Pages scientifiques 2012

In chiral magnets, calculations performed in the large-pitch limit show that spirals with a well-defined twist direction (helical order) should have a lower energy in magnets, and that blue phases cannot appear.39 Despite this result, a blue phase has been suspected to occur in the itinerant chiral magnet MnSi at high hydrostatic pressure, where a so-called partial magnetic order has been observed [3].

In collaboration with A. Hamman, T. Wolf, and H. v. Löhneysen (Karlsruhe Institute of Technology) and D. Reznik (University of Colorado), we have performed calculations starting from magnetic moments of the same fixed magnitude, with random orientations, and placed on simple cubic or B20 lattices. The orientation of each moment was optimized sequentially in random order until the total energy of the system stabilized.

A typical spin arrangement obtained by this method is shown in Figure 1 (top). The topology is the same as for blue phases of chiral liquid crystals, i.e., with moments twisting away from cylindrical axes. We tested our model via experiments on the 4F cold triple-axis neutron spectrometer and on the Papyrus small angle neutron scattering (SANS) diffractometer. The magnetic correlation length is expected to be reduced by Fe-doping similarly to thermal fluctuations.

According to our model, Fe doping should allow the blue phase to persist to lower temperatures. Experiments show that this is indeed the case: T_c decreases with Fe doping [4] and the SANS spectrum of Mn_0.8Fe_0.15Si at 1.5 K is a ring characteristic of the blue phase (Figure 1, bottom). The same result was obtained for Fe_1-xCo_xSi [5], pointing to a universality of this phenomenon. 

Our results provide a blueprint for understanding cubic chiral magnets, including MnSi, by showing that the frustrated nature of the magnetic interactions naturally and generically induces complex spin arrangements, which had so far been found only in liquid-crystal blue phases. Controlling special properties of this blue phase with impurities in thin films has a potential for new applications, particularly in electronics and magnetic memory devices [6].

Molecular magnetism is a relatively new field of research, which has attracted growing interest among physicists since the discovery, fifteen years ago, of the first “single molecule magnet” (SMM), Mn₁₂-acetate, which behaves as a magnet at the molecular scale below a blocking temperature of 3 K. The present challenges in this field consist in understanding how to control magnetic anisotropy in order to obtain SMM behavior at higher temperatures, and in designing new multifunctional materials possessing other functional properties in addition to magnetism, e.g. magnetic photoswitchable compounds. Optical magnetic switching in the solid state raises questions of fundamental interest as to the mechanisms governing phase separation into magnetic domains or continuous structural changes during photoexcitation.

Combined analysis of charge and spin densities in molecular magnets

The goal of the CEDA project, supported by the ANR (2008-2010), on the Convergence of Electron spin, charge and momentum Densities Analysis, is to combine data obtained by different techniques on the electron density in order to refine a unique model, thereby allowing the electronic structure of magnetic molecular compounds to be more accurately described. This work involves theoreticians (SPMS, ECP), x-ray (CRM2, Nancy) and neutron (LLB) diffraction specialists, and chemists (LMI, Lyon). The first step devoted to the joint refinement of charge and spin densities from x-ray and polarized neutron diffraction data sets has been completed. In this approach, the multipole mode l[1] is used to describe the densities of spin up and spin down electrons, the sum (or difference) of which is the charge (spin) density. The software called Mollynx was developed by splitting the electron density model into two spin components. Experimental data were collected by x‑ray diffraction at CRM2, and polarized neutron diffraction on 5C1 at LLB on two cubane-like [M₄O₄] complexes (M = Cu^II, Ni^II), synthesized in Lyon (LMI) (Figure 1, top). The joint refinement of these data is expected to ensure a better characterization of the metal-ligand bonds and a more complete understanding of their role in the intramolecular magnetic interactions and the magnetic anisotropy. The first analysis of the PND data alone [2] shows that the experimental spin distribution in the Cu₄ complex is mainly located in the basal planes of the Cu^II ions and visualizes the magnetic interaction pathways via the bridging oxygen atoms (Figure 1, bottom).

Studies conducted over the past few decades came to the conclusion that ferroelectricity and magnetism tend to be mutually exclusive, and interact only weakly with each other when they do coexist. The recent discovery of multiferroic materials, in which ferroelectricity and magnetism are intimately coupled, has completely changed these established views, opened routes to new technological applications, and has forced physicists to reconsider some long-accepted ideas [1].

Coupling between magnetic and electric orders makes it possible to design promising devices, such as high-speed memories with magnetically and electrically addressable states, magnetically tunable tunnel-junctions, sensors, filters, transducers, etc. For instance, the control of magnetization by an applied electric field via the magnetoelectric coupling offers an opportunity to combine the respective advantages of FeRAMs (ferroelectric random access memories) and MRAMs (magnetic random access memories) in the form of non-volatile magnetic storage bits, which can be switched by an electrical field and thus combine the merits of MRAMs in terms of access time and endurance with a low writing energy [2].

Meanwhile, a major effort is currently underway to revisit the ferroelectric-ferromagnetic dichotomy and to understand the multiferroic properties on fundamental grounds. The multiferroic compounds are scarce, as symmetry considerations impose severe constraints. Indeed, their space group should be non-centrosymmetric, to accommodate ferroelectricity, but also compatible with a magnetic ground state. Last but not least, all multiferroic materials found to date exhibit a certain amount of magnetic frustration. This, together with a strong spin- lattice coupling, seems to be a basic physical ingredient for multiferroicity. For example, the orthorhombic RMnO₃ compounds, where R is a rare-earth element (R = Eu, Gd, Tb, Dy), are archetypes of multiferroic materials. Magnetic frustration arises from competing magnetic exchange interactions, yielding an incommensurable spiral magnetique structure. As proposed in the “spin-current” model, the spin-lattice coupling is a result of the Dzyaloshinskii-Moriya interaction: minimizing the corresponding term in the Hamiltonian of the system results in a relaxation of atomic positions and, in turn, in an electric polarization [3].

In chemically ordered compounds with short-range magnetic interactions, geometrical frustration appears when all interactions cannot be satisfied simultaneously due to the lattice geometry. A well-known example is a triangle of antiferromagnetically coupled spins. This frustration results in a strong degeneracy of the ground state, since many configurations have the same energy. It now appears as a powerful ingredient for designing materials or tuning physical properties, since frustrated magnetic states may be easily switched by small changes in the energy balance. Pyrochlores R₂Ti₂O₇ (where R³⁺ is a rare earth ion) are model systems showing  exotic short-range magnetic orders called “spin liquids” and “spin ices”, with unconventional excitations akin to magnetic monopoles, and a huge sensitivity to perturbations. Various aspects have been studied in collaborations with P. Bonville and A. Forget (IRAMIS/SPEC, Saclay), G. Dhalenne and C. Decorse (ICMMO, Orsay), and H. Mutka (ILL, Grenoble).

Local susceptibility in R₂Ti₂O₇ pyrochlores (R = Tb, Ho, Yb, Er)

Rare earth titanate pyrochlores of Fd-3m space group, where the R³⁺ ions occupy the summits of corner-sharing tetrahedra and the Ti⁴⁺ ions are nonmagnetic, show very different ground states, such as spin-liquid (Tb) and spin-ice (Ho) states, short range orders, planar antiferromagnetism (Er), and 2D ferromagnetism (Yb). Their physics is governed by the exchange and dipolar interactions between the R³⁺ ions, together with the crystal-field anisotropy along the <111> axes. The study of the paramagnetic state under applied fields points to precursor effects of these ground states. Single-crystal polarized neutron diffraction gives access to the local susceptibility tensor in the Gukasov-Brown model [1], which reduces to 2 components χ_// and χ_⊥ with respect to the <111> local axis, owing to the high symmetry the R³⁺ site. This allows one to describe the complex field-induced structures using only two parameters. This tensor cannot be derived from the bulk magnetization due to the equivalent <111> axes. Measurements performed in four typical cases where the R-R interactions may be F or AF, and the crystal-field anisotropy either Ising- or XY-type, showed that the local susceptibility is highly anisotropic (Figure 1). This is the first determination of the local susceptibility in the Ho₂Ti₂O₇ spin ice with (F, Ising) behavior. Comparison with a calculation of the non-interacting susceptibility deduced from the crystal-field scheme (measured independently by inelastic neutron scattering) yields the tensor of exchange interactions, which is also found to be anisotropic [2]. As a surprising consequence, in Yb₂Ti₂O₇ with (F, XY) behavior, crystal-field and exchange anisotropies nearly compensate, yielding a Curie susceptibility and ground state fluctuations in an original “exchange spin-ice” ground state [3].

4f electrons are known to be more localized than d electrons and subject to strong spin-orbit coupling. Instabilities of otherwise magnetic 4f states occur mainly in metallic compounds as a result of hybrid­ization with conduction band states, as described in the well-known Anderson model. This approach basically describes a competition between exchange interactions and Kondo-type magnetic fluctua­tions, producing renormalized Fermi-liquid quasiparticle states, heavy effective masses and, in some cases, “heavy-fermion superconductivity”. Some materials do not fit into this framework, however, because other degrees of freedom (such as the multipole moments) are involved, or because the renormalization leads to a semiconducting ground state (“Kondo insulators”).

Unconventional ordered states: competing dipole and multipole interactions

Whereas neutron diffraction techniques remain unchallenged in solving intricate magnetic structures, systems in which spin and orbital degrees of freedom are intimately coupled require more diversified approaches. This type of situation occurs in f-electrons exhibiting various types of “multipole orders” at low temperature, such as skutterudites or hexaborides. The best-known example is CeB₆, which shows a mixed order of Ce quadrupole and octupole moments in a narrow temperature range (“phase II”: 2.3 < T < 3.2 K in zero field), prior to developing a long-range order of magnetic dipole moments. A field-induced (H || [001]) magnetic contribution ascribed to the ordered T_xyz octupole moments was observed for the first time in neutron diffraction experiments performed at LLB (collaboration with Tohoku University, Sendai and JAEA, Tokai). Substituting Ce by other lanthanide elements with different ground state multiplets (^2S+1L_J = ¹S₀, ²F_5/2, ³H₄, ⁴I_9/2 for La, Ce, Pr, and Nd, respectively), one can selectively tune the different multipole moments and study the effect on the ordering properties of the material. Studies of the (Ce,Pr)B₆, (Ce,Nd)B₆, and (La,Pr)B₆ series, carried out in a collaboration between the LLB (neutron diffraction), Hiroshima University (bulk properties), and Spring-8 (synchrotron x-ray diffraction), led to the characterization of several new ordered phases in the tem­pera­ture–magnetic field phase diagrams for different

compositions, whose relative stability ranges can be traced back to the interplay between multipole interactions of different orders and symmetries [1],[2].

Recently a new material, CeRu₂Al₁₀, became the focus of considerable attention: despite being a seemingly innocuous intermetallic Ce compound, it appears to be on the verge of a semiconducting state, which can be stabilized under a moderate pressure of ~ 1 GPa, but subsequently collapses above 3 GPa. This dramatic effect is directly connected to an elusive phase transition occurring at T₀ = 27 K, whose order parameter remains controversial. Superstructure satellites were observed in a powder diffraction experiment on G4-1 [3], and proven to be magnetic (wave vector k_AF = [1,0,0]) using neutron polarization analysis on 4F1. However, the simplest antiferromagnetic structure suggested by the neutron measurements fails to explain the recent results from Al NMR. Furthermore, INS data performed on powder (Figure 1.2) indicate that the main peak developing at about 8 meV below T₀ has a mixed (magnetic + nuclear) character. This suggests that the transition does not reduce to a conventional antiferromagnetic ordering, and more detailed investigations have been undertaken to clarify this important question.

Investigation of "nanosystems" and "heterosystems" at LLB involve studies of:

• Polymer based nanocomposites
• Metallic based nanocomposites
• Magnetic nano-objects (nanoparticles - molecular magnets - thin films)
• Confined liquids and molecules
• Microporous, zeolithes, MOFs and hybrids (guest@host)
• Structure and nanostructure of compounds of pharmaceutical and biological interest

Most benefits from experimental studies on one of the 20 world-class instruments, installed on the neutron beams from the reactor Orphée.

The transformation of solar energy into chemical energy stored in the form of hydrogen, through photoelectrochemical water splitting is a promising method with the important advantage of being environment friendly and free from carbon dioxide emission. Several metal oxide semiconductors are able to split water into hydrogen and oxygen but the efficiencies are still low. Several strategies were proposed to improve the photoelectrochemical properties like doping or nanostucturing, but the origin of the improvement is not well understood yet. In the laboratory, we study the photo-electrochemical activity of thin epitaxial oxide films in order to understand the influence of different parameters on the photoanode efficiency.

Matière active :

• Modélisation d’expériences in vivo

• Classes d’universalité

• Interactions non décomposables en paires

• Equations continues

• Nouveaux modèles

• Renormalisation non perturbative

• Criticalité auto-organisée

• Voisinages non métriques

Matière granulaire :

• Interaction particules-fluide en rotation.
• Méca. Stat. des systèmes granulaires en tambour et écoulement cisaillé

Measurements of very weak field (down to the femtotesla range) have been mainly addressed by low-Tc SQUIDS. We have proposed in 2004 the principle of a new sensor, based on spin-electronics, to offer extremely good sensitivity for measurements of magnetic fields over a wide range of frequency (dc to hundreds of MHz).

These sensors, called mixed sensors, are based on the association of a sensitive field sensor (Giant Magnetoresistance or Tunnel Magnetoresistance) and of an efficient flux-to-field transformer. The transformer is a cm size superconducting loop containg a micron size constriction. When an external field is applied perpendicular to the loop, a supercurrent is generated, to prevent the entrance of the field in the loop. This supercurrent reaches a very high density when passing through the constriction. The stray field lines due to this current can be locally hundreds to thousands times larger than the applied field. If one locates a GMR or TMR element on top or below the constriction, it will collect this locally enhanced field lines and a change in the resistance of the GMR (TMR) element can be measured (see Figure 1).

This device is fully made out of thin film technology, and can be either using a low-Tc (typically Nb) loop or a High-Tc (YBaCuO) loop. In the first case, liquid helium cooling is required whereas in the second one only nitrogen cooling is necessary.

L’immense complexité du vivant a pu faire croire que les méthodes de la physique ne pouvaient être utiles pour le comprendre. Avec le recul, il est apparu que cette difficulté était due au fait que les systèmes biologiques évoluent dans un univers impliquant un très grand nombre d’états d’énergies voisines et séparés par des barrières peu élevées : une faible contrainte imposée peut entrainer un déplacement considérables des équilibres.

Un problème typique est le repliement des protéines. Le paradoxe de Levinthal résulte simplement du dénombrement des configurations géométriques possibles d’une protéine moyenne, de l’ordre de 10 ^ 143 ! Il est donc impossible qu’elle explore toutes les configurations pour trouver la plus stable. Il est également possible qu’il existe plusieurs états métastables. In vivo, une mauvaise configuration de certaines protéines est considérée comme la cause de pathologies graves (ESB, Creutzfeld-Jacobs, Alzheimer).

Nous avons étudié une facette de ce problème en considérant un système "modèle" i.e. considérablement simplifié : un peptide composé de la répétition d’un seul acide aminé, l’acide polyglutamique. Cet homopeptide est connu pour sa propension à constituer une hélice à basse température et une chaîne aléatoire à température élevée. L’état hélicoïdal peut être facilement quantifié expérimentalement par la technique de dichroïsme circulaire. Cette transition a été modélisée dès 1959 par une théorie mettant en jeu deux paramètres énergétiques, l’énergie d’une boucle élémentaire et l’énergie de couplage entre boucles voisines ainsi que l’entropie des configurations. Ce modèle est formellement analogue au modèle d’Ising à une dimension et donc soluble. Des mesures récentes précises se sont révélées incompatibles avec ce modèle. Nous avons pu montrer que par contre les données étaient bien décrites par la considération d’un  troisième état possible et l’absence de coopérativité.