NEURAL NETS SPEED FLOW CALCULATION

Neural Nets Successfully Solve Complex Fluid Flow Problem. Much research is being done in the area of neural networks, and industry is actively seeking successful application to real world problems. We describe here a successful application. We have used neural networks to model complex coolant flow patterns, such as those encountered in design of hypersonic aircraft. Previous calculation methods, while reasonably accurate, are iterative and extremely time consuming. Our new approach uses a hierarchical neural network architecture to model coolant flow distribution in multiple heat exchanger panels. This method is direct, fast, and accurate.     

Synthesis and evaluation of hexafluorodimethylcarbinol functionalized polymers as microsensor coatings

Polystyrenes, polyacrylates, and poly(3,4-isoprenes) incorporating the hexafluorodimethylcarbinol functionality have been synthesized, characterized, and tested as microsensor coatings on a surface acoustic wave (SAW) device for dimethyl methylphosphonate (DMMP) vapor absorption sensitivity. The syntheses involved monomer functionalization and polymerization or hexafluoroacetone reaction with preformed polymer. All fluoroalcohol functionalized polymer coatings displayed sub parts per million level sensitivity with the slope of the absorption isotherm steepest at low DMMP concentrations. The order of sensitivity for the isomeric polystyrene fluoroalcohols (meta > para ? ortho) paralleled that of the relative free hydroxyl to hydrogen-bonded hydroxyl content. Strong hydrogen bonding between the fluoroalcohol polymers and DMMP vapor was observed by IR spectroscopy. Acylation of the fluoroalcohol group markedly reduced the DMMP sensitivity.     

A frequency domain ultrasonic model for the pulse-echo transducer response of an arbitrary scatterer in a fluid

A very general model formulation is presented in the frequency domain for the pulse-echo ultrasonic response of an arbitrary scatterer in a fluid. The transducer is modeled as a piston source and the scatterer can be located anywhere in the transducer wavefield. The model is computationally efficient and is shown to agree well with initial experiments.     

Multisource-sputtered ErBa2Cu3O7−x films on single-crystal and buffered crystalline substrates

High-quality superconducting films of erbium-barium-copper-oxide have been produced in a multisource sputtering system on a variety of substrates, including buffered sapphire and silicon wafers as well as various single-crystal materials. Fine-grained polycrystalline films with narrow (<4 K) resistive transitions have been grown on a number of different substrates. The use of erbium in the 12–3 compound leads to improved film morphology and to more forgiving conditions for establishing of the superconducting phase. Sputtered neutral mass spectroscopy (SNMS) yields useful information about the compositional profile of the films.     

A Microscopic study of the transformation of sphalerite particles during the roasting of zinc concentrate

The formation of zinc ferrite (ZnFe₂O₄) during the roasting of iron-bearing zinc concentrates requires substantial additional processing to recover the zinc from this compound by leaching and to eliminate the iron from the leachate. The phase changes that occur in the particles of a typical industrial zinc sulfide concentrate during roasting in a fluidized bed at 1223 K were investigated by the use of light microscopy, electron microprobe analysis, and SEM with EDS. The processes which the iron undergoes during its eventual transformation into ferrite have been clarified by examination of the phases and the morphology of partially roasted marmatitic sphalerite particles (Zn, Fe)S, and by reference to the known phase equilibria involved in the Zn-Fe-S-0 system. The oxidation of ironbearing sphalerite occurs in three stages. The first involves the selective diffusion of most of the iron to the particle surface resulting in the formation of an iron oxide shell enclosing a largely unreacted zinc sulfide kernel. In the second stage, this kernel is oxidized to form a solid solution of zinc oxide and iron oxide. The iron is initially present in the ferrous state but, with the disappearance of the sulfide kernel, is oxidized to ferric iron. In the final stage, this dissolved iron oxide and the iron oxide shell react with the surrounding zinc oxide to form the refractory spinel zinc ferrite.     

The chemical anchoring of noble metal amine precursors to silica

The effect of pretreatment on the dispersion of supported noble metal Catalyst prepared from amine precursors in basic solution have been studied. The following metal precursors were used: Pt(NH₃)₄(NO₃)₂, Pd(NH₃)₄(NO₃)₂, Ru(NH₃)₆Cl₃ and [Rh(NH₃)₅Cl]Cl₂ Pretreatment in oxygen, prior to reduction in H₂ at 400C, resulted in poor dispersions for Ru and Rh, moderate dispersions for Pd and high dispersions for Pt. Pretreatment in H₂ resulted in poor dispersions for Pd and Pt and high dispersions for Ru and Rh. Decomposition of the adsorbed Pt and Pd precursors in argon resulted in very high dispersions.     

The effect of pack-aluminisation on the microstructure of MCrAlY and the performance of thermal barrier coatings

Plasma spraying and pack-aluminising processes were combined and applied to the nickel-base superalloy Mar-M247 to improve its cyclic oxidation resistance. The performance tests of duplex ZrO₂-8 wt.%Y₂O₃/MCrAlY thermal barrier coatings (TBCs) were conducted at 1050 °C, 1075 °C, 1100 °C, 1150 °C and 1200 °C. The results of the experiments in this study showed that TBC specimens with the aluminised MCrAlY bond coat exhibited higher cyclic lives (except for the Ni-22Cr-10Al-1Y bond coat), at all the temperatures tested, than specimens on which the bond coat was not aluminised. The microstructures of the Co-29Cr-6Al-1Y, Co-32Ni-21Cr-8Al-0.5Y and Ni-22Cr-10Al-1Y bond coats with or without aluminising treatment were examined in detail using a scanning electron microscope equipped with an electron probe microanalyzer.     

Energy dissipation via radiative recombination and vibronic relaxation in σ-delocalized silicon backbone network polymers

The nature of ultrafast energy dissipation in poly(n-hexylsilyne), a prototypical σ-delocalized alkysilicon network polymer, is explored. This disordered silicon backbone material exhibits strong near-UV to visible band-edge absorption and a high quantum yield of visible emission. The time evolution of the emission band is studied over four decades of time using time-resolved luminescence as a probe. The data indicate that while there is an ‘intrinsic’ Stokes shift after photoexcitation due to kinetic energy relaxation (<10ps), thermalization within a dense band of vibronic states via intramolecular phonon-assisted hopping on a nanosecond timescale is the dominant mechanism for excited-state decay. The data can be understood in terms of theoretical predictions for energy relaxation in disordered materials.     

Off-line multiresponse optimization of electrochemical surface grinding by a multi-objective programming method

The objective of this study is to optimize electrochemical grinding (ECG) process responses simultaneously by an off-line multiresponse optimization methodology. The responses considered as objectives are side and bottom overcuts, surface finish, spindle load, total metal removal rate, and wheel wear. Materials of 304 Stainless Steel are ground by the ECG process. The process variables optimized for the above objectives include electrolyte type, wheel material, grit size, grit concentration, d.c. voltage, electrolyte flow rate, wheel speed, feed rate, and ripple effect. A simple weighting method transforms the multi-objective problem into a single-objective programming format and then, by parametric variation of the weights, the set of non-dominated optimum solutions are obtained. It is shown in this paper that the multi-objective optimization methodology can be applied for an ECG operation, and that the optimal operating conditions for any given set of weights can be obtained depending upon the objectives.