Flow equations for the one-dimensional Kondo lattice model: static and dynamic ground state properties

The one-dimensional Kondo lattice model is investigated by means of Wegner's flow equation method. The renormalization procedure leads to an effective Hamiltonian which describes a free one-dimensional electron gas and a Heisenberg chain. The localised spins of the effective model are coupled by the well-known RKKY interaction. They are treated within a Schwinger boson mean field theory which permits the calculation of static and dynamic correlation functions. In the regime of small interaction strength static expectation values agree well with the expected Luttinger liquid behaviour. The parameter K_ρ of the Luttinger liquid theory is estimated and compared to recent results from density matrix renormalization group studies.     

Low temperature properties of the Kondo insulator FeSi

In this paper we study the low temperature (T) properties of the Kondo insulator FeSi within the X-boson approach. We show that the ground state of the FeSi is metallic and highly correlated with a large effective mass; the low temperature contributions to the specific heat and the resistivity are of the Fermi-liquid type. The low temperature properties are governed by a reentrant transition into a metallic state, that occurs when the chemical potential crosses the gap and enters the conduction band, generating a metallic ground state. The movement of the chemical potential is due to the strong correlations present in the system. We consider the low temperature regime of the Kondo insulator FeSi, where the hybridization gap is completely open. In this situation we identify the two characteristic temperatures: the coherence temperature T₀ and the Kondo temperature T_KL. In the range T < T₀, we identify a regime characterized by the formation of coherent states and Fermi-liquid behavior of the low temperature properties; in the range T_KL > T > T₀, we identify a regime characterized by an activation energy. Within the X-boson approach we study those low temperature regimes although we do not try to adjust parameters to recover the experimental energy scales.     

Composite spin-triplet superconductivity in an symmetric lattice model

The two-channel Anderson lattice model which has SU (2) ⊗ SU (2) symmetry is of relevance to understanding of the magnetic, quadrupolar and superconducting phases in U_1-xTh_xBe₁₃ or Pr based skutterudite compounds such as PrFe₄P₁₂ or PrOs₄Sb₁₂. Possible unconventional superconducting phases of the model are explored. They are characterized by a composite order parameter comprising of a local magnetic or quadrupolar moment and a triplet conduction electron Cooper-pair. This binding of local degrees of freedom removes the entropy of the non Fermi-liquid normal state. We find superconducting transitions in the intermediate valence regime which are suppressed in the stable moment regime. The gap function is non analytic and odd in frequency: a pseudo-gap develops in the conduction electron density of states which vanishes as |ω| close to ω = 0. In the strong intermediate valent regime, the gap function acquires an additional -dependence. Received 28 February 2002 / Received in final form 18 April 2002 Published online 9 July 2002     

The Bethe lattice spin glass revisited

So far the problem of a spin glass on a Bethe lattice has been solved only at the replica symmetric level, which is wrong in the spin glass phase. Because of some technical difficulties, attempts at deriving a replica symmetry breaking solution have been confined to some perturbative regimes, high connectivity lattices or temperature close to the critical temperature. Using the cavity method, we propose a general non perturbative solution of the Bethe lattice spin glass problem at a level of approximation which is equivalent to a one step replica symmetry breaking solution. The results compare well with numerical simulations. The method can be used for many finite connectivity problems appearing in combinatorial optimization. Received 27 September 2000     

Some comments on Kondo lattices and the Mott transition

The so called exhaustion problem occurs when few electrons have to screen many spins in a metal with magnetic impurities. A singlet Fermi liquid ground state is possible only if all impurities are “isotropized” in such a way as to suppress their entropy. That takes a time and the corresponding energy limits the Fermi liquid range. The present note explores that issue of time and energy scales, and it concludes that is much smaller than the single impurity Kondo temperature. Similarly the relevant energy scale is proportional to the number of electrons. Recent results on the Mott metal insulator transition in infinite dimension are reconsidered in the light of these results: controversies in that respect are shown to reduce to a simple physical question, with no firm answer as to now. Received: 5 May 1998 / Received in final form and Accepted: 29 July 1998     

Low temperature Kondo reduction of quadrupolar and magnetic moments in the YbCuAl series

We present measurements in the YbCu_5-xAl_x series, down to the 50 mK range, using ¹⁷⁰Yb M?ssbauer absorption spectroscopy and magnetisation measurements. In this series, the hybridisation between the Yb 4 f electrons and the conduction electrons is known to decrease as the Al content x increases. We apply the variational solution of the impurity Kondo problem to the interpretation of our data. We show that the Kondo temperature can be derived from the measured 4 f quadrupole moment and, for the magnetically ordered compounds (), we obtain the exchange energy as a function of the Al content. Our findings are in general agreement with Doniach's model describing the onset of magnetic ordering according to the relative values of the Kondo and exchange energy scales. Received 16 April 1998     

Polymers with self-avoiding interaction in random medium: a localization-delocalization transition

In this paper we investigate the problem of a long self-avoiding polymer chain immersed in a random medium. We find that in the limit of a very long chain and when the self-avoiding interaction is weak, the conformation of the chain consists of many “blobs” with connecting segments. The blobs are sections of the molecule curled up in regions of low potential in the case of a Gaussian distributed random potential or in regions of relatively low density of obstacles in the case of randomly distributed hard obstacles. We find that as the strength of the self-avoiding interaction is increased the chain undergoes a delocalization transition in the sense that the appropriate free energy per monomer is no longer negative. The chain is then no longer bound to a particular location in the medium but can easily wander around under the influence of a small perturbation. For a localized chain we estimate quantitatively the expected number of monomers in the “blobs” and in the connecting segments. Received 13 November 2002 Published online 14 March 2003     

Localization of polymers in a finite medium with fixed random obstacles

In this paper we investigate the conformation statistics of a Gaussian chain embedded in a medium of finite size, in the presence of quenched random obstacles. The similarities and differences between the case of random obstacles and the case of a Gaussian random potential are elucidated. The connection with the density of states of electrons in a metal with random repulsive impurities of finite range is discussed. We also interpret the results obtained in some previous numerical simulations. Received 14 August 2001     

Twist free energy in a spin glass

The field theory of a short range spin glass with Gaussian random interactions, is considered near the upper critical dimension six. In the glassy phase, replica symmetry breaking is accompanied with massless Goldstone modes, generated by the breaking of reparametrization invariance of a Parisi type solution. Twisted boundary conditions are thus imposed at two opposite ends of the system in order to study the size dependence of the twist free energy. A loop-expansion is performed to first order around a twisted background. It is found, as expected but it is non trivial, that the theory does renormalize around such backgrounds, as well as for the bulk. However two main differences appear, in comparison with simple ferromagnetic transitions: (i) the loop expansion yields a (negative) anomaly in the size dependence of the free energy, thereby lifting the lower critical dimension to a value greater than two (ii) the free energy is lowered by twisting the boundary conditions. This situation is common in spin glasses, reflecting the non-positivity of mode multiplicity in replica symmetry breaking, but its physical meaning is still unclear. Received 12 April 2002 / Received in final form 30 July 2002 Published online 19 November 2002     

Renormalization group approach to spin glass systems

A renormalization group transformation suitable for spin glass models and, more generally, for disordered models, is presented. The procedure is non-standard in both the nature of the additional interactions and the coarse graining transformation, that is performed on the overlap probability measure. Universality classes are thus naturally defined on a large set of models, going from and Gaussian spin glasses to Ising and fully frustrated models, and others. The proposed analysis is tested numerically on the Edwards-Anderson model in d = 4. Good estimates of the critical index ν and of T _c are obtained, and an RG flow diagram is sketched for the first time. Received 17 November 2000     

Solvable lattice gas models of random hetero-polymers at finite density: I. Statics

We introduce -dimensional lattice gas versions of three common models of random hetero-polymers, in which both the polymer density and the density of the polymer-solvent mixture are finite. These solvable models give valuable insight into the problems related to the (quenched) average over the randomness in statistical mechanical models of proteins, without having to deal with the hard geometrical constraints occurring in finite-dimensional models. Our exact solution, which is specific to the -dimensional case, is compared to the results obtained by a saddle-point analysis and by the grand ensemble approach, both of which can also be applied to models of finite dimension. We find, somewhat surprisingly, that the saddle-point analysis can lead to qualitatively incorrect results. Received 15 June 1999 and Received in final form 14 October 1999     

Role of the d-f Coulomb interaction in intermediate valence and Kondo systems: a numerical renormalization group study

Using the numerical renormalization group method, the dependences on temperature of the magnetic susceptibility χ(T) and specific heat C(T) are obtained for the single-impurity Anderson model with inclusion of d-f the Coulomb interaction. It is shown that the exciton effects caused by this effect (charge fluctuations) can significantly change the behaviour of C(T) in comparison with the standard Anderson model at moderately low temperatures, whereas the behaviour of χ(T) remains nearly universal. The ground-state and temperature-dependent renormalizations of the effective hybridization parameter and f-level position caused by the d-f interaction are calculated, and satisfactory agreement with the Hartree-Fock approximation is derived.     

Relationship between resistivity and specific heat in a canonical non-magnetic heavy fermion alloy system: UPt5-xAu x

A comprehensive study of the relationship between the electronic specific heat coefficient () and the temperature square coefficient (A) of the electrical resistivity for a single, cubic, heavy fermion alloy system, UPt_5-xAu_x is presented. In this alloy system, whose low temperature properties are consistent with the Fermi-liquid behavior, varies by more than a factor of 10 while the corresponding A coefficient changes by a factor larger than 200. A tracks changes in fairly well, but , postulated to have a universal value for heavy fermions, is not constant and varies from about 10^-6 (x = 0, 0.5) to 10^-5 cm (mol K/mJ)² (x > 1.1), thus from a value typical of transition metals to that characteristic of other heavy fermion compounds. We have found a correlation between and magnetic characteristics such as the paramagnetic Curie-Weiss temperature and the low temperature magnetic susceptibility divided by . Received 29 January 1999     

The low temperature spectral weight transfer problem in Kondo insulators

In this paper we address the problem of the spectral weight transfer in Kondo insulators (KI). We employ the X-boson approach for the periodic Anderson model, in the U →∞ limit. We calculate the two energy gaps of the system analytically: the indirect gap, Δ_ind = E_g ≃ E_mir, present in the density of states, and the direct one Δ_dir, associated with the minimum energy necessary to produce inter-band transitions. We find that the optical behavior of the system is governed by two energy scales: one of low frequency, characterized by E_g ≃ E_mir, in the mid-infrared region (MIR), which is a reminiscent of the heavy fermion E_mir peak, that appears in Kondo insulators as a broad maximum in the MIR region and that controls the low temperature transport properties, the gap opening in optical conductivity and the formation of the Drude peak, at ω = 0, in the intermediate temperature range. The other energy scale appears at high frequencies, and is characterized by the direct gap Δ_dir. According to our results, this peak controls the anomalous redistribution of spectral weight in the optical conductivity. We apply the theory in order to study the Kondo insulator FeSi, and we calculate the optical conductivity of the system and the spectral weight transfer in the optical conductivity.     

Generic replica symmetric field-theory for short range Ising spin glasses

Symmetry considerations and a direct, Hubbard-Stratonovich type, derivation are used to construct a replica field-theory relevant to the study of the spin glass transition of short range models in a magnetic field. A mean-field treatment reveals that two different types of transitions exist, whenever the replica number n is kept larger than zero. The Sherrington-Kirkpatrick critical point in zero magnetic field between the paramagnet and replica magnet (a replica symmetric phase with a nonzero spin glass order parameter) separates from the de Almeida-Thouless line, along which replica symmetry breaking occurs. We argue that for studying the de Almeida-Thouless transition around the upper critical dimension d = 6, it is necessary to use the generic cubic model with all the three bare masses and eight cubic couplings. The critical role n may play is also emphasized. To make perturbative calculations feasible, a new representation of the cubic interaction is introduced. To illustrate the method, we compute the masses in one-loop order. Some technical details and a list of vertex rules are presented to help future renormalisation-group calculations. Received 9 October 2001     

Topological jamming and the glass transition in a frustrated system

The relationship between extended structures, glassy dynamics and an underlying critical point is examined in the context of a lattice model of fluctuating lines. Monte Carlo simulations are used to construct an effective, coarse-grained dynamics for the “order parameter” near the critical point. Analysis of the effective dynamics reveals that the critical point is associated with diverging barriers leading to the observed Vogel-Fulcher divergence of the relaxation times. A direct connection is established between the presence of extended structures and the activated dynamics. Received 15 March 2002     

Spin glasses on Bethe lattices for large coordination number

We study spin glasses on random lattices with finite connectivity. In the infinite connectivity limit they reduce to the Sherrington Kirkpatrick model. In this paper we investigate the expansion around the high connectivity limit. Within the replica symmetry breaking scheme at two steps, we compute the free energy at the first order in the expansion in inverse powers of the average connectivity (z), both for the fixed connectivity and for the fluctuating connectivity random lattices. It is well known that the coefficient of the 1/z correction for the free energy is divergent at low temperatures if computed in the one step approximation. We find that this annoying divergence becomes much smaller if computed in the framework of the more accurate two steps breaking. Comparing the temperature dependance of the coefficients of this divergence in the replica symmetric, one step and two steps replica symmetry breaking, we conclude that this divergence is an artefact due to the use of a finite number of steps of replica symmetry breaking. The 1/z expansion is well defined also in the zero temperature limit. Received 15 July 2002 Published online 31 December 2002     

Advanced memory effects in the aging of a polymer glass

A new kind of memory effect on low frequency dielectric measurements on plexiglass (PMMA) is described. These measurements show that cooling and heating the sample at constant rate give an hysteretic dependence on temperature of the dielectric constant ε. A temporary stop of cooling produces a downward relaxation of ε. Two main features are observed (i) when cooling is resumed ε goes back to the values obtained without the cooling stop (i.e. the low temperature state is independent of the cooling history) (ii) upon reheating ε keeps the memory of all the cooling stops (Advanced memory). The dependence of this effect on frequency and on the cooling rate is analyzed. The memory deletion is studied too. Finally the results are compared with those of similar experiments done in spin glasses and with the famous experiments of Kovacs. Received 24 September 2001 and Received in final form 20 November 2001     

Microphase separation at two length scales

The possibility of microphase separation at two different length scales in monodisperse AB block copolymer melts consisting of a homopolymer A block and either a linear alternating AB copolymer block (poly(A)_m-block-poly(B-alt -A)_n) or an AB comb copolymer block poly(A)_m-block-poly(A-graft-B)_n, is investigated. An analysis of the structure factor reveals that in the parameter space of n and m three different cases can be distinguished: I) The structure factor has only one minimum corresponding to the short length scale (i.e. the characteristic length of the repeating unit of the alternating or comb block). II) The structure factor has only one minimum corresponding to the long length scale (the characteristic length of the blocks). III) Two minima are present leading to a competition between microphase separation at the short and the long length scale. Depending on the choice of n and m, one of these three possibilities will occur. Received 25 August 2000