We show exactly with an SU(N) interacting model that even if the ambiguity associated with the placement of the chemical potential, μ, for a T=0 gapped system is removed by using the unique value μ(T→0), Luttinger's sum rule is violated even if the ground-state degeneracy is lifted by an infinitesimal hopping. The failure stems from the nonexistence of the Luttinger-Ward functional for a system in which the self-energy diverges. Since it is the existence of the Luttinger-Ward functional that is the basis for Luttinger's theorem which relates the charge density to sign changes of the single-particle Green function, no such theorem exists. Experimental data on the cuprates are presented which show a systematic deviation from the Luttinger count, implying a breakdown of the electron quasiparticle picture in strongly correlated electron matter.
We show exactly with an SU(N) interacting model that even if the ambiguity associated with the placement of the chemical potential, μ, for a T=0 gapped system is removed by using the unique value μ(T→0), Luttinger's sum rule is violated even if the ground-state degeneracy is lifted by an infinitesimal hopping. The failure stems from the nonexistence of the Luttinger-Ward functional for a system in which the self-energy diverges. Since it is the existence of the Luttinger-Ward functional that is the basis for Luttinger's theorem which relates the charge density to sign changes of the single-particle Green function, no such theorem exists. Experimental data on the cuprates are presented which show a systematic deviation from the Luttinger count, implying a breakdown of the electron quasiparticle picture in strongly correlated electron matter.
In this Comment, we first present general arguments showing that the absence of the Luttinger theorem (LT) for the SU(N) model of Dave, Phillips and Kane (DPK) is rooted in the non-uniqueness of the ground state (GS) of this model for 0 < n < N, where n denotes the number of particles in the GS; the validity of the Luttinger theorem for n = N/2, when N even, is accidental, a consequence of particle-hole symmetry. Consequently, by supplementing the Hamiltonian of the SU(N) model with a perturbation Hamiltonian that removes the GS degeneracy, the LT is to apply also for the SU(N) model in the limit of the coupling constant, λ, of this perturbation approaching zero, where the limit λ--> 0 is clearly to be taken subsequent to taking the zero-temperature limit of the thermal single-particle Green function in the expression for the Luttinger number N_L. We explicitly establish the validity of this statement for the case of N=4. The details of the relevant calculations being distinctly transparent, one can readily convince oneself that our observation is valid for arbitrary N. It follows that the issues raised by DPK, such as non-existence of the Luttinger-Ward functional and "breakdown of the elemental particle picture in strongly correlated electron matter", are all inessential to the observed failure of the LT. As regards the singularity of the self-energy Σ(ω) on the real ω-axis, observed by DPK, we demonstrate that this also is a direct consequence of the non-uniqueness of the GS of the SU(N) model for 0 < n < N. In the light of the above observations, we are in a position to state that to this date no case has come to light indicative of the failure of the LT under the conditions for which it has been deduced. [Abridged Abstract]
We analyze the original proof by Luttinger and Ward of the Luttinger theorem, according to which for uniform ground states of systems of (interacting) fermions, which may be metallic or insulating, the number of k points corresponding to non-negative values of G_s(k;mu) is equal to the total number of particles with spin index s in these ground states. Here G_s(k;mu) is the single-particle Green function of particles with spin index s at the chemical potential mu. For the cases where the two-body interaction potential is short-range, and in particular for lattice models, we explicitly demonstrate that this theorem is unconditionally valid, irrespective of the strength of the bare interaction potential. We arrive at this conclusion by amongst other things demonstrating that the perturbation series expansion for self-energy in terms of skeleton diagrams, as encountered in the proof of the Luttinger-Ward identity, is uniformly convergent for almost all momenta and energies. We further investigate the mechanisms underlying some reported instances of failure of the Luttinger theorem. With one exception, for all the cases considered in this paper, we show that the apparent failures of the Luttinger theorem can be attributed either to shortcomings of the employed single-particle Green functions or to misapplication of this theorem. The one exceptional case brings to light the possibility of a genuine failure of the Luttinger theorem for insulating ground states, which we show to be brought about by a false limit that in principle can be reached on taking the zero-temperature limit without the value of mu coinciding with the zero-temperature limit of the chemical potential satisfying the equation of state at finite temperatures; no such ambiguity can arise for metallic states.
Using the first-order series expansion of the function G(k;mu), in powers of mu' = mu - U/2, pertaining to the insulating ground state of a single-band Hubbard Hamiltonian at half-filling, Kokalj and Prelovsek have in a recent paper [Eur. Phys. J. B 63, 431 (2008), arXiv:arXiv:0709.0263] reported breakdown of the Luttinger theorem for the specific case where the lattice on which the Hubbard Hamiltonian is defined is a two-dimensional triangular lattice, for which the ground state is not invariant under particle-hole transformation. Here G(k;mu) is the single-particle Green function G(k;e) evaluated at e = mu, the zero-temperature limit of the chemical potential corresponding to half-filling, and U the on-site interaction energy. In this Comment we demonstrate that unless mu' = 0 (to be strictly distinguished from mu' small but non-vanishing), any finite-order series expansion for G(k;mu) in powers of mu' in general falsely signals breakdown of the Luttinger theorem. The violation of this theorem as asserted by Kokalj and Prelovsek is therefore an artifact of their first-order calculation.
Luttinger theorem relates Fermi surface volume to electron density per spin n. It holds in an arbitrary Fermi liquid but is widely believed to break when a fermionic self-energy has a pole. We show that this is not necessarily the case. As an example, we consider a fermionic system consisting of two gapped bands, originating from a self-energy Σ(ω,k) with a pole, and a narrow residual quasiparticle band crossing ω=0, with a small residue Z≪1 (a residual Fermi liquid). We consider Model I, in which Z also renormalizes quasiparticle bandwidth W to ZW and Model II, in which W remains unrenormalized. The Green's function for such a system G(ω,k) contains two zeros and three poles --- one from the quasiparticle band and two from the gapped bands. Yet, we show that in Model I, the Luttinger-Ward analysis is fully applicable for any Z>0, and the results are exactly the same as if the gapped bands did not exist. For Model II, the Luttinger-Ward analysis recovers once Z exceeds some critical value, which depends on fermionic density.
Many-body physics faces an apparent paradox concerning violations of Luttinger's theorem in strongly correlated systems. The theorem relates the electron number to the Luttinger count -- the monodromy of the logarithm of the imaginary-frequency Green's function G(iε) -- and, at finite magnetic field, connects the Hall conductivity to the magnetic-field derivative of the Luttinger count via the Streda formula. Here, we provide a non-perturbative solution to this paradox. We demonstrate that the reported violations of Luttinger's theorem do not require a breakdown of the skeleton diagram representation of physical quantities, thereby resolving a long-standing point of contention in many-body theory. Furthermore, we show that the Luttinger integral, which accounts for the mismatch between the particle number and the Luttinger count, is itself a genuine topological invariant directly tied to the Luttinger-Ward functional Φ[G] rather than to the Green's function alone. We demonstrate that this invariant, like the monodromy of lnG(iε) itself, stems exclusively from the zero-frequency discontinuity of the Matsubara trajectory.
This appendix summarizes the arguments of the papers by Luttinger and Ward to derive the Luttinger theorem that is stated in Sec. 3.6. This is not a proof that the theorem applies to all possible states of a crystal; it is the derivation of arguments that it applies to all states that can be analytically continued from some non-interacting system, i.e., a “normal state of matter” as defined in Sec. 3.4. The derivation is an example of the use of the T ≠ 0 Green's functions in App. D and the conclusions for T = 0. The Luttinger theorem is a cornerstone in the theory of condensed matter. As described qualitatively in Sec. 3.6, it requires that the volume enclosed by the Fermi surface is conserved independent of interactions, i.e., it is the same as for a system of non-interacting particles. Similarly, the Friedel sum rule is the requirement that the sum of phase shifts around an impurity is determined by charge neutrality, which was derived by Friedel [163] for non-interacting electrons. This section is devoted to a short summary of the original work of Luttinger and Ward and the extension of the arguments to the Freidel sum rule [166]. Here we explicitly indicate the chemical potential μ , since the variation from μ is essential to the arguments. There are two key points: in the interacting system the wavevector in the Brillouin zone k is conserved so that excitations can be labeled k , and the self-energy ∑ k ( ω ) is purely real at the Fermi energy ω = μ at temperature T = 0. The latter point is an essential feature of a Fermi liquid or a “normal metal,” which is justified by the argument that the phase space for scattering at T = 0 vanishes as ω → μ (see Sec. 7.5). Thus, at the Fermi energy the Green's function as a function of k is the same as for an independent-particle problem with eigenvalues. (Of course, for an interacting system at any other energy cannot be described by independent particles.) In an independent-particle system at T = 0 the occupation numbers jump from 1 to 0 as a function of k at the Fermi surface, and in the interacting system there is still a discontinuity in n k that defines the surface (Sec. 7.5).