Using Ab Initio Calculations in Designing bcc MgLi–X Alloys for Ultra‐Lightweight Applications

Body centered cubic (bcc) Mg–Li‐based alloys are a promising light‐weight structural material. In order to tailor the Mg–Li composition with respect to specific industrial requirements, systematic materials‐design concepts need to be developed and applied. Quantum‐mechanical calculations are increasingly employed when designing new alloys as they accurately predict basic thermodynamic, structural, and functional properties using only the atomic composition as input. We have therefore performed a quantum‐mechanical study using density functional theory (DFT) to systematically explore fundamental physical properties of a broad set of bcc MgLi‐based compounds. These DFT‐determined properties are used to calculate engineering parameters such as (i) the specific Young's modulus (Y/ρ) or (ii) the bulk over shear modulus ratio (B/G) which allow differentiating between brittle and ductile behavior. As we have recently shown, it is not possible to increase both specific Young's modulus, as a measure of strength, and B/G ratio, as a proxy for ductility, by changing only the composition in the binary bcc Mg–Li system. In an attempt to bypass such fundamental materials‐design limitations, a large set of MgLi–X substitutional ternaries derived from stoichiometric MgLi with CsCl structure are studied. Motivated by the fact that for Mg–Li alloys (i) 3rd row Si and Al and (ii) 4th row Zn are industrially used as alloying elements, we probe the alloying performance of the 3rd (Na, Al, Si, P, S, Cl) and 4th row transition metal (Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn) elements. The studied solutes offer a variety of properties but none is able to simultaneously improve both specific Young's modulus and ductility. Therefore, in order to explore the alloying performance of yet a broader set of solutes, we predict the bulk modulus of MgX and LiX B2‐compounds running over 40 different elements.     

The selection of effective retrofit scenarios for panel houses in urban neighborhoods based on expected energy savings and increase in market value: The Vilnius case

Some of the problems associated with assessing the retrofit effectiveness of apartment buildings in urban areas are considered. The retrofit of houses should be followed by the amelioration of their surroundings. The priority order of districts to be renovated depends on the condition of the buildings in a district and on strategic urban development programmes. In order to determine the profitability of investments in housing retrofit, a number of retrofit scenarios should be developed. The authors of this paper offer a new approach to determining the retrofit effectiveness of houses based both on expected energy savings and the increase in market value of renovated buildings. In line with the proposed approach, retrofit scenarios for apartment buildings in Vilnius were developed, i.e. retrofit investment packages for various districts were prepared and arranged in the priority order for their application according to the method of geographical analysis suggested by the authors.     

Efficient cerium-based sol-gel derived phosphors in different garnet matrices for light-emitting diodes

Efficiency of sol-gel derived Y_3−xAl₅O₁₂:Ce_x³⁺ (YAG:Ce) and Y_3−xMg₂AlSi₂O₁₂:Ce_x³⁺ (YMASG:Ce) phosphors, which are prospective for application in white light emitting diodes (LED), is studied. Sets of samples containing different cerium amount x from 0.015 to 0.06 and sintered at different temperatures (1400-1550 °C) were investigated. Importance of absorption peculiarities in agglomerates of phosphor nanocrystals is demonstrated by studying the excitation wavelength dependence of quantum efficiency and by applying PL measurements in confocal mode. Emission saturation is demonstrated to occur at higher excitation intensities than those typical for operating white LEDs.     

Improving sale performance prediction using support vector machines

In this article, an expert system based on support vector machines is developed to predict the sale performance of some insurance company candidates. The system predicts the performance of these candidates based on some scores, which are measurements of cognitive characteristics, personality, selling skills and biodata. An experiment is conducted to compare the accuracy of the proposed system with respect to previously reported systems which use discriminant functions or decision trees. Results show that the proposed system is able to improve the accuracy of a baseline linear discriminant based system by more than 10% and that also exceeds the state of the art systems by almost 5%. The proposed approach can help to reduce considerably the direct and indirect expenses of the companies.     

Scaling Law for Electrospray Droplet Formation

There has been a lot of research on liquid atomization by means of electrospray, particularly because of its many practical applications. Nevertheless, full understanding and control of the electrospray mechanisms are still incomplete. An experimental setup was developed in order to investigate the frequency characteristics of droplet formation and ejection at the tip of Taylor's cone of aqueous electrosprays. Droplet formation and oscillations were monitored using a Kodak Ektapro high-speed camera. The frequency of droplet formation at the tip of the capillary needle was analyzed. Droplet frequency formation appears to exhibit three distinct regimes with an abrupt transition from one regime to another. Droplet formation was recorded at different needle-plate electrode distances. Based on the analysis of experimental data a scaling law for droplet formation was found.     

Transistor-like behavior of single metalloprotein junctions

Single protein junctions consisting of azurin bridged between a gold substrate and the probe of an electrochemical tunneling microscope (ECSTM) have been obtained by two independent methods that allowed statistical analysis over a large number of measured junctions. Conductance measurements yield (7.3 ± 1.5) × 10(-6)G(0) in agreement with reported estimates using other techniques. Redox gating of the protein with an on/off ratio of 20 was demonstrated and constitutes a proof-of-principle of a single redox protein field-effect transistor.     

Variations in systemic and pulmonary endothelin-1 in horses with recurrent airway obstruction (heaves)

Recurrent airway obstruction (RAO) is an asthma-like condition of the horse that represents a major cause of morbidity and loss of performance. The exact pathogenesis of asthma in man is unclear but the role of endothelin (ET) is currently under investigation, thus sparking interest in the bronchoconstrictive and vasoconstrictive properties of endothelin in the equine-specific disease entity. In this study, we investigated the levels of ET-1 in systemic blood, as well as in bronchoalveolar lavage (BAL) from horses with RAO. We also studied how these values might correlate with those of lung function tests and pulmonary artery pressure. Five horses with RAO were evaluated both in remission and in crisis and compared to five control horses. RAO horses had significantly (P<0.05) higher systemic ET-1 levels than control horses. They also had a negative arteriovenous ET-1 difference that may correspond to a net uptake of ET-1 in the lung. RAO horses in crisis had increased amounts of immunoreactive ET in BAL fluid compared to normal control subjects. Additionally, the reduction in lung function seen in RAO horses in crisis was significantly correlated with lower epithelial lining fluid ET-1 levels. Our results demonstrate that endothelin may contribute to the pathogenesis of asthma.     

Pulmonary function values in Friesian and double-muscled calves during acute poikilocapnic hypoxia

In the genus Eulemur (Malagasy lemurs) karyotype diversification has occurred mainly through Robertsonian mechanisms of chromosome fusion (Rumpler et al., 1976). Eulemur coronatus is the sole species to have the largest genome size, due to a very large amount of C-heterochromatin, mostly located at the pericentromeric regions of the largest chromosomes (Warter and Rumpler, 1985). This increase in C-heterochromatin was thought to be due to DNA amplification (Ronchetti et al., 1993). The aim of this work was to investigate whether the large C-heterochromatin of Eulemur coronatus might have derived by amplification of the smaller C-heterochromatin of Eulemur macaco, a closely related species with smaller genome size. To obtain information on the overall base composition of the total genomes, on the relative interspersion of AT and GC base paris along the DNA molecule and on the structural differences in C-heterochromatin, we used a quantitative cyto-chemical approach, based on fluorescence resonance energy transfer (FRET) between DNA-specific fluorochromes (i.e. the AT-specific Hoechst 33258, and the non base-specific dye, propidium iodide). Micro-spectrofluorometry and image analysis were used to investigate both the overall FRET efficiency and its spatial distribution along the chromosome arms. FRET efficiency values of the DNA in C-heterochromatin were significantly different in the two Eulemur species, indicating a different qualitative composition of repetitive DNA. This suggests that the repetitive DNA of Eulemur coronatus cannot have originated by amplification in toto of the repetitive DNA sequences of Eulemur macaco.     

Advantages of Unbiased Support Vector Classifiers for Data Mining Applications

Many learning algorithms have been used for data mining applications, including Support Vector Classifiers (SVC), which have shown improved capabilities with respect to other approaches, since they provide a natural mechanism for implementing Structural Risk Minimization (SRM), obtaining machines with good generalization properties. SVC leads to the optimal hyperplane (maximal margin) criterion for separable datasets but, in the nonseparable case, the SVC minimizes the L ₁ norm of the training errors plus a regularizing term, to control the machine complexity. The L ₁ norm is chosen because it allows to solve the minimization with a Quadratic Programming (QP) scheme, as in the separable case. But the L ₁ norm is not truly an “error counting” term as the Empirical Risk Minimization (ERM) inductive principle indicates, leading therefore to a biased solution. This effect is specially severe in low complexity machines, such as linear classifiers or machines with few nodes (neurons, kernels, basis functions). Since one of the main goals in data mining is that of explanation, these reduced architectures are of great interest because they represent the origins of other techniques such as input selection or rule extraction. Training SVMs as accurately as possible in these situations (i.e., without this bias) is, therefore, an interesting goal. We propose here an unbiased implementation of SVC by introducing a more appropriate “error counting” term. This way, the number of classification errors is truly minimized, while the maximal margin solution is obtained in the separable case. QP can no longer be used for solving the new minimization problem, and we apply instead an iterated Weighted Least Squares (WLS) procedure. This modification in the cost function of the Support Vector Machine to solve ERM was not possible up to date given the Quadratic or Linear Programming techniques commonly used, but it is now possible using the iterated WLS formulation. Computer experiments show that the proposed method is superior to the classical approach in the sense that it truly solves the ERM problem.