microstructure informed kinetic model : application to solid explosives
When an explosive composition is subjected to an intense stress, such as a shock, the wave generated interacts with the microstructure and in particular with the defects it contains. Due to the nature of the defects, the energy can be localised, as when porosity is compacted, which can lead to the appearance of hot spots. Beyond a certain critical size, these hot spots grow as a result of the chemical decomposition of the explosive, and in some cases this can lead to the creation of a detonation wave. The role of these hot spots is therefore decisive in the initiation of solid explosives. The majority of macroscopic models used to study the shock-detonation transition (SDT) are phenomenological models calibrated on experiments (e.g. multi-strand gauge experiments) and therefore do not take into account the microstructural peculiarities specific to each explosive. It then becomes necessary to recalibrate a model for each composition, which limits any predictive capacity.
Microtomographic studies of real microstructures of explosive compositions have revealed that these deviate significantly from an average description based on a spherical pore. Through image segmentation, these microtomographs can provide essential ingredients for mesoscopic-scale simulation codes: these microstructures can be used directly as input for calculations or as a basis for generating virtual but realistic microstructures, thereby extending the accessible database given the experimental difficulties in generating this type of image in large numbers.
The computing power available today means that we can now envisage explicit simulations of realistic microstructures of explosive compositions. These simulations, in two or even three dimensions, will form the basis for the construction of a macroscopic kinetics model for modelling the shock-detonation transition. The results expected from this work are cross-disciplinary and can be transposed to all composite energetic materials. The effect of thermal or mechanical damage on the behaviour of an explosive or a solid propellant (vulnerability issues) could also benefit from this project. This more detailed knowledge of the role of microstructure (grain shape, porosity, etc.) could also improve filler manufacturing processes (e.g. ‘Very Insensitive’-RDX).
Development of theoretical Raman spectra with application on minerals from the surface of Mars
As we push the boundaries of space exploration with new missions to nearby planets, improving our investigation tools is crucial. Mars rovers have revealed a surface mineralogy unlike anything on Earth, shaped by the planet’s former hydrosphere followed by an extended dry and cold environment. For example, this favors the formation of perchlorates, or mixed silicate–salts glassy phases — minerals that are difficult to synthesize and stabilize on Earth but remain surprisingly stable on Mars. Recent Raman spectrometry data confirms their presence, highlighting an opportunity for deeper research. Understanding these minerals could offer new insights into Martian chemistry and planetary evolution.
Here we want to calculate the theoretical Raman spectra of perchlorates and other Martian minerals using the density functional perturbation theory (DFPT) as implemented in the ABINIT package. We want to obtain not only the position and the intensity of the peaks, but also the peak widths. They are necessary to correctly identify between similar spectra and to assess, by integration, the actual intensity of the peaks, which are directly comparable to experimental values on the field. These allow us to choose the representative peaks that can be used in identification and to analyze the displacement patterns associated with the vibrations. The results of our simulations will be compared and interpreted in the light of measurements performed by the current rovers on the surface of Mars.
For this, we need to implement several third- and fourth-order derivatives of the energy. This will be done as a series of DFPT terms, where the perturbations can be atomic displacements or electric fields. We will use a combination of the 2n+1 theorem and finite differences. The implementation will be done within the "Projector Augmented-Wave" approach (PAW) to DFT. The entire development effort will be integrated into the ABINIT package and made available to the entire community. ABINIT (www.abinit.org) is a highly visible international collaborative project for ab initio simulations based on DFT and DFPT. The computed spectra will be made available to the community via the WURM database.
The successful candidate will be co-advised between the IPGP (Paris) and the CEA (Bruyères-le-Chatel, S of Paris) groups. IPGP is a world-renowned geosciences research institute founded in 1921, associated with the CNRS, a component of the Université Paris Cité and employing more than 500 people. The group of Razvan Caracas is highly active in computational mineralogy, study of matter at extreme conditions, and planetology. The Quantum simulation of Matter group at CEA Bruyères-le-Chatel led by Marc Torrent is a main developer of the ABINIT package, highly active in density functional theory, projector augmented-wave, and high-performance computing.
Design and test of a PLL in FD-SOI 28nm technology
The goal of this PhD thesis is to design a Phase Locked Loop for generic use at 5 GHz. This PLL will also include a study regarding each building bloc sensitivity to radiation and thermal sensitivity regarding space environment. This is the main point of this PhD thesis because integrating a PLL in harsh environment requires an accurate knowledge of the circuit's parameters. The candidate will begin its work by analysing existing works on the FD-SOI technology (structure characteristics and impact on radiation hardening) to serve as a base for its work and design a Phase Locked Loop architecture. He will also study how to characterise each PLL building bloc variations in harsh environment (radiation and temperature).