Glass Strengthening Using Ormosil Polymeric Coatings

Ormosil polymeric coatings comprised of an epoxy resin, an amine hardener and a silane have been applied to glass samples containing large, controlled defects introduced by Vicker's indentation. The coatings can completely overcome these controlled defects. The strengthening effect is due to penetration of the defects by the coatings. Therefore, the reactions between the components of these hybrid materials and the glass substrate are of crucial importance in determining both the degree of strengthening that is achieved and the hydrolytic durability of the coatings. The maximum strengthening effect of these ormosil polymeric coatings is obtained when 25% of the active hydrogen is supplied by the silane (7.2 wt% silane). More than 7.2 wt% silane reduces coating cohesion and thus gives a reduced strengthening effect. Studies on silane primed systems show that both good coating adhesion and cohesion are required for significant strengthening. Coatings with good adhesion also have greater hydrolytic durability.     

Cerium and boron chemistry in doped borosilicate glasses examined by EELS

Spatially resolved measurements of boron coordination and cerium valency in a doped borosilicate glass with crystalline nano-precipitates are described. The fine structure of the boron K-edge and the white-line ratio of the cerium M-edge doublet were evaluated from EELS line scans. Due to high beam sensitivity it was found that reliable boron-coordination measurements in some of the glasses studied required extrapolation of results acquired after different periods of irradiation back to a zero-irradiation. However, borosilicates that contained heavy alkali atoms were found to suffer very little structural change. The Ce valency of a 4% (molar) doped alkali-borosilicate glass was found to be mixed +III/+IV in the glass matrix and purely +IV (indicative of CeO2) in the precipitates. A significant dependency of the valence results on the data processing method was found and explained.     

Mössbauer studies of phosphate glasses for the immobilisation of toxic and nuclear wastes

Various iron-containing phosphate glasses were investigated by Mössbauer spectroscopy. Iron was found to occur predominantly as Fe³⁺ in all glasses, and largely occupied sites with distorted octahedral coordination for both redox states. Using a base glass of nominal composition 60 P₂O₅–40 Fe₂O₃ (mol%), stepwise molar replacement of Fe₂O₃ by (0.67 Na₂O?×?0.33 Al₂O₃) increased the redox ratio, Fe²⁺/ΣFe, from 0.13 at 40% Fe₂O₃ to 0.25 at 10% Fe₂O₃. The centre shift increased and quadrupole splitting decreased by up to ～20% over this range, interpreted as a decrease in the average distortion of Fe sites from cubic symmetry, and an increase in average iron coordination. Literature revealed that recoil-free fraction ratio f (Fe³⁺)?/?f (Fe²⁺)?≈?1.3 in iron phosphate glasses, and this was considered when assessing redox. Mössbauer parameters of these and other glasses demonstrated a combination of structural stability and compositional flexibility which makes them so suitable for waste immobilisation.     

Heating of biological tissues by two-dimensional phased arrays with random and regular element distributions

The effect of an irregularity of the element distribution in a two-dimensional phased array upon the efficiency of heating of biological tissue is studied in an ultrasonic surgery regime. Two arrays of 256 piston elements, which either form a regular square pattern or are positioned randomly on the surface of a spherical segment, are considered as a model. The formation and the steering of a set of nine foci along the array axis and in the direction perpendicular to it are investigated. The theoretical model includes the algorithm of determining a phase set at the array elements that is optimal for the formation of foci with equal intensities and a preset geometry, as well as the calculation of acoustic and temperature fields in a tissue. The results of numerical simulation are presented for the spatial distributions of ultrasonic intensity, temperature, and the corresponding thermal dose in tissue. It is demonstrated that an irregularity of the element distribution reduces the level of secondary intensity peaks in the field produced by the array. This provides an opportunity to avoid the overheating and ablation of tissue outside the target volume, even in the case of steering with the set of foci away from the array axis within a distance of ±7 mm. A nine-foci regime is studied with the parameters necessary to produce uniform thermal ablation in a volume that is evaluated on the basis of the thermal dose distribution.     

The Effect of Electron-Donating and Electron-Withdrawing Substituents on 1H-and 13C-NMR chemical shifts of novel 7′-aryl-substituted 7′-apo-β-carotenes

The synthesis of 7′-aryl-7′-apo-β-carotenes, where aryl (Ar) is Ph, 4-NO₂C₆H₄, 4-MeOC₆H₄, 4-(MeO₂C)C₆H₄, C₆F₅, and 2,4,6-Me₃C₆H₂, is described. NMR Chemical shifts of all H- and C-atoms are presented, together with specific examples of the spectra. In contrast to ¹H chemical shifts which, except for H? C(8′) and H? C(7′), did not differ greatly from those of β,β-carotene, considerable variations in ¹³C chemical shifts were observed. Signals of the C(α) atoms of the polyene chain [C(β)? C(α)] +_n Ar were shielded, those of the C(β) atoms were deshielded, with some exceptions when n = 1; the effects decreased with increasing n.     

Lookahead scorecards for new fixed term credit products

Standard approaches to scorecard construction require that a body of data has already been collected for which the customers have known good/bad outcomes, so that scorecards can be built using this information. This requirement is not satisfied by new financial products. To overcome this lack, we describe a class of models based on using information about the length of time customers have been using the product, as well as any available information which does exist about true good/bad outcome classes. These models not only predict the probability that a new customer will go bad at some time during the product's term, but also evolve as new information becomes available. Particular choices of functional form in such models can lead to scorecards with very simple structures. The models are illustrated on a data set relating to loans.     

Mechanism of the reaction of 2-haloketones with 2-aminopyridine

Time-dependent ¹H NMR spectra of DMSO-d₆ solutions of p-substituted phenacyl bromides and 2-aminopyridine indicate that the formation of imidazo[1,2-a]pyridines occurs via two relatively long-lived intermediates, C and D, which are in equilibrium with each other. The assigned structures are in accord with chemical shifts, pK_a estimates, and substituent effects (-OMeCH₃, -H, and -NO₂) on both the equilibrium constant (C?D) and rates of reaction. The slowest step in the reaction with phenacyl bromide is conversion of the intermediate D to product E. With phenacyl chloride no intermediates are observed and initial formation of C determines the overall rate. Even through the intermediate D is already protonated, its conversion to E is subject to acid catalysis. Compared to the p-OMe substituent, the p-NO₂ group enhances the rate of formation of C and D by a factor of only 2.6. The same rate enhancement is observed in the reaction of pyridine with phenacyl bromides. Rates of reaction of a given phenacyl halide with pyridine and 2-aminopyridine are similar. It is concluded that the initial reaction is alkylation of the pyridine nitrogen atom to give C and that the other possible initial condensation product, the carbinolamine F, cannot be a kinetically significant intermediate. Reasons for preferred N-alkylation are presented. Recommendations for improved syntheses of imidazo[1,2-a]pyridines are included.     

No Statistical-Computational Gap in Spiked Matrix Models with Generative Network Priors

We provide a non-asymptotic analysis of the spiked Wishart and Wigner matrix models with a generative neural network prior. Spiked random matrices have the form of a rank-one signal plus noise and have been used as models for high dimensional Principal Component Analysis (PCA), community detection and synchronization over groups. Depending on the prior imposed on the spike, these models can display a statistical-computational gap between the information theoretically optimal reconstruction error that can be achieved with unbounded computational resources and the sub-optimal performances of currently known polynomial time algorithms. These gaps are believed to be fundamental, as in the emblematic case of Sparse PCA. In stark contrast to such cases, we show that there is no statistical-computational gap under a generative network prior, in which the spike lies on the range of a generative neural network. Specifically, we analyze a gradient descent method for minimizing a nonlinear least squares objective over the range of an expansive-Gaussian neural network and show that it can recover in polynomial time an estimate of the underlying spike with a rate-optimal sample complexity and dependence on the noise level.