Cette thèse est consacrée à l’étude d’une technique innovante de lyse de microorganismes, fondée sur l’utilisation d’impulsions haute tension en milieu aqueux. Cette technique se distingue de l’électroporation qui exploite le champ électrique produit pour dégrader la membrane cellulaire ; dans notre étude les impulsions haute tension permettent la formation d’arcs électriques produisant de multiples phénomènes physico-chimiques qui peuvent entraîner la lyse des microorganismes.L’efficacité du procédé a été évalué sur les microorganismes suivants : des microalgues productrices d’huile (Nannochloropsis gaditana et Phaeodactylum tricornutum) et des bactéries couramment utilisées comme modèles de laboratoire (Escherishia coli et Bacillus subtilis). Dans ces travaux, nous avons montré que les ondes de pression produites sont principalement responsables de la lyse.En fin d’étude, des perspectives sont explorées en vue du développement de systèmes autonomes soit dans le cadre de la bioproduction, soit dans le cadre de l’analyse cellulaire. / This thesis focuses on the study of an innovative technology for microorganisms lysis, based on high voltage pulses generated in an aqueous medium. This technology is different from electroporation which operates thanks to the electric field for damaging cell membranes ; in our study high voltage pulses generate an electric arc leading to various physicochemical phenomena supposed to lyse microorganisms.The technology efficiency is evaluated with the following microorganims : some lipid producting microalgae (Nannochloropsis gaditana and Phaeodactylum tricornutum) and classical laboratory model bacteria (Escherishia coli and Bacillus subtilis). In this work, we found that generated shock waves are mainly responsible of the cells lysis.At the end, the development of self-functioning devices is investigated either for bioproduction or for cell analysis.

<p>Multiple industry applications, including combustion, flow control, and medicine, have leveraged nanosecond pulsed plasma (NPP) discharges to create plasma generated reactive species (PGRS). The PGRS are essential to induce plasma-assisted mechanisms, but the rate of generation and permanence of these species remains complex. Many of the mechanisms surrounding plasma discharge have been discovered through experiments, but a consistent challenge of time scales limits the plasma measurements. Thus, a well-constructed model with experimental research will help elucidate complex plasma physics. The motivation of this work is to construct a feasible physical model within the additional numerical times scale limitations and computational resources. This thesis summarizes the development of a one-moment fluid model for NPP discharges, which are applied due to their efficacy in generating ionized and excited species from vacuum to atmospheric pressure. </p><p>From a pulsed power perspective, the influence of pulse parameters, such as electric field intensity, pulse shape and repetition rate, are critical; however, the effects of these parameters on PGRS remain incompletely characterized. Here, we assess the influence of pulse conditions on the electric field and PGRS computationally by coupling a quasi-one-dimensional model for a parallel plate geometry, with a Boltzmann solver (BOLSIG+) used to improve plasma species characterization. We first consider a low-pressure gas discharge (3 Torr) using a five-species model for argon. <a>We then extend to a 23 species model with a reduced set of reactions for air chemistry remaining at low pressure.</a> The foundations of a single NPP is first discussed to build upon the analysis of repeating pulses. Because many applications use multiple electric pulses (EPs) the need to examine EP parameters is necessary to optimize ionization and PGRS formation. </p><p>The major goal of this study is to understand how the delivered EP parameters scale with the generated species in the plasma. Beginning with a similar scaling study done by Paschen we examine the effects of scaling pressure and gap length when the product remains constant for the two models. This then leads to our study on the relationship of pulsed power for different voltages and pulse widths of EPs. By fixing the energy delivered to the gap for a single pulse we determine that the electron and ion number densities both increased with decreasing pulse duration; however, the rate of this increase of number densities appeared to reach a limit for 3 ns. These results suggest the feasibly of achieving comparable outputs using less expensive pulse generators with higher pulse duration and lower peak voltage. Lastly, we study these outcomes when increasing the number of pulses and discuss the effects of pulse repetition and the electron temperature.</p><p>Future work will extend this parametric study to different geometries (i.e. pin-to-plate, and pin-to pin) and ultimately incorporate this model into a high-fidelity computational fluid dynamics (CFD) model that may be compared to spectroscopic results under quiescent and flowing conditions will be discussed.<br></p>