Hopping Kinetics, Quantum Dynamics and Transport

Hamiltonians which are inaccessible in static systems can be engineered in periodically driven many-body systems, i.e., Floquet many-body systems. We propose to use interacting particles in a one-dimensional (1D) harmonic potential with periodic kicking to investigate two-dimensional topological and many-body physics. Depending on the driving parameters, the Floquet Hamiltonian of single kicked harmonic oscillator has various lattice structures in phase space. The noncommutative geometry of phase space gives rise to the topology of the system. We investigate the effective interactions of particles in phase space and find that the point-like contact interaction in quasi-1D real space becomes a long-rang Coulomb-like interaction in phase space, while the hardcore interaction in pure-1D real space becomes a confinement quark-like potential in phase space. We also find that the Floquet exchange interaction does not disappear even in the classical limit, and can be viewed as an effective long-range spin–spin interaction induced by collision. Our proposal may provide platforms to explore new physics and exotic phases by Floquet many-body engineering.

In this paper, we focus on the problem of searching for complex activities involving multiple, interacting objects in video. We examine the dynamics of formation and dispersal of groups as well as their interactions with other groups and individuals. In order to establish a general formalism, we examine activities using relative distances in phase space via pairwise analysis of all objects. This allows us to characterize interactions directly by modeling multi-object activities with the Multiple Objects, Pairwise Analysis (MOPA) feature vector, which is based upon physical models of complex interactions in phase space; specifically, we model paired motion as a damped oscillator in phase space. We model and recognize more complex interactions by characterizing pairs which are correlated in phase space as groups. We show how this model can be used for recognition of complex activities on the standard CAVIAR, VIVID, and UCR Videoweb datasets capturing a variety of problem settings.

Summary form only given. The physics of kinetic electrostatic electron nonlinear (KEEN) waves is elucidated by detailed diagnoses of Vlasov-Poisson simulations. Here we explain the scaling of the multimode non-stationary density response and the phase space partitioned dynamics as a function of drive amplitude and using long time simulations with increasing velocity resolution. In addition, by using 2x-2v simulations, two-dimensional KEEN waves are driven and shown to retain their character found in 1x-1v simulations. Different transverse widths of ponderomotive drive will be compared. Also, Lenard-Bernstein collisions are included to show that the general features of phase space partition and multimode phase locked response both persist but now with a weakly decaying amplitude in time. Thus, this work proves the robustness of KEEN wave results found previously and extends them in dimension and with the addition of velocity space collisions.KEEN-KEEN and KEEN-EPW (electron plasma wave) interactions are considered with novel resonances involving multiple harmonics of KEEN wave interacting simultaneously with an EPW at resonance. This disallows EPW trapped particle states. But at different frequency ratios, their interactions are non-resonant and they are shown to be able to coexist. These are nonlocal interactions in phase space mediated by a common, self-consistent, nonstationary, multimode electrostatic field.

In this contribution we present a mixed quantum-classical dynamical approach for the computation of vibronic absorption spectra of molecular aggregates and their nonadiabatic dynamics, taking into account the coupling between local excitations (LE) and charge-transfer (CT) states. The approach is based on an adiabatic (Ad) separation between the soft degrees of freedom (DoFs) of the system and the stiff vibrations, which are described by the quantum dynamics (QD) of wave packets (WPs) moving on the coupled potential energy surfaces (PESs) of the LE and CT states. These PESs are described with a linear vibronic coupling (LVC) Hamiltonian, parameterized by an overlap-based diabatization on the grounds of time-dependent density functional theory computations. The WPs time evolution is computed with the multiconfiguration time-dependent Hartree method, using effective modes defined through a hierarchical representation of the LVC Hamiltonian. The soft DoFs are sampled with classical molecular dynamics (MD), and the coupling between the slow and fast DoFs is included by recomputing the key parameters of the LVC Hamiltonians, specifically for each MD configuration. This method, named Ad-MD|gLVC, is applied to a perylene diimide (PDI) dimer in acetonitrile and water solutions, and it is shown to accurately reproduce the change in the vibronic features of the absorption spectrum upon aggregation. Moreover, the microscopic insight offered by the MD trajectories allows for a detailed understanding of the role played by the fluctuation of the aggregate structure on the shape of the vibronic spectrum and on the population of LE and CT states. The nonadiabatic QD predicts an extremely fast (∼50 fs) energy transfer between the two LEs. CT states have only a moderate effect on the absorption spectrum, despite the fact that after photoexcitation they are shown to acquire a fast and non-negligible population, highlighting their relevance in dictating the charge separation and transport in PDI-based optical devices.

We made a descriptive theoretical analysis for the transverse momentum dependence of elliptic flow (ν2) of neutrons and protons within the framework of isospin-dependent quantum molecular dynamics (IQMD) model. Simulations are carried out for the symmetric reactions of 124Sn+124Sn and 197Au+197Au. We demonstrate the effect of momentum dependent interactions (MDI) as well as of nucleonic cross-sections (isospin-dependent & independent) on the elliptical flow. The MDI and different forms of cross-section are found to affect the elliptical flow of both protons and neutrons considerably.

Quantum dynamics and delocalization in coherently driven one-dimensional double-well system

The Hamiltonian of a two-level system coupled to a boson mode is transformed to Bargmann's Hilbert space of analytical functions. The quantum dynamics is calculated with a suitable choice of a complete set of orthonormal functions in Bargmann space. A semiclassical approximation which is formulated as a dynamical Hamiltonian system is compared with the quantum dynamics. For the parameters under consideration, semiclassical dynamics is not at all appropriate to approximate the quantum dynamics. These parameters are suited to view the system as a two-site model of small-polaron dynamics.

Reaction dynamics in collisions of antiprotons on nuclei is investigated within the Lanzhou quantum molecular dynamics model. The reaction channels of elastic scattering, annihilation, charge exchange and inelastic collisions of antiprotons on nucleons have been included in the model. Dynamics on particle production, in particular pions, kaons, antikaons and hyperons, is investigated in collisions of p on12 C,20Ne,40 Ca and181Ta from low to high incident momenta. It is found that the annihilations of p on nucleons are of importance to the dynamics of particle production in phase space. Hyperons are mainly produced via meson induced reactions on nucleons and strangeness exchange collisions, which lead to delayed emission in antiproton-nucleus collisions.

The existence of many different dynamic large eddy simulation (LES) methods leads to questions about the theoretical foundation of dynamic LES methods. It was shown recently that the use of stochastic analysis enables a theoretically well based systematic derivation of a realizable linear dynamic model (LDM) and a realizable nonlinear dynamic model (NDM). A priori and a posteriori analyses of turbulent channel flow are used here to study the characteristic properties of these dynamic models. The LDM and NDM are compared with other dynamic models: the non-stabilized and stabilized dynamic Smagorinsky model (DSM), which is used in many applications of LES, and Wang-Bergstrom's dynamic model (WBDM), which represents an extension of the DSM. The DSM and WBDM do not represent realizable models because they are not derived as consequences of a realizable stochastic process. The comparisons reported here show that the LDM and NDM are based on a dynamic model formulation that avoids shortcomings of existing concepts. The LDM and NDM account for backscatter, and they are computationally stable without any modification. The LDM and NDM represent the instantaneous small scale structure of turbulence very well. Compared to the DSM and WBDM, respectively, the LDM and NDM are computationally more efficient.

Many interesting physical and biological phenomena can be investigated using molecular modeling techniques, either theoretically or by using computer simulation methods, such as molecular dynamics and Monte Carlo simulations. Due to the increasing power of computer processing units, these simulation methods allowed over the last decades for the dramatic increase in knowledge of the behavior of systems at the molecular level. In the first part of this thesis the foundations of molecular modeling techniques are revisited. Empirical force fields and the physical background between thermodynamics and individual particles are discussed. The applicability of molecular modeling techniques is shown by two representative cases. First, the molecular dynamics simulation method is used to understand the dynamics of specific proteins at the molecular level. This is important, because drug design efforts are increasingly laborious, especially with the paucity of available structural information. Therefore, computational methods are helpful in predicting the structure of proteins, and, more importantly, to predict conformational dynamics leading to protein activation. To that end a specific asthma-related protein, the beta2-adrenergic receptor, is investigated in atomistic detail together with the molecules that can bind to the protein to cause activation or inhibition. Clearly, molecular dynamics simulations are an important tool to provide further knowledge on the activation pathway of this protein. Although these all-atom simulations give some insight on the dynamics, the computational demand does not allow for systems much larger than several nanometers or time scales exceeding several nanoseconds. An attempt to overcome these problems is presented by the development of a coarse grained description of the transmembrane proteins. Because coarse graining reduces the number of degrees of freedom, the computational demands decrease, and larger systems can be investigated. However, to maintain the specific characteristics of transmembrane proteins, the general force field used in molecular modeling techniques needs to be extended with hydrogen bonding capabilities and helical backbone stabilization. This new coarse grained model is applicable to transmembrane proteins, and is used to investigate two independent cases: WALP-peptides and antimicrobial peptides. The first serve as a model system for both experiments and theory to investigate the interaction between transmembrane peptides and lipid membranes, whereas the latter are antibiotics whose pore-forming capacities are of great interest to act as target-specific drug candidates. From the molecular dynamics simulations of the WALP-peptides it is shown that the apparent hydrophobic mismatch between peptide and membrane can be resolved by two mechanisms (membrane thickness adaptation and peptide tilting) and that these two mechanisms occur sequential and not in parallel. In the case of the antimicrobial peptides it is shown that many of the orientations found with the molecular simulation techniques are in agreement with experimental observations. The second case to show the applicability of the molecular modeling techniques is that of the heat transfer characteristics of gas flows in nanochannels. Understanding these characteristics is important, because these very small channels are considered to be promising devices to locally cool systems (such as computer processing units) or to be used in lab-on-chip devices for at home medical diagnostics. Thus, understanding the interactions between the channel walls and the gas flow is of great importance. Unfortunately, the computational cost involved in simulating the solid wall, currently restrains the size of the systems that can be investigated using molecular dynamics simulations. Therefore, instead of the explicit modeling of the solid wall, appropriate boundary conditions are used, such as wall potentials or stochastic models. Both of these boundary conditions are examined in great detail and a new wall potential is presented. Also the investigations of a specific case of a channel with platinum walls with a noble gas (argon or xenon) in between allows to introduce a new method to compute an important heat transfer determining parameter. Furthermore, it is shown that both boundary conditions have their benefits and drawbacks, and that the use of either one depends heavily on the application under consideration. Both cases used to show the applicability of molecular modeling techniques, although very different from each other, indicate the importance of particle simulation methods. Investigating the interactions at the molecular level, and the development of new models allows for an even better understanding of underlying molecular processes.

A Busy March Meeting Is Brewing in Seattle

The isospin- and momentum-dependent interaction using the isospin-dependent quantum molecular dynamics model is improved. The momentum dependence of the interaction is fitted to the optical potentials extracted with proton-nucleus scattering data, while the isospin dependence is adjusted based on the current constraints on the symmetry energy and its density slope at normal density. The resulting parameters of the interaction are used to calculate the effective masses characterized by the nonlocality of the nuclear potential in the spatial coordinates (the so-called effective $k$ masses). It is found that this parametrization can reproduce the effective $k$ masses calculated by the Dirac-Brueckner-Hartree-Fock model. The current constraints on the total symmetry energy can accommodate a wide-range adjustment of the local symmetry energy and effective $k$-mass splitting. Four groups of parameters, which provide different density dependences of symmetry energy and effective $k$-mass splitting, are applied to investigate the isospin diffusion and neutron to proton double ratios in collisions involving $^{124}\mathrm{Sn}$ and $^{112}\mathrm{Sn}$ nuclei at 50 and 120 MeV/nucleon. Calculations confirm the sensitivity of symmetry energy to the isospin diffusion, but indicate the synchronous effects of symmetry energy and effective $k$-mass splitting on the double ratios.

We explore the capability of linear fully connected neural networks (FCNs) to learn and predict quantum dynamics governed by various equations of motion, including the time-dependent Schrödinger equation, quantum Liouville equation, and thermofield dynamics (TFD). Given the inherent linearity of these equations, linear FCNs naturally align with the fundamental structure of quantum propagators, making them a promising machine-learning approach for simulating quantum systems. Through numerical experiments on different model systems, including single-state and two-state anharmonic potentials, we demonstrate that linear FCNs without hidden layers achieve high accuracy while preserving population conservation. For finite-temperature dynamics, we find that learning TFD is more efficient than learning density matrix evolution due to the suppressed noise from the partial trace over the fictitious Hilbert space. However, as system dimensionality increases, the computational cost of linear FCNs becomes substantial due to scaling and sampling challenges, suggesting the necessity of more efficient network structures. Our results highlight the potential of physics-informed neural networks for quantum dynamics and point toward future developments in neural network architectures inspired by physical principles.

Effective interactions and elementary excitations in nuclear matter

Amorphous carbon films approximately 20 nm thick are used throughout the computer industry as protective coatings on magnetic storage disks. As storage densities increase, the role of the overcoat becomes increasingly important because of smaller spacings between the recording head and the spinning disk. Furthermore, future-generation disks call for an overcoat thickness of 5 nm or less. These small length scales and the high speed of the spinning disk (10-30 m/s) suggest that a molecular dynamics (MD) model might provide useful insight into friction and wear mechanisms when the head and disk make contact. One of the necessary inputs required to carry out such an MD model is a specification of the position of all the atoms in the simulation, i.e. a detailed model of the material microstructure. Such a detailed understanding of the microstructure of amorphous carbon overcoats does not presently exist. Neutron and electron diffraction studies demonstrate that the material is amorphous. Previous classical MD simulations yield pair distribution functions in qualitative agreement with the diffraction studies, but they all differ in detail. More recent, quantum-mechanical tight-binding MD (TBMD) studies give a better description of the interatomic interactions and the chemical hybridization (sp{sup 2}-graphite-like versus sp{sup 3}-diamond-like). However, these studies are presently limited to rather small system sizes and rapid quench rates. In this paper we present molecular dynamics simulations of the growth of amorphous carbon films deposited onto a diamond substrate using a bond-order potential model. This classical potential mimics the quantum mechanics allowing carbon to form strong chemical bonds with a variety of hybridizations. It was found that the system formed unphysical bonding configurations without an added torsional energy between sp{sup 2} hybridized carbon atoms. This torsional energy was included for all results presented here. 18 refs., 3 figs.