arXiv subjects

Manfred Salmhofer

Publications and source records attributed to Manfred Salmhofer.

54 records · Page 3Linked to original sources

arXiv2006

Decay of the Fourier transform of surfaces with vanishing curvature

We prove $L^p$-bounds on the Fourier transform of measures $μ$ supported on two dimensional surfaces. Our method allows to consider surfaces whose Gauss curvature vanishes on a one-dimensional submanifold. Under a certain non-degeneracy condition, we prove that $\whμ\in L^{4+β}$, $β>0$, and we give a logarithmically divergent bound on the $L^4$-norm. We use this latter bound to estimate almost singular integrals involving the dispersion relation, $e(p)= \sum_1^3 [1-\cos p_j]$, of the discrete Laplace operator on the cubic lattice. We briefly explain our motivation for this bound originating in the theory of random Schrödinger operators.

math-ph↗

arXiv2006

Towards the quantum Brownian motion

We consider random Schrödinger equations on $\bR^d$ or $\bZ^d$ for $d\ge 3$ with uncorrelated, identically distributed random potential. Denote by $λ$ the coupling constant and $ψ_t$ the solution with initial data $ψ_0$. Suppose that the space and time variables scale as $x\sim λ^{-2 -κ/2}, t \sim λ^{-2 -κ}$ with $0< κ\leq κ_0$, where $κ_0$ is a sufficiently small universal constant. We prove that the expectation value of the Wigner distribution of $ψ_t$, $\bE W_{ψ_{t}} (x, v)$, converges weakly to a solution of a heat equation in the space variable $x$ for arbitrary $L^2$ initial data in the weak coupling limit $λ\to 0$. The diffusion coefficient is uniquely determined by the kinetic energy associated to the momentum $v$.

math-ph↗

arXiv2005

Eliashberg equations derived from the functional renormalization group

We describe how the fermionic functional renormalization group (fRG) flow of a Cooper+forward scattering problem can be continued into the superconducting state. This allows us to reproduce from the fRG flow the fundamental equations of the Eliashberg theory for superconductivity at all temperatures including the symmetry-broken phase. We discuss possible extensions of this approach like the inclusion of vertex corrections.

cond-mat.supr-con↗

arXiv2004

Renormalization group flows into phases with broken symmetry

We describe a way to continue the fermionic renormalization group flow into phases with broken global symmetry. The method does not require a Hubbard-Stratonovich decoupling of the interaction. Instead an infinitesimally small symmetry-breaking component is inserted in the initial action, as an initial condition for the flow of the selfenergy. Its flow is driven by the interaction and at low scales it saturates at a nonzero value if there is a tendency for spontaneous symmetry breaking in the corresponding channel. For the reduced BCS model we show how a small initial gap amplitude flows to the value given by the exact solution of the model. We also discuss the emergence of the Goldstone boson in this approach.

cond-mat.str-el↗

arXiv2003

Ferromagnetism and triplet superconductivity in the two-dimensional Hubbard model

We review magnetic and superconducting instabilities in the t-t' Hubbard model on the two-dimensional square lattice as obtained with functional one-loop renormalization group techniques. Special emphasis is put on ferromagnetic and triplet superconducting tendencies that could be relevant to the triplet superconductor Sr2RuO4.

cond-mat.str-el↗

arXiv2003

On the quantum Boltzmann equation

We give a nonrigorous derivation of the nonlinear Boltzmann equation from the Schrodinger evolution of interacting fermions. The argument is based mainly on the assumption that a quasifree initial state satisfies a property called restricted quasifreeness in the weak coupling limit at any later time. By definition, a state is called restricted quasifree if the four-point and the eight-point functions of the state factorize in the same manner as in a quasifree state.

math-ph↗

arXiv2002

Flow of the quasiparticle weight in the N-patch renormalization group scheme

Using the N-patch renormalization group method we investigate the flow of the quasiparticle weight in one-dimensional, weakly two-dimensional and fully two-dimensional Hubbard models. In one dimension we reproduce the Luttinger exponent that describes the disappearance of the quasiparticle peak towards lower scales. Further we analyze effects of the band structure, interchain hopping and the flow of the interactions in quasi-one-dimensional models. For the two-dimensional case we study how the suppression of the quasiparticle weight affects the flow to strong coupling. We find that the flow to strong coupling remains essentially unchanged. This strengthens the evidence for d-wave superconductivity in the weakly repulsive Hubbard model.

cond-mat.supr-con↗

arXiv2001

Magnetic and superconducting instabilities of the Hubbard model at the van Hove filling

We use a novel temperature-flow renormalization group technique to analyze magnetic and superconducting instabilities in the two-dimensional t-t' Hubbard model for particle densities close to the van Hove filling as a function of the next-nearest neighbor hopping t'. In the one-loop flow at the van Hove filling, the characteristic temperature for the flow to strong coupling is suppressed drastically around t'_c approx. -0.33t, suggesting a quantum critical point between d-wave pairing at moderate t'>t'_c and ferromagnetism for t'<t'_c. Upon increasing the particle density in the latter regime the leading instability occurs in the triplet pairing channel.

cond-mat.str-el↗

arXiv2001

The temperature-flow renormalization group and the competition between superconductivity and ferromagnetism

We derive a differential equation for the one-particle-irreducible vertex functions of interacting fermions as a function of the temperature. Formally, these equations correspond to a Wilsonian renormalization group scheme which uses the temperature as an explicit scale parameter. Our novel method allows us to analyze the competition between superconducting and various magnetic Fermi surface instabilities in the one-loop approximation. In particular this includes ferromagnetic fluctuations, which are difficult to treat on an equal footing in conventional Wilsonian momentum space techniques. Applying the scheme to the two-dimensional t-t' Hubbard model we investigate the RG flow of the interactions at the van Hove filling with varying next-nearest neighbor hopping t'. Starting at t'=0 we describe the evolution of the flow to strong coupling from an antiferromagnetic nesting regime over a d-wave regime at moderate t' to a ferromagnetic region at larger absolute values of t'. Upon increasing the particle density in the latter regime the ferromagnetic tendencies are cut off and the leading instability occurs in the triplet superconducting pairing channel.

cond-mat.str-el↗

arXiv1999

Positivity and Convergence in Fermionic Quantum Field Theory

We derive norm bounds that imply the convergence of perturbation theory in fermionic quantum field theory if the propagator is summable and has a finite Gram constant. These bounds are sufficient for an application in renormalization group studies. Our proof is conceptually simple and technically elementary; it clarifies how the applicability of Gram bounds with uniform constants is related to positivity properties of matrices associated to the procedure of taking connected parts of Gaussian convolutions. This positivity is preserved in the decouplings that also preserve stability in the case of two-body interactions.

math-ph↗

arXiv1999

Mean field analysis of a model for superconductivity in an antiferromagnetic background

We study a lattice fermion model for superconductivity in the presence of an antiferromagnetic background, described as a fixed external staggered magnetic field. We discuss the possibility that our model provides an effective description of coexistence of antiferromagnetic correlations and superconductivity, and its possible application to high temperature superconductivity. We also argue that, under certain conditions, this model describes a variant of the periodic Anderson model for heavy fermions. Using a path integral formulation we construct mean field equations, which we study in some detail. We evaluate the superconducting critical temperature and show that it is strongly enhanced by antiferromagnetic order. We also evaluate the superconducting gap, the superconducting density of states, and the tunneling conductivity, and show that the most stable channel usually has a $d_{x^2-y^2}$-wave gap.

cond-mat.supr-con↗

arXiv1998

Truncation of a 2-dimensional Fermi surface due to quasiparticle gap formation at the saddle points

We study a two-dimensional Fermi liquid with a Fermi surface containing the saddle points $(π,0)$ and $(0,π)$. Including Cooper and Peierls channel contributions leads to a one-loop renormalization group flow to strong coupling for short range repulsive interactions. In a certain parameter range the characteristics of the fixed point, opening of a spin and charge gap and dominant pairing correlations are similar to those of a 2-leg ladder at half-filling. An increase of the electron density we argue leads to a truncation of the Fermi surface with only 4 disconnected arcs remaining.

cond-mat.str-el↗

arXiv1997

Continuous renormalization for fermions and Fermi liquid theory

I derive a Wick ordered continuous renormalization group equation for fermion systems and show that a determinant bound applies directly to this equation. This removes factorials in the recursive equation for the Green functions, and thus improves the combinatorial behaviour. The form of the equation is also ideal for the investigation of many-fermion systems, where the propagator is singular on a surface. For these systems, I define a criterion for Fermi liquid behaviour which applies at positive temperatures. As a first step towards establishing such behaviour in d ge 2, I prove basic regularity properties of the interacting Fermi surface to all orders in a skeleton expansion. The proof is a considerable simplification of previous ones.

cond-mat↗

arXiv1997

Regularity of the moving Fermi surface: the exact selfenergy

Regularity of the deformation of the Fermi surface under short-range interactions is established to all orders in perturbation theory. The proofs are based on a new classification of all graphs that are not doubly overlapping. They turn out to be generalized RPA graphs. This provides a simple extension to all orders of the regularity theorem of the Fermi surface movement proven in [FST2]. Models in which the Fermi surface is not symmetric under the reflection p to -p are included.

cond-mat.str-el↗

arXiv1997

Perturbation Theory around Non-Nested Fermi Surfaces II. Regularity of the Moving Fermi Surface: RPA contributions

Regularity of the deformation of the Fermi surface under short-range interactions is established for all contributions to the RPA self-energy (it is proven in an accompanying paper that the RPA graphs are the least regular contributions to the self-energy). Roughly speaking, the graphs contributing to the RPA self-energy are those constructed by contracting two external legs of a four-legged graph that consists of a string of bubbles. This regularity is a necessary ingredient in the proof that renormalization does not change the model. It turns out that the self--energy is more regular when derivatives are taken tangentially to the Fermi surface than when they are taken normal to the Fermi surface. The proofs require a very detailed analysis of the singularities that occur at those momenta p where the Fermi surface S is tangent to S+p. Models in which S is not symmetric under the reflection p to -p are included.

cond-mat.str-el↗

arXiv1996

Improved Power Counting and Fermi Surface Renormalization

The naive perturbation expansion for many-fermion systems is infrared divergent. One can remove these divergences by introducing counterterms. To do this without changing the model, one has to solve an inversion equation. We call this procedure Fermi surface renormalization (FSR). Whether or not FSR is possible depends on the regularity properties of the fermion self--energy. When the Fermi surface is nonspherical, this regularity problem is rather nontrivial. Using improved power counting at all orders in perturbation theory, we have shown sufficient differentiability to solve the FSR equation for a class of models with a non-nested, non-spherical Fermi surface. I will first motivate the problem and give a definition of FSR, and then describe the combination of geometric and graphical facts that lead to the improved power counting bounds. These bounds also apply to the four--point function. They imply that only ladder diagrams can give singular contributions to the four--point function.

cond-mat↗

arXiv1995

Perturbation Theory around Non-Nested Fermi Surfaces I. Keeping the Fermi Surface Fixed

The perturbation expansion for a general class of many-fermion systems with a non-nested, non-spherical Fermi surface is renormalized to all orders. In the limit as the infrared cutoff is removed, the counterterms converge to a finite limit which is differentiable in the band structure. The map from the renormalized to the bare band structure is shown to be locally injective. A new classification of graphs as overlapping or non-overlapping is given, and improved power counting bounds are derived from it. They imply that the only subgraphs that can generate $r$ factorials in the $r^{\rm th}$ order of the renormalized perturbation series are indeed the ladder graphs and thus give a precise sense to the statement that `ladders are the most divergent diagrams'. Our results apply directly to the Hubbard model at any filling except for half-filling. The half-filled Hubbard model is treated in another place.

cond-mat↗

arXiv1993

Consistent axial--like gauge fixing on hypertori

We analyze the Gribov problem for $\SU(N)$ and $\U(N)$ Yang-Mills fields on $d$-dimensional tori, $d=2,3,\ldots$. We give an improved version of the axial gauge condition and find an infinite, discrete group $\cG'=\Z^{dr}\rtimes({\Z_2}^{N-1}\rtimes\Z_2)$, where $r=N-1$ for $\GG=\SU(N)$ and $r=N$ for $\GG=\U(N)$, containing all gauge transformations compatible with that condition. This residual gauge group $\cG'$ provides (generically) all Gribov copies and allows to explicitly determine the space of gauge orbits which is an orbifold. Our results apply to Yang-Mills gauge theories either in the Lagrangian approach on $d$-dimensional space-time $T^d$, or in the Hamiltonian approach on $(d+1)$-dimensional space-time $T^d\times \R$. Using the latter, we argue that our results imply a non-trivial structure of all physical states in any Yang-Mills theory, especially if also matter fields are present.

hep-th↗