Search

Simple and direct prototyping methods are ideal for large-scale delivery of cognitive photonic hardware. Here, we choose ultrafast laser writing as a direct fabrication technique to later demonstrate all-optical synaptic-like performance along the laser-written waveguides in a chalcogenide glass. Neuronal communication protocols, such as excitatory and inhibitory responses, temporal summations, and spike-timing-dependent plasticity, are shown in the glass chip. This work manifests the potential for large-scale delivery of fully integrated photonic chips based on cognitive principles by single-step fabrication procedures. Agency for Science, Technology and Research (A*STAR) Ministry of Education (MOE) Published version M.R. and C.S. acknowledge support from the Singapore Ministry of Education (Nos. MOE2011-T3-1-005 and MOE2016-T3-1-006) and the Agency for Science, Technology and Research (A STAR, Grant No. A18A7b0058). IFN-CNR is grateful for support from the H2020 ERC project PAIDEIA (GA No. 816313) and H2020 Marie Curie ITN projects LasIonDef (GA No. 956387) and PHOTOTRAIN (GA No. 722591). A.G. is thankful for support from the Lombardy Region Project sPATIALS3 “Miglioramento delle produzioni agroalimentari e tecnologie innovative per un alimentazione piu sana, sicura e sostenibile,” cofunded by POR FESR 2014-2020 Call HUB Ricerca e Innovazione. S.M.E. is thankful for support from the CNR Short Term Mobility grant 2019.

We propose a method to dope silicon with phosphorus by means of the nuclear resonant reaction 30Si(p,γ)31P, which takes place when a natural silicon target is bombarded with a few MeV proton beam. This alternative method considerably reduces the usual target damage produced by the more commonly used direct phosphorus implantation. Partial support from the Spanish DGICYT (Project Nos. PB92-0341 and PB93-1125). N.P.K. thanks the Spanish DGICYT for financial support (SAB93-0182) during his sabbatical stay in the Universidad de Murcia.

Binary collision theory has been widely used in this context due to its computational convenience, despite its wellknown simplistic description of the materials excitation properties. 20 In the present work, we advance a MC model of electron-beam energy dissipation in multi-walled carbon nanotube (MWCNT) materials based on a quasi firstprinciples discrete-energy-loss model deduced from a realistic description of the target electronic excitations. This approach has the advantage that secondary-electron cascade generation can be explicitly simulated without the need for an arbitrary separation of plasmon and single-particle losses, since the complete excitation spectrum of MWCNTs is builtinto the model via the energy (�x) and momentum (�) dependent dielectric response function, eðx;kÞ. Then, under the constraint of physically motivated sum-rules (which preserve causality), the energy losses in single inelastic collisions can be computed in a self-consistent manner according to the properties of the so-called energy-loss-function (ELF), Im½� 1=eðx;kÞ� .

We present a simple model dielectric response function for both bulk and individual carbon nanotubes based on a parameterization of experimental optical data and analytic dispersion relations that account for dimensionality and linewidth broadening. The model is used to calculate electron inelastic mean free paths over a broad energy range of interest to various applications. Financial support for D.E., I.K., and K.K. was given by the European Union FP7 ANTICARB (Grant No. HEALTHF2-2008-201587) and for RGM and IA by the Spanish Ministerio de Ciencia e Innovación (Project Nos. FIS2006-13309-C02-01 and FIS2006-13309-C02-02).