Nanoscale Three-Dimensional Imaging of Integrated Circuits Using a Scanning Electron Microscope and Transition-Edge Sensor Spectrometer.

X-ray nanotomography is a powerful tool for the characterization of nanoscale materials and structures, but it is difficult to implement due to the competing requirements of X-ray flux and spot size. Due to this constraint, state-of-the-art nanotomography is predominantly performed at large synchrotron facilities. We present a laboratory-scale nanotomography instrument that achieves nanoscale spatial resolution while addressing the limitations of conventional tomography tools. The instrument combines the electron beam of a scanning electron microscope (SEM) with the precise, broadband X-ray detection of a superconducting transition-edge sensor (TES) microcalorimeter. The electron beam generates a highly focused X-ray spot on a metal target held micrometers away from the sample of interest, while the TES spectrometer isolates target photons with a high signal-to-noise ratio. This combination of a focused X-ray spot, energy-resolved X-ray detection, and unique system geometry enables nanoscale, element-specific X-ray imaging in a compact footprint. The proof of concept for this approach to X-ray nanotomography is demonstrated by imaging 160 nm features in three dimensions in six layers of a Cu-SiO2 integrated circuit, and a path toward finer resolution and enhanced imaging capabilities is discussed.

Noise-resilient deep learning for integrated circuit tomography.

X-ray tomography is a non-destructive imaging technique that reveals the interior of an object from its projections at different angles. Under sparse-view and low-photon sampling, regularization priors are required to retrieve a high-fidelity reconstruction. Recently, deep learning has been used in X-ray tomography. The prior learned from training data replaces the general-purpose priors in iterative algorithms, achieving high-quality reconstructions with a neural network. Previous studies typically assume the noise statistics of test data are acquired a priori from training data, leaving the network susceptible to a change in the noise characteristics under practical imaging conditions. In this work, we propose a noise-resilient deep-reconstruction algorithm and apply it to integrated circuit tomography. By training the network with regularized reconstructions from a conventional algorithm, the learned prior shows strong noise resilience without the need for additional training with noisy examples, and allows us to obtain acceptable reconstructions with fewer photons in test data. The advantages of our framework may further enable low-photon tomographic imaging where long acquisition times limit the ability to acquire a large training set.

Photon echoes using atomic frequency combs in Pr:YSO - experiment and semiclassical theory.

Photon echoes in rare-earth-doped crystals are studied to understand the challenges of making broadband quantum memories using the atomic frequency comb (AFC) protocol in systems with hyperfine structure. The hyperfine structure of Pr3+ poses an obstacle to this goal because frequencies associated with the hyperfine transitions change the simple picture of modulation at an externally imposed frequency. The current work focuses on the intermediate case where the hyperfine spacing is comparable to the comb spacing, a challenging regime that has recently been considered. Operating in this regime may facilitate storing quantum information over a larger spectral range in such systems. In this work, we prepare broadband AFCs using optical combs with tooth spacings ranging from 1 MHz to 16 MHz in fine steps, and measure transmission spectra and photon echoes for each. We predict the spectra and echoes theoretically using the optical combs as input to either a rate equation code or a density matrix code, which calculates the redistribution of populations. We then use the redistributed populations as input to a semiclassical theory using the frequency-dependent dielectric function. The two sets of predictions each give a good, but different account of the photon echoes.

Linearity Characterization and Uncertainty Quantification of Spectroradiometers via Maximum Likelihood and the Non-parametric Bootstrap.

A technique for characterizing and correcting the linearity of radiometric instruments is known by the names the "flux-addition method" and the "combinatorial technique". In this paper, we develop a rigorous uncertainty quantification method for use with this technique and illustrate its use with both synthetic data and experimental data from a "beam conjoiner" instrument. We present a probabilistic model that relates the instrument readout to a set of unknown fluxes via a set of polynomial coefficients. Maximum likelihood estimates (MLEs) of the unknown fluxes and polynomial coefficients are recommended, while a non-parametric bootstrap algorithm enables uncertainty quantification including standard errors and confidence intervals. The synthetic data represent plausible outputs of a radiometric instrument and enable testing and validation of the method. The MLEs for these data are found to be approximately unbiased, and confidence intervals derived from the bootstrap replicates are found to be consistent with their target coverage of 95%. For the polynomial coefficients, the observed coverages range from 91% to 99%. The experimental data set illustrates how a complete calibration with uncertainties can be achieved using the method plus one well-known flux level. The uncertainty contribution attributable to estimation of the instrument's nonlinear response is less than 0.025% over most of its range.

Physics-assisted generative adversarial network for X-ray tomography.

X-ray tomography is capable of imaging the interior of objects in three dimensions non-invasively, with applications in biomedical imaging, materials science, electronic inspection, and other fields. The reconstruction process can be an ill-conditioned inverse problem, requiring regularization to obtain satisfactory results. Recently, deep learning has been adopted for tomographic reconstruction. Unlike iterative algorithms which require a distribution that is known a priori, deep reconstruction networks can learn a prior distribution through sampling the training distributions. In this work, we develop a Physics-assisted Generative Adversarial Network (PGAN), a two-step algorithm for tomographic reconstruction. In contrast to previous efforts, our PGAN utilizes maximum-likelihood estimates derived from the measurements to regularize the reconstruction with both known physics and the learned prior. Compared with methods with less physics assisting in training, PGAN can reduce the photon requirement with limited projection angles to achieve a given error rate. The advantages of using a physics-assisted learned prior in X-ray tomography may further enable low-photon nanoscale imaging.

X-ray computed tomography using partially coherent Fresnel diffraction with application to an optical fiber.

A reconstruction algorithm for partially coherent x-ray computed tomography (XCT) including Fresnel diffraction is developed and applied to an optical fiber. The algorithm is applicable to a high-resolution tube-based laboratory-scale x-ray tomography instrument. The computing time is only a few times longer than the projective counterpart. The algorithm is used to reconstruct, with projections and diffraction, a tilt series acquired at the micrometer scale of a graded-index optical fiber using maximum likelihood and a Bayesian method based on the work of Bouman and Sauer. The inclusion of Fresnel diffraction removes some reconstruction artifacts and use of a Bayesian prior probability distribution removes others, resulting in a substantially more accurate reconstruction.

Deep Brain Stimulation Generator Failure due to External Defibrillation in a Patient with Essential Tremor.

There exist only two case reports to date of open cardiac defibrillation with deep brain stimulator system (DBS) implantation. We report a 64-year-old male with DBS system in place for essential tremor who underwent cardiac defibrillation after cardiac arrest. Afterwards, his device impedances were all high and his tremor symptoms returned. Both problems resolved with implantation of a new generator and required no changes to the intracranial leads or extension cables. This is significantly different from the two previous reports. One included a significantly different DBS system relying on transcutaneous RF transmission and reported a lesioning effect after cardioversion. The other utilized a modern DBS system but reported damage to the generator and intracranial leads. We report that only the generator sustained damage, and that there were no intracranial changes that occurred.

Design of a 3000-Pixel Transition-Edge Sensor X-Ray Spectrometer for Microcircuit Tomography.

Feature sizes in integrated circuits have decreased substantially over time, and it has become increasingly difficult to three-dimensionally image these complex circuits after fabrication. This can be important for process development, defect analysis, and detection of unexpected structures in externally sourced chips, among other applications. Here, we report on a non-destructive, tabletop approach that addresses this imaging problem through x-ray tomography, which we uniquely realize with an instrument that combines a scanning electron microscope (SEM) with a transition-edge sensor (TES) x-ray spectrometer. Our approach uses the highly focused SEM electron beam to generate a small x-ray generation region in a carefully designed target layer that is placed over the sample being tested. With the high collection efficiency and resolving power of a TES spectrometer, we can isolate x-rays generated in the target from background and trace their paths through regions of interest in the sample layers, providing information about the various materials along the x-ray paths through their attenuation functions. We have recently demonstrated our approach using a 240 Mo/Cu bilayer TES prototype instrument on a simplified test sample containing features with sizes of ∼ 1 µm. Currently, we are designing and building a 3000 Mo/Au bilayer TES spectrometer upgrade, which is expected to improve the imaging speed by factor of up to 60 through a combination of increased detector number and detector speed.

Analyses of protein cores reveal fundamental differences between solution and crystal structures.

There have been several studies suggesting that protein structures solved by NMR spectroscopy and X-ray crystallography show significant differences. To understand the origin of these differences, we assembled a database of high-quality protein structures solved by both methods. We also find significant differences between NMR and crystal structures-in the root-mean-square deviations of the C α atomic positions, identities of core amino acids, backbone, and side-chain dihedral angles, and packing fraction of core residues. In contrast to prior studies, we identify the physical basis for these differences by modeling protein cores as jammed packings of amino acid-shaped particles. We find that we can tune the jammed packing fraction by varying the degree of thermalization used to generate the packings. For an athermal protocol, we find that the average jammed packing fraction is identical to that observed in the cores of protein structures solved by X-ray crystallography. In contrast, highly thermalized packing-generation protocols yield jammed packing fractions that are even higher than those observed in NMR structures. These results indicate that thermalized systems can pack more densely than athermal systems, which suggests a physical basis for the structural differences between protein structures solved by NMR and X-ray crystallography.

The Mitochondrial Peptide Humanin Targets but Does Not Denature Amyloid Oligomers in Type II Diabetes.

Mitochondrially derived peptides (MDPs) such as humanin (HN) have shown a remarkable ability to modulate neurological amyloids and apoptosis-associated proteins in cells and animal models. Recently, we found that humanin-like peptides also inhibit amyloid formation outside of neural environments in islet amyloid polypeptide (IAPP) fibrils and plaques, which are hallmarks of Type II diabetes. However, the biochemical basis for regulating amyloids through endogenous MDPs remains elusive. One hypothesis is that MDPs stabilize intermediate amyloid oligomers and discourage the formation of insoluble fibrils. To test this hypothesis, we carried out simulations and experiments to extract the dominant interactions between the S14G-HN mutant (HNG) and a diverse set of IAPP structures. Replica-exchange molecular dynamics suggests that MDPs cap the growth of amyloid oligomers. Simulations also indicate that HNG-IAPP heterodimers are 10 times more stable than IAPP homodimers, which explains the substoichiometric ability of HNG to inhibit amyloid growth. Despite this strong attraction, HNG does not denature IAPP. Instead, HNG binds IAPP near the disordered NFGAIL motif, wedging itself between amyloidogenic fragments. Shielding of NFGAIL-flanking fragments reduces the formation of parallel IAPP ß-sheets and subsequent nucleation of mature amyloid fibrils. From ThT spectroscopy and electron microscopy, we found that HNG does not deconstruct mature IAPP fibrils and oligomers, consistent with the simulations and our proposed hypothesis. Taken together, this work provides new mechanistic insight into how endogenous MDPs regulate pathological amyloid growth at the molecular level and in highly substoichiometric quantities, which can be exploited through peptidomimetics in diabetes or Alzheimer's disease.

Surface force measurements and simulations of mussel-derived peptide adhesives on wet organic surfaces.

Translating sticky biological molecules-such as mussel foot proteins (MFPs)-into synthetic, cost-effective underwater adhesives with adjustable nano- and macroscale characteristics requires an intimate understanding of the glue's molecular interactions. To help facilitate the next generation of aqueous adhesives, we performed a combination of surface forces apparatus (SFA) measurements and replica-exchange molecular dynamics (REMD) simulations on a synthetic, easy to prepare, Dopa-containing peptide (MFP-3s peptide), which adheres to organic surfaces just as effectively as its wild-type protein analog. Experiments and simulations both show significant differences in peptide adsorption on CH3-terminated (hydrophobic) and OH-terminated (hydrophilic) self-assembled monolayers (SAMs), where adsorption is strongest on hydrophobic SAMs because of orientationally specific interactions with Dopa. Additional umbrella-sampling simulations yield free-energy profiles that quantitatively agree with SFA measurements and are used to extract the adhesive properties of individual amino acids within the context of MFP-3s peptide adhesion, revealing a delicate balance between van der Waals, hydrophobic, and electrostatic forces.

Multi-Energy X-Ray Tomography of an Optical Fiber: The Role of Spatial Averaging.

Using a commercial X-ray tomography instrument, we have obtained reconstructions of a graded-index optical fiber with voxels of edge length 1.05 µm at 12 tube voltages. The fiber manufacturer created a graded index in the central region by varying the germanium concentration from a peak value in the center of the core to a very small value at the core-cladding boundary. Operating on 12 tube voltages, we show by a singular value decomposition that there are only two singular vectors with significant weight. Physically, this means scans beyond two tube voltages contain largely redundant information. We concentrate on an analysis of the images associated with these two singular vectors. The first singular vector is dominant and images of the coefficients of the first singular vector at each voxel look are similar to any of the single-energy reconstructions. Images of the coefficients of the second singular vector by itself appear to be noise. However, by averaging the reconstructed voxels in each of several narrow bands of radii, we can obtain values of the second singular vector at each radius. In the core region, where we expect the germanium doping to go from a peak value at the fiber center to zero at the core-cladding boundary, we find that a plot of the two coefficients of the singular vectors forms a line in the two-dimensional space consistent with the dopant decreasing linearly with radial distance from the core center. The coating, made of a polymer rather than silica, is not on this line indicating that the two-dimensional results are sensitive not only to the density but also to the elemental composition.

Scatter Corrections in X-Ray Computed Tomography: A Physics-Based Analysis.

Fundamental limits for the calculation of scattering corrections within X-ray computed tomography (CT) are found within the independent atom approximation from an analysis of the cross sections, CT geometry, and the Nyquist sampling theorem, suggesting large reductions in computational time compared to existing methods. By modifying the scatter by less than 1 %, it is possible to treat some of the elastic scattering in the forward direction as inelastic to achieve a smoother elastic scattering distribution. We present an analysis showing that the number of samples required for the smoother distribution can be greatly reduced. We show that fixed forced detection can be used with many fewer points for inelastic scattering, but that for pure elastic scattering, a standard Monte Carlo calculation is preferred. We use smoothing for both elastic and inelastic scattering because the intrinsic angular resolution is much poorer than can be achieved for projective tomography. Representative numerical examples are given.

A post-processing-free single-photon random number generator with ultra-low latency.

The low-latency requirements of a practical loophole-free Bell test preclude time-consuming post-processing steps that are often used to improve the statistical quality of a physical random number generator (RNG). Here we demonstrate a post-processing-free RNG that produces a random bit within 2.4(2) ns of an input trigger. We use weak feedback to eliminate long-term drift, resulting in 24 hour operation with output that is statistically indistinguishable from a Bernoulli process. We quantify the impact of the feedback on the predictability of the output as less than 6.4×10-7 and demonstrate the utility of the Allan variance as a tool for characterizing non-idealities in RNGs.

Acceleration of diffraction calculations in cylindrically symmetrical optics.

We have significantly accelerated diffraction calculations using three independent acceleration devices. These innovations are restricted to cylindrically symmetrical systems. In the first case, we consider Wolf's formula for integrated flux in a circular region following diffraction of light from a point source by a circular aperture or a circular lens. Although the formula involves a double sum, we evaluate it with the effort of a single sum by use of fast Fourier transforms (FFTs) to perform convolutions. In the second case, we exploit properties of the Fresnel-Kirchhoff propagator in the Gaussian, paraxial optics approximation to achieve the propagation of a partial wave from one optical element to the next. Ordinarily, this would involve a double loop over the radial variables on each element, but we have reduced the computational cost by a factor approximately equal to the smaller number of radius values. In the third case, we reduce the number of partial waves, for which the propagation needs to be calculated, to determine the throughput of an optical system of interest in radiometry when at least one element is very small, such as a pinhole aperture. As a demonstration of the benefits of the second case, we analyze intricate diffraction effects that occur in a satellite-based solar radiometry instrument.

Regulation and aggregation of intrinsically disordered peptides.

Intrinsically disordered proteins (IDPs) are a unique class of proteins that have no stable native structure, a feature that allows them to adopt a wide variety of extended and compact conformations that facilitate a large number of vital physiological functions. One of the most well-known IDPs is the microtubule-associated tau protein, which regulates microtubule growth in the nervous system. However, dysfunctions in tau can lead to tau oligomerization, fibril formation, and neurodegenerative disease, including Alzheimer's disease. Using a combination of simulations and experiments, we explore the role of osmolytes in regulating the conformation and aggregation propensities of the R2/wt peptide, a fragment of tau containing the aggregating paired helical filament (PHF6*). We show that the osmolytes urea and trimethylamine N-oxide (TMAO) shift the population of IDP monomer structures, but that no new conformational ensembles emerge. Although urea halts aggregation, TMAO promotes the formation of compact oligomers (including helical oligomers) through a newly proposed mechanism of redistribution of water around the perimeter of the peptide. We put forth a "superposition of ensembles" hypothesis to rationalize the mechanism by which IDP structure and aggregation is regulated in the cell.

Algorithm for rapid determination of optical scattering parameters.

Preliminary experiments at the NIST Spectral Tri-function Automated Reference Reflectometer (STARR) facility have been conducted with the goal of providing the diffuse optical properties of a solid reference standard with optical properties similar to human skin. Here, we describe an algorithm for determining the best-fit parameters and the statistical uncertainty associated with the measurement. The objective function is determined from the profile log likelihood, including both experimental and Monte Carlo uncertainties. Initially, the log likelihood is determined over a large parameter search box using a relatively small number of Monte Carlo samples such as 2·104. The search area is iteratively reduced to include the 99.9999% confidence region, while doubling the number of samples at each iteration until the experimental uncertainty dominates over the Monte Carlo uncertainty. Typically this occurs by 1.28·106 samples. The log likelihood is then fit to determine a 95% confidence ellipse. The inverse problem requires the values of the log likelihood on many points. Our implementation uses importance sampling to calculate these points on a grid in an efficient manner. Ultimately, the time-to-solution is approximately six times the cost of a Monte Carlo simulation of the radiation transport problem for a single set of parameters with the largest number of photons required. The results are found to be 64 times faster than our implementation of Particle Swarm Optimization.

Comparison of One-Dimensional and Volumetric Computed Tomography Measurements of Injected-Water Phantoms.

The goal of this study was to compare volumetric analysis in computed tomography (CT) with the length measurement prescribed by the Response Evaluation Criteria in Solid Tumors (RECIST) for a system with known mass and unknown shape. We injected 2 mL to 4 mL of water into vials of sodium polyacrylate and into disposable diapers. Volume measurements of the sodium polyacrylate powder were able to predict both mass and proportional changes in mass within a 95 % prediction interval of width 12 % and 16 %, respectively. The corresponding figures for RECIST were 102 % and 82 %.

Molecularly Smooth Self-Assembled Monolayer for High-Mobility Organic Field-Effect Transistors.

Despite the need for molecularly smooth self-assembled monolayers (SAMs) on silicon dioxide surfaces (the most common dielectric surface), current techniques are limited to nonideal silane grafting. Here, we show unique bioinspired zwitterionic molecules forming a molecularly smooth and uniformly thin SAM in "water" in <1 min on various dielectric surfaces, which enables a dip-coating process that is essential for organic electronics to become reality. This monomolecular layer leads to high mobility of organic field-effect transistors (OFETs) based on various organic semiconductors and source/drain electrodes. A combination of experimental and computational techniques confirms strong adsorption (Wad > 20 mJ m-2), uniform thickness (∼0.5 or ∼1 nm) and orientation (all catechol head groups facing the oxide surface) of the "monomolecular" layers. This robust (strong adsorption), rapid, and green SAM represents a promising advancement toward the next generation of nanofabrication compared to the current nonuniform and inconsistent polysiloxane-based SAM involving toxic chemicals, long processing time (>10 h), or heat (>80 °C).

Continuous-feed optical sorting of aerosol particles.

We consider the problem of sorting, by size, spherical particles of order 100 nm radius. The scheme we analyze consists of a heterogeneous stream of spherical particles flowing at an oblique angle across an optical Gaussian mode standing wave. Sorting is achieved by the combined spatial and size dependencies of the optical force. Particles of all sizes enter the flow at a point, but exit at different locations depending on size. Exiting particles may be detected optically or separated for further processing. The scheme has the advantages of accommodating a high throughput, producing a continuous stream of continuously dispersed particles, and exhibiting excellent size resolution. We performed detailed Monte Carlo simulations of particle trajectories through the optical field under the influence of convective air flow. We also developed a method for deriving effective velocities and diffusion constants from the Fokker-Planck equation that can generate equivalent results much more quickly. With an optical wavelength of 1064 nm, polystyrene particles with radii in the neighborhood of 275 nm, for which the optical force vanishes, may be sorted with a resolution below 1 nm.