Search NASA⌕ Search

SEARCH · Search NASA

Results for “Mask”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 919 records · Page 51

Enhancing molecular design efficiency: Uniting language models and generative networks with genetic algorithms

This study examines the effectiveness of generative models in drug discovery, material science, and polymer science, aiming to overcome constraints associated with traditional inverse design methods relying on heuristic rules. Generative models generate synthetic data resembling real data, enabling deep learning model training without extensive labeled datasets. They prove valuable in creating virtual libraries of molecules for material science and facilitating drug discovery by generating molecules with specific properties. While generative adversarial networks (GANs) are explored for these purposes, mode collapse restricts their efficacy, limiting novel structure variability. To address this, we introduce a masked language model (LM) inspired by natural language processing. Although LMs alone can have inherent limitations, we propose a hybrid architecture combining LMs and GANs to efficiently generate new molecules, demonstrating superior performance over standalone masked LMs, particularly for smaller population sizes. This hybrid LM-GAN architecture enhances efficiency in optimizing properties and generating novel samples.

97 MATHEMATICS AND COMPUTING↗

Computer vision models enable mixed linear modeling to predict arbuscular mycorrhizal fungal colonization using fungal morphology

Abstract The presence of Arbuscular Mycorrhizal Fungi (AMF) in vascular land plant roots is one of the most ancient of symbioses supporting nitrogen and phosphorus exchange for photosynthetically derived carbon. Here we provide a multi-scale modeling approach to predict AMF colonization of a worldwide crop from a Recombinant Inbred Line (RIL) population derived from Sorghum bicolor and S. propinquum . The high-throughput phenotyping methods of fungal structures here rely on a Mask Region-based Convolutional Neural Network (Mask R-CNN) in computer vision for pixel-wise fungal structure segmentations and mixed linear models to explore the relations of AMF colonization, root niche, and fungal structure allocation. Models proposed capture over 95% of the variation in AMF colonization as a function of root niche and relative abundance of fungal structures in each plant. Arbuscule allocation is a significant predictor of AMF colonization among sibling plants. Arbuscules and extraradical hyphae implicated in nutrient exchange predict highest AMF colonization in the top root section. Our work demonstrates that deep learning can be used by the community for the high-throughput phenotyping of AMF in plant roots. Mixed linear modeling provides a framework for testing hypotheses about AMF colonization phenotypes as a function of root niche and fungal structure allocations.

59 BASIC BIOLOGICAL SCIENCES↗

Fabrication of compound refractive lens using soft and deep X-ray lithography beamline on Indus-2

X-ray compound refractive lenses (CRL) have been fabricated using Soft and Deep X-ray lithography (SDXRL) beamline on Indus-2. Upto 50 parabolic lenses are fabricated in PMMA using 1X X-ray mask. The mask contains of 20 micron thick gold absorbing pattern on polyimide membrane. The lens fabrication is done by exposing 500 {mu}m thick PMMA sheets to 5-12keV X-rays from Indus-2. The fabricated CRL will be suitable for generating focused X-ray beam of Indus-2 bending magnet source.

43 PARTICLE ACCELERATORS↗

Highly efficient and controllable method to fabricate ultrafine metallic nanostructures

We report a highly efficient, controllable and scalable method to fabricate various ultrafine metallic nanostructures in this paper. The method starts with the negative poly-methyl-methacrylate (PMMA) resist pattern with line-width superior to 20 nm, which is obtained from overexposing of the conventionally positive PMMA under a low energy electron beam. The pattern is further shrunk to sub-10 nm line-width through reactive ion etching. Using the patter as a mask, we can fabricate various ultrafine metallic nanostructures with the line-width even less than 10 nm. This ion tailored mask lithography (ITML) method enriches the top-down fabrication strategy and provides potential opportunity for studying quantum effects in a variety of materials.

36 MATERIALS SCIENCE↗

Analysis of rapidly synthesized guest-filled porous complexes with synchrotron radiation: practical guidelines for the crystalline sponge method

This report describes complete practical guidelines and insights for the crystalline sponge method, which have been derived through the first use of synchrotron radiation on these systems, and includes a procedure for faster synthesis of the sponges. These guidelines will be applicable to crystal sponge data collected at synchrotrons or in-house facilities, and will allow researchers to obtain reliable high-quality data and construct chemically and physically sensible models for guest structural determination. A detailed set of synthetic and crystallographic guidelines for the crystalline sponge method based upon the analysis of expediently synthesized crystal sponges using third-generation synchrotron radiation are reported. The procedure for the synthesis of the zinc-based metal–organic framework used in initial crystal sponge reports has been modified to yield competent crystals in 3 days instead of 2 weeks. These crystal sponges were tested on some small molecules, with two being unexpectedly difficult cases for analysis with in-house diffractometers in regard to data quality and proper space-group determination. These issues were easily resolved by the use of synchrotron radiation using data-collection times of less than an hour. One of these guests induced a single-crystal-to-single-crystal transformation to create a larger unit cell with over 500 non-H atoms in the asymmetric unit. This led to a non-trivial refinement scenario that afforded the best Flack x absolute stereochemical determination parameter to date for these systems. The structures did not require the use of PLATON/SQUEEZE or other solvent-masking programs, and are the highest-quality crystalline sponge systems reported to date where the results are strongly supported by the data. A set of guidelines for the entire crystallographic process were developed through these studies. In particular, the refinement guidelines include strategies to refine the host framework, locate guests and determine occupancies, discussion of the proper use of geometric and anisotropic displacement parameter restraints and constraints, and whether to perform solvent squeezing/masking. The single-crystal-to-single-crystal transformation process for the crystal sponges is also discussed. The presented general guidelines will be invaluable for researchers interested in using the crystalline sponge method at in-house diffraction or synchrotron facilities, will facilitate the collection and analysis of reliable high-quality data, and will allow construction of chemically and physically sensible models for guest structural determination.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

SU-E-T-211: Induced Release of Nanocarrier Encapsulated Chemotherapeutic Drugs Using Proton Radiotherapy Beams

Purpose: The purpose of this work was to develop and validate an open source independent MU dose calculation software for S and S IMRT based in the algorithm proposed by Kung et.al. Methods: Treatment plans were done using Iplan v4.5 BrainLAB TPS and S and S IMRT modality. A 6MV photon beam produced by a Primus linear accelerator equipped with an Optifocus MLC was used. TPS dose calculation algorithms were pencil beam and Monte Carlo. 230 IMRT treatments plans were selected for the study. The software was written under MALTLAB environment. Treatment plans were imported by the software using RTP format. Field fluences were reconstructed adding all segments.The algorithm implemented in the software calculates the dose at a reference point as the sum of primary and scatter dose. The primary dose is obtained by masking the fluence map with a circle of radius 1cm. The scatter dose is obtained through a shaped ring mask around the previous circle with a thickness of 0.5cm; the rings are increased one after another with constant thickness until cover the entire map of influence. The dosimetric parameters Sc, Sp and TPR vary depending on radio, the transmission effect of the MLC, inverse square law and dose profile are used for the calculation. Results: The average difference between measured and independent calculated dose was 0.4% ± 2.2% [−6.8%, 6.4%]. For 91% of the studied plans the difference was less than 3%. The difference between the measured and TPS dose with pencilbeam algorithm was 2.6% ± 1.41% [−2.0%, 5.6%] and Monte Carlo algorithm was 0.4% ± 1.5% [−4.9%, 3.7%]. The differences obtained are comparable to that obtained with the ionization chamber and TPS. Conclusion: The developed software is suitable for use in S and S IMRT dose calculation. This application is open and can be downloading under request.

60 APPLIED LIFE SCIENCES↗

Adaptive language model training for molecular design

Abstract The vast size of chemical space necessitates computational approaches to automate and accelerate the design of molecular sequences to guide experimental efforts for drug discovery. Genetic algorithms provide a useful framework to incrementally generate molecules by applying mutations to known chemical structures. Recently, masked language models have been applied to automate the mutation process by leveraging large compound libraries to learn commonly occurring chemical sequences (i.e., using tokenization) and predict rearrangements (i.e., using mask prediction). Here, we consider how language models can be adapted to improve molecule generation for different optimization tasks. We use two different generation strategies for comparison, fixed and adaptive. The fixed strategy uses a pre-trained model to generate mutations; the adaptive strategy trains the language model on each new generation of molecules selected for target properties during optimization. Our results show that the adaptive strategy allows the language model to more closely fit the distribution of molecules in the population. Therefore, for enhanced fitness optimization, we suggest the use of the fixed strategy during an initial phase followed by the use of the adaptive strategy. We demonstrate the impact of adaptive training by searching for molecules that optimize both heuristic metrics, drug-likeness and synthesizability, as well as predicted protein binding affinity from a surrogate model. Our results show that the adaptive strategy provides a significant improvement in fitness optimization compared to the fixed pre-trained model, empowering the application of language models to molecular design tasks.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

CHESS 2025: Spectrometer orthorectified at-sensor radiance from NEON AOP imaging spectroscopy surveys

This dataset provides Level 1 (L1) orthorectified at-sensor radiance derived from measurements collected by the Imaging Spectrometer-1 (NIS-1) onboard the NEON (National Ecological Observatory Network) Airborne Observation Platform (AOP) for the 2025 Colorado Headwaters Ecological Spectroscopy Study (CHESS). NIS-1 captures light reflected from the Earth’s surface in 426 discrete wavelength bands as raw digital numbers (DNs; Level 0). These data are then calibrated to physical units (uW/cm²·sr·nm) following the processing steps described in the NEON Imaging Spectrometer Level 1B Calibrated Radiance Algorithm Theoretical Basis Document (ATBD; Gallery 2022). The data delivered here are the primary inputs for the surface reflectance product in “Custom surface reflectance, shade masks, and equivalent water thickness maps for the Colorado Headwaters Ecological Spectroscopy Study” (Carroll et al. 2026). For intertemporal comparison, the radiance data here are most directly relatable to the v2 radiance data in “NEON AOP Imaging Spectroscopy Survey of Upper East River Colorado Watersheds: Raw-Space Radiance and Observational Variable Dataset” (Goulden et al. 2018), to which the same processing methodology was applied. Together, the radiance and reflectance data enable users to exploit the unique reflection signatures of different surface objects for land cover classification, foliar trait mapping, plant vigor assessment, water content estimation, trace-element identification, and other scientific applications. The data were acquired over three study domains in the Upper Gunnison river basin: the upper East River watershed (CRBU); Almont Triangle and Taylor Canyon (ALMO); and Upper Taylor River watershed (UPTA) between 2025-06-13 and 2025-07-15. Within each domain, data are delivered by flightline as orthorectified and calibrated hyperspectral rasters in Hierarchical Data Format version 5 (HDF5) format, with radiance values provided in uW/cm²·sr·nm on a fixed, uniform Universal Transverse Mercator (UTM) grid at 1 meter spatial resolution. The radiance rasters include all 426 NIS-1 spectral bands, along with associated quality-assurance (QA) and diagnostic and ancillary layers needed for atmospheric correction workflows. Orthorectified radiance is produced from pushbroom spectrometer observations by applying NEON’s radiometric calibration (including bad pixel masking, dark subtract, dark pedestal shift correction, electronic panel ghost correction, grating ghost correction, deblur correction and flat-fielding) and spectral calibration (using spectral response function band centers and full-width at half-maximum intensity), followed by geolocation and regridding to the fixed grid. CHESS Project Description: The Colorado Headwaters Ecological Spectroscopy Study (CHESS) comprised a multi-week airborne remote sensing and field observation campaign in the Upper Gunnison Basin, Colorado, conducted in June and July of 2025. Airborne remote sensing was conducted by the National Ecological Observatory Network Airborne Observation Platform (NEON AOP), concurrent with a field campaign run by the Rocky Mountain Biological Laboratory (RMBL), the Lawrence Berkeley National Laboratory (LBNL) and SLAC National Accelerator Laboratory Watershed Function Science Focus Area (SFA), and NASA-JPL (Jet Propulsion Laboratory) Earth Surface Mineral Dust Source Investigation (EMIT) program. Between June 10 and July 18, 2025, the NEON AOP flight team collected high-resolution aerial imaging spectroscopy and Light Detection and Ranging (LiDAR) data over three domains: the Upper East River (CRBU), Almont Triangle (ALMO), and the Upper Taylor Basin (UPTA). In coordination with the flights, a field campaign acquired ground-truth observations, including observations of vegetation composition, foliar traits, forest demography, and subsurface properties in 18 core sampling areas within the domains. Additional surface water observations were taken at over 380 point locations. All CHESS campaign datasets can be found within the CHESS ESS-DIVE data portal: https://data.ess-dive.lbl.gov/portals/chess. Funding Acknowledgement: Field and remote-sensing data acquisition was performed under a grant from the National Aeronautics and Space Administration (80NSSC24K1005). This work was also supported by the Watershed Function Science Focus Area at Lawrence Berkeley National Laboratory funded by the US Department of Energy, Office of Science, Biological and Environmental Research under Contract No. DE-AC02-05CH11231.

2018 NEON and 2025 CHESS Campaigns↗

Event Localization in Bulk Scintillator Crystals Using Coded Apertures

The localization of radiation interactions in bulk scintillators is generally limited by the size of the light distribution at the readout surface of the crystal/light-pipe system. By finding the centroid of the light spot, which is typically of order centimeters across, practical single-event localization is limited to ~2 mm/cm of crystal thickness. Similar resolution can also be achieved for the depth of interaction by measuring the size of the light spot. Through the use of near-field coded-aperture techniques applied to the scintillation light, light transport simulations show that for 3-cm-thick crystals, more than a five-fold improvement (millimeter spatial resolution) can be achieved both laterally and in event depth. At the core of the technique is the requirement to resolve the shadow from an optical mask placed in the scintillation light path between the crystal and the readout. In this paper, experimental results are presented that demonstrate the overall concept using a 1D shadow mask, a thin-scintillator crystal and a light pipe of varying thickness to emulate a 2.2-cm-thick crystal. Spatial resolutions of ~ 1 mm in both depth and transverse to the readout face are obtained over most of the crystal depth.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Diagnosing Malaria Patients with Plasmodium falciparum and vivax Using Deep Learning for Thick Smear Images

We propose a new framework, PlasmodiumVF-Net, to analyze thick smear microscopy images for a malaria diagnosis on both image and patient-level. Our framework detects whether a patient is infected, and in case of a malarial infection, reports whether the patient is infected by Plasmodium falciparum or Plasmodium vivax. PlasmodiumVF-Net first detects candidates for Plasmodium parasites using a Mask Regional-Convolutional Neural Network (Mask R-CNN), filters out false positives using a ResNet50 classifier, and then follows a new approach to recognize parasite species based on a score obtained from the number of detected patches and their aggregated probabilities for all of the patient images. Reporting a patient-level decision is highly challenging, and therefore reported less often in the literature, due to the small size of detected parasites, the similarity to staining artifacts, the similarity of species in different development stages, and illumination or color variations on patient-level. We use a manually annotated dataset consisting of 350 patients, with about 6000 images, which we make publicly available together with this manuscript. Our framework achieves an overall accuracy above 90% on image and patient-level.

60 APPLIED LIFE SCIENCES↗

Time-intensity relations in binaural unmasking.

Variations in binaural-masked threshold of 500-cps tone masked by random noise as function of simultaneous shifts in interaural amplitude ratio and time delay of tone

AUDITORY SENSATION AREA↗

MISR Level 2 TOA/Cloud Stereo parameters (MIL2TCST_V1)

The MISR Top-of-Atmosphere (TOA)/Cloud Stereo geophysical parameters include stereoscopically-derived cloud motion vectors (winds), cloud-top heights, and an accompanying cloud mask. The Stereo product geophysical parameters include a stereoscopically-derived cloud mask and cloud height on a 1.1 km grid. It also includes cloud motion vectors on a 70.4 km grid. The three types of stereo heights are: the BestWind heights are only calculated for those regions where the associated wind vectors passed the quality tests. Therefore, they have sparse coverage but since the wind correction is included, these contain our 'best guess' as to what the true heights are. The WithoutWind heights are calculated assuming a constant wind vector of zero. They have almost complete coverage and therefore form a nice 'pretty picture' of the relative cloud heights over small areas. The RawWind heights are a diagnostic product as they are calculated using all available wind vectors (even the bad ones). It is therefore recommended that one only use the Best and Without wind products. It is important to remember that the stereo matchers pick up the layer of maximum contrast, which is not necessarily the same as the highest cloud so all the stereo heights are keyed to this level of maximum contrast. Therefore, higher and thinner cirrus layers may not be detected by any of the height fields. [Location=GLOBAL] [Temporal_Coverage: Start_Date=2000-02-24; Stop_Date=] [Spatial_Coverage: Southernmost_Latitude=-90; Northernmost_Latitude=90; Westernmost_Longitude=-180; Easternmost_Longitude=180] [Data_Resolution: Latitude_Resolution=1.1 km; Longitude_Resolution=1.1 km; Horizontal_Resolution_Range=1 km - < 10 km or approximately .01 degree - < .09 degree; Temporal_Resolution=about 15 orbits/day; Temporal_Resolution_Range=Daily - < Weekly].

EOSDIS↗

MISR Level 2 TOA/Cloud Stereo parameters (MIL2TCST_V2)

The MISR Top-of-Atmosphere (TOA)/Cloud Stereo geophysical parameters include stereoscopically-derived cloud motion vectors (winds), cloud-top heights, and an accompanying cloud mask. The Stereo product geophysical parameters include a stereoscopically-derived cloud mask and cloud height on a 1.1 km grid. It also includes cloud motion vectors on a 70.4 km grid. The three types of stereo heights are: the BestWind heights are only calculated for those regions where the associated wind vectors passed the quality tests. Therefore, they have sparse coverage but since the wind correction is included, these contain our 'best guess' as to what the true heights are. The WithoutWind heights are calculated assuming a constant wind vector of zero. They have almost complete coverage and therefore form a nice 'pretty picture' of the relative cloud heights over small areas. The RawWind heights are a diagnostic product as they are calculated using all available wind vectors (even the bad ones). It is therefore recommended that one only use the Best and Without wind products. It is important to remember that the stereo matchers pick up the layer of maximum contrast, which is not necessarily the same as the highest cloud so all the stereo heights are keyed to this level of maximum contrast. Therefore, higher and thinner cirrus layers may not be detected by any of the height fields. [Temporal_Coverage: Start_Date=2000-02-24; Stop_Date=] [Spatial_Coverage: Southernmost_Latitude=-90; Northernmost_Latitude=90; Westernmost_Longitude=-180; Easternmost_Longitude=180] [Data_Resolution: Latitude_Resolution=1.1 km; Longitude_Resolution=1.1 km; Temporal_Resolution=about 15 orbits/day].

CLOUD MASK↗

MODIS Collection 6 Clear Sky Restoral (CSR): Filtering Cloud Mast 'Not Clear' Pixels

Correctly identifying cloudy pixels appropriate for the MOD06 cloud optical and microphysical property retrievals is accomplished in large part using results from the MOD35 1km cloud mask tests (note there are also two 250m subpixel cloud mask tests that can convert the 1km cloudy designations to clear sky). However, because MOD35 is by design clear sky conservative (i.e., it identifies "not clear" pixels), certain situations exist in which pixels identified by MOD35 as "cloudy" are nevertheless likely to be poor retrieval candidates. For instance, near the edge of clouds or within broken cloud fields, a given 1km MODIS field of view (FOV) may in fact only be partially cloudy. This can be problematic for the MOD06 retrievals because in these cases the assumptions of a completely overcast homogenous cloudy FOV and 1-dimensional plane-parallel radiative transfer no longer hold, and subsequent retrievals will be of low confidence. Furthermore, some pixels may be identified by MOD35 as "cloudy" for reasons other than the presence of clouds, such as scenes with thick smoke or lofted dust, and should therefore not be retrieved as clouds. With such situations in mind, a Clear Sky Restoral (CSR) algorithm was introduced in C5 that attempts to identify pixels expected to be poor retrieval candidates. Table 1 provides SDS locations for CSR and partly cloudy (PCL) pixels.

cloud mask↗

A Systematic Review of Local to Regional Yield Forecasting Approaches and Frequently Used Data Resources

Forecasting crop yields, or providing an expectation of ex-ante harvest amounts, is highly relevant to the whole agricultural production chain. Farmers can adapt their management, traders or insurers their pricing schemes, suppliers their stocks, logistic companies their routes, national authorities their food balance sheets to guide import or export and, finally, international aid organizations can mobilize reliefs. Evidence has grown in the literature that such forecasts with a meaningful lead time are possible on various geographic scales and for a broad range of crops. Here, we present a systematic review of the methods applied in end-of-season yield forecasting and three frequently used data sources: weather data, satellite data and crop masks. Our literature database comprises 362 studies (2004–2019) which were evaluated regarding methods, crops, regions, data sources, lead time and performance. Moreover, we present 24 sources of real-time and predictive weather data, 21 sources of remote sensing data and 16 crop masks. Yield forecasting in our literature sample has been performed for 44 crops in 71 countries, also including many non-staple crops, but with an apparent bias in regions and crops. Forecasting performance depends on various factors, including crop, region, method, lead time to harvest and input diversity. Our systematic review supports a broader application of locally successful approaches at larger scales by providing a comprehensive, accessible compendium of necessary information for yield forecasting. We discuss improvement potentials with respect to methodological approaches and available data sources. We additionally suggest standardization procedures for future forecasting studies and encourage studying additional crops and geographic regions. Implications of forecasts for different target groups on different scales and the adaptation towards climate change are also discussed.

Seasonal crop yield forecasting↗

Comparison of Cloud Detection Algorithms for Sentinel-2 Imagery

Accurate, automated cloud and cloud shadow detection is a key component of the processing needed to prepare optical satellite imagery for scientific analysis. Many existing cloud detection algorithms rely on temperature information to identify clouds, making detection difficult for imagers that lack a thermal band, like Sentinel-2. To get maximum benefit from Sentinel-2 products it is critical to understand which algorithms best identify clouds and their shadows in images. We examined the relative performance of five different cloud-masking algorithms (Sen2Cor, MAJA, LaSRC, Fmask and Tmask) in 6 Sentinel-2 scenes (28 total images) distributed across the Eastern Hemisphere. Expanding on these comparisons, we tested ensemble approaches to improve results. We tested three ensemble approaches to cloud and shadow classification based on the outputs of the five initial algorithms using the cloud masks in: (1) a majority prediction model; (2) a random forests model; and (3) a conditional logic model. Accuracy assessments show a trade-off between omission and commission errors in cloud detection for individual algorithms across all sites, and some algorithms are better at detecting either clouds or cloud shadows. No single algorithm outperforms the others for both clouds and shadows. Aggregating the results from multiple algorithms produces fewer undetected clouds and higher overall accuracy than any single algorithm, with as high as 2.7% improvement over the top-performing algorithm, suggesting an ensemble approach may be the most useful for processing of Sentinel-2 data.

Sentinel-2↗

Markovian Statistical Model of Cloud Optical Thickness. Part I: Theory and Examples

We present a generalization of the binary-value Markovian model previously used for statistical characterization of cloud masks to a continuous-value model describing 1D fields of cloud optical thickness (COT). This model has simple functional expressions and is specified by four parameters: the cloud fraction, the autocorrelation (scale) length, and the two parameters of the normalized probability density function of (non-zero) COT values (this PDF is assumed to have gamma-distribution form). Cloud masks derived from this model by separation between the values above and below some threshold in COT appear to have the same statistical properties as in binary-value model described in our previous publications. We demonstrate the ability of our model to generate examples of various cloud-field types by using it to statistically imitate actual cloud observations made by the Research Scanning Polarimeter (RSP) during two field experiments.

binary-value Markovian model↗

2D Exotic Optical Lattice via a Digital‐Coding Circular Airy Beam

Optical lattices have been widely used from classical to quantum physics. The tunable and scalable fabrication of lattices would be of great significance in lattice‐based multipartite applications. This work demonstrates first that a circular Airy beam (CAB), which has the peculiar properties of self‐healing and abrupt autofocusing, can be used to generate two‐dimensional (2D) optical lattices in propagation when encoded by a programmable spatial mask, resulting in the formation of large‐scale and tunable optical lattices with both axis and axial symmetry, and even high‐orbital kaleidoscope shapes. The efficient diffraction of CAB during the spatial crosstalk with the mask enables the realization of tunable lattices with rich periodicity and complexity. The study shows a flexible method to manipulate lattices with large‐scale and versatile structures for potential applications in integrated and scalable optical and photonic devices.

Sun, Peisheng↗