Showing papers by "Jožef Stefan Institute published in 2005"

Impact of the Carbon Coating Thickness on the Electrochemical Performance of LiFePO4 / C Composites

Abstract: Porous, well crystalline LiFePO 4 /C composites with different amounts of carbon have been prepared by a sol-gel technique. The thickness of carbon coatings (paintings) has been determined by high-resolution transmission electron microscopy. It is shown that carbon coating thickness can be controlled by the amount of carbon and it has an impact on the obtained reversible capacity. Furthermore, it is shown that atomic ratio of nonactive Fe(III) phase (presumably Fe 3 P) in as-synthesized LiFePO 4 /C composites depends on the amount of carbon in the composite. Using Mossbauer spectroscopy, we have confirmed that the nonactive Fe(III) remains nearly unchanged in the composite during cycling. The lowest amount of carbon in LiFePO 4 /C composites obtained from citrate anion as a gelling agent was 3.2 wt % and this particular amount corresponds to the carbon coating thickness of about 1-2 nm. The reversible capacity obtained from the above-mentioned composite delivers close to 80% of the theoretical capacity at room temperature with a current density of 170 mA/g (C/1 rate).

Mercury in environmental samples: Speciation, artifacts and validation

Abstract: Speciation of mercury compounds in environmental samples requires rigorous analytical procedures at each stage of sample collection, treatment and measurement. Sampling, the first step, is generally only critical for water samples. The best materials for water-sample storage and processing are Pyrex and Teflon (PTFE or FEP) after a severe cleaning procedure. Extraction of the Hg species from its matrix, the second step, requires an aggressive treatment, such as acid extraction (mostly combined with solvent extraction), distillation or alkaline extraction. Extraction is one of the most critical steps, and, for biota and sediments, almost certainly the most critical, because two conflicting issues need to be addressed – obtaining adequate recovery, and preventing losses. Extraction efficiency and validation are discussed as well as methylation artifacts, especially when the distillation method is applied to sediments. Separation of the Hg species, the third step, includes derivatization followed by gas chromatography (GC), new GC improvements and liquid chromatography (LC), which has, since the development of more sensitive detectors, resulted in wider applications in environmental studies. Detection of the Hg species is the fourth and last step in the analytical procedure. The development of commercial, relatively inexpensive, extremely sensitive, selective cold vapor atomic fluorescence spectrometry (CV-AFS) instrumentation in the late 1980s and 1990s made this the most popular detector for the laboratories working on the biogeochemical cycling of Hg. In recent years, the use of inductively coupled plasma mass spectrometry (ICP-MS) in speciation analysis has increased tremendously. Besides its high sensitivity and selectivity, ICP-MS offers the opportunity to perform speciated isotope dilution mass spectrometry (SID-MS). Finally, as other measurement fields, speciation analysis requires suitable reference materials to be available for the verification of accuracy and to meet quality-assurance needs. We also address these important aspects.

Dynamic imaging of protease activity with fluorescently quenched activity-based probes

Abstract: Protease activity is tightly regulated in both normal and disease conditions. However, it is often difficult to monitor the dynamic nature of this regulation in the context of a live cell or whole organism. To address this limitation, we developed a series of quenched activity-based probes (qABPs) that become fluorescent upon activity-dependent covalent modification of a protease target. These reagents freely penetrate cells and allow direct imaging of protease activity in living cells. Targeted proteases are directly identified and monitored biochemically by virtue of the resulting covalent tag, thereby allowing unambiguous assignment of protease activities observed in imaging studies. We report here the design and synthesis of a selective, cell-permeable qABP for the study of papain-family cysteine proteases. This probe is used to monitor real-time protease activity in live human cells with fluorescence microscopy techniques as well as standard biochemical methods.

Dynamic systems identification with Gaussian processes

Abstract: This paper describes the identification of nonlinear dynamic systems with a Gaussian process (GP) prior model. This model is an example of the use of a probabilistic non-parametric modelling approach. GPs are flexible models capable of modelling complex nonlinear systems. Also, an attractive feature of this model is that the variance associated with the model response is readily obtained, and it can be used to highlight areas of the input space where prediction quality is poor, owing to the lack of data or complexity (high variance). We illustrate the GP modelling technique on a simulated example of a nonlinear system.

Similarities and Specificities of Fungal Keratinolytic Proteases: Comparison of Keratinases of Paecilomyces marquandii and Doratomyces microsporus to Some Known Proteases

Abstract: Based on previous screening for keratinolytic nonpathogenic fungi, Paecilomyces marquandii and Doratomyces microsporus were selected for production of potent keratinases. The enzymes were purified and their main biochemical characteristics were determined (molecular masses, optimal temperature and pH for keratinolytic activity, N-terminal amino acid sequences). Studies of substrate specificity revealed that skin constituents, such as the stratum corneum, and appendages such as nail but not hair, feather, and wool were efficiently hydrolyzed by the P. marquandii keratinase and about 40% less by the D. microsporus keratinase. Hydrolysis of keratin could be increased by the presence of reducing agents. The catalytic properties of the keratinases were studied and compared to those of some known commercial proteases. The profile of the oxidized insulin B-chain digestion revealed that both keratinases, like proteinase K but not subtilisin, trypsin, or elastase, possess broad cleavage specificity with a preference for aromatic and nonpolar amino acid residues at the P-1 position. Kinetic studies were performed on a synthetic substrate, succinyl-Ala-Ala-Pro-Phe-p-nitroanilide. The keratinase of P. marquandii exhibited the lowest Km among microbial keratinases reported in the literature, and its catalytic efficiency was high in comparison to that of D. microsporus keratinase and proteinase K. All three keratinolytic enzymes, the keratinases of P. marquandii and D. microsporus as well as proteinase K, were significantly more active on keratin than subtilisin, trypsin, elastase, chymotrypsin, or collagenase.

Synthesis of materials within reverse micelles

Abstract: Reverse micelles as nanosized aqueous droplets existing at certain compositions of water-in-oil microemulsions are widely used today in the synthesis of many types of nanoparticles. However, without a rich conceptual network that would correlate the properties and compositions of reverse micellar microemulsions to the properties of to-be-obtained particles, the design procedures in these cases usually rely on a trial-and-error approach. As like every other science, what is presently known is merely the tip of the iceberg compared to the uninvestigated vastness still lying below. The aim of this article is to present readers with most of the major achievements from the field of materials synthesis within reverse micelles since the first such synthesis was performed in 1982 until today, to possibly open up new perspectives of viewing the typical problems that nowadays dominate the field, and to hopefully initiate the observation and generation of their actual solutions. We intend to show that by refining th...

Alkaline-earth doping in (K,Na)NbO3 based piezoceramics

Abstract: In this contribution the effect of alkaline-earth (AE) dopants Mg2+, Ca2+, Sr2+ and Ba2+ on sinterability and functional response of KNN is studied. Ceramic powders with the general formula (K0.5Na0.5)1−2yAEyNbO3, y = 0.005 were prepared by classical ceramic processing. The values of 94–95% of theoretical density (TD) are obtained for stoichiometric KNN after sintering at 1115 °C for 2 h. Calcium and strontium doping promotes densification while magnesium doping inhibits it. In the case of Ba-doping a secondary phase forms. The dielectric constant and piezo d33 coefficient determined by Berlincourt piezo d33-meter of Ca and Sr-doped KNN ceramics are 500 and 95 pC/N, respectively, while the values for undoped KNN are slightly lower.

EFFTRAN : A Monte Carlo efficiency transfer code for gamma-ray spectrometry

Abstract: An implementation of the efficiency transfer method for cylindrical samples in gamma-ray spectrometry is presented, based on a Monte Carlo integration of the interaction probabilities of gamma rays over the detector and sample volumes. This approach makes it possible for the computer code to be relatively straightforward to write and the speed of the calculation to be adequate for routine on-line analysis of gamma-ray spectra. Virtual components of the detector system are introduced to prevent any erroneous or double counting of gamma-ray paths, which simplifies the tracking algorithm.

Microarray data mining with visual programming

Abstract: Summary: Visual programming offers an intuitive means of combining known analysis and visualization methods into powerful applications. The system presented here enables users who are not programmers to manage microarray and genomic data flow and to customize their analyses by combining common data analysis tools to fit their needs. Availability: http://www.ailab.si/supp/bi-visprog Contact: blaz.zupan@fri.uni-lj.si Supplementary information: http://www.ailab.si/supp/bi-visprog

Impact of LiFePO4 ∕ C Composites Porosity on Their Electrochemical Performance

Abstract: Fe(III) citrate was used as a source for synthesis of microsized porous LiFePO 4 /C particles. All samples, prepared either by solid-state or by sol-gel techniques, are phase-pure triphylite phases, which, however, have different morphology highly influenced by the type of synthesis and synthesis parameters. Their common feature is porosity due to thermal decomposition of citrate anion. The impact of particle porosity on the electrochemical behavior is discussed in terms of qualitative results obtained from scanning electron microscopy (SEM) micrographs and in terms of quantitative results obtained from N 2 adsorption isotherms. The best electrochemical behavior (above 140 mAh/g at C/2 rate during continuous cycling) was obtained with composites prepared at a relatively high heating rate (above 5 K/min). This suggests that interlaced pores were formed inside particles. A strong correlation between the electrochemical results and the heating rate was observed, which could easily be explained based on SEM micrographs and on some trends found in porosity measurements. The latter reveal the main difference between samples prepared by solid-state and by sol-gel techniques.

Determination of Valence States of Chromium in Calcium Chromates by Using X‐ray Absorption Near‐Edge Structure (XANES) Spectroscopy

Abstract: Calcium chromates with the empirical formulas Ca 10 Cr 6 -O 25 , Ca 3 Cr 2 O 8 and Ca 5 Cr 3 O 12 , which form at temperatures >900°C at CaO:Cr 2 O 3 molar ratios of ≥3 in an oxidation atmosphere have been synthesized in the pure state. X-ray absorption near-edge structure (XANES) spectroscopy has been used to determine the average valence state of chromium in the samples. The presence of unusual chromium valence states, 4+ and 5+, which was proposed via X-ray diffractometry studies, is strongly supported.

Semi-automatic construction of topic ontologies

Abstract: In this paper, we review two techniques for topic discovery in collections of text documents (Latent Semantic Indexing and K-Means clustering) and present how we integrated them into a system for semi-automatic topic ontology construction. The OntoGen system offers support to the user during the construction process by suggesting topics and analyzing them in real time. It suggests names for the topics in two alternative ways both based on extracting keywords from a set of documents inside the topic. The first set of descriptive keyword is extracted using document centroid vectors, while the second set of distinctive keyword is extracted from the SVM classification model dividing documents in the topic from the neighboring documents.

A novel type of proximity focusing RICH counter with multiple refractive index aerogel radiator

Abstract: A proximity focusing ring imaging Cherenkov detector, with the radiator consisting of two or more aerogel layers of different refractive indices, has been tested in 1 – 4 GeV / c pion beams at KEK. Essentially, a multiple refractive index aerogel radiator allows for an increase in Cherenkov photon yield on account of the increase in overall radiator thickness, while avoiding the simultaneous degradation in single photon angular resolution associated with the increased uncertainty of the emission point. With the refractive index of consecutive layers suitably increasing in the downstream direction, one may achieve overlapping of the Cherenkov rings from a single charged particle. The impact of the observed improvement on the π / K separation at the upgraded Belle detector is discussed.

Discrete component analysis

Abstract: This article presents a unified theory for analysis of components in discrete data, and compares the methods with techniques such as independent component analysis, non-negative matrix factorisation and latent Dirichlet allocation. The main families of algorithms discussed are a variational approximation, Gibbs sampling, and Rao-Blackwellised Gibbs sampling. Applications are presented for voting records from the United States Senate for 2003, and for the Reuters-21578 newswire collection.

Feature selection for dimensionality reduction

Abstract: Dimensionality reduction is a commonly used step in machine learning, especially when dealing with a high dimensional space of features. The original feature space is mapped onto a new, reduced dimensionally space. The dimensionality reduction is usually performed either by selecting a subset of the original dimensions or/and by constructing new dimensions. This paper deals with feature subset selection for dimensionality reduction in machine learning. We provide a brief overview of the feature subset selection techniques that are commonly used in machine learning. Detailed description is provided for feature subset selection as commonly used on text data. For illustration, we show performance of several methods on document categorization of real-world data.

Constraint based induction of multi-objective regression trees

Abstract: Constrained based inductive systems are a key component of inductive databases and responsible for building the models that satisfy the constraints in the inductive queries. In this paper, we propose a constraint based system for building multi-objective regression trees. A multi-objective regression tree is a decision tree capable of predicting several numeric variables at once. We focus on size and accuracy constraints. By either specifying maximum size or minimum accuracy, the user can trade-off size (and thus interpretability) for accuracy. Our approach is to first build a large tree based on the training data and to prune it in a second step to satisfy the user constraints. This has the advantage that the tree can be stored in the inductive database and used for answering inductive queries with different constraints. Besides size and accuracy constraints, we also briefly discuss syntactic constraints. We evaluate our system on a number of real world data sets and measure the size versus accuracy trade-off.

Proceedings of the 3rd international workshop on Link discovery

Abstract: The LinkKDD-2005 workshop aims to bring together a diverse group of researchers and industry practitioners to advance the state of the art in link discovery. Recently, there has been increasing interest in developing information technology for Link Discovery (LD). LD research studies and develops data mining techniques for extracting valuable patterns linking together seemingly unrelated items. LD, rooted in fields such as discreet mathematics, graph theory, social science, pattern analysis, link analysis and spatial databases, is relevant to a wide range of research topics that have been developed in past decades. Successful LD systems will discover the hidden structure of organizations, relate groups, identify fraudulent behaviour, model group activity and provide early detection of emerging threats. The broader context of this workshop invites both theoretical and applied contributions to LD spanning techniques from Data Mining, Machine Learning, Information Retrieval, Natural Language Processing, Social Networks Analysis, and general Graph Theory.Typical characteristics of link discovery problems are:.. Data is heterogeneous, arising from multiple sources;.. Data and patterns sought include representations of people, organizations, objects, actions and events, each of which has its own set of attributes, and particular types of relations linking them;.. The structure may include temporal, spatial, organizational, and/or transactional patterns;.. A relatively low number of observations for each entity can be recorded and the overall sample is typically small relative to the size of the population;.. The data becomes available over time, so the timing of when to make a decision based on the analysis is a central issue.LD problems are found in various areas such as homeland security, social network analysis, fraud detection, recommendation systems, and user modelling. The interdisciplinary nature of link discovery promotes a concerted effort from various researchers. The purpose of this workshop is to provide a forum to foster such interactions, discuss the new achievements and identify future research directions in link discovery.

Selective etching of metallic single-wall carbon nanotubes with hydrogen plasma.

Abstract: We present Raman scattering and scanning tunnelling microscopy (STM) measurements on hydrogen plasma etched single-wall carbon nanotubes (SWNTs). Interestingly, both the STM and Raman spectroscopy show that the metallic SWNTs are dramatically altered and highly defected by the plasma treatment. In addition, structural characterizations show that metal catalysts are detached from the ends of the SWNT bundles. For semiconducting SWNTs we observe no feature of defects or etching along the nanotubes. Raman spectra in the radial breathing mode region of plasma-treated SWNT material show that most of the tubes are semiconducting. These results show that hydrogen plasma treatment favours etching of metallic nanotubes over semiconducting ones and therefore could be used to tailor the electronic properties of SWNT raw materials.

Lysosomal Cysteine Proteases: Structural Features and their Role in Apoptosis

Abstract: Among the variety of proteolytic enzymes enormous progress has been seen recently in our understanding of lysosomal cysteine proteases, also known as cysteine cathepsins. These enzymes play a crucial role in diverse biological processes in physiological and pathological states, including genetic diseases. In the present review, their properties and structural features that are important to an understanding of their biological function are presented. Special emphasis is given to the newly discovered role of lysosomal cathepsins in apoptotic pathways. IUBMB Life, 57: 347-353, 2005

Data sparsity issues in the collaborative filtering framework

Abstract: With the amount of available information on the Web growing rapidly with each day, the need to automatically filter the information in order to ensure greater user efficiency has emerged. Within the fields of user profiling and Web personalization several popular content filtering techniques have been developed. In this chapter we present one of such techniques – collaborative filtering. Apart from giving an overview of collaborative filtering approaches, we present the experimental results of confronting the k-Nearest Neighbor (kNN) algorithm with Support Vector Machine (SVM) in the collaborative filtering framework using datasets with different properties. While the k-Nearest Neighbor algorithm is usually used for collaborative filtering tasks, Support Vector Machine is considered a state-of-the-art classification algorithm. Since collaborative filtering can also be interpreted as a classification/regression task, virtually any supervised learning algorithm (such as SVM) can also be applied. Experiments were performed on two standard, publicly available datasets and, on the other hand, on a real-life corporate dataset that does not fit the profile of ideal data for collaborative filtering. We conclude that the quality of collaborative filtering recommendations is highly dependent on the sparsity of available data. Furthermore, we show that kNN is dominant on datasets with relatively low sparsity while SVM-based approaches may perform better on highly sparse data.