La découverte en 2008 des supraconducteurs à base de fer a ouvert un nouveau champ d'investigation de la supraconductivité à haute température critique. En particulier, la phase nématique de ces matériaux pourrait jouer un rôle prépondérant dans le mécanisme de la supraconductivité. Nous avons étudié le composé FeSe par spectroscopie Raman, à pression ambiante et sous pression hydrostatique. Celui-ci ne possède pas d'ordre magnétique statique à pression ambiante, ce qui en fait un composé de choix pour l'étude de l'ordre nématique. Nous avons observé les fluctuations nématiques de charge. Leur évolution dans la phase tétragonale prouve l'existence d'une instabilité nématique d'origine électronique, qui gouverne la transition structurale. Dans la phase orthorhombique, le comportement des phonons souligne le rôle du couplage spin-phonon dans la transition nématique. Par ailleurs, la forme de la réponse Raman supraconductrice est compatible avec l'existence de deux gaps de symétrie s, dont un est anisotrope. Sous pression hydrostatique, les fluctuations nématiques s'atténuent rapidement. Le point critique quantique électronique associé se situe à très basse pression, peu avant l'apparition de l'ordre magnétique. Les fluctuations nématiques disparaissent complètement vers 2 GPa, quand la transition structurale passe de second ordre à premier ordre. C'est également proche de cette pression que se produit une anomalie dans le comportement des phonons, qui indique une modification de la structure électronique du système. Nos mesures révèlent en outre l'existence d'un pseudogap. Sa température d'apparition chute simultanément à la disparition de la phase magnétique, quand la température critique de supraconductivité atteint son maximum. Enfin, la réponse Raman de l'état supraconducteur à 7.8 GPa montre une signature claire d'un gap plein. / The discovery in 2008 of the iron-based superconductors opened a new field of investigation of high-temperature superconductivity. In particular, the nematic phase of these materials may play a major role in the mecanism of superconductivity. We studied the FeSe compound using Raman spectroscopy, at ambient pressure and under hydrostatic pressure. This material does not display any static magnetic order at ambient pressure and is therefore an excellent choice to study the nematic order. We observed the charge nematic fluctuations. Their evolution in the tetragonal phase proves the existence of an electronic nematic instability, which drives the structural transition. In the orthorhombic phase, the behaviour of the phonons underlines the role of the spin-phonon coupling in the nematic transition. Besides, the shape of the superconducting Raman response is compatible with the existence of two s-wave gaps, one of which is anisotropic. Under hydrostatic pressure, the nematic fluctuations reduce rapidly. The associated electronic quantum critical point is situated at very low pressure, just before the appearance of magnetic order. The nematic fluctuations completely disappear around 2 GPa, when the structural transition changes from second order to first order. An anomaly of the phonons also occurs close to this pressure, which indicates a modification of the electronic structure of the system. Our measurements additionally reveal the existence of a pseudogap. Its temperature of appearance reduces significantly simultaneously to the disappearance of magnetic order, when the critical temperature of superconductivity reaches its maximum. Finally, the Raman response in the superconducting state at 7.8 GPa shows a clear signature of a full gap.

This thesis presents experimental studies of the electronic and magnetic properties of several iron-based unconventional superconductors, primarily using the techniques of magnetotransport and torque magnetometry in high magnetic fields and synchrotron-based angle-resolved photo-emission spectroscopy (ARPES). Superconductivity in the iron-based superconductors is always found in proximity to a magnetic phase, and the details of the electronic structure and Fermi surface are also important in determining the strength of interactions, and ultimately superconductivity. This motivates the experimental studies of electronic, magnetic and superconducting properties of Fe-based superconductors presented in this thesis. First, quantum oscillation measurements using high-field torque magnetometry are used to provide a partial determination of the Fermi surface of superconducting LiFeAs. The data are compared with density functional theory calculations, finding strong mass enhancements on the observed electron bands, however the hole bands are not observed. A large portion of this thesis concerns experiments on FeSe, which uniquely has a structural transition but is not magnetically ordered at any temperature. High field magnetotransport measurements show quantum oscillations, revealing small quasi-two dimensional Fermi surfaces, and it is argued that both hole and electron pockets are observed. The low-temperature Fermi surface consisting of one hole pocket and two electron pockets is also deduced from low-field magnetotransport. ARPES studies show that both hole and electron pockets undergo a significant elongation when cooling through the structural transition at ~90 K, interpreted as the result of orbital order. Measurements of the resistivity anisotropy above the structural transition are used to show that the structural distortion is electronically-driven. By combining these data sets, a complete picture of the symmetry-broken electronic structure of FeSe is constructed. The final chapter concerns another iron-based superconductor with a more complex crystal structure, the so-called ``10-3-8" phase, and in particular finds an unusual field-induced magnetic transition.

La supraconductivité non-conventionnelle a récemment été observée à proximité d'un point critique induit sous pression dans CePt$_2$In$_7$ et dans FeSe. Le premier est un fermion lourd tandis que le deuxième fait parti de la famille des supraconducteurs à base de Fer. Cette thèse a pour objectif de contribuer à la compréhension de ces systèmes à électrons fortement corrélés en étudiant les évolutions des structures cristallographiques et magnétiques, ainsi que les surfaces de Fermi sous conditions extrêmes.Tout d'abord, nous présentons une étude de diffraction de neutrons dans la phase magnétique de CePt$_2$In$_7$. Une seule structure magnétique, avec comme vecteur de propagation $textbf{Q} = (0.5,0.5,0.5)$ et $0.45~mu_B$ par atome de Cérium à 2 K, a été détectée en dessous de $T_N = 5.5$ K.Ensuite, des mesures de torque sous champ pulsé de CePt$_2$In$_7$ ne montrent aucun changement des surfaces de Fermi jusqu'à 70 T, bien au-dessus du point critique quantique induit sous champ, attendu à $55-60$ T selon la littérature. Cependant, ces mesures révèlent une claire anomalie métamagnétique à 47 T, très peu dépendante de la température et de l'orientation du champ ainsi qu'une chute des masses effectives vers 50 T. Nous suggérons que ces deux derniers éléments sont la manifestation d'un changement de valence des atomes de Ce de l'ordre de 0.06 électron par atomes de Cérium.L'étude des surfaces de Fermi de CePt$_2$In$_7$ sous pression a nécessité un développement instrumental à partir d'un circuit résonant à base d'une diode tunnel, combinée avec une cellule de pression de type Bridgman. Nous montrons qu'il est tout à fait possible de sonder les surfaces de Fermi à la fois sous champ magnétique et sous pression avec cette technique. Cependant, la fragilité du système résonnant nous amène à considérer des améliorations en vue de fiabiliser la mesure.Enfin, nous analysons une expérience de diffraction de rayons X sur un échantillon de FeSe sous pression hydrostatique. A 20 K, nous quantifions un durcissement de l'axe cristallographique $c$ qui s'opère à 1.9 GPa. A 50 K, outre le passage d'une maille orthorhombique à tétragonale à 1 GPa, nous mettons en évidence qu'une symétrie monoclinique s'installe à partir de 2 GPa, où, selon la littérature, une phase antiferromagnétique apparaît. / Unconventional superconductivity was recently observed in the vicinity of a pressure-induced quantum critical point in CePt$_2$In$_7$ and FeSe. The former is a heavy fermion compound, while the latter is an iron-based superconductor. This PhD thesis aims at improving our understanding of the physics of these newly discovered strongly correlated electron systems. This is achieved by experimental investigation of the evolution of crystal and magnetic structures, as well as of the Fermi surfaces under extreme conditions of high magnetic fields, high pressure, and low temperatures.We have investigated the magnetic structure of CePt$_2$In$_7$ by neutron diffraction. We observed only one magnetic propagation vector $textbf{Q} = (0.5, 0.5, 0.5)$ below $T_N = 5.5$ K. The magnetic moment is estimated at $0.45~mu_B$ per cerium atom at 2 K.Our torque measurements on CePt$_2$In$_7$ in pulsed fields suggest that the Fermi surfaces remain unchanged up to 70 T. This is well above the field-induced quantum critical point, which is expected to occur at $55-60$ T according to previous studies. However, a clear metamagnetic-like anomaly is found at 47 T. The anomaly is almost temperature and field-angle independent. Furthermore, a sudden drop of the effective mass is observed at about the same field. We suggest that the last two observations are most naturally accounted for by a valence crossover, where the cerium valence changes by about 0.06, the order of magnitude expected in Ce-based compounds.In order to study the Fermi surfaces of CePt$_2$In$_7$ under pressure, we developed a tunnel diode oscillator combined with a Bridgman-type pressure cell. We have demonstrated that this set-up is suitable for measuring quantum oscillations both at high magnetic fields and under high pressure. However, the working conditions of the oscillator have to be improved in order to obtain a more reliable system.Finally, we performed an X-ray diffraction experiment on FeSe under hydrostatic pressure. At 20 K, we found a change of the bulk modulus along the $c$ axis at 1.9 GPa. At 50 K, the orthorhombic to tetragonal crystallographic phase transition occurs at 1 GPa. We found evidences that this is followed by the emergence of a monoclinic symmetry above 2 GPa, where an antiferromagnetic phase was previously reported.

<p>The 1D Wigner crystal is a long sought after strongly correlated quantum state. Here we present electronic transport data of asymmetric quantum point contacts (QPC) tuned to the spin-incoherent regime, which provides evidence for achieving the 1D Wigner state. Our result can be distinguished in several particularly noticeable ways. First, we utilize an asymmetric point contact geometry that is simple to fabricate and has not been studied previously. We are able to tune to the conductance anomalies simply by asymmetrically applying voltages to the gates. Second, we observe clear suppression of the first plateau and direct jumps to the second in these asymmetric QPCs at liquid helium temperatures (4.2 K). Such conductance behavior is indicative of Wigner crystal row formation.</p> <p>This thesis suggests that the novel geometry and gating scheme allows for a novel way to search for strongly correlated electronic behavior in quasi-1D quantum wires. A key finding is the importance of asymmetric QPCs for observation of anomalous transport characteristics. We have observed a strongly developed e<super>2</super>/h feature under asymmetric voltage gating and zero applied magnetic field. Such a feature is attributed to enhanced spin energies in the system. We believe the asymmetric design allows for a relaxing of the 1D confinement so that a quasi-1D electron conformation develops, which in turn allows for various possible magnetic states. In addition, by optimally tuning the confinement potential, we observe an unexpected suppression of the 2e<super>2</super>/h plateau. This provides further evidence for unusual electron arrangements in the asymmetric quantum point contact.</p> <p>I also discuss transport studies on the new FeSe superconductor. Our collaboration discovered the superconducting &beta;-FeSe compound with a Tc approximately 8 K. The crystal lattice structure of &beta;-FeSe is by far the simplest of the Fe superconductors. One of the most interesting observations regarding FeSe is that the crystal structure undergoes a structural transition at approximately 105 K from tetragonal to orthorhombic (or triclinic) symmetry. We believe this structural transition to be closely related to the origin of superconductivity in this class of materials.</p> <p>Transport studies also seem to support this claim. From Hall effect measurements of bulk FeSe, we find that FeSe is likely a two band (electron and hole) superconductor, which suggests it is quite different from the cuprates, and that very unconventional superconducting mechanisms are at play. The temperature dependence of the Hall coefficient is measured, and found to rapidly increase below 105 K. This suggests the scattering time related to hole bands dominate the transport at low temperature. As there is no magnetic ordering observed at low temperature, we do not expect the scattering from random Fe magnetic impurities to play a significant role in the enhanced hole scattering times. Thus, we speculate that this change is related to the structural transition observed.</p> / Dissertation

In this thesis, the results of analyzing the electronic structure of two iron-based superconductors: FeSe and LiFeAs are presented. To access the electronic structure, angle-resolved photoemission spectroscopy was used. In our analysis, we focus on the structure of the superconducting gap and the influence of nematicity on the electronic structure. We have revealed changes in the electronic structure of FeSe caused by nematicity in all parts of the Brillouin zone. A scale of these changes is smaller than it was believed earlier. Also, we have observed an anomalous shift of the dispersions in opposite directions with temperature in this material. We have observed anisotropic superconducting gap on all sheets of the Fermi surfaces of both: FeSe and LiFeAs. We have shown that in LiFeAs, rotational symmetry is broken in the superconducting state, which manifests not only in the gap symmetry but also in the shapes of the Fermi surfaces sheets. This result indicates a realization of a novel phenomenon of superconductivity-induced nematicity:1 Iron-based superconductors 1.1 Introduction to iron-based superconductors 1.2 LiFeAs - special iron-based superconductor 1.3 FeSe - structurally simplest iron-based superconductor 2 Angle-Resolved Photoemission 3 Temperature evolution of the electronic structure of FeSe 3.1 Effects of nematicity from low-temperature measurements 3.2 Temperature dependent shift of the dispersions 3.3 Discussion and conclusions 4 Three-dimensional superconducting gap in FeSe 4.1 Superconducting gap on the electron-like pockets 4.2 Superconducting gap on the hole-like pocket 4.3 Discussion and conclusions 5 Superconductivity-induced nematicity in LiFeAs 5.1 Superconducting gap 5.2 Nematicity 5.3 Discussion and conclusions Summary

La proximité omniprésente de l’ordre magnétique et supraconducteur dans les supraconducteurs non conventionnels implique l’importance de comprendre le magnétisme dans ces matériaux. Dans ce contexte, cette thèse porte sur l’étude du magnétisme dans trois supraconducteurs non conventionnels. Les excitations magnétiques dans le système d’oxychlorure de cuivre de l’élément léger Ca2CuO2Cl2 ont été étudiées en fonction du dopage et de la température en utilisant principalement la diffusion inélastique résonante aux rayons X. L’effet de la pression sur le magnétisme dans epsilon-fer et le beta-FeSe a été étudié en utilisant la spectroscopie d’émission des rayons X et la diffraction des neutrons sur poudre. / The ubiquitous proximity of magnetic and superconducting order in unconventional superconductors implies the importance of understanding magnetism in these materials. In this context, this thesis concerns the study of magnetism in three unconventional superconductors. The magnetic excitations in the light element copper oxychloride system Ca2CuO2Cl2 were studied as a function of doping and temperature using primarily resonant inelastic x-ray scattering. The effect of pressure on magnetism in epsilon-iron and beta-FeSe was studied using x-ray emission spectroscopy and neutron powder diffraction.

Recent resonant inelastic x-ray scattering (RIXS) experiments have detected a significant high-energy spin-excitation anisotropy in the nematic phase of the enigmatic iron-based superconductor FeSe, whose origin remains controversial. We apply an itinerant model previously used to describe the spin-excitation anisotropy as measured by neutron scattering measurements, with magnetic fluctuations included within the RPA approximation. The calculated RIXS cross section exhibits overall agreement with the RIXS data, including the high energy spin-excitation anisotropy