arXiv subjects

Chuan Li

Publications and source records attributed to Chuan Li.

125 records · Page 7Linked to original sources

arXiv2017

$4π$ periodic Andreev bound states in a Dirac semimetal

Electrons in a Dirac semimetals possess linear dispersion in all three spatial dimensions, and form part of a developing platform of novel quantum materials. Bi$_{1-x}$Sb$_x$ supports a three-dimensional Dirac cone at the Sb-induced band inversion point. Nanoscale phase-sensitive junction technology is used to induce superconductivity in this Dirac semimetal. Radio frequency irradiation experiments reveal a significant contribution of 4$π$-periodic Andreev bound states to the supercurrent in Nb-Bi$_{0.97}$Sb$_{0.03}$-Nb Josephson junctions. The conditions for a substantial $4π$ contribution to the supercurrent are favourable because of the Dirac cone's topological protection against backscattering, providing very broad transmission resonances. The large g-factor of the Zeeman effect from a magnetic field applied in the plane of the junction, allows tuning of the Josephson junctions from 0 to $π$ regimes.

cond-mat.supr-con↗

arXiv2016

Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networks

This paper proposes Markovian Generative Adversarial Networks (MGANs), a method for training generative neural networks for efficient texture synthesis. While deep neural network approaches have recently demonstrated remarkable results in terms of synthesis quality, they still come at considerable computational costs (minutes of run-time for low-res images). Our paper addresses this efficiency issue. Instead of a numerical deconvolution in previous work, we precompute a feed-forward, strided convolutional network that captures the feature statistics of Markovian patches and is able to directly generate outputs of arbitrary dimensions. Such network can directly decode brown noise to realistic texture, or photos to artistic paintings. With adversarial training, we obtain quality comparable to recent neural texture synthesis methods. As no optimization is required any longer at generation time, our run-time performance (0.25M pixel images at 25Hz) surpasses previous neural texture synthesizers by a significant margin (at least 500 times faster). We apply this idea to texture synthesis, style transfer, and video stylization.

cs.CV↗

arXiv2016

Full range of proximity effect probed with Superconductor/Graphene/Superconductor junctions

The high tunability of the density of states of graphene makes it an ideal probe of quantum transport in different regimes. In particular, the supercurrent that can flow through a non-superconducting (N) material connected to two superconducting electrodes, crucially depends on the lenghth of the N relative to the superconducting coherence length. Using graphene as the N material we have investigated the full range of the superconducting proximity effect, from short to long diffusive junctions. By combining several S/graphene/S samples with different contacts and lengths, and measuring their gate-dependent critical currents ($ I_c $) and normal state resistance $ R_N $, we compare the product $eR_NI_c$ to the relevant energies, the Thouless energy in long junctions and the superconducting gap of the contacts in short junctions, over three orders of magnitude of Thouless energy. The experimental variations strikingly follow a universal law, close to the predictions of the proximity effect both in the long and short junction regime, as well as in the crossover region, thereby revealing the interplay of the different energy scales. Differences in the numerical coefficients reveal the crucial role played by the interfacial barrier between graphene and the superconducting electrodes, which reduces the supercurrent in both short and long junctions. Surprisingly the reduction of supercurrent is independent of the gate voltage and of the nature of the electrodes. A reduced induced gap and Thouless energy are extracted, revealing the role played by the dwell time in the barrier in the short junction, and an effective increased diffusion time in the long junction. We compare our results to the theoretical predictions of Usadel equations and numerical simulations which better reproduce experiments with imperfect NS interfaces.

cond-mat.mes-hall↗

arXiv2016

Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesis

This paper studies a combination of generative Markov random field (MRF) models and discriminatively trained deep convolutional neural networks (dCNNs) for synthesizing 2D images. The generative MRF acts on higher-levels of a dCNN feature pyramid, controling the image layout at an abstract level. We apply the method to both photographic and non-photo-realistic (artwork) synthesis tasks. The MRF regularizer prevents over-excitation artifacts and reduces implausible feature mixtures common to previous dCNN inversion approaches, permitting synthezing photographic content with increased visual plausibility. Unlike standard MRF-based texture synthesis, the combined system can both match and adapt local features with considerable variability, yielding results far out of reach of classic generative MRF methods.

cs.CV↗

arXiv2016

Signature of gate-tunable magnetism in graphene grafted with Pt-porphyrins

Inducing magnetism in graphene holds great promises, such as controlling the exchange interaction with a gate electrode and generating exotic magnetic phases. Coating graphene with magnetic molecules or atoms has so far mostly lead to decreased graphene mobility. In the present work, we show that Pt-porphyrins adsorbed on graphene lead to an enhanced mobility and to gate-dependent magnetism. We report that porphyrins can be donor or acceptor, depending on graphene s initial doping. The porphyrins transfer charge and ionize around the charged impurities on graphene, decreasing the graphene doping and increasing its mobility. In addition, ionized porphyrins carry a magnetic moment. Using the sensitivity of mesoscopic transport to magnetism, in particular the superconducting proximity effect and conductance fluctuations, we explore the magnetic order induced in graphene by the interacting magnetic moments of the ionized porphyrins. Among the signatures of magnetism, we find two-terminal-magnetoresistance fluctuations with an odd component, a tell-tale sign of time reversal symmetry breaking at zero field, that does not exist in uncoated graphene sample. When graphene is connected to superconducting electrodes, the induced magnetism leads to a gate-voltage-dependent suppression of the supercurrent, modified magnetic interference patterns, and gate-voltage-dependent magnetic hysteresis. The magnetic signatures are greatest for long superconductor graphene superconductor junctions and for samples with the highest initial doping, compatible with a greater number of ionized and thus magnetic porphyrins. Our findings suggest that long-range magnetism is induced through graphene by the ionized porphyrins magnetic moment. This magnetic interaction is controlled by the density of carriers in graphene, a tunability that could be exploited in spintronic applications.

cond-mat.mes-hall↗

arXiv2015

Predicting Neighbor Distribution in Heterogeneous Information Networks

Recently, considerable attention has been devoted to the prediction problems arising from heterogeneous information networks. In this paper, we present a new prediction task, Neighbor Distribution Prediction (NDP), which aims at predicting the distribution of the labels on neighbors of a given node and is valuable for many different applications in heterogeneous information networks. The challenges of NDP mainly come from three aspects: the infinity of the state space of a neighbor distribution, the sparsity of available data, and how to fairly evaluate the predictions. To address these challenges, we first propose an Evolution Factor Model (EFM) for NDP, which utilizes two new structures proposed in this paper, i.e. Neighbor Distribution Vector (NDV) to represent the state of a given node's neighbors, and Neighbor Label Evolution Matrix (NLEM) to capture the dynamics of a neighbor distribution, respectively. We further propose a learning algorithm for Evolution Factor Model. To overcome the problem of data sparsity, the learning algorithm first clusters all the nodes and learns an NLEM for each cluster instead of for each node. For fairly evaluating the predicting results, we propose a new metric: Virtual Accuracy (VA), which takes into consideration both the absolute accuracy and the predictability of a node. Extensive experiments conducted on three real datasets from different domains validate the effectiveness of our proposed model EFM and metric VA.

cs.SI↗

arXiv2014

Superconducting nanowires by electron-beam-induced deposition

Superconducting nanowires can be fabricated by decomposition of an organometallic gas using a focused beam of Ga ions. However, physical damage and unintentional doping often results from the exposure to the ion beam, motivating the search for a means to achieve similar structures with a beam of electrons instead of ions. This has so far remained an experimental challenge. We report the fabrication of superconducting tungsten nanowires by electron-beam-induced-deposition, with critical temperature of 2.0 K and critical magnetic field of 3.7 T, and compare them with superconducting wires made with ions. This work opens up new possibilities for the realization of nanoscale superconducting devices, without the requirement of an ion beam column.

cond-mat.mes-hall↗

arXiv2014

Proton activity of the Sun in current solar cycle 24

We present a study of 7 large solar proton events (SPEs) of current solar cycle 24 (from 2009 January up to date). They were recorded by GOES spacecraft with highest proton fluxes over 200 pfu for energies $>$10 MeV. In situ particle measurements show that: (1) The profiles of the proton fluxes are highly dependent of the locations of their solar sources, namely flares or coronal mass ejections (CMEs); (2) The solar particle release (SPR) times fall in the decay phase of the flare emission, and are in accordance with the times when the CMEs travel to an average height of 7.9 solar radii; (3) The time differences between the SPR and the flare peak are also dependent of the locations of the solar active regions (ARs). The results tend to support the concept of proton acceleration by the CME-driven shock, even though there exists a possibility of particle acceleration at flare site with subsequent perpendicular diffusion of accelerated particles in the interplanetary magnetic field (IMF). We derive the integral time-of-maximum (TOM) spectra of solar protons in two forms: a single power-law distribution and a power law broken with an exponential tail. It is found that the unique Ground Level Enhancement (GLE) event on 2012 May 17 displays a hardest spectrum and a largest broken energy that may explain why the this event could extend to relativistic energy.

astro-ph.SR↗

arXiv2014

Waiting time distribution of solar energetic particle events modeled with a non-stationary Poisson process

We present a study of the waiting time distributions (WTDs) of solar energetic particle (SEP) events observed with the spacecraft $WIND$ and $GOES$. Both the WTDs of solar electron events (SEEs) and solar proton events (SPEs) display a power-law tail $\sim Δt^{-γ}$. The SEEs display a broken power-law WTD. The power-law index is $γ_{1} =$ 0.99 for the short waiting times ($<$70 hours) and $γ_{2} =$ 1.92 for large waiting times ($>$100 hours). The break of the WTD of SEEs is probably due to the modulation of the corotating interaction regions (CIRs). The power-law index $γ\sim$ 1.82 is derived for the WTD of SPEs that is consistent with the WTD of type II radio bursts, indicating a close relationship between the shock wave and the production of energetic protons. The WTDs of SEP events can be modeled with a non-stationary Poisson process which was proposed to understand the waiting time statistics of solar flares (Wheatland 2000; Aschwanden $\&$ McTiernan 2010). We generalize the method and find that, if the SEP event rate $λ= 1/Δt$ varies as the time distribution of event rate $f(λ) = A λ^{-α}exp(-βλ)$, the time-dependent Poisson distribution can produce a power-law tail WTD $\sim Δt^{α- 3}$, where $0 \leq α< 2$.

astro-ph.SR↗

arXiv2014

Magnetic field resistant quantum interferences in bismuth nanowires based Josephson junctions

We investigate proximity induced superconductivity in micrometer-long bismuth nanowires con- nected to superconducting electrodes with a high critical field. At low temperature we measure a supercurrent that persists in magnetic fields as high as the critical field of the electrodes (above 11 T). The critical current is also strongly modulated by the magnetic field. In certain samples we find regular, rapid SQUID-like periodic oscillations occurring up to high fields. Other samples ex- hibit less periodic but full modulations of the critical current on Tesla field scales, with field-caused extinctions of the supercurrent. These findings indicate the existence of low dimensionally, phase coherent, interfering conducting regions through the samples, with a subtle interplay between orbital and spin contributions. We relate these surprising results to the electronic properties of the surface states of bismuth, strong Rashba spin-orbit coupling, large effective g factors, and their effect on the induced superconducting correlations.

cond-mat.mes-hall↗

arXiv2014

Strain superlattices and macroscale suspension of Graphene induced by corrugated substrates

We investigate the organized formation of strain, ripples and suspended features in macroscopic CVD-prepared graphene sheets transferred onto a corrugated substrate made of an ordered arrays of silica pillars of variable geometries. Depending on the aspect ratio and sharpness of the corrugated array, graphene can conformally coat the surface, partially collapse, or lay, fakir-like, fully suspended between pillars over tens of micrometers. Upon increase of pillar density, ripples in collapsed films display a transition from random oriented pleats emerging from pillars to ripples linking nearest neighboring pillars organized in domains of given orientation. Spatially-resolved Raman spectroscopy, atomic force microscopy and electronic microscopy reveal uniaxial strain domains in the transferred graphene, which are induced and controlled by the geometry. We propose a simple theoretical model to explain the transition between suspended and collapsed graphene. For the arrays with high aspect ratio pillars, graphene membranes stays suspended over macroscopic distances with minimal interaction with pillars tip apex. It offers a platform to tailor stress in graphene layers and open perspectives for electron transport and nanomechanical applications.

cond-mat.mes-hall↗

arXiv2013

Minimal Ward-Takahashi Vertices and Pion Light Cone Distribution Amplitudes from Gauge Invariant, Nonlocal, Dynamical Quark Model

The gauge-invariant, nonlocal, dynamical quark model is proved to generate the minimal vertices which satisfy the Ward-Takahashi identities. In the chiral limit, the momentum-dependent quark self-energy results in a flat-like form with some end point delta-funtions for the light-cone pion distribution amplitudes, similarly found in the Nambu Jona-Lasino model with constant constituent mass. The leading order nonzero pion and current quark masses corrections lead concave type asymptotic-like form modifications to twist-2 pion distribution amplitude with end point pillars and twist-3 tensor pion distribution amplitude above the flat-like form backgrounds. A by-product of our investigation shows that the variable u appearing in pion light-cone distribution amplitudes is just the standard Feynman parameter in the Feynman parameter integrals; also chiral perturbation works well for these amplitudes.

hep-ph↗

arXiv2013

Amending the Vafa-Witten Theorem

The strong version of the Vafa-Witten theorem is shown may not to hold because the zero condensate from a direct computation of the order parameter is found to be a result on the symmetric vacuum. The validity of the Vafa-Witten theorem relies then on its weak version, that the Goldstone boson is absent in vector-like gauge theories with vanishing θ-angle. The existence of a charged ρ-meson condensate, which violates electromagnetic gauge symmetry, is consistent with this weak version of the Vafa-Witten theorem when applied to strong magnetic fields in QCD.

hep-th↗

arXiv2012

Superconducting proximity effect through graphene from zero field to the Quantum Hall regime

We investigate the superconducting proximity effect through graphene in the long diffusive junction limit, at low and high magnetic field. The interface quality and sample phase coherence lead to a zero resistance state at low temperature, zero magnetic field, and high doping. We find a striking suppression of the critical current near graphene\rq{}s charge neutrality point, which we attribute to specular reflexion of Andreev pairs at the interface of charge puddles. This type of reflexion, specific to the Dirac band structure, had up to now remained elusive. At high magnetic field the use of superconducting electrodes with high critical field enables the investigation of the proximity effect in the Quantum Hall regime. Although the supercurrent is not directly detectable in our two wire configuration, interference effects are visible which may be attributed to the injection of Cooper pairs into edge states.

cond-mat.mes-hall↗

arXiv2009

Computation of the p6 order chiral Lagrangian coefficients from the underlying theory of QCD

We present results of computing the p6 order low energy constants in the normal part of chiral Lagrangian both for two and three flavor pseudo-scalar mesons. This is a generalization of our previous work on calculating the p4 order coefficients of the chiral Lagrangian in terms of the quark self energy Sigma(p2) approximately from QCD. We show that most of our results are consistent with those we can find in the literature.

hep-ph↗

arXiv2001

Ramping Control of HLS Storage Ring

HLS (Hefei Light Source) is a second generation synchrotron radiation source. After injected into the storage ring at energy of 200MeV, the electrons will be ramped to 800MeV. During ramping, the magnetic fields of the bend, quadruple and sextuple magnets are ramped up synchronously to keep the working point of the storage ring unchanged. This process is carefully controlled by HLS control system, which is built under EPICS. Several measures are taken to insure the synchronization and linearity of the ramping.

physics.acc-ph↗

arXiv2001

Injection Control System of HLS Storage Ring

The injection control system of Hefei Light Source (HLS) storage ring is a subsystem of the upgraded HLS control system, which is based upon EPICS. Three programmable logic controllers (PLCs) are used as device controllers, which control one septum modulator and four kicker modulators of HLS storage ring. An Industrial PC is used as Input/Output Controller (IOC) and it connects the PLCs with serial communication (RS232 mode) over fibre. A PC with Linux is used as operator interface (OPI), operator application are running on it. The control system was completed in July 2000. The commissioning shows that the control system is reliable and easy operational.

physics.acc-ph↗