Open Access
Add to BibSonomy
Abstract
The limited pattern area of periodic nanostructures limits the development of practical devices. This study introduces an X-ray interference lithography (XIL) stitching technique to fabricate a large-area (1.5 cm × 1.5 cm) two-dimensional photonic crystal (PhC) on the YAG: Ce scintillator, which functions as an encoder in a high numerical aperture optical encoding imaging system to effectively capture high-frequency information. An X-ray imaging experiment revealed a substantial 7.64 dB improvement in the signal-to-noise ratio (SNR) across a large field of view (2.6 mm × 2.6 mm) and achieved comparable or superior image quality with half the exposure dose. These findings have significant implications for advancing practical applications of X-ray imaging.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
The growing maturity of nanofabrication has reshaped the landscape of numerous fields [1–3]. Recent years have seen a tremendous interest in fabricating highly ordered nanostructures to improve device performances [4–8]. Enhanced practical performance requires the patterned area of the periodic nanostructure to satisfy the application needs [9]. Current methods for preparing periodic nanostructures include direct writing [10], interference lithography [11], and self-assembly [2]. Electron-beam lithography (EBL), a direct writing method that is traditionally implemented for fabricating 2D nanostructures [12], faces challenges in preparing large-area samples owing to its high cost and low efficiency. Chemical self-assembly, which enables the production of large-area 2D nanostructures, is limited in achieving perfect periodicity beyond several tens of microns [2]. Interference lithography stands out among these methods for generating highly ordered structures due to the strict periodicity of the interference pattern. Furthermore, compared with traditional laser-based interference lithography, synchrotron-based soft X-ray interference lithography (XIL) enhances the pattern resolution, which is attributed to the high brilliance and good coherence of the adopted source. Numerous soft XIL instruments have been constructed globally [13–16], with the continuous development of XIL technology [17,18]. With this technological progress, the XIL stitching technique has emerged as a powerful method for producing large-area periodic 2D nanostructures with a consistent periodic orientation for the fabrication of practical devices [19,20].
Scintillators, luminescent materials that emit visible light following absorption of ionizing radiation, are an important component of lens-coupled indirect X-ray imagers, which have applications in medical imaging and high-energy physics experiments [21,22]. Unfortunately, many inorganic scintillators exhibit a high refractive index (n = 1.8–2.5) for visible fluorescence [23], which leads to total reflection and a subsequent reduction in the outgoing photon yield. Specifically, the total internal reflection of outgoing fluorescence at large angles results in the loss of high-spatial-frequency information and image detail [24]. Recently, a high-spatial-frequency spectrum-enhanced reconstruction (HSFER) method was proposed to enhance crucial information pertaining to image characteristics [25]. The HSFER method involves fabrication of a large-area 2D photonic crystal (PhC) on the output surface of a scintillator that functions as an encoder. The optical information is encoded using a 2D high-density PhC and decoded using an iterative method. In this process, both mid-to-high- and high-frequency information is captured, both of which contribute to high signal-to-noise ratio (SNR) images with fidelity even under low-dose X-ray irradiation and improve noise-equivalent resolution. To fulfill diverse imaging requirements, such as a large field of view and small magnification, the HSFER method relies on the fabrication of large-area 2D PhCs.
Single-exposure patterned areas are typically limited to approximately 1 mm2 as determined by the XIL mask area. An innovative large-area exposure stitching technology was developed at the Shanghai Synchrotron Radiation Facility (SSRF), which applies an order-sorting aperture (OSA) with an in-situ monitoring scheme in the XIL system [25]. The novel technology enables the fabrication of periodic nanostructures with areas extending to cm2, which can be subsequently transferred onto different substrates through etching or lift-off processing from the photoresist pattern. To promote the practical application of the HSFER method in optimizing X-ray imaging, we fabricated a large-area (1.5 cm × 1.5 cm) 2D PhC on the YAG: Ce scintillator, which functions as an encoder in a high numerical aperture optical encoding imaging system. The performed X-ray imaging experiments demonstrated a substantial 7.64 dB improvement in SNR at 2.6 mm × 2.6 mm field of view. Furthermore, our proposed framework achieves comparable or superior image quality with only half the exposure dose, which has significant implications for nondestructive testing in biological imaging.
2. Methods
2.1 XIL stitching technique
Figure 1 illustrates the fundamental principles of the XIL stitching technique. An X-ray coherent beam of wavelength λ is diffracted by a grating mask to form periodic patterns through multi-beam interference, which are recorded on the photoresist. In the case of a four-grating mask with period Pg, the recorded nanoholes have a period of ${P}_{g}{/}\sqrt {2} $. In conventional XIL experiments, a zeroth-order diffracted beam is also recorded near the central interference region, which can affect the stitching of the interference patterns. An OSA was employed to block the zero-order diffraction of the mask grating to tightly integrate the single-exposure regions. To obtain clear imaging on a charge-coupled device (CCD), the mask and OSA were collimated with a small number of higher-order harmonic X-rays prior to exposure. The quantum efficiency of the CCD was optimized to monitor the real-time position of the zeroth-order spot (Fig. 1(b)). Subsequently, an OSA of specified size was employed to precisely block the zero-order spot, ensuring comprehensive detection of the interference region (Fig. 1(c)).
Fig. 1. (a) Schematic representation of the XIL stitching technique. (b) The position of the zero-order spot observed on the CCD. (c) The position of the interference region when the zero-order spot was completely blocked by the OSA.
Two crucial considerations are noteworthy for this XIL method. First, owing to the small angle of the first-order diffraction and the thickness of the OSA itself (300 µm – 400 µm), the tolerance for the moving distance δ of the OSA is highly constrained along the beam direction. δ can be estimated as follows:
Where Pg is the period of the grating mask, G is the distance between the grating pairs, D is the width of the grating, λ is the wavelength of the X-ray coherent beam. In the stitching experiment, a photon energy of 140 eV, surpassing the typical 92.5 eV energy, was selected to augment the tolerance of the OSA movement. Consequently, a series of masks featuring permalloy blocking layers has been devised to deal with higher photon energies [26]. However, to ensure that the zero-order light spot is completely blocked, the horizontal dimension of the OSA size (A) must surpass the interference pattern area while being smaller than G. If the OSA size is equal to or smaller than the mask grating size, it may block at least one diffracted beam from different mask gratings, causing inconsistencies between the edge and center portions of the patterns.2.2 High-spatial-frequency spectrum enhanced reconstruction method
In indirect X-ray imaging, a scintillator converts penetrating X-rays into visible light that is captured by a detector for imaging purposes. Figure 2 illustrates the two key steps of the imaging approach proposed in this paper: 1) A large-scale 2D nanostructure array was etched onto the output surface of the scintillator to enhance the extraction of high-frequency information. This modification enables the extraction of fluorescence that would otherwise be confined within the scintillator owing to total internal reflection. 2) By analyzing the influence of the large-scale 2D nanostructure array on the point spread function (PSF) of the system, the PSF of the system can be utilized for image spectrum reconstruction to yield a high-SNR image that accurately preserves high-frequency information.
Fig. 2. Schematic diagram of the HSFER method. The image within the scintillator is extracted using a large-area photonic crystal encoder, followed by decoding through a post-processing algorithm.
In the absence of the nanostructure array on the output surface of the scintillator, the output light cone is composed of output photons with angles less than the total reflection angle θc. However, with the presence of the nanostructure array, the evanescent wave on the output surface can be converted into a propagating wave, even when the output light surpasses the total reflection angle. This conversion was facilitated by the diffraction effect of the nanostructured array, which allowed light extraction.
where λ0 is the luminous wavelength in the vacuum, G0 = 2π / d is the inverse lattice vector of the PhC, and d is the period of the photonic crystal. P is an integer that indicates the visible light wave level in the scintillation body. ${K}_{{/{/}}}^ {\textrm{P}}$ is the internal light wave vector of scintillation, whose value is expressed as where nsc is the refractive index of the scintillator (nYAG: Ce = 1.82), θm is the angle between the luminous and normal directions at the light source point, and $m$ is the light wave order in the scintillation body.This proposed method improves the light extraction efficiency, broadens the output light cone, and significantly increases the amount of object light-field information obtained at the detection end. The performance of a scintillator-based detector is intricately linked to both luminescence conversion efficiency and light-extraction efficiency. Under equal luminescence conversion efficiencies, PhCs incorporated into scintillators effectively enhance light extraction. While the 2D nanostructure array increases the number of extracted photons, the light wave undergoes diffraction and is reflected by the top and bottom surfaces, forming multiple virtual object points. Factors such as fluorescence wavelength distribution, different emission point positions, and different wave vector directions defocus the image. All these influencing factors were fully considered and transformed into specific model parameters, and the Richardson-Lucy deconvolution algorithm was employed in image decoding to achieve high-fidelity reconstruction of the image.
2.3 2D PhC nanostructures fabrication
The fabrication process of the 2D PhC nanostructures on the YAG: Ce crystals is illustrated in Fig. 3. To transfer the periodic nanoholes on the YAG: Ce crystal, a 300 nm-thick silicon nitride layer was chemically deposited using vapor deposition on a 200 µm thick YAG: Ce crystal to bypass the challenges associated with etching the YAG crystal. PMMA (MicroChem PMMA A4, 950 k) was spin-coated onto a double-sided polished silicon wafer at 1500 rpm for 45 s, resulting in an approximately 300 nm thick layer. This thickness is instrumental in providing protection against subsequent non-etched areas of silicon nitride. Subsequently, pre-baking was conducted at 180 °C for 1.5 min. In the XIL experiment, the Si wafer was moved incrementally to stitch the exposed areas together. To compensate for the decrease in diffraction efficiency at the edge of the field, the step size in the stitching was chosen to be slightly smaller than the size of the mask grating. This is critical for the fabrication of large-scale uniform periodic nanostructures. Otherwise, an uncompensated exposure dose may result in blank areas with no patterns between fields. Significantly, with our current EBL equipment, it takes 11 hours to fabricate a 1 mm2 2D PhC nanostructure. In contrast, employing the XIL stitching method allows us to produce a 2.25 cm2 2D PhC nanostructure in just 12 hours, marking an efficiency improvement exceeding 200-fold.
Next, PMMA was developed in a mixture of methyl isobutyl ketone (MIBK, Aladdin Biochemical Technology, Shanghai) diluted 1:3 in IPA for 45 s, rinsed with alcohol for 30 s, and finally dried with a gentle N2 flow. Subsequently, the Si3N4 layer was etched using an inductively coupled plasma (ICP) etch system (Plasmalab ICP 180, Oxford Instruments, Bristol). Finally, the YAG: Ce crystal was soaked in acetone (Sinopharm Chemical Reagent Co., Ltd. Shanghai, China) for 10 min to remove the PMMA photoresist, and then dried with nitrogen.
3. Results and discussion
3.1 2D PhC nanostructure characteristics
Figure 4(a) shows the XIL stitching exposure results of the PMMA photoresist. The gap between fields measured less than 5 µm in both the horizontal and vertical directions, as depicted in Fig. 4(b) and Fig. 4(c). Notably, periodic nanostructures were also present in the gap area (Fig. 4(d)) with the same period as the nanostructures in the central portion of the field. Furthermore, the final appearance of the device presented in Fig. 4(e) reveals a remarkable total structural area of up to 1.5 cm × 1.5 cm. This extensive nanostructure area is sufficiently large for optical coding imaging of X-ray scintillator imagers. Atomic force microscopy (AFM) images (Fig. 4 f and 4 g) showcase a 2D nano-PhC with a lattice constant of 300 nm and a period nanostructure height of approximately 150 nm.
Fig. 4. (a). Large-area result exposure on the photoresist obtained using the XIL stitching technique; (b). Horizontal stitching gap; (c). Vertical stitching gap; (d). Nanostructures in the gap area; (e). Periodic nanostructures transferred onto a YAG: Ce crystal with a total area of 1.5 cm × 1.5 cm; (f). AFM imaging of the periodic nanostructures; (g) the corresponding height curve.
Figure 5 shows a comparison of the luminescence spectra of the YAG: Ce with and without PhC coverage under X-ray irradiation. Compared with plain YAG: Ce for reference, the YAG: Ce covered with the PhC exhibited a significantly increased light output throughout the spectrum. An enhancement of 155% was achieved at the emission peak of 525 nm. In addition, we integrated and compared X-ray excited luminescence spectra. The results demonstrate an approximately 2.3-fold intensity gain in the 2D PhC-modified YAG: Ce compared to the unmodified one, further emphasizing the enhancement achieved by the 2D PhC nanostructure.
Fig. 5. Comparison of luminescence spectra for YAG: Ce with and without PhC coverage under X-ray irradiation.
The 2D nanostructure array enhanced the diffraction of the output light field inside the scintillator as well as the extracted light-field information by the entire imaging system.
3.2 Imaging experiment
After fabrication of the 2D PhC on the YAG: Ce crystal, imaging experiments were conducted using resolution targets and zebrafish samples. Experimental data were obtained at the BL13HB and BL09B beamlines in the SSRF with an energy of 15 keV, imaging magnification of 5×, and a pixel resolution of 1.3 µm × 1.3 µm. Figure 6 provides a comparative analysis of the imaging effects for resolution targets under a large field of view (2.6 mm × 2.6 mm) with a 2 s long exposure time. The imaging results of the resolution target clearly demonstrate the advantages of utilizing soft X-ray large-area stitching technology, without which the PhC-modified scintillator can only image a small portion of the resolution target. However, the modified scintillator enables the entire resolution target area to be imaged while remaining high reconstruction image quality. While the theoretical maximum field of view for the imaging system is 1.5 cm × 1.5 cm, the size limitations of the sCMOS camera in our experiment limit the presented images to the millimeter scale. To investigate the enhancement effect of the proposed method on the image SNR, all images were normalized in terms of their contrast and brightness values and compared with images obtained using an unmodified scintillator. Figure 6(a) shows a significantly improved SNR and notably reduced noise level in images from the experimental group, obtained by fabricating large-area modified structures for image information encoding behind the scintillator coupled with digital computer deconvolution decoding, as compared to the control group images.
Fig. 6. (a) A comparison of the imaging effects for resolution targets under a large field of view (2.6 mm × 2.6 mm) with a 2 s long exposure time. The upper panel displays the image utilizing an unmodified scintillator, while the lower panel shows the image obtained through deconvolution after encoding with the PhC-modified scintillator. Scale bar: 500 µm. (b) A comparison of mid-high frequency features in the resolution target image composed of line pairs at intervals of 5 µm and 3 µm. The image in the red triangle frame represents the experimental group, and the black triangle frame represents the control group. (c) A comparison of the PSD information of the PhC-modified and unmodified images.
The adoption of large-area X-ray stitching technology has considerably broadened the imaging field of view, allowing the encoding-decoding technique to extend beyond specific micro areas and enhancing imaging across the overall resolution target. Figure 6(b) shows mid-high frequency features in the resolution target image, composed of alternating line pairs at 5 µm and 3 µm distance. Using large-area X-ray encoding and decoding technology, effective noise reduction and enhancement of mid-high-frequency feature information were observed. For 5 µm spaced line pairs, the intensity cross section shows an average 2.15-fold direct feature enhancement. The intensity cross section of the 3 µm line pairs shows seven distinct peaks, while the intensity troughs representing line brightness changes are indistinguishable in the control group cross-section. Figure 6(c) shows the PSD of the PhC-modified and unmodified images. The results show dramatic enhancement to the SNR using the proposed large-area X-ray encoding and decoding scheme within the spatial frequency range of (145 µm)−1 to (3 µm)−1, with a maximum amplification of spatial information in the mid-high frequency range of up to 7.64 dB.
Figure 7 compares zebrafish images from the 200 ms experimental group and the 100 ms control group. The results indicate that the large-area PhC-modified scintillator encoding and decoding scheme significantly reduces the high-frequency noise in the images, which results in clearer image details and a noticeable improvement in the SNR (highlighted by the red box). The enhanced image details and textures, as indicated by the enlargements in Fig. 7(c) and 7(d), imply that with half the exposure dose, similar or even superior image quality and SNR can be achieved. We employed the no-reference image quality assessment methods, perceptual image quality evaluator (PIQE [27]), and natural image quality evaluator (NIQE [28]), which are standard assessment methods for cases in which reference images are not available, to assess the quality of the images produced by the different imaging frameworks depicted in Fig. 6 and Fig. 7.
Fig. 7. Comparison of zebrafish images from the 200 ms experimental group and the 100 ms control group. (a) zebrafish images from the 200 ms experimental group. (b) zebrafish images from the 100 ms control group. Images (c) and (d) show enlargements of (a) and (b) to illustrate image details and textures.
The detailed outcomes of these assessments are presented in Table 1. Because lower values indicate higher image quality and fidelity, we can conclude that the large-area PhC-modified scintillator encoding and decoding scheme robustly improves the fidelity and quality of the images captured in large field-of-view imaging scenarios.
4. Conclusion
Soft X-ray interference lithography (XIL), which leverages the advantages of short wavelengths and high throughput, offers high efficiency and resolution in the fabrication of periodic nanostructures. In our effort to advance the practical application of X-ray imaging technology, we employed soft XIL stitching techniques to create a large-area (1.5 cm × 1.5 cm) 2D nanostructured PhC on the output surface of a YAG: Ce scintillator, which serves as an encoder for a high numerical aperture optical imaging system. In the stitching experiment, we employed a high-harmonic online observation scheme and precisely adjusted the position of an OSA of specified size with an extremely small tolerance. This facilitated the precise blocking of zero-order light and the unimpeded transmission of first-order diffracted light. By accurately controlling the movement step of the sample stage, we seamlessly stitched the interference patterns of 1 mm2 regions using a single exposure image. Notably, intact nanostructures with consistent orientations were maintained even within the microscale stitching gaps, which provided seamless integration of the patterns.
Subsequently, we validated the performance of the prepared X-ray imager at the BL13W and BL09B beamlines of the SSRF. Experimental imaging data demonstrated that, compared to plain YAG: Ce, the YAG: Ce covered with photonic crystals exhibited reduced signal noise and significantly improved imaging contrast and clarity within a large field of view (2.6 mm × 2.6 mm). The PSD results indicated a substantial increase in the SNR within the spatial frequency range of (145 µm)−1 to (3 µm)−1, achieving a maximum amplification of spatial information in the mid-high frequency range reaching up to 7.64 dB. In addition, to demonstrate the applicability of this imaging enhancement technique to specific micro areas, we conducted a comparative analysis of the features of the line pairs in a resolution target. The distinct visibility of the seven peaks in the intensity distribution of the images of the experimental group confirmed the effectiveness of the proposed technology. Furthermore, we compared the imaging quality of zebrafish in the experimental and control groups at different doses, confirming that the X-ray scintillator imaging device modified with a large-area nanophotonic crystal achieved comparable or superior image quality using only half the exposure dose. This finding has significant implications for non-destructive testing in biological imaging and advances the practical applications of X-ray imaging. We utilized the no-reference image quality assessment methods PIQE and NIQE to determine whether the encoding and decoding scheme of the large-area PhC-modified scintillator consistently enhanced the fidelity and quality of images obtained in large field-of-view imaging scenarios.
Funding
National Key Research and Development Program of China (2021YFA1601003, 2017YFA0206002, 2017YFA0403400); National Natural Science Foundation of China (11775291).
Acknowledgements
The authors are grateful to BL08U1B, BL13HB, and BL09B at the SSRF for the PhC fabrication and imaging experiments. Financial support was provided by the National Key R&D Program of China (2021YFA1601003,2017YFA0206002 and 2017YFA0403400) and National Natural Science Foundation of China (11775291).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper may be obtained from the authors upon reasonable request.
References
1. H. Altug, S.-H. Oh, S. A. Maier, et al., “Advances and applications of nanophotonic biosensors,” Nat. Nanotechnol. 17(1), 5–16 (2022). [CrossRef]
2. Z. Chai, A. Childress, and A. A. Busnaina, “Directed Assembly of Nanomaterials for Making Nanoscale Devices and Structures: Mechanisms and Applications,” ACS Nano 16(11), 17641–17686 (2022). [CrossRef]
3. P. Li, S. Chen, H. Dai, et al., “Recent advances in focused ion beam nanofabrication for nanostructures and devices: fundamentals and applications,” Nanoscale 13(3), 1529–1565 (2021). [CrossRef]
4. M. E. Stewart, C. R. Anderton, L. B. Thompson, et al., “Nanostructured Plasmonic Sensors,” Chem. Rev. 108(2), 494–521 (2008). [CrossRef]
5. H. Ko, S. Singamaneni, and V. V. Tsukruk, “Nanostructured Surfaces and Assemblies as SERS Media,” Small 4(10), 1576–1599 (2008). [CrossRef]
6. L. Wen, Z. Wang, Y. Mi, et al., “Designing Heterogeneous 1D Nanostructure Arrays Based on AAO Templates for Energy Applications,” Small 11(28), 3408–3428 (2015). [CrossRef]
7. S. F. Du, “Recent Advances in Electrode Design Based on One-Dimensional Nanostructure Arrays for Proton Exchange Membrane Fuel Cell Applications,” Engineering 7(1), 33–49 (2021). [CrossRef]
8. Z. J. Wang, D. W. Cao, R. Xu, et al., “Realizing ordered arrays of nanostructures: A versatile platform for converting and storing energy efficiently,” Nano Energy 19, 328–362 (2016). [CrossRef]
9. Y. Li and G. A. Somorjai, “Nanoscale Advances in Catalysis and Energy Applications,” Nano Lett. 10(7), 2289–2295 (2010). [CrossRef]
10. K. K. B. Hon, L. Li, and I. M. Hutchings, “Direct writing technology-Advances and developments,” CIRP Annals 57(2), 601–620 (2008). [CrossRef]
11. N. Mojarad, J. Gobrecht, and Y. Ekinci, “Interference lithography at EUV and soft X-ray wavelengths: Principles, methods, and applications,” Microelectronic Engineering 143, 55–63 (2015). [CrossRef]
12. S. Okazaki, “High resolution optical lithography or high throughput electron beam lithography: The technical struggle from the micro to the nano-fabrication evolution,” Microelectronic Engineering 133, 23–35 (2015). [CrossRef]
13. J. Wallace, Y. C. Cheng, A. Isoyan, et al., “A novel EUV exposure station for nanotechnology studies,” Nucl. Instrum. Methods Phys. Res., Sect. A 582(1), 254–257 (2007). [CrossRef]
14. N. Mojarad, D. Fan, J. Gobrecht, et al., “Broadband interference lithography at extreme ultraviolet and soft x-ray wavelengths,” Opt. Lett. 39(8), 2286 (2014). [CrossRef]
15. Y. Fukushima, Y. Yamaguchi, T. Iguchi, et al., “Development of interference lithography for 22 nm node and below,” Microelectronic Engineering 88(8), 1944–1947 (2011). [CrossRef]
16. K. Kim, J. W. Lee, B. G. Park, et al., “Investigation of correlative parameters to evaluate EUV lithographic performance of PMMA,” RSC Adv. 12(5), 2589–2594 (2022). [CrossRef]
17. N. Mojarad, M. Hojeij, L. Wang, et al., “Single-digit-resolution nanopatterning with extreme ultraviolet light for the 2.5 nm technology node and beyond,” Nanoscale 7(9), 4031–4037 (2015). [CrossRef]
18. S. Yang, C. Xue, J. Zhao, et al., “Parallel direct writing achromatic talbot lithography: a method for large-area arbitrary sub-micron periodic nano-arrays fabrication,” Nanotechnology 30(31), 315301 (2019). [CrossRef]
19. N. Mojarad, D. Kazazis, and Y. Ekinci, “Fabrication of high aspect ratio and tilted nanostructures using extreme ultraviolet and soft x-ray interference lithography,” Journal of Vacuum Science & Technology B. 39(4), 1 (2021). [CrossRef]
20. C. Xue, J. Zhao, Y. Wu, et al., “Fabrication of large-area high-aspect-ratio periodic nanostructures on various substrates by soft X-ray interference lithography,” Applied Surface Science 425, 553–557 (2017). [CrossRef]
21. H. M. Thirimanne, K. D. G. I. Jayawardena, A. J. Parnell, et al., “High sensitivity organic inorganic hybrid X-ray detectors with direct transduction and broadband response,” Nat. Commun. 9(1), 2926 (2018). [CrossRef]
22. B. Liu, Z. M. Qi, M. Gu, et al., “First-principles study of oxygen vacancies in Lu2SiO5,” J. Phys.: Condens. Matter 19(43), 436215 (2007). [CrossRef]
23. M. Moszynski, M. Kapusta, M. Mayhugh, et al., “Absolute light output of scintillators,” IEEE Trans. Nucl. Sci. 44(3), 1052–1061 (1997). [CrossRef]
24. N. V. Klassen and B. Ille, “Mechanism of additional losses of the emitted light in optically anisotropic scintillators like lead tungstate,” NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH SECTION A-ACCELERATORS SPECTROMETERS DETECTORS AND ASSOCIATED EQUIPMENT 486, 426 (2002).
25. H. Xia, Y. Wu, L. Zhang, et al., “Great enhancement of image details with high fidelity in a scintillator imager using an optical coding method,” Photonics Res. 8(7), 1079 (2020). [CrossRef]
26. J. Zhao, S. Yang, C. Xue, et al., “The recent development of soft x-ray interference lithography in SSRF,” Int. J. Extrem. Manuf. 2(1), 012005 (2020). [CrossRef]
27. N. Venkatanath, D. Praneeth, M. C. Bh, et al., “BLIND IMAGE QUALITY EVALUATION USING PERCEPTION BASED FEATURES,” 2015 TWENTY FIRST NATIONAL CONFERENCE ON COMMUNICATIONS (NCC); 2015. [CrossRef]
28. A. Mittal, R. Soundararajan, and A. C. Bovik, “Making a “Completely Blind” Image Quality Analyzer,” ISPL. 20, 209 (2013).
