The goal of this talk is to present the basis of the visual system in mammals and more precisely how long-range connections in the primary visual cortex can influence local cortical activity as seen in optical imaging experiments for example. It is known that the representation of contours in the primary visual cortex is locally organised around “pinwheels” and possesses an approximate Euclidean invariance when only local connections are considered. We will then briefly show how to use these ‘ingredients' to reproduce visual hallucinatory patterns.

Les écoulements lents de suspensions de particules solides non-browniennes sont omniprésents dans les problèmatiques industrielles (polymères chargés, béton frais, moulage des propergols) et les phénomènes naturels (écoulements sédimentaires marins, glissements de terrain). Même lorsque le fluide suspendant est simple, ces suspensions exhibent une rhéologie complexe, et ce d'autant plus que la fraction de solide est importante. Ainsi, la viscosité du matériau augmente avec la fraction solide jusqu'au blocage complet, tandis que des contraintes normales anisotropes se développent. D'autre part, lorsque la suspension est cisaillée, les particules sont animées d'un mouvement stochastique, alors même que le mouvement brownien, d'origine thermique, est complètement négligeable. Ceci induit par exemple à l'échelle macroscopique des gradients de fraction volumique solide pour peu que le taux de cisaillement soit hétérogène..

Depuis les travaux d'Einstein en 1906 sur la viscosité des suspensions diluées, de nombreux études expérimentales, théoriques et numériques ont été dévolues à la compréhension de ces différents phénomènes. Les descriptions théoriques se sont longtemps focalisées sur les interactions hydrodynamiques, compliquées car elles mettent en jeu des échelles très différentes, à longue portée d'une part, mais aussi dans le régime de lubrification où les interactions entre paires de particules divergent lorsque leurs surfaces se rapprochent. Cependant, la question des interactions directes entre particules, en lien avec la microstructure induite par le cisaillement; s'est posée de façon croissante, jusqu'à devenir centrale actuellement.

Dans cet exposé, j'aborderai quelques aspects récents de la physique des écoulements de suspensions non-browniennes en m'appuyant sur des expériences et des simulations. J'évoquerai notamment comment des mesures rhéologiques permettent d'explorer la connection entre microstructure et réponse rhéologique. Des mesures directes de la microstructure confirment également l'existence de contacts directs entre particules via les rugosités de surface. A ce jour, peu de travaux ont été consacrés à l'influence de la loi de contact entre particules sur la rhéologie. Je montrerai l'apport des simulations numériques dans ce domaine. En particulier, la prise en compte d'une friction solide entre particules permet de réconcilier quantitativement les mesures expérimentales et les calculs numériques.

The 2014 Nobel Prize in physics has been attributed to the three inventors of the first efficient blue LED which has enabled bright and energy-saving white light sources. In the mid-1990s, I. Akasaki, H. Amano and S. Nakamura found the way to achieve the epitaxial growth and the doping of high quality nitride semiconductors that is Gallium Nitride (GaN) and its alloys with Aluminium and Indium. Since then, GaN is considered as the material of choice for optoelectronic applications as well as for high frequency devices. However the growth of this wide-bandgap material remains challenging: the lack of an adapted substrate results in a very large density of defects, in particular dislocations, stacking faults and cracks. Under these circumstances, the growth of nanostructures and in particular nanowires appears as an appealing means of circumventing the problem. First, these nanostructures present a low density of structural defects and an efficient elastic relaxation of strain by their lateral free-facets[Hug13], which should enable fundamentally more efficient devices. Second, their 1D geometry improves the light extraction and absorption efficiency making them suitable candidates for optoelectronic applications such as solar cells, LEDs or lasers. Furthermore, their high surface-to-volume ratio results in a high detection sensitivity and improved response time, which explains the important development of Nanowire-based sensors [Pea11][Hsu11][Pan13]. In this talk, some examples of application using GaN nanowires will be described and the activities done at CRHEA on the growth by metalorganic vapour phase epitaxy (MOVPE) and the characterization of GaN-based nanowires will be presented. Their structural and optical properties will be discussed based on transmission electron microscopy and cathodoluminescence experiments.

[1] H. Hugues, P.A. Shields, F. Sacconi, M. Mexis, M. Auf der Maur, M. Cooke, M. Dineen, A. Di Carlo, D. W. E. Allsopp and J. Zúñiga-Pérez, J. Appl. Phys. 114, 084307 (2013)

[2] S. J. Pearton, C. Y. Chang, B. H. Chu, C-F. Lo, F. Ren, W. Chen and J. Guo, IEEE J. Select. Topics Quantum Electron. 17, 1092 (2011)

[3] C-W. Hsu, A. Ganguly, C-P. Chen, C-C Kuo, P. P. Paskov, P. O. Holtz, L-C Chen and K-H Chen, J. Appl. Phys. 109, 053523 (2011)

[4] C. Pan, L. Dong, G. Zhu, S. Niu, R. Yu, Q. Yang, Y. Liu and Z. L. Wang, Nature Photon. 7, 752 (2013)

In this joint work with M.-S. Abdelouahab and L. O. Chua, we define a common mathematical framework which includes the modelling of the behavior of the new electronic circuit elements with memory recently discovered. In this general frame we extend Ohm’s law to any case.

Circuit elements that store information without the need of a power source would represent a paradigm change in electronics, allowing for low-power computation and storage.

One such circuit element is the memory-resistor (memristor for short) which was postulated by Leon O. Chua, from Berkeley University (U.S.A.) in 1971 by analyzing mathematical relations between pairs of fundamental circuit variables. Considering that resistor links voltage to intensity (Ohm’s law), capacitor associates voltage to charge and inductor intensity to magnetic flux (defined mathematically as the time integral of the voltage, which needs not have a magnetic flux interpretation), a fourth element is missing: the one linking the charge and the flux. While resistors, capacitors, inductors are macroscopic elements, there is no obvious macroscopic device corresponding to memristors. In the years following Chua’s work, little research was directed towards the memristor concept until the birth of nanotechnology some years ago. In 2008 a group at the Helwett-Packard (HP) lab managed to construct a physical component acting as a memristor (in 2008) using thin films of TiO2. Co-developers HP and Hynix plan to bring memristor technology to market in summer 2014. The technology is eventually expected to replace flash, DRAM and even hard drives.

Turning back to his theoretical research on this device, 40 years after his genuine publication, L. O. Chua extended the notion of memristive systems to capacitive and inductive elements, namely, capacitors and inductors whose properties depend on state and history of the system.

Besides the discovery of these new electronic elements, by means of the mathematical concept of fractional derivative and the Laplace transform, some authors around ’90 eventually extended the generalized Ohm’s law (which links voltage to intensity in circuits encompassing resistors, capacitors and inductors) to any impedance, introducing the paradigm of fractance.

The most important conceptual difficulty raised by the generalization of fractance to memory elements, is the fact that it is necessary to use an interpolation of two parameters instead of one. Using such a two-parameter interpolation, we introduce Memfractance, a common mathematical framework which encompasses electronic circuit elements with or without memory (resistors, capacitors, inductors, memresistors, memcapacitors and meminductors). A new element, the memfractor, interpolates all these devices. Ultimately, in this new paradigm we generalize Ohm’s law allowing to model the behavior of every electronic device which can be built using nanotechnology.

Moreover, it has been recently shown and experimentally demonstrated that many nano scale existing memristive devices can behave chaotically. We will present some new results on chaotic dynamics of memfractor.

A standard laser has usually two ingredients: a gain medium providing light amplification by stimulated emission, and an optical cavity providing feedback. However, there is currently a strong interest in more exotic laser systems, especially "mirrorless" lasers, in which the electromagnetic feedback is provided either by disorder (multiple scattering in the fain medium) or by order (multiple Bragg reflection). These mechanisms correspond, respectively, to "random lasers" and "photonic crystal lasers".

I will discuss two experiments, one demonstrating random lasing, and one demonstrating the one-dimensional version of photonic crystal lasing, in which both the gain and the feedback mechanisms are provided by a very dilute atomic vapor containing a few billion atoms, which are laser-cooled to microkelvin temperature.