Theoretical studies based on first-principles theory are presented for a number of different magnetic systems. The first part of the thesis concerns spin dynamics and the second part concerns properties of magnetic multilayers. The theoretical treatment is based on electronic structure calculations performed by means of density functional theory. A method is developed for simulating atomistic spin dynamics at finite temperatures, which is based on solving the equations of motion for the atomic spins by means of Langevin dynamics. The method relies on a mapping of the interatomic exchange interactions from density functional theory to a Heisenberg Hamiltonian. Simulations are performed for various magnetic systems and processes beyond the reach of conventional micromagnetism. As an example, magnetization dynamics in the limit of large magnetic and anisotropy fields is explored. Moreover, the method is applied to studying the dynamics of systems with complex atomic order such as the diluted magnetic semiconductor MnGaAs and the spin glass alloy CuMn. The method is also applied to a Fe thin film and a Fe/Cr/Fe trilayer system, where the limits of ultrafast switching are explored. Current induced magnetization dynamics is investigated by calculating the current induced spin-transfer torque by means of density functional theory combined with the relaxation time approximation and semi-classical Boltzmann theory. The current induced torque is calculated for the helical spin-density waves in Er and fcc Fe, where the current is found to promote a rigid rotation of the magnetic order. Properties of magnetic multilayers composed of magnetic and nonmagnetic layers are investigated by means of the Korringa-Kohn-Rostocker interface Green's function method. Multilayer properties such as magnetic moments, interlayer exchange coupling and ordering temperatures are calculated and compared with experiments, with focus on understanding the influence of interface quality. Moreover, the influence on the interlayer exchange coupling of alloying the nonmagnetic spacer layers with small amounts of a magnetic impurity is investigated.

The understanding of new emerging unconventional ground states is a great challenge for experimental and theoretical solid-state physicists. New ground states are developing, where different energy scales compete, leading to a high sensitivity of the system to external tuning parameters like doping, pressure or magnetic field. The exploration of superconductivity proved to be a fascinating and challenging scientific undertaking. Discovered by H. Kammerlingh Onnes in 1911, prior to the development of the quantum theory of matter, superconductivity was defying a microscopic theory for more than four decades until the BCS theory was formulated in 1957 by J. Bardeen, L. N. Cooper and J. R. Schrieffer. Superconductivity of most of the simple metals or metallic alloys is well described within the frame of the BCS scenario, however, in the last thirty years numerous new superconducting materials were found to exhibit exotic properties not accounted for by the BCS theory. Among them are included the high-Tc compounds, the heavy-fermion superconductors and as well the organic superconductors. It was the purpose of this work to probe different facets of superconductivity in heavy-fermion and in low-dimensional metallic compounds. This dissertation is divided into six chapters. After this introduction, in Chapter 1 we will outline the basic theoretical concepts later needed for the analysis of the experimental results. In Chapter 2 we briefly introduce the experimental techniques with a special focus on the new pressure cells developed during this thesis and used for the measurements presented in Chapters 3 to 5. In Chapter 3 the possible realization of the inhomogeneous superconducting FFLO state in CeCoIn5 is studied by specific heat measurements under hydrostatic pressure, while in Chapter 4 the results of AC specific heat experiments on UBe13 under uniaxial pressure are presented. The ambient pressure properties as well as results obtained by resistivity measurements under hydrostatic pressure on the one-dimensional metallic compounds TlxV6S8 are discussed in Chapter 5. At the end, Chapter 6 summarizes and concludes this thesis.

The charge-density-wave (CDW) material orthorhombic tantalum trisulfide (TaS3) is a quasi-one dimensional material that forms long ribbon shaped crystals, and exhibits unique physical behavior. We have measured the dependence of the hysteretic voltage-induced torsional strain (VITS) in TaS3, which was first discovered by Pokrovskii et. al. in 2007, on temperature and applied torque. Our experimental results shows that the application of torque to the crystal could also change the VITS time constant, magnitude, and sign. This suggests that the VITS is a consequence of residual torsional strain originally present in the sample which twists the polarizations of the CDW when voltage is applied. This polarization twist then results in torque on the crystal. Another group of materials that may attract interest is that of small-molecule soluble organic semiconductors. Due to their assumed small phonon thermal conductivities and higher charge carrier mobilities, which will increase their seebeck coefficients with doping as compared to polymers, the small-molecule organic materials are promising for thermoelectric applications. In our experiments, we have measured the interlayer thermal conductivity of rubrene (C42H28), using ac-calorimetry. For rubrene, we find that the interlayer thermal conductivity, ≈ 0.7 mW/cm·K, is several times smaller than the (previously measured) in-plane value. Also, we have measured the interlayer and in-plane thermal conductivities of 6,13-bis((triisopropylsilyl)ethynyl) pentacene (TIPS-Pn). The in-plane value is comparable to that of organic metals with excellent π-orbital overlap. The interlayer (c-axis) thermal diffusivity is at least an order of magnitude larger than the in-plane, and this unusual anisotropy implies very strong dispersion of optical modes in the interlayer direction, presumably due to interactions between the silyl-containing side groups. Similar values for both in-plane and interlayer conductivities have been observed for several other functionalized pentacene semiconductors with related structures.

No description available.

The study of the instabilities in boiling water reactors is of significant importance to the safety withwhich they can be operated, as they can cause damage to the reactor posing risks to both equipmentand personnel. The instabilities that concern this paper are progressive growths in the oscillatingpower of boiling-water reactors. As thermal power is oscillatory is important to be able to identifywhether or not the power amplitude is stable. The main focus of this paper has been the development of a neural network estimator of these insta-bilities, fitting a non-linear model function to data by estimating it’s parameters. In doing this, theambition was to optimize the networks to the point that it can deliver near ”best-guess” estimationsof the parameters which define these instabilities, evaluating the usefulness of these networks whenapplied to problems like this. The goal was to design both MLP(Multi-Layer Perceptron) and SVR/KRR(Support Vector Regres-sion/Kernel Rigde Regression) networks and improve them to the point that they provide reliableand useful information about the waves in question. This goal was accomplished only in part asthe SVR/KRR networks proved to have some difficulty in ascertaining the phase shift of the waves.Overall, however, these networks prove very useful in this kind of task, succeeding with a reasonabledegree of confidence to calculating the different parameters of the waves studied.

The interplay between different self-organized electronically ordered states and their relation to unconventional electronic properties like superconductivity constitutes one of the most exciting challenges of modern condensed matter physics. In the present thesis this issue is thoroughly investigated for the prototypical layered material 1T-TaS2 both experimentally and theoretically. At first the static charge density wave order in 1T-TaS2 is investigated as a function of pressure and temperature by means of X-ray diffraction. These data indeed reveal that the superconductivity in this material coexists with an inhomogeneous charge density wave on a macroscopic scale in real space. This result is fundamentally different from a previously proposed separation of superconducting and insulating regions in real space. Furthermore, the X-ray diffraction data uncover the important role of interlayer correlations in 1T-TaS2. Based on the detailed insights into the charge density wave structure obtained by the X-ray diffraction experiments, density functional theory models are deduced in order to describe the electronic structure of 1T-TaS2 in the second part of this thesis. As opposed to most previous studies, these calculations take the three-dimensional character of the charge density wave into account. Indeed the electronic structure calculations uncover complex orbital textures, which are interwoven with the charge density wave order and cause dramatic differences in the electronic structure depending on the alignment of the orbitals between neighboring layers. Furthermore, it is demonstrated that these orbital-mediated effects provide a route to drive semiconductor-to-metal transitions with technologically pertinent gaps and on ultrafast timescales. These results are particularly relevant for the ongoing development of novel, miniaturized and ultrafast devices based on layered transition metal dichalcogenides. The discovery of orbital textures also helps to explain a number of long-standing puzzles concerning the electronic self-organization in 1T-TaS2 : the ultrafast response to optical excitations, the high sensitivity to pressure as well as a mysterious commensurate phase that is commonly thought to be a special phase a so-called “Mott phase” and that is not found in any other isostructural modification.

<div>In order to better understand and quantify the effect of instabilities in systems utilizing flow boiling heat transfer, the present study explores dynamic results for pressure drop, mass velocity, thermodynamic equilibrium quality, and heated wall temperature to ascertain and analyze the dominant modes in which they oscillate. Flow boiling experiments are conducted for a range of mass velocities with both subcooled and saturated inlet conditions in vertical upflow, vertical downflow, and horizontal flow orientations. High frequency pressure measurements are used to investigate the influence of individual flow loop components (flow boiling module, pump, pre-heater, condenser, etc.) on dynamic behavior of the fluid, with fast Fourier transforms of the same used to provide critical frequency domain information. Conclusions from this analysis are used to isolate instabilities present within the system due to physical interplay between thermodynamic and hydrodynamic effects. Parametric analysis is undertaken to better understand the conditions under which these instabilities form and their impact on system performance. Several prior stability maps are presented, with new stability maps provided to better address contextual trends discovered in the present study.</div><div>Further, this study utilizes experimental results for vertical upflow boiling of FC-72 in a rectangular channel with finite inlet quality to investigate Density Wave Oscillations (DWOs) and assess their potential impact on design of two-phase systems for future space missions. High-speed flow visualization image sequences are presented and used to directly relate the cyclical passage of High and Low Density Fronts (HDFs and LDFs) to dominant low-frequency oscillations present in transient pressure signals commonly attributed to DWOs. A methodology is presented to determine frequency and amplitude of DWO induced pressure oscillations, which are then plotted for a wide range of relevant operating conditions. Mass velocity (flow inertia) is seen to be the dominant parameter influencing frequency and amplitude of DWOs. Amplitude of pressure oscillations is at most 7% of the time-averaged pressure level for current operating conditions, meaning there is little risk to space missions. Reconstruction of experimental pressure signals using a waveform defined by frequency and amplitude of DWO induced pressure fluctuations is seen to have only moderate agreement with the original signal due to the oversimplifications of treating DWO induced fluctuations as perfectly sinusoidal in nature, assuming they occur at a constant frequency value, and neglecting other transient flow features. This approach is nonetheless determined to have potential value for use as a boundary condition to introduce DWOs in two-phase flow simulations should a model be capable of accurately predicting frequency and amplitude of oscillation.</div><div>Additionally, this study presents a new mechanistic model for Density Wave Oscillations (DWOs) in vertical upflow boiling using conclusions drawn from analysis of flow visualization images and transient experimental results as a basis from which to begin modeling. Counter to many prior studies attributing DWOs to feedback effects between flow rate, pressure drop, and flow enthalpy causing oscillations in position of the bulk boiling boundary, the present instability mode stems primarily from body force acting on liquid and vapor phases in a separated flow regime leading to liquid accumulation in the near-inlet region of the test section, which eventually departs and moves along the channel, acting to re-wet liquid film along the channel walls and re-establish annular, co-current flow. This process was modeled by dividing the test section into three distinct control volumes and solving transient conservation equations for each, yielding predictions of frequencies at which this process occurs as well as amplitude of associated pressure oscillations. Values for these parameters were validated against an experimental database of 236 FC-72 points and show the model provides good predictive accuracy and capably captures the influence of parametric changes to operating conditions.</div><div>Also, this study shows analysis of pressure signals in condensing systems reveal the presence of relevant oscillatory phenomena during flow condensation as well, which may impact performance in applications concerned with precise system control. Towards this end, the present study presents results for oscillatory behavior observed in pressure measurements during flow condensation of FC-72 in a smooth circular tube in vertical upflow, vertical downflow, and horizontal flow orientations. Dynamic behavior observed within the test section is determined to be independent of other components within the flow loop, allowing it to be isolated and interpreted as resulting from physical aspects of two-phase flow with condensation. The presence of a peak oscillatory mode (one of significantly larger amplitude than any others present) is seen for 72% of</div><div>vertical upflow test cases, 61% of vertical downflow, and 54% of horizontal flow. Relative intensities of this peak oscillatory mode are evaluated through calculation of Q Factor for the corresponding frequency response peak. Frequency and amplitude of peak oscillatory modes are also evaluated. Overall, vertical upflow is seen to exhibit the most significant oscillatory behavior, although in its maximum case amplitude is only seen to be 7.9% of time-averaged module inlet pressure, indicating there is little safety risk posed by oscillations under current operating conditions. Flow visualization image sequences for each orientation are also presented and used to draw parallels between physical characteristics of condensate film behavior under different operating conditions and trends in oscillatory behavior detected in pressure signals</div><div>Further, the present work outlines a new methodology utilizing temperature and pressure measurements to identify condensation flow regimes. For vertical upflow condensation, amplitude of dynamic temperature and pressure oscillations are shown to clearly indicate transition from counter-current flow regimes (i.e., falling film, oscillating film, flooding) to annular, co-current flow (climbing film flow regime). In horizontal flow condensation, standard deviation between multiple thermocouple measurements distributed around the tube circumference was calculated at all axial (stream-wise) measurement locations. High values of standard deviation are present for stratified flow (stratified flow, wavy-stratified, plug flow), while axisymmetric flow regimes (i.e., slug flow, annular flow) yield significantly lower values. Successful development of this technique represents a valuable contribution to literature as it allows condensation flow regime to be identified without the often-costly restriction of designing a test section to allow optical access. Identified flow regimes in both vertical upflow and horizontal flow orientations are compared to regime maps commonly found in the literature in pursuit of optimum performing maps.</div><div>Finally, the present study aims to better analyze the influence of body force on flow condensation heat transfer by conducting tests at multiple orientations in Earth’s gravity. Dielectric FC-72 is condensed in a smooth stainless-steel tube with 7.12 mm diameter and 574.55 mm condensing length by counterflow of cooling water across the outer surface of the tube. Test conditions span FC-72 mass velocities of 50.3 – 360.3 kg/m2s, test section inlet pressures of 127.0 – 132.1 kPa, and test section inlet thermodynamic equilibrium qualities of 0.13 – 1.15. A subset of data gathered corresponding to axisymmetric, annular condensation heat transfer is identified and a detailed methodology for data reduction to calculate heat transfer coefficient presented. Uncertainty analysis is also presented and indicates channel average heat transfer coefficients are calculated within ±3.6% to ±26.7% (depending on operating conditions). Analysis of parametric trends for condensation heat transfer reveals the dominant influence of mass velocity (flow inertia), secondary influence of vapor mass fraction (thermodynamic equilibrium quality), and strong dependence on orientation (body force) at low mass velocities. At higher mass velocities results for all orientations investigated begin to converge, indicating body force independent annular condensation heat transfer is achieved. Separated Flow Model predictions of vertical downflow condensation heat transfer provide reasonable agreement with experimental results, evidence by a Mean Absolute Error (MAE) of 31.2%. Evaluation of condensation heat transfer correlations for horizontal flow reveal most correlations struggle for cases with high liquid content. Specific correlations are identified for superior accuracy in predicting the measured data.</div>

La formation des ondes de densité de charge (ODC) est un phénomène de brisure de symétrie qui apparaît dans systèmes électroniques, et particulièrement dans les conducteurs quasi-unidimensionnels. Elle est observée aussi bien dans les matériaux très anisotropes que les isotropes comme par exemple les supraconducteurs pnictures. L'ODC peut être vue comme une déformation sinusoïdale de la densité électronique et de la modulation du réseau, ou également comme un cristal de singulets électroniques. Dans un état d'ODC, les cellules élémentaires peuvent être modifiées en absorbant ou en rejetant des paires d'électrons. Un tel processus passe par des configurations topologiquement non triviales: des solitons et des dislocations, ou plus généralement des vortex d'ODC. Ces états inhomogènes peuvent être obtenus expérimentalement dans des nano-produits appelés ''mésa-jonctions'', et observés à l'aide d'un microscope à effet tunnel ou d’une radiographie par micro-diffraction. Afin de simuler ces expériences, nous avons réalisé un programme modélisant les états stationnaires d'ODC ainsi que leur dynamique transitoire à travers des géométries restreintes auxquelles sont appliquées une tension ou un courant. Le modèle prend en compte plusieurs champs en interaction non linéaire: le paramètre d'ordre complexe d'ODC, la distribution de champ électrique, ainsi que la densité et le courant des autres porteurs de charge. Nous avons utilisé une approche de type Ginzburg-Landau ainsi qu'une extension basée sur une propriété d'invariance chirale. A l'aide de ce programme, nous avons observé la création progressive de dislocations statiques dans les jonctions. La dynamique transitoire est alors très riche avec notamment des créations, des annihilations et des balayages de vortex multiples. Des chutes de tension apparaissent au centre des dislocations, créant ainsi des jonctions tunnel microscopiques à travers lesquelles transitent des paires électron-trou. Les résultats qualitatifs obtenus sont en très bon accord avec les observations expérimentales. Ce model peut aussi être étendu à tout type de cristaux électronique comme les cristaux de Wigner dans les hétéro-jonctions et les nano-fils, les ODC dans les composés de chaîne, les ondes de densité de spin dans les conducteurs organiques, ou encore les structures de bandes dans les oxydes dopés. La reconstruction des ODC dans les jonctions tunnel peut aussi trouver son importance dans l'étude des effets de champs sur les matériaux fortement corrélés induisant des brisures spontanées de symétries. / Formation of charge density waves (CDW) is a symmetry breaking phenomenon found in electronic systems, which is particularly common in quasi-one-dimensional conductors. It is widely observed from highly anisotropic materials to isotropic ones like the superconducting pnictides. The CDW is seen as a sinusoidal deformation of coupled electronic density and lattice modulation; it can be also viewed as a crystal of singlet electronic pairs. In the CDW state, the elementary units can be readjusted by absorbing or rejecting pairs of electrons. Such a process should go via topologically nontrivial configurations: solitons and dislocations - the CDW vortices. An experimental access to these inhomogeneous CDW states came from studies of nano-fabricated mesa-junctions, from the STM and from the X-ray micro-diffraction. Following these requests, we have performed a program of modeling stationary states and of their transient dynamic for the CDW in restricted geometries under applied voltage or at passing normal currents. The model takes into account multiple fields in mutual nonlinear interactions: the two components of the CDW complex order parameter, and distributions of the electric field, the density and the current of normal carriers. We were using the Ginzburg-Landau type approach and also we have derived its extension based on the property of the chiral invariance. We observed the incremental creation of static dislocations within the junctions. The transient dynamics is very rich showing creation, annihilation and sweeping of multiple vortices. The dislocations cores concentrate the voltage drops thus providing self-tuned microscopic junctions where the tunneling creation of electron-hole pairs can take place. The results obtained from this model agree with experiment observations. The methods can be extended to other types of charge organization known under the general name of the Electronic Crystal. It takes forms of Wigner crystals at hetero-junctions and in nano-wires, CDWs in chain compounds, spin density waves in organic conductors, and stripes in doped oxides. The studied reconstruction in junctions of the CDW may be relevant also to modern efforts of the field-effect transformations in strongly correlated materials with a spontaneous symmetry breaking.

Formation of charge density waves (CDW) is a symmetry breaking phenomenon found in electronic systems, which is particularly common in quasi-one-dimensional conductors. It is widely observed from highly anisotropic materials to isotropic ones like the superconducting pnictides. The CDW is seen as a sinusoidal deformation of coupled electronic density and lattice modulation; it can be also viewed as a crystal of singlet electronic pairs. In the CDW state, the elementary units can be readjusted by absorbing or rejecting pairs of electrons. Such a process should go via topologically nontrivial configurations: solitons and dislocations - the CDW vortices. An experimental access to these inhomogeneous CDW states came from studies of nano-fabricated mesa-junctions, from the STM and from the X-ray micro-diffraction. Following these requests, we have performed a program of modeling stationary states and of their transient dynamic for the CDW in restricted geometries under applied voltage or at passing normal currents. The model takes into account multiple fields in mutual nonlinear interactions: the two components of the CDW complex order parameter, and distributions of the electric field, the density and the current of normal carriers. We were using the Ginzburg-Landau type approach and also we have derived its extension based on the property of the chiral invariance. We observed the incremental creation of static dislocations within the junctions. The transient dynamics is very rich showing creation, annihilation and sweeping of multiple vortices. The dislocations cores concentrate the voltage drops thus providing self-tuned microscopic junctions where the tunneling creation of electron-hole pairs can take place. The results obtained from this model agree with experiment observations. The methods can be extended to other types of charge organization known under the general name of the Electronic Crystal. It takes forms of Wigner crystals at hetero-junctions and in nano-wires, CDWs in chain compounds, spin density waves in organic conductors, and stripes in doped oxides. The studied reconstruction in junctions of the CDW may be relevant also to modern efforts of the field-effect transformations in strongly correlated materials with a spontaneous symmetry breaking.

Cette thèse présente une étude expérimentale de l'interaction entre la supraconductivité et une onde de densité de charge (ODC). Dans la théorie standard, la température critique d’un matériau supraconducteur est favorisée principalement par deux paramètres : une grande densité d’états au niveau de Fermi (nF), et un fort couplage électron-phonon. Cependant, un fort couplage électron-phonon favorise aussi l’apparition d’une ODC, ce qui réduit nF et rivalise ainsi avec la supraconductivité.Notre démarche a consisté à étudier deux composés où supraconductivité et ODC coexistent, et dans lesquels on peut faire disparaître l’ODC grâce à un paramètre externe : pression ou substitution. Le premier composé, 2H-NbSe2, présente une ODC en dessous de 33 K à pression ambiante. Celle-ci coexiste avec la supraconductivité en dessous de 7 K. Sous pression, l’ODC disparaît au-dessus de 4.6 GPa, sans que la température critique varie notablement. L’ODC disparaît aussi en remplaçant le sélénium par du soufre : 2H-NbS2 est ainsi un supraconducteur sans ODC (Tc = 6 K), et peut donc servir de composé témoin pour une étude comparative. Dans le second composé, Lu5Ir4Si10, une ODC est présente en dessous de 77 K à pression ambiante. Celle-ci disparaît sous pression au-dessus de 2 GPa, tandis que la température critique saute simultanément de 4 à 9 K. Pour étudier ces composés, j’ai utilisé trois techniques expérimentales : la mesure de la dispersion des phonons à basse température (300-2 K) et sous pression (0-16 GPa) par diffusion inélastique des rayons X, la mesure de la dépendance en température de la longueur de pénétration magnétique grâce à un oscillateur à diode tunnel et la mesure des champs critiques via des microsondes Hall.Dans la première partie, je présente la dépendance en température de la dispersion des phonons dans 2H-NbS2. Nous observons la présence d’un phonon mou dont l’énergie reste toujours positive, même extrapolée à température nulle. Ce composé est ainsi à la limite d'une instabilité ODC. De plus, nous montrons qu’il est relativement unique, car seuls les effets anharmoniques empêchent l’amollissement complet des phonons. Je présente ensuite la dépendance en température et en pression de la dispersion des phonons dans 2H-NbSe2. Ces expériences montrent qu’un mode de phonon mou persiste jusqu’à 16 GPa, même quand l'état à température nulle n'est pas l’ODC. La dépendance en température de ce phonon mou est alors similaire à celle de 2H-NbS2. Dans les deux composés, ces phonons mous semblent liés à la présence d'un couplage électron-phonon à la fois fort et anisotrope. Nous suggérons qu’il s’agit d’un élément essentiel pour expliquer leurs propriétés supraconductrices.Dans la seconde partie, je mesure l'anisotropie et la dépendance en température de la longueur de pénétration magnétique dans l’état supraconducteur de 2H-NbS2 et Lu5Ir4Si10. La dépendance en température de la densité superfluide dans 2H-NbS2 confirme la présence d'un gap supraconducteur réduit dont l'amplitude est très proche de celle mesurée dans 2H-NbSe2. Les phonons mous et le gap réduit étant présents dans 2H-NbS2 et 2H-NbSe2, nous prouvons expérimentalement qu'il faut raisonner en termes de renforcement de la supraconductivité par les phonons mous plutôt qu'en termes d’interaction avec l'état fondamental (ODC ou métal). Nous proposons que ce renforcement soit lié à l'anisotropie du couplage électron-phonon.En revanche, cet effet n’est pas général aux composés où supraconductivité et ODC coexistent. Les propriétés supraconductrices de Lu5Ir4Si10 sont en effet bien décrites par le modèle BCS couplage faible. Ceci est peut être lié aux caractéristiques de l’ODC : la présence d’une hystérésis montre que la transition ODC est du premier ordre. D’autre part, les mesures de diffraction X sous pression et à basse température révèlent que cette ODC est multiple : en plus de la périodicité 1/7, nous observons une seconde périodicité de 1/20. / This thesis presents an experimental study of the interaction between superconductivity and a charge density wave (CDW). In the standard theory, the critical temperature of a superconductor is principally enhanced by two parameters: a large density of states at the Fermi level (nF) and a strong electron-phonon coupling. However, a strong electron-phonon coupling also favors the appearance of a CDW, which reduces nF and therefore competes with superconductivity.Our strategy was to study two compounds in which superconductivity and CDW coexist, and in which the CDW can be suppressed through an external parameter: pressure or substitution. The first compound is 2H-NbSe2, it presents a CDW below 33 K at ambient pressure. This CDW coexists with superconductivity below 7 K. Under pressure, the CDW disappears above 4.6 GPa, meanwhile the critical temperature slowly changes. The CDW also disappears when replacing selenium by sulfur: 2H-NbS2 is a superconductor without CDW (Tc=6 K), it can therefore serve as a “test compound” for a comparative study. The second compound is Lu5Ir4Si10, it presents a CDW below 77 K at ambient pressure. Under pressure, this CDW disappears above 2 GPa, meanwhile the critical temperature abruptly jumps from 4 to 9 K.For this study, I used three experimental techniques: inelastic x-ray scattering at low temperature (300-2 K) and under pressure (0-16 GPa) to measure the dispersion of phonons, a tunnel diode oscillator to measure the temperature dependence of the magnetic penetration depth, and Hall microprobes to measure the first and second critical fields. In the first part, I present the temperature dependence of the phonon dispersion in 2H-NbS2. We observe a soft phonon that always remains at positive energies, even extrapolated to zero temperature. Thus, this compound is on the verge of CDW instability. It is also relatively unique, since we show anharmonicity is the only effect that prevents the complete softening of the phonons.Then I present the temperature and pressure dependence of the phonon dispersion in 2H-NbSe2. These experiments show that a soft phonon persists up to 16 GPa, even if the ground state is not a CDW. The temperature dependence of this soft phonon is then similar to that of 2H-NbS2. In both compounds, these soft modes seem to be related to the strength and anisotropy of the electron-phonon coupling. We suggest this is a fundamental element to explain their superconducting properties.In the second part, I measure the anisotropy and temperature dependence of the magnetic penetration depth in the superconducting state of 2H-NbS2 and Lu5Ir4Si10. The temperature dependence of the superfluid density in 2H-NbS2 confirms the presence of a reduced superconducting gap. Its amplitude is very similar to the one measured in 2H-NbSe2. The soft modes and the reduced gap being present in both 2H-NbSe2 and 2H-NbS2, we prove experimentally that the enhancement of superconductivity is related to the soft modes rather than to the nature of the ground state (CDW or metal). We suggest this enhancement is due the anisotropy of the electron-phonon coupling.However, this effect is not general to all compounds where superconductivity and CDW coexist. The superconducting properties of Lu5Ir4Si10 are indeed well fitted by the BCS model in the weak coupling limit. This may be related to the characteristics of the CDW: the presence of hysteresis shows that the CDW transition is first order. In addition, under pressure and at low temperature, x-ray diffraction measurements indicate that the CDW is multiple: aside from the periodicity of 1/7, we observe a second periodicity of 1/20.