Abstract: The autoregressive neural networks are emerging as a powerful computational tool to solve relevant problems in classical and quantum mechanics. One of their appealing functionalities is that, after they have learned a probability distribution from a dataset, they allow exact and efficient sampling of typical system configurations. Here we employ a neural autoregressive distribution estimator (NADE) to boost Markov chain Monte Carlo (MCMC) simulations of a paradigmatic classical model of spin-glass theory, namely, the two-dimensional Edwards-Anderson Hamiltonian. We show that a NADE can be trained to accurately mimic the Boltzmann distribution using unsupervised learning from system configurations generated using standard MCMC algorithms. The trained NADE is then employed as smart proposal distribution for the Metropolis-Hastings algorithm. This allows us to perform efficient MCMC simulations, which provide unbiased results even if the expectation value corresponding to the probability distribution learned by the NADE is not exact. Notably, we implement a sequential tempering procedure, whereby a NADE trained at a higher temperature is iteratively employed as proposal distribution in a MCMC simulation run at a slightly lower temperature. This allows one to efficiently simulate the spin-glass model even in the low-temperature regime, avoiding the divergent correlation times that plague MCMC simulations driven by local-update algorithms. Furthermore, we show that the NADE-driven simulations quickly sample ground-state configurations, paving the way to their future utilization to tackle binary optimization problems.

Abstract: HAADF-STEM tomography is a widely used experimental technique for analyzing nanometer-scale crystalline structures of a large variety of materials in three dimensions. Unfortunately, the acquisition of conventional HAADF-STEM tilt series can easily take up one hour or more, depending on the complexity of the experiment. It is therefore far from straightforward to investigate samples that do not withstand long acquisition or to acquire large amounts of tilt series during a single TEM experiment. The latter would lead to the ability to obtain statistically meaningful 3D data, or to perform in situ 3D characterizations with a much shorter time resolution. Various HAADF-STEM acquisition strategies have been proposed to accelerate the tomographic acquisition and reduce the required electron dose. These methods include tilting the holder continuously while acquiring a projection “movie” and a hybrid, incremental, methodology which combines the benefits of the conventional and continuous technique. However, until now an experimental evaluation has been lacking. In this paper, the different acquisition strategies will be experimentally compared in terms of speed, resolution and electron dose. This evaluation will be performed based on experimental tilt series acquired for various metallic nanoparticles with different shapes and sizes. We discuss the data processing involved with the fast HAADF-STEM tilt series and provide a general guideline when which acquisition strategy should be preferentially used.

Abstract: The flexoelectric and electronic properties of zig-zag graphene nanoribbons are explored under mechanical bending using state of the art first principles calculations. A linear dependence of the bending induced out of plane polarization on the applied strain gradient is found. The inferior flexoelectric properties of graphene nanoribbons can be improved by more than two orders of magnitude by hydrogen and fluorine functionalization (CH and CF nanoribbons). A large out of plane flexoelectric effect is predicted for CF nanoribbons. The origin of this enhancement lies in the electro-negativity difference between carbon and fluorine atoms, which breaks the out of plane charge symmetry even for a small strain gradient. The flexoelectric effect can be further improved by co-functionalization with hydrogen and fluorine (CHF Janus-type nanoribbon), where a spontaneous out of plane dipole moment is formed even for flat nanoribbons. We also find that bending can control the charge localization of valence band maxima and therefore enables the tuning of the hole effective masses and band gaps. These results present an important advance towards the understanding of flexoelectric and electronic properties of hydrogen and fluorine functionalized graphene nanoribbons, which can have important implications for flexible electronic applications. (C) 2020 Elsevier Ltd. All rights reserved.

Abstract: Dry powder coating is an effective approach to protect the surfaces of layered cathode active materials (CAMs) in lithium-ion batteries. Previous investigations indicate an incorporation of lithium ions in fumed Al2O3, ZrO2, and TiO2 coatings on LiNi0.7Mn0.15Co0.15O2 during cycling, improving the cycling performance. Here, this coating approach is transferred for the first time to fumed ternary LiAlO2, Li4Zr3O8, and Li4Ti5O12 and directly compared with their lithium-free equivalents. All materials could be processed equally and their nanostructured small aggregates accumulate on the CAM surfaces to quite homogeneous coating layers with a certain porosity. The LiNixMnyCozO2 (NMC) coated with lithium-containing materials shows an enhanced improvement in overall capacity, capacity retention, rate performance, and polarization behavior during cycling, compared to their lithium-free analogues. The highest rate performance was achieved with the fumed ZrO2 coating, while the best long-term cycling stability with the highest absolute capacity was obtained for the fumed LiAlO2-coated NMC. The optimal coating agent for NMC to achieve a balanced system is fumed Li4Ti5O12, providing a good compromise between high rate capability and good capacity retention. The coating agents prevent CAM particle cracking and degradation in the order LiAlO2 ≈ Al2O3 > Li4Ti5O12 > Li4Zr3O8 > ZrO2 > TiO2. A schematic model for the protection and electrochemical performance enhancement of high-nickel NMC with fumed metal oxide coatings is sketched. It becomes apparent that physical and chemical characteristics of the coating significantly influence the performance of NMC. A high degree of coating-layer porosity is favorable for the rate capability, while a high coverage of the surface, especially in vulnerable grain boundaries, enhances the long-term cycling stability and improves the cracking behavior of NMCs. While zirconium-containing coatings possess the best chemical properties for high rate performances, aluminum-containing coatings feature a superior chemical nature to protect high-nickel NMCs.

Abstract: Single crystals with two different morphologies, black and brown, of C60 with about 12% C70 have been studied by synchrotron X-ray diffraction. Integrated intensities from 8 reflections show that the charge distribution of the C60 cluster is shell-like giving rise to an oscillatory behaviour in the diffracted intensities. From the intensities, the distance between diametrically opposite carbon atoms has been determined to be (7.24 +/- 0.22) angstrom. The thickness of the charge distribution of the shell can be estimated as 3.06 angstrom and the cavity within a C60 cluster to have a diameter of about 4.18 angstrom. The finite longitudinal width of the diffraction peaks from black crystals indicates a lack of long-range crystalline order. The rocking curves are about 7-degrees broad and depict an unusually smooth behaviour. This may be indicative of a glassy or hexatic phase. High-resolution electron microscopy allows small crystallites with a relatively well-defined orientation relationship to be identified. In contrast, brown crystals have long-range order.

Abstract: We analyzed ultra-thin ZrO2/SiO2/Si gate dielectrics under post-deposition anneals in dry O-2 at temperatures from 500 to 700 degreesC. TOF-SIMS profiling of ZrO2/SiO2/Si stacks is hampered by many sputter induced artifacts. The depletion of oxygen leads to a decrease in SIMS intensities. However, preferential sputtering is accompanied by transport of the depleted species towards the surface. Due to recoil implantation oxygen gets piled-up near the ZrO2/SiO2 interface. Either normal or radiation-enhanced diffusion transports oxygen back to the surface. Simultaneously also segregation of zirconium towards and through the interface is observed, resulting in a large zirconium tail in the underlying silicon substrate. (C) 2002 Elsevier Science B.V. All rights reserved.