Inertial Alfvén waves occur in plasmas where the Alfvén speed is greater than the electron thermal speed and the scale of wave field structure across the background magnetic field is comparable to the electron skin depth. Such waves have an electric field aligned with the background magnetic field that can accelerate electrons. It is likely that electrons are accelerated by inertial Alfvén waves in the auroral magnetosphere and contribute to the generation of auroras. While rocket and satellite measurements show a high level of coincidence between inertial Alfvén waves and auroral activity, definitive measurements of electrons being accelerated by inertial Alfvén waves are lacking. Continued uncertainty stems from the difficulty of making a conclusive interpretation of measurements from spacecraft flying through a complex and transient process. A laboratory experiment can avoid some of the ambiguity contained in spacecraft measurements. Experiments have been performed in the Large Plasma Device (LAPD) at UCLA. Inertial Alfvén waves were produced while simultaneously measuring the suprathermal tails of the electron distribution function. Measurements of the distribution function use resonant absorption of whistler mode waves. During a burst of inertial Alfvén waves, the measured portion of the distribution function oscillates at the Alfvén wave frequency. The phase space response of the electrons is well-described by a linear solution to the Boltzmann equation. Experiments have been repeated using electrostatic and inductive Alfvén wave antennas. The oscillation of the distribution function is described by a purely Alfvénic model when the Alfvén wave is produced by the inductive antenna. However, when the electrostatic antenna is used, measured oscillations of the distribution function are described by a model combining Alfvénic and non-Alfvénic effects. Indications of a nonlinear interaction between electrons and inertial Alfvén waves are present in recent data.

The Sun is a star, which due to its proximity has a tremendous influence on Earth. Since its very first days mankind tried to "understand the Sun", and especially in the 20th century science has uncovered many of the Sun's secrets by using high resolution observations and describing the Sun by means of models. As an active star the Sun's activity, as expressed in its magnetic cycle, is closely related to the sunspot numbers. Flares play a special role, because they release large energies on very short time scales. They are correlated with enhanced electromagnetic emissions all over the spectrum. Furthermore, flares are sources of energetic particles. Hard X-ray observations (e.g., by NASA's RHESSI spacecraft) reveal that a large fraction of the energy released during a flare is transferred into the kinetic energy of electrons. However the mechanism that accelerates a large number of electrons to high energies (beyond 20 keV) within fractions of a second is not understood yet. The thesis at hand presents a model for the generation of energetic electrons during flares that explains the electron acceleration based on real parameters obtained by real ground and space based observations. According to this model photospheric plasma flows build up electric potentials in the active regions in the photosphere. Usually these electric potentials are associated with electric currents closed within the photosphere. However as a result of magnetic reconnection, a magnetic connection between the regions of different magnetic polarity on the photosphere can establish through the corona. Due to the significantly higher electric conductivity in the corona, the photospheric electric power supply can be closed via the corona. Subsequently a high electric current is formed, which leads to the generation of hard X-ray radiation in the dense chromosphere. The previously described idea is modelled and investigated by means of electric circuits. For this the microscopic plasma parameters, the magnetic field geometry and hard X-ray observations are used to obtain parameters for modelling macroscopic electric components, such as electric resistors, which are connected with each other. This model demonstrates that such a coronal electric current is correlated with large scale electric fields, which can accelerate the electrons quickly up to relativistic energies. The results of these calculations are encouraging. The electron fluxes predicted by the model are in agreement with the electron fluxes deduced from the measured photon fluxes. Additionally the model developed in this thesis proposes a new way to understand the observed double footpoint hard X-ray sources. / Die Sonne ist ein Stern, der aufgrund seiner räumlichen Nähe einen großen Einfluss auf die Erde hat. Seit jeher hat die Menschheit versucht die "Sonne zu verstehen" und besonders im 20. Jahrhundert gelang es der Wissenschaft viele der offenen Fragen mittels Beobachtungen zu beantworten und mit Modellen zu beschreiben. Die Sonne ist ein aktiver Stern, dessen Aktivität sich in seinem magnetischen Zyklus ausdrückt, welcher in enger Verbindung zu den Sonnenfleckenzahlen steht. Flares spielen dabei eine besondere Rolle, da sie hohe Energien auf kurzen Zeitskalen freisetzen. Sie werden begleitet von erhöhter Strahlungsemission über das gesamte Spektrum hinweg und setzen darüber hinaus auch energetische Teilchen frei. Beobachtungen von harter Röntgenstrahlung (z.B. mit der RHESSI Raumsonde der NASA) zeigen, dass ein großer Teil der freigesetzten Energie in die kinetische Energie von Elektronen transferiert wird. Allerdings ist nach wie vor nicht verstanden, wie die Beschleunigung der vielen Elektronen auf hohe Energien (jenseits von 20 keV) in Bruchteilen einer Sekunde erfolgt. Die vorliegende Arbeit präsentiert ein Model für die Erzeugung von energetischen Elektronen während solarer Flares, das auf mit realen Beobachtungen gewonnenen Parametern basiert. Danach bauen photosphärische Plasmaströmungen elektrische Spannungen in den aktiven Regionen der Photosphäre auf. Für gewöhnlich sind diese Potentiale mit elektrischen Strömen verbunden, die innerhalb der Photosphäre geschlossen sind. Allerdings kann infolge von magnetischer Rekonnektion eine magnetische Verbindung in der Korona aufgebaut werden, die die Regionen von magnetisch unterschiedlicher Polarität miteinander verbindet. Wegen der deutlich höheren koronalen elektrischen Leitfähigkeit, kann darauf die photosphärische Spannungsquelle über die Korona geschlossen werden. Das auf diese Weise generierte elektrische Feld führt nachfolgend zur Erzeugung eines hohen elektrischen Stromes, der in der dichten Chromosphäre harte Röntgenstrahlung generiert. Die zuvor erläuterte Idee wird mit elektrischen Schaltkreisen modelliert und untersucht. Dafür werden die mikroskopischen Plasmaparameter, die Geometrie des Magnetfeldes und Beobachtungen der harten Röntgenstrahlung verwendet, um makroskopische elektronische Komponenten, wie z.B. elektrische Widerstände zu modellieren und miteinander zu verbinden. Es wird gezeigt, dass der auftretende koronale Strom mit hohen elektrischen Feldern verbunden ist, welche Elektronen schnell auf hohe relativistische Energien beschleunigen können. Die Ergebnisse dieser Berechnungen sind ermutigend. Die vorhergesagten Elektronenflüsse stehen im Einklang mit aus gemessenen Photonenflüssen gewonnenen Elektronenflüssen. Zudem liefert das Model einen neuen Ansatz für das Verständnis der harten Röntgendoppelquellen in den Fußpunkten.

Our dynamic Sun manifests its activity by different phenomena: from the 11-year cyclic sunspot pattern to the unpredictable and violent explosions in the case of solar flares. During flares, a huge amount of the stored magnetic energy is suddenly released and a substantial part of this energy is carried by the energetic electrons, considered to be the source of the nonthermal radio and X-ray radiation. One of the most important and still open question in solar physics is how the electrons are accelerated up to high energies within (the observed in the radio emission) short time scales. Because the acceleration site is extremely small in spatial extent as well (compared to the solar radius), the electron acceleration is regarded as a local process. The search for localized wave structures in the solar corona that are able to accelerate electrons together with the theoretical and numerical description of the conditions and requirements for this process, is the aim of the dissertation. Two models of electron acceleration in the solar corona are proposed in the dissertation: I. Electron acceleration due to the solar jet interaction with the background coronal plasma (the jet--plasma interaction) A jet is formed when the newly reconnected and highly curved magnetic field lines are relaxed by shooting plasma away from the reconnection site. Such jets, as observed in soft X-rays with the Yohkoh satellite, are spatially and temporally associated with beams of nonthermal electrons (in terms of the so-called type III metric radio bursts) propagating through the corona. A model that attempts to give an explanation for such observational facts is developed here. Initially, the interaction of such jets with the background plasma leads to an (ion-acoustic) instability associated with growing of electrostatic fluctuations in time for certain range of the jet initial velocity. During this process, any test electron that happen to feel this electrostatic wave field is drawn to co-move with the wave, gaining energy from it. When the jet speed has a value greater or lower than the one, required by the instability range, such wave excitation cannot be sustained and the process of electron energization (acceleration and/or heating) ceases. Hence, the electrons can propagate further in the corona and be detected as type III radio burst, for example. II. Electron acceleration due to attached whistler waves in the upstream region of coronal shocks (the electron--whistler--shock interaction) Coronal shocks are also able to accelerate electrons, as observed by the so-called type II metric radio bursts (the radio signature of a shock wave in the corona). From in-situ observations in space, e.g., at shocks related to co-rotating interaction regions, it is known that nonthermal electrons are produced preferably at shocks with attached whistler wave packets in their upstream regions. Motivated by these observations and assuming that the physical processes at shocks are the same in the corona as in the interplanetary medium, a new model of electron acceleration at coronal shocks is presented in the dissertation, where the electrons are accelerated by their interaction with such whistlers. The protons inflowing toward the shock are reflected there by nearly conserving their magnetic moment, so that they get a substantial velocity gain in the case of a quasi-perpendicular shock geometry, i.e, the angle between the shock normal and the upstream magnetic field is in the range 50--80 degrees. The so-accelerated protons are able to excite whistler waves in a certain frequency range in the upstream region. When these whistlers (comprising the localized wave structure in this case) are formed, only the incoming electrons are now able to interact resonantly with them. But only a part of these electrons fulfill the the electron--whistler wave resonance condition. Due to such resonant interaction (i.e., of these electrons with the whistlers), the electrons are accelerated in the electric and magnetic wave field within just several whistler periods. While gaining energy from the whistler wave field, the electrons reach the shock front and, subsequently, a major part of them are reflected back into the upstream region, since the shock accompanied with a jump of the magnetic field acts as a magnetic mirror. Co-moving with the whistlers now, the reflected electrons are out of resonance and hence can propagate undisturbed into the far upstream region, where they are detected in terms of type II metric radio bursts. In summary, the kinetic energy of protons is transfered into electrons by the action of localized wave structures in both cases, i.e., at jets outflowing from the magnetic reconnection site and at shock waves in the corona. / Die Sonne ist ein aktiver Stern, was sich nicht nur in den allseits bekannten Sonnenflecken, sondern auch in Flares manifestiert. Während Flares wird eine große Menge gespeicherter, magnetischer Energie in einer kurzen Zeit von einigen Sekunden bis zu wenigen Stunden in der Sonnenkorona freigesetzt. Dabei werden u.a. energiereiche Elektronen erzeugt, die ihrerseits nichtthermische Radio- und Röntgenstrahlung, wie sie z.B. am Observatorium für solare Radioastronomie des Astrophysikalischen Instituts Potsdam (AIP) in Tremsdorf und durch den NASA-Satelliten RHESSI beobachtet werden, erzeugen. Da diese Elektronen einen beträchtlichen Anteil der beim Flare freigesetzten Energie tragen, ist die Frage, wie Elektronen in kurzer Zeit auf hohe Energien in der Sonnenkorona beschleunigt werden, von generellem astrophysikalischen Interesse, da solche Prozesse auch in anderen Sternatmosphären und kosmischen Objekten, wie z.B. Supernova-Überresten, stattfinden. In der vorliegenden Dissertation wird die Elektronenbeschleunigung an lokalen Wellenstrukturen im Plasma der Sonnenkorona untersucht. Solche Wellen treten in der Umgebung der magnetischen Rekonnektion, die als ein wichtiger Auslöser von Flares angesehen wird, und in der Nähe von Stoßwellen, die infolge von Flares erzeugt werden, auf. Generell werden die Elektronen als Testteilchen behandelt. Sie werden durch ihre Wechselwirkung mit den elektrischen und magnetischen Feldern, die mit den Plasmawellen verbunden sind, beschleunigt. Infolge der magnetischen Rekonnektion als Grundlage des Flares werden starke Plasmaströmungen (sogenannte Jets) erzeugt. Solche Jets werden im Licht der weichen Röntgenstrahlung, wie z.B. durch den japanischen Satelliten YOHKOH, beobachtet. Mit solchen Jets sind solare Typ III Radiobursts als Signaturen von energiereichen Elektronenstrahlen in der Sonnenkorona verbunden. Durch die Wechselwirkung eines Jets mit dem umgebenden Plasma werden lokal elektrische Felder erzeugt, die ihrerseits Elektronen beschleunigen können. Dieses hier vorgestellte Szenarium kann sehr gut die Röntgen- und Radiobeobachtungen von Jets und den damit verbundenen Elektronenstrahlen erklären. An koronalen Stoßwellen, die infolge Flares entstehen, werden Elektronen beschleunigt, deren Signatur man in der solaren Radiostrahlung in Form von sogenannten Typ II Bursts beobachten kann. Stoßwellen in kosmischen Plasmen können mit Whistlerwellen (ein spezieller Typ von Plasmawellen) verbunden sein. In der vorliegenden Arbeit wird ein Szenarium vorgestellt, das aufzeigt, wie solche Whistlerwellen an koronalen Stoßwellen erzeugt werden und durch ihre resonante Wechselwirkung mit den Elektronen dieselben beschleunigen. Dieser Prozess ist effizienter als bisher vorgeschlagene Mechanismen und kann deshalb auch auf andere Stoßwellen im Kosmos, wie z.B. an Supernova-Überresten, zur Erklärung der dort erzeugten Radio- und Röntgenstrahlung dienen.

No description available.

Laser plasma wakefield accelerators have seen tremendous progress in the last years, now capable of producing electron beams in the GeV energy range. The inherent few-femtoseconds short bunch duration of these accelerators leads to ultra-high peak-currents. Reducing the energy spread found in these accelerators, while scaling their output to hundreds of kiloampere peak current would stimulate the next generation of radiation sources covering high-field THz, high-brightness X-ray and -ray sources, compact free-electron lasers and laboratory-size beam-driven plasma accelerators. At such high currents, an accelerator operates in the beam loaded regime where the accelerating field is strongly modified by the self-fields of the injected bunch, potentially deteriorating key beam parameters. However, if appropriately controlled, the beam loading effect can be employed to improve the accelerator’s performance, specifically to reduce the energy spread. In this thesis the beam-loading effect is systematically studied at a quasi-monoenergetic nanocoulomb-class laser wakefield accelerator. For this purpose, a tailored scheme of the self-truncated ionisation injection process is introduced for the non-linear bubble regime. This scheme facilitates stable and tunable injection of high-charge electron bunches within a short and limited time-frame, ensuring low energy spread right after injection. Employing a three millimetres gas-jet acceleration medium and a moderate 150 TW short pulse laser system as driver, unprecedented charges of up to 0.5 nC within a quasi-monoenergetic peak and energies of ~0.5 GeV are achieved. Studying the beam loading mechanism, it is demonstrated that at the optimal loading condition, i.e. at a specific amount of injected charge, performance of the accelerator is optimised with a minimisation of the energy spread. At a relative energy spread of only 15%, the associated peak current is around 10 kA, while scaling this scheme to operate with a petawatt driver laser promises peak-currents up to 100 kA.

Cette thèse porte sur le rayonnement X bêtatron généré par des électrons accélérés par sillage laser plasma dans des tubes capillaires diélectriques. En l’état actuel de la technologie des impulsions laser multi-térawatts, on peut produire des faisceaux ayant une intensité crête élevée, de l’ordre de 1018 W/cm2 dans le plan focal. Une telle impulsion laser se propageant au sein d’un gaz sous-dense conduit à des phénomènes d’interaction laser-plasma non-linéaires, tels que la création d’une bulle de plasma, i.e. une bulle ne contenant aucun électron, suivant le laser. La séparation spatiale des charges en résultant crée des champs électriques très élevés au sein de la bulle, de l'ordre de 100 GV/m, ce qui offre la possibilité d'accélérer des électrons jusqu'au GeV après seulement quelques centimètres d’interaction. En outre, un rayonnement synchrotron ultra-bref, appelé rayonnement bêtatron, est produit lors de l’accélération des électrons puisque ces derniers, soumis au champ électrique radial de la bulle plasma, ont une trajectoire oscillante. Cette thèse présente des résultats expérimentaux sur la génération et l'optimisation de faisceaux d'électrons et de leur rayonnement X, en particulier lorsque le tube capillaire est utilisé pour recueillir l'énergie du halo laser dans le plan focal facilitant l’autofocalisation du laser sur de longues distances. Des faisceaux d’électrons de quelques dizaines de picocoulomb, avec une énergie maximale allant jusqu’à 300 MeV, et dont le spectre est soit piqué à haute énergie soit exponentiellement décroissant, ont été produits dans des tubes capillaires de 10 mm de long avec l’installation laser du Lund Laser Center (LLC, en Suède) par une impulsion laser de 40 fs d’un 16 TW Ti: Saphir. Un rayonnement bêtatron a également été mesuré, il se compose de de photons X dont l’énergie est comprise entre 1 et 10 keV et atteint une luminosité maximale d’environ 1021 photons/s/mm²/mrad²/0.1%BW. Cela équivaut à environ 30 fois l’intensité des faisceaux générés dans le cas des jets de gaz de longueur 2 mm ne disposant pas de guidage optique externe. La compensation des fluctuations de pointé laser permet de minimiser les fluctuations des propriétés du faisceau d’électrons. On obtient des faisceaux d'électrons dont les fluctuations tir-a-tir sont de 1 mrad en pointé, de quelques pourcents en énergie et d’environ 20% RMS en charge. La fluctuation en charge du faisceau, qui peut être considérée comme relativement grande, s’avère être principalement corrélée à la fluctuation en puissance du laser. De plus, il a été montré que le rayonnement bêtatron pouvait être utilisé pour caractériser le processus d'accélération des électrons en caractérisant le nombre moyen d'oscillations bêtatron effectuées par les électrons à l'intérieur de la bulle plasma. La taille typique des sources de rayonnement X (dimension pour laquelle l’intensité gaussienne est égale à 1/e² de la valeur crête) est estimée à ~ 2.5 µm en utilisant un modèle de diffraction de Fresnel induite par une lame de rasoir. Cela correspond à une émittance RMS normalisée pour le faisceau d'électrons d’environ 0,83π mm.mrad. Des simulations tridimensionnelles particle-in-cell (PIC) ont été effectuées et confirment les résultats expérimentaux. Elles indiquent également que les paquets d'électrons générés ainsi que les flashs X directionnels sont ultra-brefs : ~ 10 fs. / This dissertation addresses electron acceleration and the associated betatron X-ray radiation generated by laser wakefield inside dielectric capillary tubes. Focusing the state-of-the-art multi-terawatt laser pulses, high peak intensity, of the order of 1018 W/cm2, can be achieved in the focal plane, where a plasma bubble free of electron is formed just behind the laser. Owing to space charge separation ultrahigh electric fields, of the order of 100 GV/m, occur inside the plasma bubble, providing the possibility to accelerate electrons up to GeV-class over merely a centimetre-scale distance. Furthermore, ultra-short synchrotron-like X-ray radiation, known as betatron radiation, is produced simultaneously when the accelerated electrons are transversely wiggled by the radial electric field inside the plasma bubble. This thesis reports experimental results on the generation and optimization of electron and X-ray beams, particularly when a capillary tube is used to collect the energy of laser halos in the focal plane to facilitate the laser keeping self-focused over a long distance. Employing the 40 fs, 16 TW Ti:sapphire laser at the Lund Laser Centre (LLC) in Sweden, either peaked or widely-spread accelerated electron spectra with a typical beam charge of tens of pC were measured with a maximum energy up to 300 MeV in 10 mm long capillary tubes. Meanwhile, betatron X-ray radiation consisting of 1-10 keV photons was measured with a peak brightness of the order of 1021 photons/s/mm2/mrad2/0.1%BW, which is around 30 times higher than that in the case of a 2 mm gas jet without external optical guiding. When the laser pointing fluctuation is compensated, exceptionally reproducible electron beams are obtained with fluctuations of only 1 mrad RMS in beam pointing, a few percent in electron energy, and around 20% RMS in beam charge. The relatively large instability of beam charge is found to be essentially correlated to laser power fluctuation. Moreover, betatron radiation is able to provide the diagnostics about electron acceleration process and average number of betatron oscillations fulfilled by electrons inside the plasma bubble. The typical X-ray source size (waist of Gaussian distribution at 1/e2 intensity) is quantified to be ~2.5 μm using Fresnel diffraction induced by a razor blade, which furthermore yields the corresponding normalized RMS emittance of electron beam 0.83π mm mrad. Three dimensional particle-in-cell (PIC) modelings are in good agreement with the experimental findings. The PIC simulations also reveal the generated electron bunches (or X-ray bursts) have pulse durations as short as 10 fs.

Accurate single-shot visualization of laser wakefield structures can improve our fundamental understanding of plasma-based accelerators. Previously, frequency domain holography (FDH) was used to visualize weakly nonlinear sinusoidal wakes in plasmas of density n[subscript e] < 0.6 × 10¹⁹/cm³ that produced few or no relativistic electrons. Here, I address the more challenging task of visualizing highly nonlinear wakes in plasmas of density n[subscript e] ~ 1 to 3× 10¹⁹/cm³ that can produce high-quality relativistic electron beams. Nonlinear wakes were driven by 30 TW, 30 fs, 800 nm pump pulses. When bubbles formed, part of a 400 nm, co-propagating, overlapping probe pulse became trapped inside them, creating a light packet of plasma wavelength dimensions--that is, an optical "bullet"--that I reconstruct by FDH methods. As ne increased, the bullets first appeared at 0.8 × 10¹⁹/cm³, the first observation of bubble formation below the electron capture threshold. WAKE simulations confirmed bubble formation without electron capture and the trapping of optical bullets at this density. At n[subscript] >1× 10¹⁹/cm³, bullets appeared with high shot-to-shot stability together with quasi-monoenergetic relativistic electrons. I also directly observed the temporal walk-off of the optical bullet from the beam-loaded plasma bubble revealed by FDH phase shift data, providing unprecedented visualization of the electron injection and beam loading processes. There are five chapters in this thesis. Chapter 1 introduces general laser plasma- based accelerators (LPA). Chapter 2 discusses the FDH imaging technique, including the setup and reconstruction process. In 2006, Dr. N. H. Matlis used FDH to image a linear plasma wakefield. His work is also presented in Chapter 2 but with new analyses. Chapter 3, the main part of the thesis, discusses the visualization of LPAs in the bubble regime. Chapter 4 presents the concept of frequency domain tomography. Chapter 5 suggests future directions for research in FDH. / text

This dissertation addresses electron acceleration and the associated betatron X-ray radiation generated by laser wakefield inside dielectric capillary tubes. Focusing the state-of-the-art multi-terawatt laser pulses, high peak intensity, of the order of 1018 W/cm2, can be achieved in the focal plane, where a plasma bubble free of electron is formed just behind the laser. Owing to space charge separation ultrahigh electric fields, of the order of 100 GV/m, occur inside the plasma bubble, providing the possibility to accelerate electrons up to GeV-class over merely a centimetre-scale distance. Furthermore, ultra-short synchrotron-like X-ray radiation, known as betatron radiation, is produced simultaneously when the accelerated electrons are transversely wiggled by the radial electric field inside the plasma bubble. This thesis reports experimental results on the generation and optimization of electron and X-ray beams, particularly when a capillary tube is used to collect the energy of laser halos in the focal plane to facilitate the laser keeping self-focused over a long distance. Employing the 40 fs, 16 TW Ti:sapphire laser at the Lund Laser Centre (LLC) in Sweden, either peaked or widely-spread accelerated electron spectra with a typical beam charge of tens of pC were measured with a maximum energy up to 300 MeV in 10 mm long capillary tubes. Meanwhile, betatron X-ray radiation consisting of 1-10 keV photons was measured with a peak brightness of the order of 1021 photons/s/mm2/mrad2/0.1%BW, which is around 30 times higher than that in the case of a 2 mm gas jet without external optical guiding. When the laser pointing fluctuation is compensated, exceptionally reproducible electron beams are obtained with fluctuations of only 1 mrad RMS in beam pointing, a few percent in electron energy, and around 20% RMS in beam charge. The relatively large instability of beam charge is found to be essentially correlated to laser power fluctuation. Moreover, betatron radiation is able to provide the diagnostics about electron acceleration process and average number of betatron oscillations fulfilled by electrons inside the plasma bubble. The typical X-ray source size (waist of Gaussian distribution at 1/e2 intensity) is quantified to be ~2.5 μm using Fresnel diffraction induced by a razor blade, which furthermore yields the corresponding normalized RMS emittance of electron beam 0.83π mm mrad. Three dimensional particle-in-cell (PIC) modelings are in good agreement with the experimental findings. The PIC simulations also reveal the generated electron bunches (or X-ray bursts) have pulse durations as short as 10 fs.

La plasmonique étudie le couplage entre le rayonnement électromagnétique et les oscillations collectives des électrons dans un matériel. Les plasmons de surface (SPs), notamment, ont la capacité de concentrer le champ électromagnétique sur des distances micrométriques, ce qui les rend intéressants pour le développement des dispositifs photoniques les plus novateurs. 'Etendre l'excitation de SPs au régime de champs élevés, où les électrons oscillent à des vitesses relativistes, ouvre des perspectives stimulantes pour la manipulation de la lumière laser ultra-intense et le développement de sources de rayonnement énergétiques et à courte durée. En fait, l'excitation de modes résonnants du plasma est l'une des stratégies possibles pour transférer efficacement l'énergie d'une impulsion laser ultra-puissante à une cible solide, cela étant parmi les défis actuels dans la physique de l’interaction laser-matière à haute intensité. Dans le cadre de ces deux sujets, ce travail de thèse démontre la possibilité d'exciter de façon résonnante des plasmons de surface avec des impulsions laser ultra-intenses. Elle étudie comment ces ondes peuvent à la fois accélérer de paquets d'électrons relativistes le long de la surface de la cible mais aussi augmenter la génération d'harmoniques d'ordre élevé de la fréquence laser. Ces deux processus ont été caractérisés avec de nombreuses expériences et simulations numériques. En utilisant un schéma d’interaction standard de la plasmonique classique, les SPs sont excités sur des cibles dont la surface présente une modulation périodique régulière à l'échelle micrométrique (cibles réseau). Dans ce cas, les propriétés de l'émission d'électrons tout comme celles des harmoniques permettent d’envisager leur utilisation dans des application pratiques. En réussissant à dépasser les principaux problèmes conceptuels et techniques qui jusqu'au présent avaient empêché l'application d'effets plasmoniques dans le régime de champs élevés, ces résultats apportent un intérêt nouveau à l'exploration de la Plasmonique Relativiste. / Plasmonics studies how the electromagnetic radiation couples with the collective oscillations of the electrons within a medium. Surface plasmons (SPs), in particular, have a well-established role in the development of forefront photonic devices, as they allow for strong enhancement of the local EM field over sub-micrometric dimensions. Promoting the SP excitation to the high-field regime, where the electrons quiver at relativistic velocities, would open stimulating perspectives for the both the manipulation of ultra-intense laser light and the development of energetic, short radiation sources. Indeed, the excitation of resonant plasma modes is a possible strategy to efficiently deliver the energy of a high-power laser to a solid target, this being among the current challenges in the physics of highly-intense laser-matter interaction. Gathering these topics, this thesis demonstrates the opportunity of resonant surface plasmon excitation at ultra-high laser intensities by studying how such waves accelerate bunches of relativistic electrons along the target surface and how they enhance the generation of high-order harmonics of the laser frequency. Both these processes have been investigated with numerous experiments and extensive numerical simulations. Adopting a standard configuration from classical plasmonics, SPs are excited on solid, wavelength-scale grating targets. In their presence, both electron and harmonic emissions exhibit remarkable features that support the conception of practical applications. Putting aside some major technical and conceptual issues discouraging the applicability of plasmonic effects in the high-field regime, these results are expected to mark new promises to the exploration of Relativistic Plasmonics.

No description available.