Publications

We propose an algorithm which combines the beneficial aspects of two different methods for studying finite-temperature quantum systems with tensor networks. One approach is the ancilla method, which gives high-precision results but scales poorly at low temperatures. The other method is the minimally entangled typical thermal state (METTS) sampling algorithm which scales better than the ancilla method at low temperatures and can be parallelized, but requires many samples to converge to a precise result. Our proposed hybrid of these two methods purifies physical sites in a small central spatial region with partner ancilla sites, sampling the remaining sites using the METTS algorithm. Observables measured within the purified cluster have much lower sample variance than in the METTS approach, while sampling the sites outside the cluster reduces their entanglement and the computational cost of the algorithm. The sampling steps of the algorithm remain straightforwardly parallelizable. The hybrid approach also solves an important technical issue with METTS that makes it difficult to benefit from quantum number conservation. By studying S=1 Heisenberg ladder systems, we find the hybrid method converges more quickly than both the ancilla and METTS algorithms at intermediate temperatures and for systems with higher entanglement.

We describe a new method (\textsc{CompaSO}) for identifying groups of particles in cosmological N-body simulations. \textsc{CompaSO} builds upon existing spherical overdensity (SO) algorithms by taking into consideration the tidal radius around a smaller halo before competitively assigning halo membership to the particles. In this way, the \textsc{CompaSO} finder allows for more effective deblending of haloes in close proximity as well as the formation of new haloes on the outskirts of larger ones. This halo-finding algorithm is used in the \textsc{AbacusSummit} suite of N-body simulations, designed to meet the cosmological simulation requirements of the Dark Energy Spectroscopic Instrument (DESI) survey. \textsc{CompaSO} is developed as a highly efficient on-the-fly group finder, which is crucial for enabling good load-balancing between the GPU and CPU and the creation of high-resolution merger trees. In this paper, we describe the halo-finding procedure and its particular implementation in \Abacus{Abacus}, accompanying it with a qualitative analysis of the finder. {We test the robustness of the \textsc{CompaSO} catalogues before and after applying the cleaning method described in an accompanying paper and demonstrate its effectiveness by comparing it with other validation techniques.} We then visualise the haloes and their density profiles, finding that they are well fit by the NFW formalism. Finally, we compare other properties such as radius-mass relationships and two-point correlation functions with that of another widely used halo finder, \textsc{ROCKSTAR}.

We propose a systematic and constructive way to determine the exchange-correlation potentials of density-functional theories including vector potentials. The approach does not rely on energy or action functionals. Instead, it is based on equations of motion of current quantities (force balance equations) and is feasible both in the ground-state and the time-dependent settings. This avoids, besides differentiability and causality issues, the optimized-effective-potential procedure of orbital-dependent functionals. We provide straightforward exchange-type approximations for different density functional theories that for a homogeneous system and no external vector potential reduce to the exchange-only local-density and Slater Xα approximations.

We analyze the instability of an unpolarized uniform quantum plasma consisting of two oppositely charged fermionic components with varying mass ratios against charge and spin density waves. Using density functional theory, we treat each component with the local spin density approximation and a rescaled exchange-correlation functional. Interactions between different components are treated with a mean-field approximation. In both two and three dimensions, we find leading unstable charge density wave modes in the second-order expansion of the energy functional, which would induce the transition to quantum liquid crystals. The transition point and the length of the wave vector are computed numerically. Discontinuous ranges of the wave vector are found for different mass ratios between the two components, indicating exotic quantum phase transitions. Phase diagrams are obtained, and a scaling relation is proposed to generalize the results to two-component fermionic plasmas with any mass scale. We discuss the implications of our results and directions for further improvement in treating quantum plasmas.

Classical probability distributions on sets of sequences can be modeled using quantum states. Here, we do so with a quantum state that is pure and entangled. Because it is entangled, the reduced densities that describe subsystems also carry information about the complementary subsystem. This is in contrast to the classical marginal distributions on a subsystem in which information about the complementary system has been integrated out and lost. A training algorithm based on the density matrix renormalization group (DMRG) procedure uses the extra information contained in the reduced densities and organizes it into a tensor network model. An understanding of the extra information contained in the reduced densities allow us to examine the mechanics of this DMRG algorithm and study the generalization error of the resulting model. As an illustration, we work with the even-parity dataset and produce an estimate for the generalization error as a function of the fraction of the dataset used in training.

We show that the recently proposed cooling-by-doping mechanism allows one to efficiently prepare interesting nonequilibrium states of the Hubbard model. Using nonequilibrium dynamical mean field theory and a particle-hole symmetric setup with dipolar excitations to full and empty bands we produce cold photodoped Mott insulating states with a sharp Drude peak in the optical conductivity, a superconducting state in the repulsive Hubbard model with an inverted population, and
η
-paired states in systems with a large density of doublons and holons. The reshuffling of entropy into full and empty bands not only provides an efficient cooling mechanism, it also allows one to overcome thermalization bottlenecks and slow dynamics that have been observed in systems cooled by the coupling to boson baths.

We investigate the effect of nonlocal interactions on the photo-doped Mott insulating state of the two-dimensional Hubbard model using a nonequilibrium generalization of the dynamical cluster approximation. In particular, we compare the situation where the excitonic states are lying within the continuum of doublon-holon excitations to a set-up where the excitons appear within the Mott gap. In the first case, the creation of nearest-neighbor doublon-holon pairs by excitations across the Mott gap results in enhanced excitonic correlations, but these excitons quickly decay into uncorrelated doublons and holons. In the second case, photo-excitation results in long-lived excitonic states. While in a low-temperature equilibrium state, excitonic features are usually not evident in single-particle observables such as the photoemission spectrum, we show that the photo-excited nonequilibrium system can exhibit in-gap states associated with the excitons. The comparison with exact-diagonalization results for small clusters allows us to identify the signatures of the excitons in the photo-emission spectrum.

Controlling phase transitions in transition metal oxides remains a central feature of both technological and fundamental scientific relevance. A well-known example is the metal-insulator transition which has been shown to be highly controllable while a less well understood aspect of this phenomenon is the length scale over which the phases can be established. To gain further insight into this issue, we have atomically engineered an artificially phase separated system through fabricating epitaxial superlattices consisting of SmNiO3 and NdNiO3, two materials undergoing a metal-to-insulator transition at different temperatures. By combining advanced experimental techniques and theoretical modeling, we demonstrate that the length scale of the metal-insulator transition is controlled by the balance of the energy cost of the domain wall between a metal and insulator and the bulk energetics. Notably, we show that the length scale of this effect exceeds that of the physical coupling of structural motifs, introducing a new paradigm for interface-engineering properties that are not available in bulk.

Compressive isothermal turbulence is known to have a near lognormal density probability distribution function (PDF) with a width that scales with the sonic Mach number and nature of the turbulent driving (solenoidal vs compressive). However, the physical processes that mold the extreme high and low density structures in a turbulent medium can be different, with the densest structures being composed of strong shocks that evolve on shorter timescales than the low density fluid. The density PDF in a turbulent medium exhibits deviations from lognormal due to shocks, that increases with the sonic Mach number, which is often ignored in analytic models for turbulence and star formation. We develop a simple model for turbulence by treating it as a continuous Markov process, which explains both the density PDF and the transient timescales of structures as a function of density, using a framework developed in n Scannapieco & Safarzadeh (2018). Our analytic model depends on only a single parameter, the effective compressive sonic Mach number, and successfully describes the non-lognormal behavior seen in both 1D and 3D simulations of supersonic and subsonic compressive isothermal turbulence. The model quantifies the non-lognormal distribution of density structures in turbulent environments, and has application to star forming molecular clouds and star formation efficiencies.

We study the stability properties of explicit marching schemes for second-kind Volterra integral equations that arise when solving boundary value problems for the heat equation by means of potential theory. It is well known that explicit finite difference or finite element schemes for the heat equation are stable only if the time step $\Delta t$ is of the order $O(\Delta x^2)$, where $\Delta x$ is the finest spatial grid spacing. In contrast, for the Dirichlet and Neumann problems on the unit ball in all dimensions $d\ge 1$, we show that the simplest Volterra marching scheme, i.e., the forward Euler scheme, is unconditionally stable. Our proof is based on an explicit spectral radius bound of the marching matrix, leading to an estimate that an $L^2$-norm of the solution to the integral equation is bounded by $c_dT^{d/2}$ times the norm of the right hand side. For the Robin problem on the half space in any dimension, with constant Robin (heat transfer) coefficient $\kappa$, we exhibit a constant $C$ such that the forward Euler scheme is stable if $\Delta t < C/\kappa^2$, independent of any spatial discretization. This relies on new lower bounds on the spectrum of real symmetric Toeplitz matrices defined by convex sequences. Finally, we show that the forward Euler scheme is unconditionally stable for the Dirichlet problem on any smooth convex domain in any dimension, in $L^\infty$-norm.