Browsing by Author "Liu, Jinn-Liang"

Now showing 1 - 6 of 6

A model for semiconductor quantum dot molecule based on the current spin density functional theory
(2006-11-01)
Liu, Jinn-Liang
;
Chen, Jen-Hao
;
Voskoboynikov, O.
;
應用數學系
;
電子工程學系及電子研究所
Based on the current spin density functional theory, a theoretical model of three vertically aligned semiconductor quantum dots is proposed and numerically studied. This quantum dot molecule (QDM) model is treated with realistic hard-wall confinement potential and external magnetic field in three-dimensional setting. Using the effective-mass approximation with band nonparabolicity, the many-body Hamiltonian results in a cubic eigenvalue problem from a finite difference discretization. A self-consistent algorithm for solving the Schrodinger-Poisson system by using the Jacobi-Davidson method and GMRES is given to illustrate the Kohn-Sham orbitals and energies of six electrons in the molecule with some magnetic fields. It is shown that the six electrons residing in the central dot at zero magnetic field can be changed to such that each dot contains two electrons with some feasible magnetic field. The Forster-Dexter resonant energy transfer may therefore be generated by two individual QDMs. This may motivate a new paradigm of Fermionic qubits for quantum computing in solid-state systems. (C) 2006 Elsevier B.V. All rights reserved.
Deep learning and control algorithms of direct perception for autonomous driving
(1970-01-01)
Lee, Der-Hau
;
Chen, Kuan-Lin
;
Liou, Kuan-Han
;
Liu, Chang-Lun
;
Liu, Jinn-Liang
We propose an end-to-end machine learning model that integrates multi-task (MT) learning, convolutional neural networks (CNNs), and control algorithms to achieve efficient inference and stable driving for self-driving cars. The CNN-MT model can simultaneously perform regression and classification tasks for estimating perception indicators and driving decisions, respectively, based on the direct perception paradigm of autonomous driving. The model can also be used to evaluate the inference efficiency and driving stability of different CNNs on the metrics of CNN's size, complexity, accuracy, processing speed, and collision number, respectively, in a dynamic traffic. We also propose new algorithms for controllers to drive a car using the indicators and its short-range sensory data to avoid collisions in real-time testing. We collect a set of images from a camera of The Open Racing Car Simulator in various driving scenarios, train the model using this dataset, test it in unseen traffics, and find that it outperforms earlier models in highway traffic. The stability of end-to-end learning and self driving depends crucially on the dynamic interplay between CNN and control algorithms. The source code and data of this work are available on our website, which can be used as a simulation platform to evaluate different learning models on equal footing and quantify collisions precisely for further studies on autonomous driving.
Magnetic quantum number dependence of hydrogen photoelectron spectra under circularly polarized pulse in barrier suppression ionization regime
(2018-08-01)
Jheng, Shih-Da
;
Jiang, Tsin-Fu
;
Chen, Jen-Hao
;
Liu, Jinn-Liang
;
物理研究所
Photoelectron spectra and transition mechanisms of a hydrogen atom in the n = 2 state under circularly polarized pulses in the barrier suppression ionization (BSI) regime, found by solving the time-dependent Schrodinger equation, are presented in detail. The photoelectron peak emerges from low-energy, in contrast to being centered at higher energy in the case of ionization from the ground state. We show that the magnetic quantum number dependence of photoelectron angular distributions on initial states yields 3- and 2-band structures for 2p(-) and 2p(0) in zenith angle, respectively, and 1-band structures for both 2p(+) and 2s. The intermediate transition pathways from these initial states to the corresponding final states are also revealed.
Poisson-Fermi Modeling of the Ion Exchange Mechanism of the Sodium/Calcium Exchanger
(2016-03-17)
Liu, Jinn-Liang
;
Hsieh, Hann-jeng
;
Eisenberg, Bob
;
應用數學系
;
Department of Applied Mathematics
The ion exchange mechanism of the sodium/calcium exchanger (NCX) crystallized by Liao et al. in 2012 is studied using the Poisson Fermi theory developed by Liu and Eisenberg in 2014. A cycle of binding and unbinding is proposed to account for the Na+/Ca2+ exchange function of the NCX molecule. Outputs of the theory include electric and steric fields of ions with different sizes, correlations of ions of different charges, and polarization of water, along with number densities of ions, water molecules, and interstitial voids. We calculate the electrostatic and steric potentials of the four binding sites in NCX, i.e., three Nal" binding sites and one Ca2+ binding site, with protein charges provided by the software PDB2PQR. The energy profiles of Na+ and Ca2+ ions along their respective Na+ and Ca2+ pathways in experimental conditions enable us to explain the fundamental mechanism of NCX that extrudes intracellular Ca2+ across the cell membrane against its chemical gradient by using the downhill gradient of Nat Atomic and numerical details of the binding sites are given to illustrate the 3 Na+:1 Ca2+ stoichiometry of NCX. The protein NCX is a catalyst. It does not provide (free) energy for transport. All energy for transport in our model comes from the ions in surrounding baths.
Study of the circular dichroism of He+ ion
(2019-12-01)
Chen, Jen-Hao
;
Liu, Jinn-Liang
;
Jheng, Shih-Da
;
Cheng, Szu-Cheng
;
Jiang, Tsin-Fu
We study the mechanism of circular dichroism (CD) in He+ under XUV and NIR pulses in the experimental condition presented by Ilchen et al. [Phys. Rev. Lett. 118, 013002 (2017)] and in a range of pulse intensities and durations by solving the time-dependent Schrodinger equation (TDSE) in momentum space. The method enables investigation of the ionization and intermediate transition probabilities of He+ in detail. We show that the lowest-order perturbation theory (LOPT) is qualitatively sufficient to describe experimental results and find that ionization favors co- and counter-rotating pulses for NIR intensities in and beyond LOPT regimes, respectively. In the experimental pulse duration, we show that a sub-peak along the first above-threshold ionization peak exists in the counter-rotating case due to one-photon absorption from excited states, which cannot be obtained by solving TDSE with a short duration of 22 fs, since the energy band width is too wide to resolve this substructure. We present the variation of CD values with NIR intensities. (C) 2019 Optical Society of America
The effect of band nonparabolicity on modeling few-electron ground states of charge-tunable InAs/GaAs quantum dot
(2009-01-01)
Hsieh, Yi-Chern
;
Chen, Jen-Hao
;
Tseng, Shi-Chang
;
Liu, Jinn-Liang
;
應用數學系
Based on the current spin density functional theory and the nonparabolic effective mass approximation, a three-dimensional model is presented to Study the few-electron system in InAs/GaAs quantum dot for which the electron spectra have been obtained from the capacitance-voltage (CV) measurements by Miller et al. [Phys. Rev. B 56 (1997) 6764]. The model is an extension of the single-electron model proposed by Filikhin et al. [Solid State Comm. 140 (2006) 483]. Our results call quantitatively well interpret the experimental CV data. It is shown that the energy differences between the parabolic and nonparabolic approximations are comparable with the exchange-correlation energies, Moreover, the nonparabolic effect is shown to be increasingly more significant than that of parabolic case in higher magnetic fields. It is also more pronounced for larger number of electrons. (C) 2008 Elsevier B.V. All rights reserved.