Search for: All records

Surface structures on radio-frequency (RF) superconductors are crucially important in determining their interaction with the RF field. Here we investigate the surface compositions, structural profiles, and valence distributions of oxides, carbides, and impurities on niobium (Nb) and niobium–tin (Nb₃Sn)in situunder different processing conditions. We establish the underlying mechanisms of vacuum baking and nitrogen processing in Nb and demonstrate that carbide formation induced during high-temperature baking, regardless of gas environment, determines subsequent oxide formation upon air exposure or low-temperature baking, leading to modifications of the electron population profile. Our findings support the combined contribution of surface oxides and second-phase formation to the outcome of ultra-high vacuum baking (oxygen processing) and nitrogen processing. Also, we observe that vapor-diffused Nb₃Sn contains thick metastable oxides, while electrochemically synthesized Nb₃Sn only has a thin oxide layer. Our findings reveal fundamental mechanisms of baking and processing Nb and Nb₃Sn surface structures for high-performance superconducting RF and quantum applications.

 

Workbench-size particle accelerators, enabled by Nb₃Sn-based superconducting radio-frequency (SRF) cavities, hold the potential of driving scientific discovery by offering a widely accessible and affordable source of high-energy electrons and x-rays. Thin-film Nb₃Sn RF superconductors with high quality factors, high operation temperatures, and high-field potentials are critical for these devices. However, surface roughness, non-stoichiometry, and impurities in Nb₃Sn deposited by conventional Sn-vapor diffusion prevent them from reaching their theoretical capabilities. Here we demonstrate a seed-free electrochemical synthesis that pushes the limit of chemical and physical properties in Nb₃Sn. Utilization of electrochemical Sn pre-deposits reduces the roughness of converted Nb₃Sn by five times compared to typical vapor-diffused Nb₃Sn. Quantitative mappings using chemical and atomic probes confirm improved stoichiometry and minimized impurity concentrations in electrochemically synthesized Nb₃Sn. We have successfully applied this Nb₃Sn to the large-scale 1.3 GHz SRF cavity and demonstrated ultra-low BCS surface resistances at multiple operation temperatures, notably lower than vapor-diffused cavities. Our smooth, homogeneous, high-purity Nb₃Sn provides the route toward high efficiency and high fields for SRF applications under helium-free cryogenic operations.

 

Biobanks often contain several phenotypes relevant to diseases such as major depressive disorder (MDD), with partly distinct genetic architectures. Researchers face complex tradeoffs between shallow (large sample size, low specificity/sensitivity) and deep (small sample size, high specificity/sensitivity) phenotypes, and the optimal choices are often unclear. Here we propose to integrate these phenotypes to combine the benefits of each. We use phenotype imputation to integrate information across hundreds of MDD-relevant phenotypes, which significantly increases genome-wide association study (GWAS) power and polygenic risk score (PRS) prediction accuracy of the deepest available MDD phenotype in UK Biobank, LifetimeMDD. We demonstrate that imputation preserves specificity in its genetic architecture using a novel PRS-based pleiotropy metric. We further find that integration via summary statistics also enhances GWAS power and PRS predictions, but can introduce nonspecific genetic effects depending on input. Our work provides a simple and scalable approach to improve genetic studies in large biobanks by integrating shallow and deep phenotypes.

 

Understanding and controlling protein motion at atomic resolution is a hallmark challenge for structural biologists and protein engineers because conformational dynamics are essential for complex functions such as enzyme catalysis and allosteric regulation. Time-resolved crystallography offers a window into protein motions, yet without a universal perturbation to initiate conformational changes the method has been limited in scope. Here we couple a solvent-based temperature jump with time-resolved crystallography to visualize structural motions in lysozyme, a dynamic enzyme. We observed widespread atomic vibrations on the nanosecond timescale, which evolve on the submillisecond timescale into localized structural fluctuations that are coupled to the active site. An orthogonal perturbation to the enzyme, inhibitor binding, altered these dynamics by blocking key motions that allow energy to dissipate from vibrations into functional movements linked to the catalytic cycle. Because temperature jump is a universal method for perturbing molecular motion, the method demonstrated here is broadly applicable for studying protein dynamics.

 

Beta-phase gallium oxide ([Formula: see text]-Ga 2 O 3 ) is a promising semiconductor for high frequency, high temperature, and high voltage applications. In addition to the [Formula: see text]-phase, numerous other polymorphs exist and understanding the competition between phases is critical to control practical devices. The phase formation sequence of Ga 2 O 3 , starting from amorphous thin films, was determined using lateral-gradient laser spike annealing at peak temperatures of 500–1400 °C on 400 μs to 10 ms timescales, with transformations characterized by optical microscopy, x-ray diffraction, and transmission electron microscopy (TEM). The resulting phase processing map showed the [Formula: see text]-phase, a defect-spinel structure, first nucleating under all annealing times for temperatures from 650 to 800 °C. The cross-sectional TEM at the onset of the [Formula: see text]-phase formation showed nucleation near the film center with no evidence of heterogeneous nucleation at the interfaces. For temperatures above 850 °C, the thermodynamically stable [Formula: see text]-phase was observed. For anneals of 1–4 ms and temperatures below 1200 °C, small randomly oriented grains were observed. Large grains were observed for anneals below 1 ms and above 1200 °C, with anneals above 4 ms and 1200 °C resulting in textured films. The formation of the [Formula: see text]-phase prior to [Formula: see text]-phase, coupled with the observed grain structure, suggests that the [Formula: see text]-phase is kinetically preferred during thermal annealing of amorphous films, with [Formula: see text]-phase subsequently forming by nucleation at higher temperatures. The low surface energy of the [Formula: see text]-phase implied by these results suggests an explanation for the widely observed [Formula: see text]-phase inclusions in [Formula: see text]-phase Ga 2 O 3 films grown by a variety of synthesis methods. 

We report the use of suboxide molecular-beam epitaxy ( S-MBE) to grow β-Ga 2 O 3 at a growth rate of ∼1 µm/h with control of the silicon doping concentration from 5 × 10 16 to 10 19  cm −3 . In S-MBE, pre-oxidized gallium in the form of a molecular beam that is 99.98% Ga 2 O, i.e., gallium suboxide, is supplied. Directly supplying Ga 2 O to the growth surface bypasses the rate-limiting first step of the two-step reaction mechanism involved in the growth of β-Ga 2 O 3 by conventional MBE. As a result, a growth rate of ∼1 µm/h is readily achieved at a relatively low growth temperature ( T sub ≈ 525 °C), resulting in films with high structural perfection and smooth surfaces (rms roughness of <2 nm on ∼1 µm thick films). Silicon-containing oxide sources (SiO and SiO 2 ) producing an SiO suboxide molecular beam are used to dope the β-Ga 2 O 3 layers. Temperature-dependent Hall effect measurements on a 1 µm thick film with a mobile carrier concentration of 2.7 × 10 17  cm −3 reveal a room-temperature mobility of 124 cm 2  V −1  s −1 that increases to 627 cm 2  V −1  s −1 at 76 K; the silicon dopants are found to exhibit an activation energy of 27 meV. We also demonstrate working metal–semiconductor field-effect transistors made from these silicon-doped β-Ga 2 O 3 films grown by S-MBE at growth rates of ∼1 µm/h. 

We report the use of suboxide molecular-beam epitaxy (S-MBE) to grow β-Ga2O3 at a growth rate of ∼1 μm/h with control of the silicon doping concentration from 5 × 1016 to 1019 cm−3 . In S-MBE, pre-oxidized gallium in the form of a molecular beam that is 99.98% Ga2O, i.e., gallium suboxide, is supplied. Directly supplying Ga2O to the growth surface bypasses the rate-limiting frst step of the two-step reaction mechanism involved in the growth of β-Ga2O3 by conventional MBE. As a result, a growth rate of ∼1 μm/h is readily achieved at a relatively low growth temperature (Tsub ≈ 525 ○C), resulting in flms with high structural perfection and smooth surfaces (rms roughness of <2 nm on ∼1 μm thick flms). Silicon-containing oxide sources (SiO and SiO2) producing an SiO suboxide molecular beam are used to dope the β-Ga2O3 layers. Temperature-dependent Hall effect measurements on a 1 μm thick flm with a mobile carrier concentration of 2.7 × 1017 cm−3 reveal a room-temperature mobility of 124 cm2 V−1 s −1 that increases to 627 cm2 V −1 s−1 at 76 K; the silicon dopants are found to exhibit an activation energy of 27 meV. We also demonstrate working metal–semiconductor feld-effect transistors made from these silicon-doped β-Ga2O3 flms grown by S-MBE at growth rates of ∼1 μm/h. 

Abstract One of the core challenges in applying machine learning and artificial intelligence to medicine is the limited availability of annotated medical data. Unlike in other applications of machine learning, where an abundance of labeled data is available, the labeling and annotation of medical data and images require a major effort of manual work by expert clinicians who do not have the time to annotate manually. In this work, we propose a new deep learning technique (SLIVER-net), to predict clinical features from 3-dimensional volumes using a limited number of manually annotated examples. SLIVER-net is based on transfer learning, where we borrow information about the structure and parameters of the network from publicly available large datasets. Since public volume data are scarce, we use 2D images and account for the 3-dimensional structure using a novel deep learning method which tiles the volume scans, and then adds layers that leverage the 3D structure. In order to illustrate its utility, we apply SLIVER-net to predict risk factors for progression of age-related macular degeneration (AMD), a leading cause of blindness, from optical coherence tomography (OCT) volumes acquired from multiple sites. SLIVER-net successfully predicts these factors despite being trained with a relatively small number of annotated volumes (hundreds) and only dozens of positive training examples. Our empirical evaluation demonstrates that SLIVER-net significantly outperforms standard state-of-the-art deep learning techniques used for medical volumes, and its performance is generalizable as it was validated on an external testing set. In a direct comparison with a clinician panel, we find that SLIVER-net also outperforms junior specialists, and identifies AMD progression risk factors similarly to expert retina specialists.