X-ray diffraction (XRD) data acquisition and analysis is among the most time-consuming steps in the development cycle of novel thin-film materials. We propose a machine-learning-enabled approach to predict crystallographic dimensionality and space group from a limited number of thin-film XRD patterns. We overcome the scarce-data problem intrinsic to novel materials development by coupling a supervised machine learning approach with a model-agnostic, physics-informed data augmentation strategy using simulated data from the Inorganic Crystal Structure Database (ICSD) and experimental data. As a test case, 115 thin-film metal-halides spanning 3 dimensionalities and 7 space-groups are synthesized and classified. After testing various algorithms, we develop and implement an all convolutional neural network, with cross-validated accuracies for dimensionality and space-group classification of 93% and 89%, respectively. We propose average class activation maps, computed from a global average pooling layer, to allow high model interpretability by human experimentalists, elucidating the root-causes of misclassification. Finally, we systematically evaluate the maximum XRD pattern step size (data acquisition rate) before loss of predictive accuracy occurs, and determine it to be 0.16° 2, which enables an XRD pattern to be obtained and classified in 5.5 minutes or less.

We study heating and heat dissipation of a single C60 molecule in the junction of a scanning tunneling microscope (STM) by measuring the electron current required to thermally decompose the fullerene cage. The power for decomposition varies with electron energy and reflects the molecular resonance structure. When the STM tip contacts the fullerene the molecule can sustain much larger currents. Transport simulations explain these effects by molecular heating due to resonant electronphonon coupling and molecular cooling by vibrational decay into the tip upon contact formation.The paradigm of molecular electronics is the use of a single molecule as an electronic device [1]. This concept is sustained on the basis that a single molecule (or a molecular thin film) should withstand the flow of electron current densities as large as 10 10 A/m 2 without degrading. A fraction of these electrons heat the molecular junction through inelastic scattering with the molecule [2]. The temperature at the junction is a consequence of an equilibrium between heating due to electron flow and heat dissipation out of the junction. The former is dominated by the coupling of electronic molecular states with molecular vibrons [2,3,4]. The latter depends on the strength of the vibrational coupling between the "hot" molecular vibrons and the bath degrees of freedom of the "cold" electrodes.Theoretical studies predicted that current-induced heating in molecular junctions can be large enough to affect the reliability of molecular devices [2]. However, experimental access to this information is very limited. Recent studies of the thermally activated force during molecular detachment from a lead [5,6] and of structural fluctuation during attachment to it [7] reveal that the temperature of a molecular junction can reach several hundred degrees under normal working conditions, thus revealing that present devices work on the limit of practical operability [8]. Heat dissipation away from the junction becomes an important issue.In this work, we characterize the mechanisms of heating and heat dissipation induced by the flow of current across a single molecule. Our approach is based on detecting the limiting electron current inducing molecular decomposition at varying applied source-drain bias (i.e. the maximum power one molecule can sustain). We use a low temperature scanning tunneling microscope (STM) to control the flow of electrons through a single C 60 molecule at an increasing rate until the molecule decomposes. By comparing the power applied for decomposition (P dec ) in tunneling regime and in contact with the STM tip we find that it depends significantly on two factors: i) P dec decreases when molecular resonances participate in the transport, evidencing that they enhance the heating; ii) P dec increases as the molecule is contacted to the source and drain electrodes, revealing the heat dissipation by phonon coupling to the leads. A good contact between the single-molecule (SM) device and the leads is hence an important requirement for its ope...

In this work we analyze the coupled piezoelectric and semiconductive behavior of vertically aligned ZnO nanowires under uniform compression. The screening effect on the piezoelectric field caused by the free carriers in vertically compressed zinc oxide nanowires (NWs) has been computed by means of both analytical considerations and finite element calculations. We predict that, for typical geometries and donor concentrations, the length of the NW does not significantly influence the maximum output piezopotential because the potential mainly drops across the tip, so that relatively short NWs can be sufficient for high-efficiency nanogenerators, which is an important result for wet-chemistry fabrication of low-cost, CMOS- or MEMS-compatible nanogenerators. Furthermore, simulations reveal that the dielectric surrounding the NW influences the output piezopotential, especially for low donor concentrations. Other parameters such as the applied force, the sectional area and the donor concentration have been varied in order to understand their effects on the output voltage of the nanogenerator.

No abstract

The main task of this research was to evaluate canopy temperature and Crop Water Stress Index (CWSI) by assessing genotype variability of maize performance for different water regimes. To that end, three hundred tropical and subtropical maize hybrids with different phenology in terms of date of anthesis were evaluated. The influence of phenology on the change in canopy temperatures and CWSI was not equal during the three dates of measurement. At the end of vegetative growth (82 days after sowing, DAS) and at the blister stage (DAS 97), a high significant difference in temperatures and CWSI (P < 0.001) were obtained between the early‐ and late‐maturity genotypes. During anthesis (DAS 89), phenology had a significant effect (P < 0.01) only for the well‐watered genotypes, while under water‐stress conditions, no differences were found between early and late genotypes in terms of canopy temperature and CWSI. High significant differences (P < 0.001) in stomatal conductance (gs) between early and late genotypes for different treatments were observed. A relationship (R2 = 0.62) between gs and canopy temperature was obtained. Under a water‐stress canopy, temperature was measured at anthesis, which was negatively correlated with grain yield of the early (r = −0.55)‐ and late (r = −0.46)‐maturity genotypes in the water‐stressed condition.

No abstract

No abstract

In this work we analyze the role played by molecular resonances on the thermal behavior of single molecule junction. We consider a fullerene molecule on a Cu͑110͒ substrate when current is driven across using an scanning tunnel microscope tip. Calculations based on density-functional theory and nonequilibrium Green's functions have predicted a lowering of molecular temperature whenever a resonance gets close to the injection window. This behavior, due on the interplay of emission and absorption phonon processes, can take place ͑at T 0 =0 K͒ for bias E L − ប Ͻ V Ͻ E L + ប where E L and ប are the lowest unoccupied molecular orbital ͑LUMO͒ and phonon energy, respectively. We further show that the decay channel into electron-hole excitations is very effective in keeping the molecule stable, whereas the direct decay of the vibron into contact phonons gives only a marginal correction. Furthermore, we have investigated the possibility to cool the molecule below the environment temperature. We found that this phenomena is favorite for modes with energy ប Ͻ k b T 0 and ប Ͻ E L − E F .