Optimal Monotonicity-Preserving Perturbations of a Given Runge–Kutta Method

We propose a fast algorithm for evaluating the moments of Bingham distribution. The calculation is done by piecewise rational approximation, where interpolation and Gaussian integrals are utilized. Numerical tests show that the algorithm reaches the maximum absolute error less than \(5\times 10^{-8}\) remarkably faster than adaptive numerical quadrature. We apply the algorithm to a model for liquid crystals with the Bingham distribution to examine the defect patterns of rod-like molecules confined in a sphere, and find a different pattern from the Landau-de Gennes theory.     

Oil-free protein isolates from full-fat,dehulled mustard flour by microfiltration

New microfiltration (MF)-based aqueous processes for the extraction of oil-free protein isolates from full-fat, dehulled mustard flour have been developed. The processes used hydrophilic MF membranes to separate oil and protein. The most successful processing sequence consisted of alkaline extraction of oil and protein from the flour at pH 11, centrifugation to remove undissolved solids, and an initial microfiltration step to separate oil and dissolved protein. This was followed by proteolytic enzyme treatment of the retentate and a second-stage MF step to recover further protein. The proteolytic enzyme treatment was designed to break aggregated proteins into smaller fractions to permit their passage through the MF membrane. The permeates from the two microfiltration steps were ultrafiltered to concentrate the protein and remove antinutritional compounds and highly fragmented peptides. With this process, 60% of the protein originally in the flour was separated from the oil phase. Approximately 40% of the protein present in the flour was recovered in the form of two oil-free protein isolates—a soluble protein isolate containing 91% protein and a precipitated protein isolate containing 100% protein. The process proves the potential of MF as an effective method of extracting oil-free protein isolates from oilseeds without the use of organic solvents. Based on a presentation at the 16th Annual Meeting of the Canadian Section of AOCS held in Winnipeg, Manitoba, Canada, September 29–October 1, 2001.     

BoostingTree: parallel selection of weak learners in boosting, with application to ranking

Boosting algorithms have been found successful in many areas of machine learning and, in particular, in ranking. For typical classes of weak learners used in boosting (such as decision stumps or trees), a large feature space can slow down the training, while a long sequence of weak hypotheses combined by boosting can result in a computationally expensive model. In this paper we propose a strategy that builds several sequences of weak hypotheses in parallel, and extends the ones that are likely to yield a good model. The weak hypothesis sequences are arranged in a boosting tree, and new weak hypotheses are added to promising nodes (both leaves and inner nodes) of the tree using some randomized method. Theoretical results show that the proposed algorithm asymptotically achieves the performance of the base boosting algorithm applied. Experiments are provided in ranking web documents and move ordering in chess, and the results indicate that the new strategy yields better performance when the length of the sequence is limited, and converges to similar performance as the original boosting algorithms otherwise.     

Preparation of PTFE/graphene nanocomposites by compression moulding and free sintering: A guideline

This work was aimed at preparing polytetrafluoroethylene (PTFE) nanocomposites filled with graphene nanoplatelets and investigating how the graphene nanoplatelets and the preparation techniques influenced the physical properties. Graphene was incorporated up to 4 vol % of the total PTFE system by dry and solvent assisted blending. The powder compaction was evaluated using the Kawakita/Ludde model to describe the compressibility of the powder blends. The nanocomposite billets were prepared using cold compression moulding by applying preform pressures between 12.7 and 140 MPa and the preform billets were sintered at 380°C using a specific sintering cycle. The changes in the physical dimensions, billet mass, density, and void content of the billets, pre and post sintering, were analyzed with experimental design to evaluate the influence of the precompaction pressure and graphene loading. From the evaluation it was concluded that the ideal compaction pressure was at 12.7 MPa and the solvent assisted blending was superior to the mechanical blending method. Furthermore, the compression creep tests confirmed the ideal processing temperature and graphene loading range to improve the mechanical properties. © 2016 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2016 , 133, 43369.     

Thermoplastic elastomers based on recycled high‐density polyethylene,ethylene–propylene–diene monomer rubber,and ground tire rubber

High‐performance thermoplastic elastomers (TPEs), based on recycled high‐density polyethylene (HDPE^R), olefinic type ethylene–propylene–diene monomer rubber (EPDM), and ground tire rubber (GTR) treated with bitumen, were prepared by using dynamic vulcanization technology, and their structure–property relationships were investigated. It was established that special pretreatment of GTR by bitumen confers outstanding mechanical properties on the resulting TPEs. TPEs, containing GTR pretreated by bitumen, exhibit thermal behavior similar to that of the HDPE^R/EPDM basic blend in the temperature region up to about 340°C. Rheological measurements showed that bitumen acts as an effective plasticizer for the GTR‐containing TPEs. SEM, DSC, and DMTA results revealed improved adhesion between the particles of GTR treated by bitumen and the surrounding thermoplastic matrix, compared to that of the untreated GTR particles. It was concluded that bitumen acts as an effective devulcanizing agent in the GTR treatment stage. In the following steps of TPE production, bitumen acts simultaneously as a curing agent for the rubber components (EPDM/GTR) and as a compatibilizer for the blend components. GTR‐containing TPEs, prepared by extrusion technology, were reprocessed (by passing through the extruder six times) without any observable changes in their tensile properties, thermal stability, and melt viscosity. © 2004 Wiley Periodicals, Inc. J Appl Polym Sci 95: 659–671, 2005     

pCBT/MWCNT Nanocomposites Prepared by In situ Polymerization of CBT After Solid‐Phase High‐Energy Ball Milling of CBT with MWCNT

pCBT/MWCNT nanocomposites were prepared by in situ polymerization of CBT after solid‐phase HEBM of the polymerization catalyst containing CBT with MWCNT. The crystallinity and crystallization behavior of the pCBT nanocomposites were studied by WAXS and DSC. The MWCNTs did not affect the crystallinity of the isothermally produced pCBT significantly, but acted as nucleation agents during the crystallization of pCBT from its melt. pCBT/MWCNT nanocomposites were subjected to DMTA, static flexure, and dynamic Charpy impact tests. The flexural modulus, strength, and impact strength from these tests all went through a maximum as a function of the MWCNT content. Optimum properties were found in the MWCNT range of 0.25–0.5 wt.‐%.

     

Destabilization of Yellow Mustard Emulsion Using Organic Solvents

An aqueous extraction process was developed consisting of aqueous contact with dehulled yellow mustard flour to recover protein followed by dissolution of the released emulsion in dimethylformamide (DMF) or isopropyl alcohol (IPA) to recover the released oil in the form of single-phase oil–solvent miscellae suitable for industrial applications. Only some 38 ± 3 % of the oil in the yellow mustard emulsion was extracted using DMF even at high weight ratios since DMF is widely miscible with water, preventing separation of the oil from the emulsion. A ternary phase diagram of DMF/oil/water was prepared and confirmed the limited solubility of the oil in DMF in the presence of water. The use of 31:1 IPA:oil weight ratio could effectively recover over 94 % of the oil in the emulsion; however, multiple-stage treatment of the emulsion was proven to be more efficient with lower volumes of IPA required to achieve high oil extraction yields. The results suggest that the optimal conditions for multiple-stage process were four stages using 2:1 IPA:oil weight ratio, with 96 ± 1 % oil recovery from the emulsion.     

Thickening effects of interactions between engine oil additives

Additives of low and high molecular weight used as performance enhancers in engine oil formulations can interact through their polar functional groups to form associates and complexes. These often have a significant thickening effect that could substitute for viscosity and viscosity‐index improvers. We therefore investigated the stability of these associates to thermal degradation, oxidation, and mechanical decomposition, in an attempt to seek information about their feasibility.