moved on the surface of the HR coated samples. The scratching sound signals were represented by the red pulses in the plot.