arXiv subjects

Junwei Liu

Publications and source records attributed to Junwei Liu.

At least 91 records · Page 5Linked to original sources

arXiv2021

Chiral Interface States and Related Quantized Transport in Disordered Chern Insulators

In this Letter, we study an Anderson-localization-induced quantized transport in disordered Chern insulators (CIs). By investigating the disordered CIs with a step potential, we find that the chiral interface states emerge along the interfaces of the step potential, and the energy range for such quantized transport can be manipulated through the potential strength. Furthermore, numerical simulations on cases with a multi-step potential demonstrate that such chiral state can be spatially shifted by varying the Fermi energy, and the energy window for quantized transport is greatly enlarged. Experimentally, such chiral interface states can be realized by imposing transverse electric field, in which the energy window for quantized transport is much broader than the intrinsic band gap of the corresponding CI. These phenomena are quite universal for disordered CIs due to the direct phase transition between the CI and the normal insulator.

cond-mat.mes-hall↗

arXiv2020

Heuristic machinery for thermodynamic studies of SU(N) fermions with neural networks

The power of machine learning (ML) provides the possibility of analyzing experimental measurements with an unprecedented sensitivity. However, it still remains challenging to probe the subtle effects directly related to physical observables and to understand physics behind from ordinary experimental data using ML. Here, we introduce a heuristic machinery by using machine learning analysis. We use our machinery to guide the thermodynamic studies in the density profile of ultracold fermions interacting within SU($N$) spin symmetry prepared in a quantum simulator. Although such spin symmetry should manifest itself in a many-body wavefuction, it is elusive how the momentum distribution of fermions, the most ordinary measurement, reveals the effect of spin symmetry. Using a fully trained convolutional neural network (NN) with a remarkably high accuracy of $\sim$94$\%$ for detection of the spin multiplicity, we investigate how the accuracy depends on various less-pronounced effects with filtered experimental images. Guided by our machinery, we directly measure a thermodynamic compressibility from density fluctuations within the single image. Our machine learning framework shows a potential to validate theoretical descriptions of SU($N$) Fermi liquids, and to identify less-pronounced effects even for highly complex quantum matter with minimal prior understanding.

cond-mat.quant-gas↗

arXiv2020

Strain Tunable Semimetal-Topological-Insulator Transition in Monolayer 1T'-WTe2

A quantum spin hall insulator(QSHI) is manifested by its conducting edge channels that originate from the nontrivial topology of the insulating bulk states. Monolayer 1T'-WTe2 exhibits this quantized edge conductance in transport measurements, but because of its semimetallic nature, the coherence length is restricted to around 100 nm. To overcome this restriction, we propose a strain engineering technique to tune the electronic structure, where either a compressive strain along a axis or a tensile strain along b axis can drive 1T'-WTe2 into an full gap insulating phase. A combined study of molecular beam epitaxy and in-situ scanning tunneling microscopy/spectroscopy then confirmed such a phase transition. Meanwhile, the topological edge states were found to be very robust in the presence of strain.

cond-mat.mtrl-sci↗

arXiv2019

Effects of Hexagonal Boron Nitride Encapsulation on the Electronic Structure of Few-layer MoS$_2$

The hexagonal boron nitride (hBN) encapsulation has been widely used in the electronics applications of 2D materials to improve device performance by protecting 2D materials against contamination and degradation. It is often assumed that hBN layers as a dielectric would not affect the electronic structure of encapsulated 2D materials. Here we studied few-layer MoS$_2$ encapsulated in hBN flakes by using a combination of theoretical and experimental Raman spectroscopy. We found that after the encapsulation the out-of-plane A$_{1g}$ mode is upshifted, while the in-plane E$_{2g}^1$ mode is downshifted. The measured downshift of the E$_{2g}^1$ mode does not decrease with increasing the thickness of MoS$_2$, which can be attributed to tensile strains in bilayer and trilayer MoS$_2$ caused by the typical experimental process of the hBN encapsulation. We estimated the strain magnitude and found that the induced strain may cause the K-Q crossover in the conduction band of few-layer MoS$_2$, so greatly modifies its electronic properties as an n-type semiconductor. Our study suggests that the hBN encapsulation should be used with caution, as it may affect the electronic properties of encapsulated few-layer 2D materials.

cond-mat.mtrl-sci↗

arXiv2019

All Optical Neural Network with Nonlinear Activation Functions

Artificial neural networks (ANNs) have now been widely used for industry applications and also played more important roles in fundamental researches. Although most ANN hardware systems are electronically based, optical implementation is particularly attractive because of its intrinsic parallelism and low energy consumption. Here, we propose and demonstrate fully-functioned all optical neural networks (AONNs), in which linear operations are programmed by spatial light modulators and Fourier lenses, and optical nonlinear activation functions are realized with electromagnetically induced transparency in laser-cooled atoms. Moreover, all the errors from different optical neurons here are independent, thus the AONN could scale up to a larger system size with final error still maintaining in a similar level of a single neuron. We confirm its capability and feasibility in machine learning by successfully classifying the order and disorder phases of a typical statistic Ising model. The demonstrated AONN scheme can be used to construct various ANNs of different architectures with the intrinsic parallel computation at the speed of light.

physics.optics↗

arXiv2019

Band engineering of a magnetic thin film rare earth monopnictide

Realizing quantum materials in few atomic layer morphologies is a key to both observing and controlling a wide variety of exotic quantum phenomena. This includes topological electronic materials, where the tunability and dimensionality of few layer materials have enabled the detection of $Z_2$, Chern, and Majorana phases. Here, we report the development of a platform for thin film correlated, topological states in the magnetic rare-earth monopnictide ($RX$) system GdBi synthesized by molecular beam epitaxy. This material is known from bulk single crystal studies to be semimetallic antiferromagnets with Neel temperature $T_N =$ 28 K and is the magnetic analog of the non-$f$-electron containing system LaBi proposed to have topological surface states. Our transport and magnetization studies of thin films grown epitaxially on BaF$_2$ reveal that semimetallicity is lifted below approximately 8 crystallographic unit cells while magnetic order is maintained down to our minimum thickness of 5 crystallographic unit cells. First-principles calculations show that the non-trivial topology is preserved down to the monolayer limit, where quantum confinement and the lattice symmetry give rise to a $C=2$ Chern insulator phase. We further demonstrate the stabilization of these films against atmospheric degradation using a combination of air-free buffer and capping procedures. These results together identify thin film $RX$ materials as potential platforms for engineering topological electronic bands in correlated magnetic materials.

cond-mat.str-el↗

arXiv2019

In-Plane Ferroelectric Tunnel Junction

The ferroelectric material is an important platform to realize non-volatile memories. So far, existing ferroelectric memory devices utilize out-of-plane polarization in ferroelectric thin films. In this paper, we propose a new type of random-access memory (RAM) based on ferroelectric thin films with the in-plane polarization called "in-plane ferroelectric tunnel junction". Apart from non-volatility, lower power usage and faster writing operation compared with traditional dynamic RAMs, our proposal has the advantage of faster reading operation and non-destructive reading process, thus overcomes the write-after-read problem that widely exists in current ferroelectric RAMs. The recent discovered room-temperature ferroelectric IV-VI semiconductor thin films is a promising material platform to realize our proposal.

physics.app-ph↗

arXiv2019

The pseudo-Landau-level representation of twisted bilayer graphene: band topology and the implications on the correlated insulating phase

We propose that the electronic structure of twisted bilayer graphene (TBG) can be understood as Dirac fermions coupled with opposite pseudo magnetic fields generated by the moiré pattern. The two low-energy flat bands from each monolayer valley originate from the two zeroth pseudo Landau levels of Dirac fermions under such opposite effective magnetic fields, which have opposite sublattice polarizations and carry opposite Chern numbers $\pm1$, giving rise to helical edge states in the gaps below and above the low-energy bulk bands near the first magic angle. We argue that small Coulomb interactions would split the eight-fold degeneracy (including valley and physical spin) of these zeroth pseudo Landau levels, and may lead to insulating phases with non-vanishing Chern numbers at integer fillings. Besides, we show that all the high-energy bands below or above the flat bands are also topologically nontrivial in the sense that for each valley the sum of their Berry phases is quantized as $\pmπ$. Such quantized Berry phases give rise to nearly flat edge states, which are dependent on truncations on the moiré length scale. Our work provides a complete and clear picture for the electronic structure and topological properties of TBG, and has significant implications on the natrue of the correlated insulating phase observed in experiments.

cond-mat.mes-hall↗

arXiv2018

de Haas-van Alphen effect of correlated Dirac states in kagome metal Fe3Sn2

The field of topological electronic materials has seen rapid growth in recent years, in particular with the increasing number of weakly interacting systems predicted and observed to host topologically non-trivial bands. Given the broad appearance of topology in such systems, it is expected that correlated electronic systems should also be capable of hosting topologically non-trivial states. Interest in correlated platforms is heightened by the prospect that collective behavior therein may give rise to new types of topological states and phenomena not possible in non-interacting systems. However, to date only a limited number of correlated topological materials have been definitively reported due to both the challenge in calculation of their electronic properties and the experimental complexity of correlation effects imposed on the topological aspects of their electronic structure. Here, we report a de Haas-van Alphen (dHvA) study of the recently discovered kagome metal Fe$_3$Sn$_2$ mapping the massive Dirac states strongly coupled to the intrinsic ferromagnetic order. We observe a pair of quasi-two-dimensional Fermi surfaces arising from the massive Dirac states previously detected by spectroscopic probes and show that these band areas and effective masses are systematically modulated by the rotation of the ferromagnetic moment. Combined with measurements of Berry curvature induced Hall conductivity, we find that along with the Dirac fermion mass, velocity, and energy are suppressed with rotation of the moment towards the kagome plane. These observations demonstrate that strong coupling of magnetic order to electronic structure similar to that observed in elemental ferromagnets can be extended to topologically non-trivial electronic systems, suggesting pathways for connecting topological states to robust spintronic technologies.

cond-mat.str-el↗

arXiv2018

Transport evidence of asymmetric spin-orbit coupling in few$-$layer superconducting 1Td$-$MoTe$_2$

Two-dimensional (2D) transition metal dichalcogenides (TMDCs) MX2 (M=W, Mo, Nb, and X=Te, Se, S) with strong spin-orbit coupling (SOC) possess plenty of novel physics including superconductivity. Due to the Ising SOC, monolayer NbSe$_2$ and gated MoS$_2$ of 2H structure can realize the Ising superconductivity phase, which manifests itself with in-plane upper critical field far exceeding Pauli paramagnetic limit. Surprisingly, we find that a few-layer 1Td structure MoTe$_2$ also exhibits an in-plane upper critical field ($H_{c2,//}$) which goes beyond the Pauli paramagnetic limit. Importantly, the in-plane upper critical field shows an emergent two-fold symmetry which is different from the isotropic $H_{c2,//}$ in 2H structure TMDCs. We show that this is a result of an asymmetric SOC in 1Td structure TMDCs. The asymmetric SOC is very strong and estimated to be on the order of tens of meV. Our work provides the first transport evidence of a new type of asymmetric SOC in TMDCs which may give rise to novel superconducting and spin transport properties. Moreover, our findings mostly depend on the symmetry of the crystal and apply to a whole class of 1Td TMDCs such as 1Td-WTe$_2$ which is under intense study due to its topological properties.

cond-mat.supr-con↗

arXiv2018

Self-learning Monte Carlo with Deep Neural Networks

Self-learning Monte Carlo (SLMC) method is a general algorithm to speedup MC simulations. Its efficiency has been demonstrated in various systems by introducing an effective model to propose global moves in the configuration space. In this paper, we show that deep neural networks can be naturally incorporated into SLMC, and without any prior knowledge, can learn the original model accurately and efficiently. Demonstrated in quantum impurity models, we reduce the complexity for a local update from $ \mathcal{O}(β^2) $ in Hirsch-Fye algorithm to $ \mathcal{O}(β\ln β) $, which is a significant speedup especially for systems at low temperatures.

cond-mat.str-el↗

arXiv2018

Symmetry Enforced Self-Learning Monte Carlo Method Applied to the Holstein Model

Self-learning Monte Carlo method (SLMC), using a trained effective model to guide Monte Carlo sampling processes, is a powerful general-purpose numerical method recently introduced to speed up simulations in (quantum) many-body systems. In this work, we further improve the efficiency of SLMC by enforcing physical symmetries on the effective model. We demonstrate its effectiveness in the Holstein Hamiltonian, one of the most fundamental many-body descriptions of electron-phonon coupling. Simulations of the Holstein model are notoriously difficult due to the combination of the typical cubic scaling of fermionic Monte Carlo and the presence of extremely long autocorrelation times. Our method addresses both bottlenecks. This enables simulations on large lattices in the most difficult parameter regions, and evaluation of the critical point for the charge density wave transition at half-filling with high precision. We argue that our work opens a new research area of quantum Monte Carlo (QMC), providing a general procedure to deal with ergodicity in situations involving Hamiltonians with multiple, distinct low energy states.

cond-mat.str-el↗

arXiv2017

Observation of Ultrahigh Mobility Surface States in a Topological Crystalline Insulator by Infrared Spectroscopy

Topological crystalline insulators (TCIs) possess metallic surface states protected by crystalline symmetry, which are a versatile platform for exploring topological phenomena and potential applications. However, progress in this field has been hindered by the challenge to probe optical and transport properties of the surface states owing to the presence of bulk carriers. Here we report infrared (IR) reflectance measurements of a TCI, (001) oriented $Pb_{1-x}Sn_{x}Se$ in zero and high magnetic fields. We demonstrate that the far-IR conductivity is unexpectedly dominated by the surface states as a result of their unique band structure and the consequent small IR penetration depth. Moreover, our experiments yield a surface mobility of 40000 $cm^{2}/(Vs)$, which is one of the highest reported values in topological materials, suggesting the viability of surface-dominated conduction in thin TCI crystals. These findings pave the way for exploring many exotic transport and optical phenomena and applications predicted for TCIs.

cond-mat.mes-hall↗

arXiv2017

Massive Dirac fermions in a ferromagnetic kagome metal

The kagome lattice is a two-dimensional network of corner-sharing triangles known as a platform for exotic quantum magnetic states. Theoretical work has predicted that the kagome lattice may also host Dirac electronic states that could lead to topological and Chern insulating phases, but these have evaded experimental detection to date. Here we study the d-electron kagome metal Fe$_3$Sn$_2$ designed to support bulk massive Dirac fermions in the presence of ferromagnetic order. We observe a temperature independent intrinsic anomalous Hall conductivity persisting above room temperature suggestive of prominent Berry curvature from the time-reversal breaking electronic bands of the kagome plane. Using angle-resolved photoemission, we discover a pair of quasi-2D Dirac cones near the Fermi level with a 30 meV mass gap that accounts for the Berry curvature-induced Hall conductivity. We show this behavior is a consequence of the underlying symmetry properties of the bilayer kagome lattice in the ferromagnetic state with atomic spin-orbit coupling. This report provides the first evidence for a ferromagnetic kagome metal and an example of emergent topological electronic properties in a correlated electron system. This offers insight into recent discoveries of exotic electronic behavior in kagome lattice antiferromagnets and may provide a stepping stone toward lattice model realizations of fractional topological quantum states.

cond-mat.mtrl-sci↗

arXiv2017

Self-Learning Monte Carlo Method: Continuous-Time Algorithm

The recently-introduced self-learning Monte Carlo method is a general-purpose numerical method that speeds up Monte Carlo simulations by training an effective model to propose uncorrelated configurations in the Markov chain. We implement this method in the framework of continuous time Monte Carlo method with auxiliary field in quantum impurity models. We introduce and train a diagram generating function (DGF) to model the probability distribution of auxiliary field configurations in continuous imaginary time, at all orders of diagrammatic expansion. By using DGF to propose global moves in configuration space, we show that the self-learning continuous-time Monte Carlo method can significantly reduce the computational complexity of the simulation.

cond-mat.str-el↗

arXiv2016

Self-Learning Monte Carlo Method

Monte Carlo simulation is an unbiased numerical tool for studying classical and quantum many-body systems. One of its bottlenecks is the lack of general and efficient update algorithm for large size systems close to phase transition or with strong frustrations, for which local updates perform badly. In this work, we propose a new general-purpose Monte Carlo method, dubbed self-learning Monte Carlo (SLMC), in which an efficient update algorithm is first learned from the training data generated in trial simulations and then used to speed up the actual simulation. We demonstrate the efficiency of SLMC in a spin model at the phase transition point, achieving a 10-20 times speedup.

cond-mat.str-el↗

arXiv2016

Self-Learning Determinantal Quantum Monte Carlo Method

Self-learning Monte Carlo method [arXiv:1610.03137, 1611.09364] is a powerful general-purpose numerical method recently introduced to simulate many-body systems. In this work, we implement this method in the framework of determinantal quantum Monte Carlo simulation of interacting fermion systems. Guided by a self-learned bosonic effective action, our method uses a cumulative update [arXiv:1611.09364] algorithm to sample auxiliary field configurations quickly and efficiently. We demonstrate that self-learning determinantal Monte Carlo method can reduce the auto-correlation time to as short as one near a critical point, leading to $\mathcal{O}(N)$-fold speedup. This enables to simulate interacting fermion system on a $100\times 100$ lattice for the first time, and obtain critical exponents with high accuracy.

cond-mat.str-el↗

arXiv2016

Self-Learning Monte Carlo Method in Fermion Systems

We develop the self-learning Monte Carlo (SLMC) method, a general-purpose numerical method recently introduced to simulate many-body systems, for studying interacting fermion systems. Our method uses a highly-efficient update algorithm, which we design and dub "cumulative update", to generate new candidate configurations in the Markov chain based on a self-learned bosonic effective model. From general analysis and numerical study of the double exchange model as an example, we find the SLMC with cumulative update drastically reduces the computational cost of the simulation, while remaining statistically exact. Remarkably, its computational complexity is far less than the conventional algorithm with local updates.

cond-mat.str-el↗