The cationic polymerization of α-methylstyrene in liquid sulfur dioxide initiated by the HI/I2 system

The polymerization of α-methylstyrene (αMeSty) initiated by HI/I₂ or HI in the presence of liquid sulfur dioxide has been investigated. The number-average molecular weight increased with the monomer concentration for reactions initiated by the HI/I₂ system. I₂ also participates in the initiation process, increasing the number-average polymer chain at higher monomer concentration. HI alone is also able to initiate the polymerization of αMeSty in the presence of SO₂. With this initiator, transfer reaction can be minimized in systems containing low amount of SO₂. Received: 19 December 1996/Revised: 27 January 1997/Accepted 29 January 1997     

Blends of i‐PP and SBS. I. Influence of the in situ compatibilization on the morphology

This article concerns the in situ compatibilization of immiscible isotactic polypropylene/butadiene‐styrene‐butadiene triblock copolymer blends (i‐PP/SBS) by means of a reactive mixture. For this purpose, maleated PP (PP‐MAH) and SBS (SBS‐MAH) were used as functionalized polymers and 4,4′‐diaminediphenylmethane was used as a coupling agent between maleated polymers, resulting in a graft copolymer. Binary blends i‐PP/SBS, nonreactive ternary blends i‐PP/PP‐MAH/SBS, and reactive ternary blends i‐PP/PP‐MAH/SBS‐MAH with varying diamine and anhydride molar ratios were prepared. Torque measurements suggest a graft copolymerization during the melt blending for ternary reactive blends, but the extension of the grafting does not vary with the diamine and anhydride molar ratio, but with the elastomer concentration. The morphology of the blends was investigated by scanning electron microscopy. The morphology of binary and ternary nonreactive blends is similar, exhibiting elastomer domains disperse in the i‐PP matrix, whose sizes increase with elastomer concentration. On the other hand, the elastomer domain size in the ternary reactive blends is practically independent of the blends composition and of the diamine and anhydride molar ratio. © 2002 Wiley Periodicals, Inc. J Appl Polym Sci 85: 847–855, 2002     

DSC, dielectric and dynamic mechanical behavior of two- and three-component ordered polyurethanes

Liquid crystalline (LC) polyurethanes were made from two diisocyanates (flexible HMDI and stiff TDI) (DIs), mesogenic diol (D) and a polybutadiene-diol (B) with stoichiometric ratios of reactive hydroxy (OH) and isocyanate (NCO) groups ((NCO)_DI/((OH)_D+(OH)_B)=1/1). Two- (B/DIs, D/DIs) and three-component ((D+B)/DIs, D/B=1/1 by weight) polymers were prepared and their dielectric, dynamic mechanical and DSC behavior was investigated. For neat B, the glass transition temperature T_gB (∼−46 °C) was detected. Two-component B/HMDI and B/TDI polymers have exhibited a homogeneous structure with the glass transition temperatures T_gU∼−9 and 2 °C. On the other hand, for D/DI polymers on cooling from the melt and subsequent heating the glass transitions at T_gU∼41 °C (D/HMDI) and 58 °C (D/TDI) together with nematic and smectic mesophases were found. In three-component systems, additional glass transitions at T_gB∼−41 °C (B/D/HMDI) and −31 °C (B/D/TDI) were observed. This means that the polymers exhibit a distinct two-phase structure with soft polybutadiene (B) and hard polyurethane (D/DI) phases. In hard polyurethane phase, the glass transitions at T_gU and LC mesophases similar to those found in two-component D/DI polyurethanes were detected. Dielectric and dynamic mechanical results correlate well with DSC measurements.     

Synthesis, characterization and crystal structure of a novel thiocyanate–ruthenium(II) complex

The cis-[Ru(dppb)(Me-bipy)(NCS)₂], dppb = 1,4-bis (diphenylphosphino)butane, Me-bipy = 4,4′-dimethyl-2,2′-bipyridine, and NCS = thiocyanate, was synthesized and characterized by spectroscopic and electrochemical techniques and its structure was determined by crystal X-ray analysis. The crystal structure reveals that the coordination geometry around the Ru(II) center is distorted octahedron where two molecules of thiocyanate are bonded to the ruthenium through nitrogen atom in cis orientation. The half-wave formal potential value E_1/2 = 0.8 V (versus Ag/AgCl) observed is considerable higher than that for the cis-[RuCl₂(dppb)(Me-bipy)] complex, E_1/2 = 0.6 V (versus Ag/AgCl), well illustrating the strong π-acceptor effect the NCS ligand toward the backbonding interaction with the Ru(II) metal center. The MLCT absorption bands of the thiocyanate complex present a higher molar absorptivity (about 12%) compared with the cis-[RuCl₂(dppb)(Me-bipy)] complex, in the same experimental conditions. These properties make the complex potentially promising for the photosensitization process.     

The Influence of the Acid Sites on the Methylamines Synthesis with Cu-HZSM-5 Zeolite

There is a high industrial demand for MMA and DMA, however, the production of TMA is favored with the commercial catalysts. This work tested the methanol amination reaction with HZSM-5 and Cu/HZSM-5 zeolites (%Cu = 1.2, 3.7 and 6.0). Regarding the MMA selectivity, light differences were observed among Cu/HZSM-5 catalysts and pure zeolite, although the catalysts containing 3.7% copper was slightly more selective to MMA than the other metallic catalysts studied. From these results it is suggested that the metal loading present in the catalysts leads to a modification in the nature of the acidic sites, in particular, the selectivity depends on ratio Brönsted/Lewis acid sites. This modifications would facilitate the amines desorption during the reaction.     

Electrochemical behaviour of heat-treated AlZnMg alloys in chloride solutions containing sulphate

The electrochemical corrosion and passivation of Al5Zn1.7Mg0.23Cu0.053Nb alloys, submitted to different heat treatments (cold-rolled, annealed, quenched and aged, and quenched in two steps and aged), in sulphate-containing chloride solutions, has been studied by means of cyclic polarization, electrochemical impedance spectroscopy (EIS), scanning electron microscopy (SEM), energy-dispersive X-ray (EDX), and X-ray photoelectron spectroscopy (XPS). The cyclic polarization curves showed that sulphate addition to the chloride solution produced a poor reproducible shift of the breakdown potential to more positive potentials. The repassivation potentials, much more reproducible, and practically separating the passive from the pitting potential region, were slightly displaced in the negative direction with that addition. When the alloys were potentiodynamically polarized in the passive potential region, sulphate was incorporated in the oxide film, thus precluding chloride ingress. In addition, Zn depletion was favoured, whereas Mg losses were avoided. Different equivalent circuits corresponding to different alloys and potentials in the passive and pitting regions were employed to account for the electrochemical processes taking place in each condition. This work shows that sulphate makes these alloys more sensitive to corrosion, increasing the fracture properties of the surface layer and favouring the pitting attack over greater areas than chloride alone.     

Effect of the addition of Cr and Nb on the microstructure and electrochemical corrosion of heat-treatable Al-Zn-Mg alloys

The effect of the addition of Cr and Nb on the microstructure and the electrochemical corrosion of the weldable, high-strength and stress corrosion cracking (SCC) resistant Al-5%Zn–1.67%Mg–0.23%Cu alloy (H) has been studied. Combined additions of the alloying elements, J (with Nb), L (with Cr) and O (with Cr and Nb) and different heat treatments, ST (cold-rolled), A (annealed), F (quenched), B (quenched and aged) and C (quenched in two steps and aged), to obtain different microstructures and hardness have been performed. To correlate the electrochemical corrosion with the microstructure of the specimens, corrosion potential (E _cor) measurements in different chloride solutions were performed and optical microscopy, SEM, TEM and EDX were applied. In chloride solutions containing dissolved O₂ or H₂O₂, the present alloys were polarized up to the pitting attack. It was shown that theE _cor measurements were very sensitive to the alloy composition and heat treatment, increasing in the order H < J < L < O < Al (for a given heat treatment) and F < A ST < B < C (for a given alloy). The MgZn₂ precipitates of the annealed (A) and cold-rolled (ST) specimens were dissolved in chloride solutions containing oxidizing agents and pitting attack was shown to develop in the cavities where the precipitates were present. In the specimens B and C, the compositions of the precipitate free zones was found to be equal to that of the matrix solid solution and preferential intergranular attack was not evident, this being in agreement with their SCC resistance. The addition of Cr and Nb increased the pitting corrosion resistance. The effects of Cr and Nb were additive, that of Cr being predominant, either, in theE _cor shift or in the increase in the pitting corrosion resistance.     

Sense representations for Portuguese: experiments with sense embeddings and deep neural language models

Sense representations have gone beyond word representations like Word2Vec, GloVe and FastText and achieved innovative performance on a wide range of natural language processing tasks. Although very useful in many applications, the traditional approaches for generating word embeddings have a strict drawback: they produce a single vector representation for a given word ignoring the fact that ambiguous words can assume different meanings. In this paper, we explore unsupervised sense representations which, different from traditional word embeddings, are able to induce different senses of a word by analyzing its contextual semantics in a text. The unsupervised sense representations investigated in this paper are: sense embeddings and deep neural language models. We present the first experiments carried out for generating sense embeddings for Portuguese. Our experiments show that the sense embedding model (Sense2vec) outperformed traditional word embeddings in syntactic and semantic analogies task, proving that the language resource generated here can improve the performance of NLP tasks in Portuguese. We also evaluated the performance of pre-trained deep neural language models (ELMo and BERT) in two transfer learning approaches: feature based and fine-tuning, in the semantic textual similarity task. Our experiments indicate that the fine tuned Multilingual and Portuguese BERT language models were able to achieve better accuracy than the ELMo model and baselines.

     

Learning to rank academic experts in the DBLP dataset

Expert finding is an information retrieval task that is concerned with the search for the most knowledgeable people with respect to a specific topic, and the search is based on documents that describe people's activities. The task involves taking a user query as input and returning a list of people who are sorted by their level of expertise with respect to the user query. Despite recent interest in the area, the current state‐of‐the‐art techniques lack in principled approaches for optimally combining different sources of evidence. This article proposes two frameworks for combining multiple estimators of expertise. These estimators are derived from textual contents, from graph‐structure of the citation patterns for the community of experts and from profile information about the experts. More specifically, this article explores the use of supervised learning to rank methods, as well as rank aggregation approaches, for combining all of the estimators of expertise. Several supervised learning algorithms, which are representative of the pointwise, pairwise and listwise approaches, were tested, and various state‐of‐the‐art data fusion techniques were also explored for the rank aggregation framework. Experiments that were performed on a dataset of academic publications from the Computer Science domain attest the adequacy of the proposed approaches.