Heat Transport in Quantum Spin Chains: the Relevance of Integrability
Jinshan Wu,Mona Berciu
2009-01-01
Abstract:Heat transport in quantum spin chains, in particular, under which conditions normal (diffusive) transport is observed, is still not understood despite considerable effort.1–14 For example, it was conjectured that integrability leads to anomalous (ballistic) transport,8 but it was also argued that integrable but gaped chains have normal conductivity.10,11 Others have argued that only the spin conductivity is normal in this case, while the thermal conductivity is still anomalous.12 A consensus as to what the necessary criteria are for normal versus anomalous conductivity is still missing. Most of this work3–10 studied infinite and/or periodic chains, and used the Kubo formula15,16 where finite (zero) Drude weight signals anomalous (normal) transport. For integrable systems, the Kubo formula predicts anomalous heat transport. In fact, full integrability is not even necessary; all that is needed is a finite overlap between the heat current operator and a conserved quantity.4,14 The derivation of the Kubo formula for finite systems requires dealing with the currents between the chain’s ends and the thermal baths it is connected to. These boundary currents may change the form of the Kubo formula.17,18 For infinite systems, one may argue that they can be ignored, as they are a boundary effect.15,16 However, it is obviously not so for finite systems, where coupling to baths should be taken into consideration explicitly.18 Furthermore, “integrability” of the chain connected to baths may be lost even if the isolated chain is integrable since the terms describing the coupling to the baths lead to a nonvanishing commutator between the heat current operator and the total Hamiltonian. Both of these facts might invalidate the main argument for anomalous transport based on the Kubo formula. Studies that explicitly consider the effects of the baths on a finite chain, while fewer, also give contradictory results. Using the local-operator Lindblad equation, Prosen et al. showed that an integrable gaped (Jz > Jxy) XXZ chain has normal spin conductivity, while its energy transport is anomalous;12 whereas, using the same approach, Michel et al. claim that, for Jz > 1.6Jxy , spin chains have normal energy transport.11 Such conflicting results exist not only in theoretical studies, but also in experiments. Anomalous heat transport observed experimentally in systems described by integrable models, such as (Sr,Ca)14Cu24O41, Sr2CuO3 and CuGeO3, seems to validate the conjecture linking anomalous transport to integrability, but Ref. 22 finds normal transport in Sr2CuO3 at high temperatures. Even if there was no conflict between results based on the local-operator Lindblad equation, such an approach is likely less reliable than one based on the Redfield equation. For example, only the latter leads to the proper Boltzmann distribution if the baths are not biased. There are, in fact, already several such studies based on the Redfield equation.23,24 However, these are for spin systems that are either noninteracting (thus, trivially integrable) or are nonintegrable. We call a spin chain “noninteracting” if it can be mapped, for example through a Jordan-Wigner transformation, to a Hamiltonian for noninteracting spinless fermions. Anomalous transport is found for the former and normal transport is found for the latter. Disordered spin systems have also been found to have normal transport.25 There are no examples of clean, interacting but integrable systems investigated via the Redfield equation. Here, we systematically investigate many models of finite spin chains coupled to thermal baths using the Redfield equation.23–27 Our results suggest that integrability is not a sufficient condition for anomalous heat transport. Note that, in this context, we call a model “integrable” if it is so when the system is isolated (i.e., not connected to baths). We find anomalous transport at all temperatures only in models that can be mapped onto homogeneous noninteracting fermionic models. All other models we investigated exhibit normal heat transport, whether they are integrable or not (however, as discussed below, at low temperatures, their heat transport may become anomalous in certain conditions). We therefore conjecture that the existence of a mapping of the spin Hamiltonian to one of noninteracting fermions is the criterion determining anomalous transport, at least for finite-size systems. The paper is organized as follows. Section II defines the problem and its Redfield equation. In Sec. III, we discuss the numerical methods we use and, in Sec. IV, we define the quantities we use to characterize thermal and spin transport. We then proceed to present our numerical results in Sec. V, and then conclude.