Selective activation of memristive interfaces in TaO x -based devices by controlling oxygen vacancies dynamics at the nanoscale.

The development of novel devices for neuromorphic computing and non-traditional logic operations largely relies on the fabrication of well controlled memristive systems with functionalities beyond standard bipolar behavior and digital ON-OFF states. In the present work we demonstrate for Ta2O5-based devices that it is possible to selectively activate/deactivate two series memristive interfaces in order to obtain clockwise or counter-clockwise multilevel squared remanent resistance loops, just by controlling both the electroforming process and the (a)symmetry of the applied stimuli, and independently of the nature of the used metallic electrodes. Based on our thorough characterization, analysis and modeling, we show that the physical origin of this electrical behavior relies on controlled oxygen vacancies electromigration between three different nanoscopic zones of the active Ta2O5-x layer: a central one and two quasi-symmetric interfaces with reduced TaO2-h(y) layers. Our devices fabrication process is rather simple as it implies the room temperature deposition of only one CMOS compatible oxide-Ta-oxide-and one metal, suggesting that it might be possible to take advantage of these properties at low cost and with easy scability. The tunable opposite remanent resistance loops circulations with multiple-analogic-intermediate stable states allows mimicking the adaptable synaptic weight of biological systems and presents potential for non-standard logic devices.

True nature of an archetypal self-assembly system: mobile Au-thiolate species on Au(111).

Alkanethiol self-assembled monolayer (SAM) phases on Au(111) have been assumed to involve direct S head group bonding to the substrate. Using x-ray standing wave experiments, we show the thiolate actually bonds to gold adatoms; self-organization in these archetypal SAM systems must therefore be governed by the movement of these Au-S-R moieties on the surface between two distinct local hollow sites on the surface. The results of recent ab initio total energy calculations provide strong support for this description, and a rationale for the implied significant molecular mobility in these systems.

Taking on the curse of dimensionality in joint distributions using neural networks.

The curse of dimensionality is severe when modeling high-dimensional discrete data: the number of possible combinations of the variables explodes exponentially. In this paper, we propose a new architecture for modeling high-dimensional data that requires resources (parameters and computations) that grow at most as the square of the number of variables, using a multilayer neural network to represent the joint distribution of the variables as the product of conditional distributions. The neural network can be interpreted as a graphical model without hidden random variables, but in which the conditional distributions are tied through the hidden units. The connectivity of the neural network can be pruned by using dependency tests between the variables (thus reducing significantly the number of parameters). Experiments on modeling the distribution of several discrete data sets show statistically significant improvements over other methods such as naive Bayes and comparable Bayesian networks and show that significant improvements can be obtained by pruning the network.

[Validation and characterization of chromatographic sorbent for protein purification]. / Validation et caractérisation de supports chromatographiques appliqués à la purification protéique.

The purification of biomolecules for human therapeutic use by liquid chromatography have been widely extended since 10 years. The preparation of such molecules have to be in accordance with quality requirements and some of them still have to be defined because of their high specificity. As a consequence, each purification step must be validated to guarantee a constant product quality and the absence of any contaminant issued from the solid phases employed. In this context, we developed several techniques allowing the characterization of sorbents and their eventual degradation products in order to determine the level of toxicity of the potential leachables in relation with their mode of use. This procedure could be generalized in the future according to the constraints due to the fractionation process itself.

Biosynthesis of the third component of complement (C3) by the human monocytic-cell line U-937. Induction by phorbol myristate acetate.

Phorbol myristate acetate (PMA)-stimulated human monocyte-like cells (U-937) were found to synthesize the third component of complement (C3), as shown by enzyme-linked immunosorbent assay and immunoprecipitation from [35S]methionine-labelled culture supernatants. C3 synthesis occurred at a rate of about 160 ng of C3/24 h per 10(6) cells on day 7 after addition of PMA; it was blocked by cycloheximide treatment and was restored after removal of the inhibitor. SDS/polyacrylamide-gel-electrophoretic analysis of the immunoprecipitated protein showed that the size and subunit structure of the newly synthesized C3 were identical with those of plasma C3, and that a single-chain intracellular precursor was present in the cell lysates. Haemolytic assays showed that the synthesized C3 fully expressed functional activity in early culture within 4 h. After longer culture, a loss of haemolytic activity was observed. The possibility that newly secreted C3 is cleaved by U-937 cells themselves was suggested.