Modeling the short-circuit current density of polymer solar cells based on P3HT:PCBM blend

We have investigated the short-circuit current density of organic solar cells based on poly (3-hexylthiophene)(P3HT)/6,6-phenyl C61-butyric acid methyl ester (PCBM) blend. In order to model charge collection efficiencies with respect to short circuit density in such blends, a full optical modeling of the cell is performed. From the distribution of the electromagnetic field, we compute the rate of exciton generation. This exciton generation rate is used as input in the transport equations of holes and electrons. Charge densities at steady state are obtained as solutions are used for computing short-circuit current densities generated in the cell. The dependence of short-circuit current densities versus the thickness of the blend is analyzed and compared with our experimental data and with data extracted from the literature.     

Single-molecule dynamics in a self-assembled 2D molecular sieve

A two-dimensional molecular sieve has been realized. It consists of a host matrix of molecularly engineered building blocks self-assembled at the liquid-solid interface. The simultaneous size- and shape-dependent dynamics of different guest molecules is observed in situ, in real time with submolecular resolution using a scanning tunneling microscope both at the liquid-solid interface and under vacuum. The temperature-dependent dynamics reveals that the diffusion proceeds through thermally activated channeling between single-molecule surface cavities.     

New one-pot synthesis of Au and Ag nanoparticles using green rust reactive particle as a micro-reactor

A new, simple, and fast one-pot synthesis of supported Au or Ag nanoparticles is implemented, for which a reactive Fe(II)-bearing green rust inorganic particle is used as an individual micro-reactor acting as both the reducing agent and support for the resulting metal nanoparticles. The mechanism involves both the solid-state oxidation of the green rust support (sulfate or carbonate) and the reduction-precipitation of soluble metal precursor. The resulting nanohybrids have a platy inorganic part supporting about one to ten nanoparticles with sizes in the 20 to 120 nm range.     

Sorption properties toward substituted phenolic derivatives in water using macroporous polyamines containing β-cyclodextrin

Macroporous polyamines containing β-cyclodextrin (β-CD) are used for the recovery of various organic pollutants from aqueous solutions. Several insoluble polyamines with various degrees of β-CD were prepared. Studies concerning the sorption kinetics, the effect of pollutant concentration, pH of the solutions, and β-CD content are presented here. The results of sorption experiments show that these sorbents exhibit high sorption capacities toward substituted phenolic derivatives. The mechanism of sorption is both physical adsorption in the resin polymer network through acid–base interactions, ion exchange, and hydrogen bonding due to the amino groups and the formation of an inclusion complex due to the β-CD molecules through host–guest interactions, and hydrophobic interactions such as guest–guest interactions. © 1999 John Wiley & Sons, Inc. J Appl Polym Sci 73: 2903–2910, 1999     

Dynamic mechanical properties of very thin adhesive joints

The effective mechanical properties of a polyurethane adhesive (oligoetherdiol, ‐triol, MDI) in gold joints (bond line thickness, d_P = 35–550 µm) are studied in the linear deformation range by dynamic mechanical analysis in shear mode. These properties depend on d_P: thin ones possess a higher dynamic glass transition temperature and show a narrower glass transition than the thick ones. The storage modulus rises with decreasing d_P for the rubbery plateau. The results attest mechanical interphases in the polyurethane with increased crosslink density and reduced cooperative mobility than in bulk. © 2015 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2015 , 132, 42058.     

Wundt's laboratory at Leipzig in 1891.

This article describes W. Wundt's laboratory at Leipzig in 1891 as viewed by a Belgian psychologist, J. J. Van Biervliet (1859-1945). Although few French-speaking psychologists worked in Wundt's laboratory, several of those who did reports wrote on how students were trained there. Van Biervliet decided to visit Wundt's laboratory at Leipzig in order to strengthen the foundation of his own laboratory at the University of Ghent and to become familiar with Wundt's experimental techniques. A translation of J. J. Van Biervliet's (1892) article "Experimental Psychology. Wundt's Institute at Leipzig" is presented here as one of the first and most complete articles in French describing the functioning of Wundt's laboratory. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Variable‐Wavelength Quick Scanning Nanofocused X‐Ray Microscopy for In Situ Strain and Tilt Mapping

Compression of micropillars is followed in situ by a quick nanofocused X‐ray scanning microscopy technique combined with 3D reciprocal space mapping. Compared to other attempts using X‐ray nanobeams, it avoids any motion or vibration that would lead to a destruction of the sample. The technique consists of scanning both the energy of the incident nanofocused X‐ray beam and the in‐plane translations of the focusing optics along the X‐ray beam. Here, the approach by imaging the strain and lattice orientation of Si micropillars and their pedestals during in situ compression is demonstrated. Varying the energy of the incident beam instead of rocking the sample and mapping the focusing optics instead of moving the sample supplies a vibration‐free measurement of the reciprocal space maps without removal of the mechanical load. The maps of strain and lattice orientation are in good agreement with the ones recorded by ordinary rocking‐curve scans. Variable‐wavelength quick scanning X‐ray microscopy opens the route for in situ strain and tilt mapping toward more diverse and complex materials environments, especially where sample manipulation is difficult.     

Structured prediction with reinforcement learning

We formalize the problem of Structured Prediction as a Reinforcement Learning task. We first define a Structured Prediction Markov Decision Process (SP-MDP), an instantiation of Markov Decision Processes for Structured Prediction and show that learning an optimal policy for this SP-MDP is equivalent to minimizing the empirical loss. This link between the supervised learning formulation of structured prediction and reinforcement learning (RL) allows us to use approximate RL methods for learning the policy. The proposed model makes weak assumptions both on the nature of the Structured Prediction problem and on the supervision process. It does not make any assumption on the decomposition of loss functions, on data encoding, or on the availability of optimal policies for training. It then allows us to cope with a large range of structured prediction problems. Besides, it scales well and can be used for solving both complex and large-scale real-world problems. We describe two series of experiments. The first one provides an analysis of RL on classical sequence prediction benchmarks and compares our approach with state-of-the-art SP algorithms. The second one introduces a tree transformation problem where most previous models fail. This is a complex instance of the general labeled tree mapping problem. We show that RL exploration is effective and leads to successful results on this challenging task. This is a clear confirmation that RL could be used for large size and complex structured prediction problems.