Nucleation and growth of nickel by electrodeposition under galvanostatic conditions

The first stages of nickel electrodeposition on amorphous carbon and polycrystalline silver substrates have been studied under galvanostatic conditions. An island growth mode is observed on the two types of substrate. For equivalent conditions the cluster density is higher on silver substrates than on carbon substrates. Dissolution of some nickel clusters is observed a short time after the onset of electrolysis on carbon substrates. This is confirmed by the evolution of the current efficiency and the evolution of the metal surface cluster density as a function of the quantity of electricity.     

A hyperheuristic approach to examination timetabling problems: benchmarks and a new problem from practice

Many researchers studying examination timetabling problems focus on either benchmark problems or problems from practice encountered in their institutions. Hyperheuristics are proposed as generic optimisation methods which explore the search space of heuristics rather than direct solutions. In the present study, the performance of tournament-based hyperheuristics for the exam timetabling problem are investigated. The target instances include both the Toronto and ITC 2007 benchmarks and the examination timetabling problem at KAHO Sint-Lieven (Ghent, Belgium). The Toronto and ITC 2007 benchmarks are post-enrolment-based examination timetabling problems, whereas the KAHO Sint-Lieven case is a curriculum-based examination timetabling problem. We drastically improve the previous (manually created) solution for the KAHO Sint-Lieven problem by generating a timetable that satisfies all the hard and soft constraints. We also make improvements on the best known results in the examination timetabling literature for seven out of thirteen instances for the To ronto benchmarks. The results are competitive with those of the finalists of the examination timetabling track of the International Timetabling Competition.     

Dehydration of aldoximes on rare earth exchanged (La<Superscript>3+</Superscript>, Ce<Superscript>3+</Superscript>, RE<Superscript>3+</Superscript>, Sm<Superscript>3+</Superscript>) Na–Y zeolites: A facile route for the synthesis of nitriles

Rare earth exchanged Na–Y zeolites, H-mordenite, K-10 montmorillonite clay and amorphous silica-alumina were effectively employed for the continuous synthesis of nitriles. Dehydration of benzaldoxime and 4-methoxybenzaldoxime were carried out on these catalysts at 473 K. Benzonitrile (dehydration product) was obtained in near quantitative yield with benzaldoxime whereas; 4-methoxybenzaldoxime produces both Beckmann rearrangement (4-methoxyphenylformamide) as well as dehydration products (4-methoxy benzonitrile) in quantitative yields. The production of benzonitrile was near quantitative under heterogeneous reaction conditions. The optimal protocol allows nitriles to be synthesized in good yields through the dehydration of aldoximes. Time on stream (TOS) studies show decline in the activity of the catalysts due to neutralization of acid sites by the basic reactant and product molecules and water formed during the dehydration of aldoximes.     

Epoxy‐tert‐butyl poly(cyanoarylene ether) blends: Phase morphology,fracture toughness,and mechanical properties

Tert‐butyl hydroquinone–based poly(cyanoarylene ether) (PENT) was synthesized by the nucleophilic aromatic substitution reaction of 2,6‐dichlorobenzonitrile with tert‐butyl hydroquinone using N‐methyl‐2‐pyrrolidone (NMP) as solvent in the presence of anhydrous potassium carbonate in a nitrogen atmosphere at 200°C. PENT‐toughened diglycidyl ether of bisphenol A epoxy resin (DGEBA) was developed using 4,4′‐diaminodiphenyl sulfone (DDS) as the curing agent. Scanning electron micrographs revealed that all blends had a two‐phase morphology. The morphology changed from dispersed PENT to a cocontinuous structure with an increase in PENT content in the blends from 5 to 15 phr. The viscoelastic properties of the blends were investigated using dynamic mechanical thermal analysis. The storage modulus of the blends was less than that of the unmodified resin, whereas the loss modulus of the blends was higher than that of the neat epoxy. The tensile strength of the blends improved slightly, whereas flexural strength remained the same as that of the unmodified resin. Fracture toughness was found to increase with an increase in PENT content in the blends. Toughening mechanisms like local plastic deformation of the matrix, crack path deflection, crack pinning, ductile tearing of thermoplastic, and particle bridging were evident from the scanning electron micrographs of failed specimens from the fracture toughness measurements. The thermal stability of the blends were comparable to that of the neat resin. © 2006 Wiley Periodicals, Inc. J Appl Polym Sci 100: 3536–3544, 2006     

Ensemble classification based on generalized additive models

Generalized additive models (GAMs) are a generalization of generalized linear models (GLMs) and constitute a powerful technique which has successfully proven its ability to capture nonlinear relationships between explanatory variables and a response variable in many domains. In this paper, GAMs are proposed as base classifiers for ensemble learning. Three alternative ensemble strategies for binary classification using GAMs as base classifiers are proposed: (i) GAMbag based on Bagging, (ii) GAMrsm based on the Random Subspace Method (RSM), and (iii) GAMens as a combination of both. In an experimental validation performed on 12 data sets from the UCI repository, the proposed algorithms are benchmarked to a single GAM and to decision tree based ensemble classifiers (i.e. RSM, Bagging, Random Forest, and the recently proposed Rotation Forest). From the results a number of conclusions can be drawn. Firstly, the use of an ensemble of GAMs instead of a single GAM always leads to improved prediction performance. Secondly, GAMrsm and GAMens perform comparably, while both versions outperform GAMbag. Finally, the value of using GAMs as base classifiers in an ensemble instead of standard decision trees is demonstrated. GAMbag demonstrates performance comparable to ordinary Bagging. Moreover, GAMrsm and GAMens outperform RSM and Bagging, while these two GAM ensemble variations perform comparably to Random Forest and Rotation Forest. Sensitivity analyses are included for the number of member classifiers in the ensemble, the number of variables included in a random feature subspace and the number of degrees of freedom for GAM spline estimation.     

Application of single-level,pointwise algebraic,and smoothed aggregation multigrid methods to direct numerical simulations of incompressible turbulent flows

Single- and multi-level iterative methods for sparse linear systems are applied to unsteady flow simulations via implementation into a direct numerical simulation solver for incompressible turbulent flows on unstructured meshes. The performance of these solution methods, implemented in the well-established SAMG and ML packages, are quantified in terms of computational speed and memory consumption, with a direct sparse LU solver (SuperLU) used as a reference. The classical test case of unsteady flow over a circular cylinder at low Reynolds numbers is considered, employing a series of increasingly fine anisotropic meshes. As expected, the memory consumption increases dramatically with the considered problem size for the direct solver. Surprisingly, however, the computation times remain reasonable. The speed and memory usage of pointwise algebraic and smoothed aggregation multigrid solvers are found to exhibit near-linear scaling. As an alternative to multi-level solvers, a single-level ILUT-preconditioned GMRES solver with low drop tolerance is also considered. This solver is found to perform sufficiently well only on small meshes. Even then, it is outperformed by pointwise algebraic multigrid on all counts. Finally, the effectiveness of pointwise algebraic multigrid is illustrated by considering a large three-dimensional direct numerical simulation case using a novel parallelization approach on a large distributed memory computing cluster.