Scientific Direction
Transfer of knowledge to industry
Scientific Direction
Département des Technologies des NanoMatériaux (LITEN)
Laboratoire des Eco-procédés et EnVironnement
01-10-2019
SL-DRT-19-0854
olivier.lebaigue@cea.fr
Post-functionalization (particularly under supercritical CO2 impregnation in our proposal) allows us to give new properties to a polymer by core impregnation with selected molecules. In this process, supercritical CO2 has two effects: it swells the polymer and transports additional molecules. Hydrophobia, thermal or mechanical reinforcement, improvement of electrical conductivity, colouring, UV resistance... are new targeted properties. A large part of the thesis will be devoted to experimental research and new measurement techniques. In addition, the measurement results will also feed into a multi-scale multi-physical (and chemical) modelling process. Once carefully validated, these physico-chemical models are expected to accurately predict the different stages of existing processes and help to develop new processes.
DM2I (LIST)
Laboratoire Capteurs Diamants
01-10-2019
SL-DRT-19-0856
samuel.saada@cea.fr
According to its nonstandard thermal conductivity combined with outstanding electronic properties, diamond is an ultimate material for power electronics. However, these properties are highly dependent on the material crystalline quality. Epitaxial diamond on iridium is currently an attractive material to fabricate diamond wafers of high crystalline quality. Such a wafer sector is not yet available. The quality of heteroepitaxial diamond was successively improved. Recently, a Japanese group reported a dislocation density of 10^7 / cm2 for a film, 60 microns thick, obtained by lateral growth [1]. At CEA LIST, heteroepitaxial diamond films are grown on Ir/SrTiO3/Si(001) pseudo-substrates of 1 cm2. The Bias Enhanced Nucleation method (BEN) is performed in a plasma-assisted chemical vapor deposition (CVD) reactor connected to a UHV surface analysis chamber. The crystalline quality of films 300 microns thick is at the state of art with a mosaicity of 0.6° and an in-plane misorientation of 0.7° [2, 3]. From local measurements in Cathodoluminescence, the dislocation density was estimated to 4 × 10^6 / cm2. These films are homogeneous on 1 cm2 surfaces. The main objective of this PhD is to better control and reduce the density of structural defects in heteroepitaxial diamond grown on 1 cm2 pseudo-substrates applying an innovative growth strategy recently patented by our laboratory [4]. The involved experimental parameters will be determined and the crystalline quality and structural defects will be finely characterized for different heteroepitaxial film thicknesses by X-Ray diffraction, Raman spectroscopy and cathodoluminescence. The best films will be used as substrates to grow boron doped diamond epilayers (p-type doping). Electrical characteristics will be then measured in collaboration with GEEPS (Paris-Saclay University). [1] Ichikawa et al, High crystalline quality heteroepitaxial diamond using grid-patterned nucleation and growth on Ir, Diam. Relat. Mater. (2019) doi.org/10.1016/j.diamond.2019.01.027 [2] Lee et al, Epitaxy of iridium on SrTiO3/Si (001): A promising scalable substrate for diamond heteroepitaxy, Diam. Relat. Mater. 66 (2016) 67. [3] Bensalah et al, Mosaicity, dislocations and strain in heteroepitaxial diamond grown on iridium, Diam. Relat. Mater. 66 (2016) 188. [4] Delchevalrie, Arnault, Saada (décembre 2018).
Département Ingénierie Logiciels et Systèmes (LIST)
Laboratoire pour la Sûreté du Logiciel
01-10-2019
SL-DRT-19-0857
yves.lhuillier@cea.fr
The development of embedded systems is subject to the consequences of more and more advanced integration. These consequences are manifested by increasingly complex architectures (multi-core ...), but also more and more sensitive (to faults). These systems, which are difficult to test (System On Chip), also have potential security vulnerabilities. All these integration-related impacts must therefore be taken into account in a strict context of information system reliability and security (SSI) requirements. The classical development methods show their limits vis-à-vis these new architectures inducing significant risks in the reliability of the final system. In almost all cases, the software is developed under the assumption of a stable and secure hardware, this being generally ensured by mechanisms of redundancies more and more expensive. To reduce these risks, the use of virtual platforms makes it possible to develop software that is less sensitive to potential errors. The aim of the thesis is to study the testability of embedded systems by an approach based on fault injection by simulation, via virtual platforms. The use of simulation makes it possible to execute a program and to disturb certain data precisely. The analysis of the divergences with a perfect execution, makes it possible to check the level of criticality of each of the operations carried out by an executable and to adapt the development accordingly.
Département Systèmes
Laboratoire Antennes, Propagation, Couplage Inductif
01-09-2019
SL-DRT-19-0873
raffaele.derrico@cea.fr
The framework of this thesis is that of the detection and imaging of objects in a lossy, and possibly layered, environment using multi-antenna wide-band radar technologies. In these systems, one of the options is to use a large number of antennas to improve the location accuracy, and another one consists in using high-resolution array signal processing approaches. The objective of this thesis is to propose co-located multi-antenna solutions, and their correponding signal processing methods, capable of exploiting MIMO techniques and beamforming in penetrating radar applications, i.e. using wide band signals. The study will start with a state of the art concerning multi-antenna radar systems and the implementation of a (simplified) propagation model, both in the wideband HR localization framework. Then, the PhD student will perform measurement campaigns using the CEA LETI channel sounder. This experimental phase will involve an evolution of the propagation model including the wave sphericity , the polarization the medium heterogeneity and potential dispersivity. The student will propose multi-antenna solutions by optimizing the layout and number of radiating elements, by developping performing signal processing techniques, and will produce a proof of concept. The PhD student will be part of the Antenna, Propagation and Inductive Coupling Laboratory at CEA-LETI, in Grenoble (France). He/she will benefit of the state of the art facilities (channel sounders, emulator, OTA setup, and electromagnetic simulator). The PhD thesis will be carried out in collaboration with Université de Rennes 1, under the supervision of Prof. Laurent FERRO-FAMIL. The student will have the opportunity to spend training periods in both institutes. Application: The position is open to outstanding students with Master of Science, ?école d'ingénieur? or equivalent. The student should have specialization in the field of telecommunications, radar, microwave and/or signal processing. The application must necessarily include a CV, cover letter and grades for the last two years of study.
Département d'Optronique (LETI)
Laboratoire pour la Visualisation et l'Eclairage
01-10-2019
SL-DRT-19-0876
christophe.martinez@cea.fr
The PhD is forming part of a research activity in our laboratory about retinal projection for Augmented Reality (AR) applications. Current AR devices (smart glasses) are limited in their design by the constraints given by the optical systems. The use of the device in a social context is still difficult to be implemented commercially because of the lack of aesthetic that discourages the user. To overcome these limitations related to the pupil management in the axial optical systems, CEA Leti is investigating the use of pixelated holograms, distributed on the surface of the glass and addressed by a grid of single mode waveguides and electrodes. The PhD will focus on the development of the technological brick related to the activation of the holograms. The work will consist in the recording of pixelated holograms on the holographic printer of the laboratory. A series of optical set-up will then be implemented to demonstrate the dynamic addressing of the holographic function. These set-ups will concern first simple free space propagation with the use of DLP and will progressively migrate to integrated photonics devices in relation to our development in Silicon Photonics and Liquid Cristal Devices. The use of these set-ups concerns retinal projection but the PhD student will also be in charge to investigate other domains of application (3Dimaging/sensing, Lidar technologies, ?).
Département Thermique Biomasse et Hydrogène (LITEN)
Laboratoire Systèmes Solaires Haute Température
01-10-2019
SL-DRT-19-0877
nicolas.lamaison@cea.fr
Heating demand represents a major part of the total energy consumption (about 40% in France for example). In parallel, residential and tertiary cold demand keeps increasing (+80% for the next 10 years at the European scale). Moreover, the ever-increasing integration of photovoltaic or wind electricity may eventually lead to over voltage and thus instabilities of the power grid. Finally, it is well proven that energy storage is more efficient and cheaper using thermal rather than electrical means. Thus, smart synergies must be found between these two energy carrier. In that context, the present PhD thesis aims at studying the combined production of cold and heat for District Thermal Networks (DTN) using a thermal/electrical co-optimization framework. First, we will focus on the different production/conversion technologies available for the thermal and electric carriers such as simultaneous heating and cooling heat pumps (SHC-HP), absorption heat pumps (A-HP), electrical resistance, etc? and for storage. We will define the advantages and breakdowns of each system depending on the domain of application. The most promising system in terms of applicability will be then studied in details in a second phase. We will particularly develop a sizing methodology, based on MILP methods (Mixed Integer Linear Programming). With the latter, the potential conflictual objectives of production (heat vs cold) will be accounted for in the problem formulation and thus in the sized system. To do so, a physico-mathematical systematic approach accounting for uncertainties will be implemented. Finally, we will study the system from an operational point of view both numerically and experimentally, notably to evaluate the real level of performance of the sized system.
