Search

The effect of scattering of conduction electrons by dynamical spin fluctuations on the thermopower in metals near a thermal phase transition into an antiferromagnetic phase is considered. We are interested in a transition at room temperature, as has been studied in a heterostructure involving layers of IrMn. We show that the electrical resistivity exhibits a narrow but low peak at the transition, which may be difficult to detect on top of the main contributions induced by phonons and impurities. By contrast, the thermopower is found to exhibit a prominent peak both as a function of temperature $T$ for fixed layer thickness ${t}_{\text{AFM}}$ and as a function of ${t}_{\text{AFM}}$ for fixed $T.$ We conjecture that the transition temperature ${T}_{c}$ is a function of both ${t}_{\text{AFM}}$ and the Fermi energy ${\ensuremath{\epsilon}}_{F}$. Both dependencies give rise to a sharp peak of the thermopower as a function of $T$ or ${t}_{\text{AFM}}$ near the transition. The estimated magnitude and width of the peak for the case of either three- or two-dimensional longitudinal spin fluctuations is in good agreement with experiment.

By means of spin current, the flow of spin angular momentum, we find a regime of "spin treacle" in a frustrated magnetic system. To establish its existence, we have performed spin transport measurements in nanometer-scale spin glasses. At temperatures high enough that the magnetic moments fluctuate at high frequencies, the spin Hall angle, the conversion yield between spin current and charge current, is independent of temperature. The spin Hall angle starts to decrease at a certain temperature $T^{*}$ and completely vanishes at a lower temperature. We argue that the latter corresponds to the spin freezing temperature $T_{\rm f}$ of the nanometer-scale spin glass, where the direction of conduction electron spin is randomized by the exchange coupling with the localized moments. The present experiment \textit{quantitatively} verifies the existence of a distinct "spin treacle" between $T_{\rm f}$ and $T^{*}$. We have also quantified a time scale of fluctuation of local magnetic moments in the spin treacle from the spin relaxation time of conduction electrons., 8 pages, 9 figures

We measure the mode-resolved direction of the precessional motion of the magnetic order, i.e., magnon polarization, via the chiral term of inelastic polarized neutron scattering spectra. The magnon polarization is a unique and unambiguous signature of magnets and is important in spintronics, affecting thermodynamic properties such as the magnitude and sign of the spin Seebeck effect. However, it has never been directly measured in any material until this work. The observation of both signs of magnon polarization in Y_{3}Fe_{5}O_{12} also gives direct proof of its ferrimagnetic nature. The experiments agree very well with atomistic simulations of the scattering cross section.

Motivated by the complex phase diagram of MnWO4, we investigate competition between anisotropy, magnetic field, and helicity for the anisotropic next-nearest-neighbor Heisenberg model. Apart from two competing exchanges, which favor a spiral magnetic structure, the model features the bi-axial single-ion anisotropy. The model is treated in the real-space mean-field approximation and the phase diagram containing various incommensurate and commensurate states is obtained for different field orientations. We discuss similarities and differences of the theoretical phase diagram and the experimental diagram of MnWO4., Comment: 5 pages, as accepted

We investigate the possibility of ultracold neutron (UCN) storage in quantum states defined by the combined potentials of the Earth's gravity and the neutron optical repulsion by a horizontal surface of liquid helium. We analyse the stability of the lowest quantum state, which is most susceptible to perturbations due to surface excitations, against scattering by helium atoms in the vapor and by excitations of the liquid, comprised of ripplons, phonons and surfons. This is an unusual scattering problem since the kinetic energy of the neutron parallel to the surface may be much greater than the binding energies perpendicular. The total scattering time constant of these UCNs at 0.7 K is found to exceed one hour, and rapidly increasing with decreasing temperature. Such low scattering rates should enable high-precision measurements of the scheme of discrete energy levels, thus providing improved access to short-range gravity. The system might also be useful for neutron beta-decay experiments. We also sketch new experimental concepts for level population and trapping of UCNs above a flat horizontal mirror., 18 pages, 3 figures

We perform a systematical analysis of the spin Hall effect (SHE) in the Cu alloys doped with a series of 5 d elements, by the combined approach of density functional theory and Hartree–Fock approximation. We find that not only the spin orbit interactions (SOI) in both the 5 d and 6 p orbitals, but also the local correlations in the 5 d orbitals of the impurities, are decisive on the sign of the spin Hall angle (SHA). Including all of these three factors properly, we predict the SHA for each alloy in the series. The signs of CuIr and CuPt are sensitive to perturbation of the local correlations. This observation is favorable for controlling the sign of the transverse spin Hall voltage.

Spin relaxation in a quantum Hall ferromagnet, where filling is $\nu=1, 1/3, 1/5,...$, can be considered in terms of spin wave annihilation/creation processes. Hyperfine coupling with the nuclei of the GaAs matrix provides spin non-conservation in the two-dimensional electron gas and determines spin relaxation in the quantum Hall system. This mechanism competes with spin-orbit coupling channels of spin-wave decay and can even dominate in a low-temperature regime where $T$ is much smaller than the Zeeman gap. In this case the spin-wave relaxation process occurs non-exponentially with time and does not depend on the temperature. The competition of different relaxation channels results in crossovers in the dominant mechanism, leading to non-monotonic behavior of the characteristic relaxation time with the magnetic field. We predict that the relaxation times should reach maxima at $B\simeq 18\,$T in the $\nu=1$ Quantum Hall system and at $B\simeq 12\,$T for that of $\nu=1/3\,$. We estimate these times as $\sim10\,-\,30\,\mu$s and $\sim2\,-\,5\,\mu$s, respectively., Comment: 26 pages, 1 figure

We consider sweeping a system through a Landau-Zener avoided-crossing, when that system is also coupled to an environment or noise. Unsurprisingly, we find that decoherence suppresses the coherent oscillations of quantum superpositions of system states, as superpositions decohere into mixed states. However, we also find an effect we call "Lamb-assisted coherent oscillations", in which a Lamb shift exponentially enhances the coherent oscillation amplitude. This dominates for high-frequency environments such as super-Ohmic environments, where the coherent oscillations can grow exponentially as either the environment coupling or temperature are increased. The effect could be used as an experimental probe for high-frequency environments in such systems as molecular magnets, solid-state qubits, spin-polarized gases (neutrons or He3) or Bose-condensates., 4 Pages & 4 Figs - New version: introduction extended & citations added

The direct electron spin resonance (ESR) absorption between a singlet ground state and the triplet excited states of spin gap systems is investigated. Such an absorption, which is forbidden by the conservation of the total spin quantum number in isotropic Hamiltonians, is allowed by the Dzyaloshinskii-Moriya interaction. We show a selection rule in the presence of this interaction, using the exact numerical diagonalization of the finite cluster of the quasi-one-dimensional bond-alternating spin system. The selection rule is also modified into a suitable form in order to interpret recent experimental results on CuGeO$_3$ and NaV$_2$O$_5$., 5 pages, Revtex, with 6 eps figures, to appear in J. Phys. Soc. Jpn. Vol. 69 No. 11 (2000)

We have performed a numerical investigation of the ground state properties of the frustrated quantum Heisenberg antiferromagnet on the square lattice (``$J_1-J_2$ model''), using exact diagonalization of finite clusters with 16, 20, 32, and 36 sites. Using a finite-size scaling analysis we obtain results for a number of physical properties: magnetic order parameters, ground state energy, and magnetic susceptibility (at $q=0$). For the unfrustrated case these results agree with series expansions and quantum Monte Carlo calculations to within a percent or better. In order to assess the reliability of our calculations, we also investigate regions of parameter space with well-established magnetic order, in particular the non-frustrated case $J_2 0.68$. There thus is a region in parameter space without any form of magnetic order. Including the 16 site cluster, or analyzing the independently calculated magnetic susceptibility we arrive at the same conclusion, but with modified values for the range of existence of the nonmagnetic region. We also find numerical values for the spin-wave velocity and the spin stiffness. The spin-wave velocity remains finite at the magnetic-nonmagnetic transition, as expected from the nonlinear sigma, 34 pages, RevTeX 3.0, figures appended in uuencoded form

We calculate the level statistics by finding the eigenvalue spectrum for a variety of one-dimensional many-body models, namely the Heisenberg chain, the t-J model and the Hubbard model. In each case the generic behaviour is GOE, however at points corresponding to models known to be exactly integrable Poisson statistics are found, in agreement with an argument we outline., Latex file (8 pages), figures available on request

A finite size scaling analysis of the spectral function and of the optical conductivity of a single hole moving in an antiferromagnetic background is performed. It is shown that both the low energy quasiparticle peak and the broad higher energy structure are robust with increasing cluster size from $4\times 4$ to $\sqrt{26}\times\sqrt{26}$ sites. In the abscence of spin fluctuations, for most static or dynamical quantities saturation occurs when the size exceeds a characteristic size $N_c(J_z)$. Typically, 16 and 26 site clusters give reliable results for $J_z>0.75$ and $J_z>0.3$ respectively. The hole optical mass is shown to be very large ($>20$) in agreement with the small bandwidth. Due to the energy gap to flip a spin in the vicinity of a hole, a small gap $\propto J_z$ separates the low energy delta-function from the rest of the spectrum in the dynamical correlation functions. On the other hand, with $J_\perp$ this gap seems to disappear with increasing system size as one would expect since the spin waves are gapless in the thermodynamic limit. The large momentum dependence of the quasiparticle weight in the isotropic case is inconsistent with a string picture but agrees well with the self-consistent Born approximation. An accurate estimation of the higher energy part of the spectral functions of the t--J model can be made for momenta close to $(0,0)$ or $(��,��)$, 33 pages (without figures)

Following the technological advances of the Industrial Revolution, classical thermodynamics was developed in the 19th century in order to understand the conversion of heat into work in newly designed machines. The works of Boltzmann brought another conceptual revolution with statistical mechanics. He demonstrated the microscopical origin of the laws of thermodynamics which actually only describe the macroscopic behaviour of systems in which local thermalization is faster than all other timescales. However, following the growing interest for nanotechnologies, it is now possible to manipulate microscopic systems in which thermalization is slower than the timescales for electron flow. A major technological advance in this field stems from the use of quantum dots, nanoscale devices which confine electrons on such small scales that they spread on discrete energy levels. It is then essential to take into account quantum effects for the study of this type of systems, that is to say to design theoretical tools combining thermodynamics and quantum mechanics.Problems of quantum thermodynamics are often tackled in the framework of the theory of open quantum systems. The general idea of this formalism is to study the dynamics of a “small” quantum system when it is coupled to another much bigger representing the environment. One can then show that the time evolution of the small system can be described by a master equation in the limit where it is weakly coupled to the environment. However, it intuitively seems that the power output of machine would be higher in the context of strong coupling.For problems of electronic transport, the Landauer-Büttiker formalism allows to describe the strong-coupling regime. In this framework, electrons are assumed to solely undergo elastic scattering processes in the central system. All the thermoelectric properties of the machine can then be characterized thanks to the transmission properties of the scatterer. However, this formalism has an important limitation; it ignores the band structure of the reservoirs.Here we have chosen to adopt a different viewpoint to tackle the strong-coupling regime by studying an exactly soluble model. We therefore analyze the Fano-Anderson model describing a discrete level coupled to a continuum. We are particularly interested by the influence of the reservoirs’ band structure. One can indeed show that, under certain conditions, discrete bound states appear in the band gaps of the reservoirs. This state play an important rôle on the dynamics of the discrete at long times: their contribution depends on the initial preparation of the system and gives rise to persistent oscillations of the occupation of the discrete level.We start by deriving the exact solution of the model especially focusing on its long-time behaviour. We then analyze two special cases. First, we study the transport properties of a single-level quantum dot coupled to a semiconductor with single a band gap. A bound state appears in this gap when the coupling to the reservoir exceeds a critical value. We show that this greatly affects the transport properties of the device. We then study the case of reservoirs described by a tight-binding model which density of states consists of a single finite-range energy band. We show that a discrete level coupled to such reservoir behaves like a many-level system as its local density of states and transmission function exhibits multiple resonances.; Suivant les progrès technologiques de la révolution industrielle, la thermodynamique classique a été développée au XIXème siècle dans le but de comprendre la conversion de la chaleur en travail intervenant dans les machines thermiques nouvellement élaborées. Les travaux de Boltzmann apportèrent une autre révolution conceptuelle avec la physique statistique. Il démontra l’origine microscopique des lois de la thermodynamique, celles-ci ne décrivant en fait que le comportement macroscopique de systèmes pour lesquels la thermalisation locale est plus rapide que toutes les autres échelles de temps. Cependant, conséquemment à l’intérêt grandissant pour les nanotechnologies, il est aujourd’hui possible de manipuler des systèmes microscopiques pour lesquels la thermalisation est plus lente que les échelles de temps associés aux flux d’électrons. Une avancée technologique majeure dans ce domaine provient de l’utilisation de boîtes quantiques, des dispositifs nanométriques permettant de confiner les électrons sur des distances si petites qu’ils se répartissent sur des niveaux d’énergie discrets. Il est alors évidemment indispensable de prendre en compte les effets quantiques pour l’étude de ce type de systèmes, c’est-à-dire de concevoir des outils théoriques alliant thermodynamique et mécanique quantique.Les problèmes de thermodynamique quantique sont souvent abordés dans le cadre de la théorie des systèmes quantiques ouverts. L’idée générale de ce formalisme est d’étudier la dynamique d’un « petit » système quantique lorsqu’il est couplé à un autre système supposé bien plus « gros » et représentant l’environnement. On démontre alors que l’évolution temporelle du petit système peut être décrite par une équation maîtresse dans la limite où il est faiblement couplé à l’environnement. Cependant, il semble intuitivement qu’une machine pourra délivrer une puissance plus importante dans un contexte de fort couplage-Pour les problèmes de transport électronique, le formalisme de Landauer-Büttiker permet de décrire le régime de fort couplage. Dans ce cadre, les électrons sont supposés ne subir que des processus de diffusion élastique dans le système central. Toutes les propriétés thermoélectriques de la machine peuvent alors être caractérisées grâce aux propriétés de transmission du diffuseur. Cependant, ce formalisme souffre aussi d’une importante limitation, la structure de bandes des réservoirs étant ignorée.Ici nous avons choisi d’adopter un point de vue différent pour aborder le régime de fort couplage en étudiant un modèle exactement résoluble. Nous analysons donc le modèle de Fano-Anderson décrivant un niveau discret couplé à un continuum. Nous nous intéressons particulièrement à l’influence de la densité d’états des réservoirs. On démontre en effet que, sous certaines conditions, des états liés discrets apparaissent dans les bandes interdites des réservoirs. Ces états jouent un rôle prépondérant sur la dynamique du niveau discret à temps longs : leur contribution dépend de la préparation initiale du système et peut donner lieu à des oscillations permanentes de l’occupation du niveau discret.Nous commençons par expliciter la solution exacte du modèle en nous concentrant particulièrement son comportement à temps longs. Nous analysons ensuite deux cas particuliers. En premier lieu, nous nous intéressons aux propriétés de transport d’une boîte quantique à un niveau couplée à un semi-conducteur présentant une unique bande interdite. Un état lié apparaît dans cette bande lorsque le couplage au réservoir dépasse une valeur critique ce qui affecte fortement les propriétés de transport du système. Nous étudions ensuite le cas de réservoirs décrit par un modèle de liaisons fortes dont la densité d’états ne comporte qu’une bande finie d’énergie. Nous montrons qu’un niveau discret couplé à un tel réservoir se comporte comme un système à plusieurs niveaux, sa densité d’états locale et sa transmission présentant de multiples résonances.

Following the technological advances of the Industrial Revolution, classical thermodynamics was developed in the 19th century in order to understand the conversion of heat into work in newly designed machines. The works of Boltzmann brought another conceptual revolution with statistical mechanics. He demonstrated the microscopical origin of the laws of thermodynamics which actually only describe the macroscopic behaviour of systems in which local thermalization is faster than all other timescales. However, following the growing interest for nanotechnologies, it is now possible to manipulate microscopic systems in which thermalization is slower than the timescales for electron flow. A major technological advance in this field stems from the use of quantum dots, nanoscale devices which confine electrons on such small scales that they spread on discrete energy levels. It is then essential to take into account quantum effects for the study of this type of systems, that is to say to design theoretical tools combining thermodynamics and quantum mechanics.Problems of quantum thermodynamics are often tackled in the framework of the theory of open quantum systems. The general idea of this formalism is to study the dynamics of a “small” quantum system when it is coupled to another much bigger representing the environment. One can then show that the time evolution of the small system can be described by a master equation in the limit where it is weakly coupled to the environment. However, it intuitively seems that the power output of machine would be higher in the context of strong coupling.For problems of electronic transport, the Landauer-Büttiker formalism allows to describe the strong-coupling regime. In this framework, electrons are assumed to solely undergo elastic scattering processes in the central system. All the thermoelectric properties of the machine can then be characterized thanks to the transmission properties of the scatterer. However, this formalism has an important limitation; it ignores the band structure of the reservoirs.Here we have chosen to adopt a different viewpoint to tackle the strong-coupling regime by studying an exactly soluble model. We therefore analyze the Fano-Anderson model describing a discrete level coupled to a continuum. We are particularly interested by the influence of the reservoirs’ band structure. One can indeed show that, under certain conditions, discrete bound states appear in the band gaps of the reservoirs. This state play an important rôle on the dynamics of the discrete at long times: their contribution depends on the initial preparation of the system and gives rise to persistent oscillations of the occupation of the discrete level.We start by deriving the exact solution of the model especially focusing on its long-time behaviour. We then analyze two special cases. First, we study the transport properties of a single-level quantum dot coupled to a semiconductor with single a band gap. A bound state appears in this gap when the coupling to the reservoir exceeds a critical value. We show that this greatly affects the transport properties of the device. We then study the case of reservoirs described by a tight-binding model which density of states consists of a single finite-range energy band. We show that a discrete level coupled to such reservoir behaves like a many-level system as its local density of states and transmission function exhibits multiple resonances., Suivant les progrès technologiques de la révolution industrielle, la thermodynamique classique a été développée au XIXème siècle dans le but de comprendre la conversion de la chaleur en travail intervenant dans les machines thermiques nouvellement élaborées. Les travaux de Boltzmann apportèrent une autre révolution conceptuelle avec la physique statistique. Il démontra l’origine microscopique des lois de la thermodynamique, celles-ci ne décrivant en fait que le comportement macroscopique de systèmes pour lesquels la thermalisation locale est plus rapide que toutes les autres échelles de temps. Cependant, conséquemment à l’intérêt grandissant pour les nanotechnologies, il est aujourd’hui possible de manipuler des systèmes microscopiques pour lesquels la thermalisation est plus lente que les échelles de temps associés aux flux d’électrons. Une avancée technologique majeure dans ce domaine provient de l’utilisation de boîtes quantiques, des dispositifs nanométriques permettant de confiner les électrons sur des distances si petites qu’ils se répartissent sur des niveaux d’énergie discrets. Il est alors évidemment indispensable de prendre en compte les effets quantiques pour l’étude de ce type de systèmes, c’est-à-dire de concevoir des outils théoriques alliant thermodynamique et mécanique quantique.Les problèmes de thermodynamique quantique sont souvent abordés dans le cadre de la théorie des systèmes quantiques ouverts. L’idée générale de ce formalisme est d’étudier la dynamique d’un « petit » système quantique lorsqu’il est couplé à un autre système supposé bien plus « gros » et représentant l’environnement. On démontre alors que l’évolution temporelle du petit système peut être décrite par une équation maîtresse dans la limite où il est faiblement couplé à l’environnement. Cependant, il semble intuitivement qu’une machine pourra délivrer une puissance plus importante dans un contexte de fort couplage-Pour les problèmes de transport électronique, le formalisme de Landauer-Büttiker permet de décrire le régime de fort couplage. Dans ce cadre, les électrons sont supposés ne subir que des processus de diffusion élastique dans le système central. Toutes les propriétés thermoélectriques de la machine peuvent alors être caractérisées grâce aux propriétés de transmission du diffuseur. Cependant, ce formalisme souffre aussi d’une importante limitation, la structure de bandes des réservoirs étant ignorée.Ici nous avons choisi d’adopter un point de vue différent pour aborder le régime de fort couplage en étudiant un modèle exactement résoluble. Nous analysons donc le modèle de Fano-Anderson décrivant un niveau discret couplé à un continuum. Nous nous intéressons particulièrement à l’influence de la densité d’états des réservoirs. On démontre en effet que, sous certaines conditions, des états liés discrets apparaissent dans les bandes interdites des réservoirs. Ces états jouent un rôle prépondérant sur la dynamique du niveau discret à temps longs : leur contribution dépend de la préparation initiale du système et peut donner lieu à des oscillations permanentes de l’occupation du niveau discret.Nous commençons par expliciter la solution exacte du modèle en nous concentrant particulièrement son comportement à temps longs. Nous analysons ensuite deux cas particuliers. En premier lieu, nous nous intéressons aux propriétés de transport d’une boîte quantique à un niveau couplée à un semi-conducteur présentant une unique bande interdite. Un état lié apparaît dans cette bande lorsque le couplage au réservoir dépasse une valeur critique ce qui affecte fortement les propriétés de transport du système. Nous étudions ensuite le cas de réservoirs décrit par un modèle de liaisons fortes dont la densité d’états ne comporte qu’une bande finie d’énergie. Nous montrons qu’un niveau discret couplé à un tel réservoir se comporte comme un système à plusieurs niveaux, sa densité d’états locale et sa transmission présentant de multiples résonances.

Water is omnipresent and plays a decisive role in a myriad of processes.However, it is often found hidden in tiny cells, pores, or channels. Insuch cases, the usual “bulk” features of water are modified by thelimited available space and the interactions of individual moleculeswith the confining material. Elucidating the properties of water in suchconfined states is critical and general understanding can only beachieved through models. While water confined in model hard materialssuch as carbon nanotubes is well documented, we found that there existno general model to study water confined in soft materials, althoughthis has been an active research topic for decades and despite thenumerous models specific to one biomolecule or polymer that have beendeveloped. In this thesis, we present a numerical model of waterconfined in soft self-assembled environments, and we provide anunderstanding of how the interplay between water and the confiningmatrix affects the structure of the assemblies and transport propertiesof water. Our model confining matrix is composed of ionic surfactants.This versatile model is able to self-assemble to a wide variety ofconfining geometries.We focus on the role of interfaces in shaping the nanometer scalestructure, and nanosecond scale transport properties. This work is adeparture from the traditional approach to the problem of transport ofwater confined in soft nanomaterials. We show that the usual hypothesisof diffusive water transport does not hold due to trapping of moleculesat the interface with the confining matrix. Instead, we support apicture where transport is sub-diffusive, and we highlight the role ofthe length-scale of the confinement and of its topological features. Wefind that this rationale explains experimental results for waterconfined in synthetic materials, and that it is compatible with recentadvances in the understanding of biological water., L'eau est partout et joue un rôle déterminant dans une multitude deprocessus. Cependant, on la trouve souvent au sein de minusculescellules, pores, ou canaux. En de tels cas, les proprietés“macroscopiques” de l'eau sont modifiées par les restrictions spatialeset les interactions entre les molécules d'eau et le matériau confinant.Elucider les propriétés de l'eau en confinement est crucial, et unecompréhension générale peut seulement être obtenue à traversl'utilisation de modèles. Alors que l'eau confinée dans des matériauxdurs tels que les nanotubes de carbone est bien documentée, nous n'avonspas trouvé de modèle général pour l'étude de l'eau confinée a desmatériaux mous, et ce en dépit de décénies de recherches sur de nombreuxmodèles spécifiques à une biomolécule ou un polymère en particulier.Dans cette thèse, nous présentons un modèle numérique d'eau confinéedans des géométries molles, générées par auto-assemblage. Nouscomprenons la manière dont les interactions réciproques entre l'eau etla matrice confinante déterminent la structure des assemblages et lespropriétés de transport de l'eau. Nous avons choisi un modèle desurfactant ioniques, matériaux très versatiles qui sont capables des'auto-assembler en diverses géométries confinantes.Nous nous concentrons sur l'effet des interfaces sur la formation de lananostructure et sur les propriétés de transport à l'échelle de lananoseconde. Nous nous distancons de l'approche traditionnelle auproblème du transport de l'eau dans des nanomatériaux. Nous montrons quel'hypothèse habituelle du transport diffusif est invalide car la matriceconfinante piège les molécules d'eau à l'interface. Nous proposons deremplacer cette hypothèse par celle du transport sous-diffusif, et nousmettons en évidence le rôle de l'échelle de taille et des propriétéstopologiques du confinement. Nous montrons que cette approche expliquedes résultats expérimentaux pour léau confinée dans des matériaux desynthèse, et qu'elle est compatible avec les développements récents liésà l'eau biologique.