One of the phenomena associated with quantum integrable systems is the possibility of persistent currents, i.e. currents which do not decay away entirely, but have some portion that continues to flow undiminished and indefinitely. These residual currents are shown to be the conserved part of the current operator, and calculable from the conservation laws of the system. In a particular system, previous attempts to calculate a known residual current from the conservation laws have failed. A numerical investigation is undertaken, and this disparity with the formal results is resolved by the inclusion of a previously overlooked conservation law. An important corollary to these results is that requiring the mutual commutativity of the conservation laws of a quantum integrable system, previously assumed by analogy with the classical case, is an unnecessary and potentially disastrous restriction. Methods of generating the local conservation laws of a quantum integrable system are investigated, and the current method of using a Boost operator is shown to be subtly flawed. The method is discovered to implicitly require additional knowledge in the form of Hamiltonian identities in order to avoid otherwise unphysical terms. A new method is proposed based on the idea that the logarithm of the Transfer matrix of a system generates these local conservation laws. The method is applicable to a wide class of systems whose Lax operator obeys a certain condition, and the majority of the work required to generate the local conservation laws is entirely general and thus only needs to be done once. This new method is then applied to two quite different spin-chain Hamiltonians, the XXZ and Hubbard models, and shown to successfully generate all of the known local conservation laws of these models and some new ones.
One of the phenomena associated with quantum integrable systems is the possibility of persistent currents, i.e. currents which do not decay away entirely, but have some portion that continues to flow undiminished and indefinitely. These residual currents are shown to be the conserved part of the current operator, and calculable from the conservation laws of the system. In a particular system, previous attempts to calculate a known residual current from the conservation laws have failed. A numerical investigation is undertaken, and this disparity with the formal results is resolved by the inclusion of a previously overlooked conservation law. An important corollary to these results is that requiring the mutual commutativity of the conservation laws of a quantum integrable system, previously assumed by analogy with the classical case, is an unnecessary and potentially disastrous restriction. Methods of generating the local conservation laws of a quantum integrable system are investigated, and the current method of using a Boost operator is shown to be subtly flawed. The method is discovered to implicitly require additional knowledge in the form of Hamiltonian identities in order to avoid otherwise unphysical terms. A new method is proposed based on the idea that the logarithm of the Transfer matrix of a system generates these local conservation laws. The method is applicable to a wide class of systems whose Lax operator obeys a certain condition, and the majority of the work required to generate the local conservation laws is entirely general and thus only needs to be done once. This new method is then applied to two quite different spin-chain Hamiltonians, the XXZ and Hubbard models, and shown to successfully generate all of the known local conservation laws of these models and some new ones.
In this thesis, I study local charges in the one-dimensional Hubbard model, derive their explicit expressions, and show their completeness.Quantum integrable lattice models can be exactly solvable via the Bethe Ansatz method, and the characteristic feature is that they possess an extensive number of local charges.Their existence has been well understood through the quantum inverse scattering method; local charges are generated from the transfer matrix.Alternatively, one can obtain local charges by using the Boost operator; the local charges are recursively generated from the charges with a smaller support.Although the existence of local charges and the formalism to obtain them are well known, it is practically difficult to derive general expressions of the local charges.There is no universal understanding of the structure of the local charges applicable to all integrable models, and there are few integrable models whose local charges are fully understood.Understanding the structure of local charges has notably advanced for some integrable spin chains.The structures of the local charges are now well understood for some integrable spin chains, such as the Heisenberg chain, the spin-1/2 XYZ chain, and the models represented by the Temperley-Lieb algebra.The situation is quite different for integrable electron systems such as the one-dimensional Hubbard model.There have been no cases in which the structure of the local charges is completely understood for integrable electron systems.One of the difficulties in the electron systems is the absence of the Boost operator, which comes from the violation of the Lorentz invariance due to the spin-charge separation.This makes the study of their local charges even more challenging, and their investigation is highly desirable.In this thesis, I study the local charges {Q k } in the one-dimensional Hubbard model, which is an integrable system used for theoretical studies of non-perturbative effects in strongly correlated electron systems.I obtained their explicit expressions in a closed form.An expression of the k-local charge Q k for k > 5, where k denotes the support, had been unknown in previous studies.I find that Q k is a linear combination of a particular kind of products of the hopping terms.I introduce a diagrammatic notation to represent these products efficiently, which enables the prediction of the general formula of the local charges.These linear combinations have nontrivial coefficients for k > 5, which were not known in previous studies.Some of the coefficients were proved to be identical to the generalized Catalan numbers, which also appear in the local charges of the Heisenberg chain.I derive the recursion equation for these coefficients.I also prove that the obtained local charges are exhaustive.Thus, any local charge is written i ii as a linear combination of {Q k }.Although the completeness of the local charges is a natural conjecture in the transfer-matrix formulation, its proof has not been presented for almost all cases.I show that the local charges generated by the transfer matrix are represented as linear combinations of my results {Q k }.Lastly, I emphasize that this is the first study demonstrating general explicit expressions of the local charges for an electron system that lacks the boost operator.One can use my result to construct the generalized current operators, which have an essential role in the study of the factorization of correlation functions, and generalized hydrodynamics, a novel way to study the nonequilibrium dynamics in integrable models.v Contents 6.1.3One-local charges . . . . . .
We examine the nature, number, and interrelation of conservation laws in the one-dimensional Hubbard model. In previous work by Shastry [Phys. Rev. Lett. 56, 1529 (1986); 56, 2334 (1986); 56, 2453 (1986); J. Stat. Phys. 50, 57 (1988)], who studied the model on a large but finite number of lattice sites (Na), only Na+1 conservation laws, corresponding to Na+1 operators that commute with themselves and the Hamiltonian, were explicitly identified, rather than the $\ensuremath{\sim}{2N}_{a}$ conservation laws expected from the solvability and integrability of the model. Using a pseudoparticle approach related to the thermodynamic Bethe ansatz, we discover an additional Na+1 independent conservation laws corresponding to nonlocal, mututally commuting operators, which we call transfer-matrix currents. Further, for the model defined in the whole Hilbert space, we find there are two other independent commuting operators (the squares of the \ensuremath{\eta}-spin and spin operators) so that the total number of local plus nonlocal commuting conservation laws for the one-dimensional Hubbard model is 2Na+4. Finally, we introduce an alternative set of 2Na+4 conservation laws which assume particularly simple forms in terms of the pseudoparticle and Yang-particle operators. This set of mutually commuting operators lends itself more readily to calculations of physically relevant correlation functions at finite energy or frequency than the previous set.
One dimensional systems sometimes show pathologically slow decay of currents. This robustness can be traced to the fact that an integrable model is nearby in parameter space. In integrable models some part of the current can be conserved, explaining this slow decay. Unfortunately, although this conservation law is formally anticipated, in practice it has been difficult to find in concrete cases, such as the Heisenberg model. We investigate this issue both analytically and numerically and find that the appropriate conservation law can be a non-analytic combination of the known local conservation laws and hence is invisible to elementary assumptions.
In integrable one-dimensional quantum systems an infinite set of local conserved quantities exists which can prevent a current from decaying completely. For cases like the spin current in the XXZ model at zero magnetic field or the charge current in the attractive Hubbard model at half filling, however, the current operator does not have overlap with any of the local conserved quantities. We show that in these situations transport at finite temperatures is dominated by a diffusive contribution with the Drude weight being either small or even zero. For the XXZ model we discuss in detail the relation between our results, the phenomenological theory of spin diffusion, and measurements of the spin-lattice relaxation rate in spin chain compounds. Furthermore, we study the Haldane-Shastry model where a conserved spin current exists.
We prove that, for translationally invariant quantum spin chains with finite-range interactions, the existence of a specific conservation law implies the presence of infinitely many local conserved quantities, i.e., integrability. This shows that the standard hierarchy of local conserved quantities arising for nearest-neighbor Hamiltonians obtained via the canonical Yang-Baxter construction is already encoded in the lowest nontrivial conservation law, known as the Reshetikhin condition. Combined with recent rigorous results on nonintegrability, our theorem strongly restricts the possibility of partially integrable systems that admit only a finite but large number of local conserved quantities. Our work establishes a rigorous foundation for the systematic identification of new integrable models and deepens the algebraic understanding of conservation-law structures in quantum spin chains.
Within a strong coupling expansion, we construct local quasi-conserved operators for a class of Hamiltonians that includes both integrable and non-integrable models. We explicitly show that at the lowest orders of perturbation theory the structure of the operators is independent of the system details. Higher order contributions are investigated numerically by means of an ab initio method for computing the time evolution of local operators in the Heisenberg picture. The numerical analysis suggests that the quasi-conserved operators could be approximations of a quasi-local conservation law, even if the model is non-integrable.