Accueil
Brain learning simulated via electronic replica memory (Vol. 46 No. 4)
A new study shows how a new way of controlling electronic systems endowed with a memory can provide insights into the way associative memories are formed by mimicking synapses.
Scientists are attempting to mimic the memory and learning functions of neurons found in the human brain. To do so, they investigated the electronic equivalent of the synapse, the bridge, making it possible for neurons to communicate with each other. Specifically, they rely on an electronic circuit simulating neural networks using memory resistors. Such devices, dubbed memristor, are well-suited to the task because they display a resistance, which depends on their past states, thus producing a kind of electronic memory. The authors have developed a novel adaptive-control approach for such neural networks, presented in this study. Potential applications are in pattern recognition as well as fields such as associative memories and associative learning.
H. Zhao, L. Li, H. Peng, J. Kurths, J. Xiao and Y. Yang, Anti-synchronization for stochastic memristor-based neural networks with non-modeled dynamics via adaptive control
approach, Eur. Phys. J. B 88, 109 (2015)
[Abstract]
Does that “green” plasticiser make my PVC flexible enough for you? (Vol. 46 No. 4)
A study of an eco-friendly solvent helping to make PVC plastic more flexible reveals the molecular-level interaction of hydrogen bonds between the two ingredients.
What gives plastic objects their flexibility and reduces their brittleness is the concentration of plasticiser. For example, a chemical solvent of the phthalate family called DOP is often used. The trouble is there are concerns that phthalates present health risks. So there is a demand for more alternatives. Now, the authors have examined the effect of using DEHHP, a new eco-friendly plasticiser, used in combination with PVC. For a plasticiser to work, there has to be adequate hydrogen bonding with the plastic. By combining experiments and simulations, the team revealed why the polymer-solvent hydrogen bonding interaction's strength decreases with dilution at a molecular level—which is a phenomenon also observed in the DOP-PVC combination. These findings have been published in the present work.
Y. Liu, R. Zhang, X. Wang, P. Sun, W. Chen, J. Shen and G. Xue, Hydrogenation induced deviation of temperature and concentration dependences of polymer-solvent interactions in poly(vinyl chloride) and a new eco-friendly plasticizer, Eur. Phys. J. Plus 130, 116 (2015)
[Abstract]
Organic nanoparticles, more lethal to tumours (Vol. 46 No. 4)
© Mediteraneo / Fotolia
Carbon-based nanoparticles could be used to sensitize cancerous tumours to proton radiotherapy and induce more focused destruction of cancer cells, a new study shows.
Radiotherapy used in cancer treatment is a promising treatment method, albeit rather indiscriminate. Indeed, it affects neighbouring healthy tissues and tumours alike. Researchers have thus been exploring the possibilities of using various radio-sensitizers; these nanoscale entities focus the destructive effects of radiotherapy more specifically on tumour cells. In a study published recently, the authors have now shown that the production of low-energy electrons by radio-sensitizers made of carbon nanostructures hinges on a key physical mechanism referred to as plasmons—collective excitations of so-called valence electrons; a phenomenon already documented in rare metal sensitizers. This reseach may lead to the development of novel types of sensitizers composed of metallic and carbon-based parts.
A. Verkhovtsev, S. McKinnon, P. de Vera, E. Surdutovich, S. Guatelli, A. V. Korol, A. Rosenfeld and V. Solov’yov,, Comparative analysis of the secondary electron yield from carbon nanoparticles and pure water medium, Eur. Phys. J. D 69, 116 (2015)
[Abstract]
Fragmentation of random trees (Vol. 46 No. 4)
Networks are ubiquitous, appearing in the study of subjects as diverse as gene-protein interactions, power grids, and algorithms.
The function of a network is closely linked to its structure. For instance, in biochemical reaction networks, removal of a species or reaction can dramatically change the output of the system. Evolving networks often undergo degradation, making it important to understand how the structure breaks apart when components are randomly removed, also revealing how resilient a network is to attacks.
We studied the fragmentation of a random tree, a network formed by repeatedly attaching new nodes to an existing node chosen uniformly randomly.
We present exact equations governing the evolution of fragment sizes after a fraction of the nodes are removed at random, along with asymptotic solutions. For very large trees, fragment size distribution decays as a power law, with an exponent of 1+1/m, m being the fraction of remaining nodes. This implies that a few very large fragments coexist with many small ones (see figure).
Our findings reveal unusual fragmentation kinetics, where the fragment size distribution is characterized by a time-dependent exponent, and can provide insight into other fragmentation processes where dynamic parameters are observed.
Z. Kalay and E. Ben-Naim, Fragmentation of random trees, J. Phys. A, Math. Theor., 48, 045001 (2015)
[Abstract]
