arXiv subjects

Mathieu Luisier

Publications and source records attributed to Mathieu Luisier.

At least 73 records · Page 4Linked to original sources

arXiv2019

Optimizing the Data Movement in Quantum Transport Simulations via Data-Centric Parallel Programming

Designing efficient cooling systems for integrated circuits (ICs) relies on a deep understanding of the electro-thermal properties of transistors. To shed light on this issue in currently fabricated FinFETs, a quantum mechanical solver capable of revealing atomically-resolved electron and phonon transport phenomena from first-principles is required. In this paper, we consider a global, data-centric view of a state-of-the-art quantum transport simulator to optimize its execution on supercomputers. The approach yields coarse- and fine-grained data-movement characteristics, which are used for performance and communication modeling, communication-avoidance, and data-layout transformations. The transformations are tuned for the Piz Daint and Summit supercomputers, where each platform requires different caching and fusion strategies to perform optimally. The presented results make ab initio device simulation enter a new era, where nanostructures composed of over 10,000 atoms can be investigated at an unprecedented level of accuracy, paving the way for better heat management in next-generation ICs.

cs.CE↗

arXiv2019

A Data-Centric Approach to Extreme-Scale Ab initio Dissipative Quantum Transport Simulations

The computational efficiency of a state of the art ab initio quantum transport (QT) solver, capable of revealing the coupled electro-thermal properties of atomically-resolved nano-transistors, has been improved by up to two orders of magnitude through a data centric reorganization of the application. The approach yields coarse-and fine-grained data-movement characteristics that can be used for performance and communication modeling, communication-avoidance, and dataflow transformations. The resulting code has been tuned for two top-6 hybrid supercomputers, reaching a sustained performance of 85.45 Pflop/s on 4,560 nodes of Summit (42.55% of the peak) in double precision, and 90.89 Pflop/s in mixed precision. These computational achievements enable the restructured QT simulator to treat realistic nanoelectronic devices made of more than 10,000 atoms within a 14$\times$ shorter duration than the original code needs to handle a system with 1,000 atoms, on the same number of CPUs/GPUs and with the same physical accuracy.

cs.CE↗

arXiv2019

Electron transport through metal/MoS2 interfaces: edge- or area-dependent process?

In ultra-thin two-dimensional (2-D) materials, the formation of ohmic contacts with top metallic layers is a challenging task that involves different processes than in bulk-like structures. Besides the Schottky barrier height, the transfer length of electrons between metals and 2-D monolayers is a highly relevant parameter. For MoS$_2$, both short ($\le$30 nm) and long ($\ge$0.5 $μ$m) values have been reported, corresponding to either an abrupt carrier injection at the contact edge or a more gradual transfer of electrons over a large contact area. Here we use \textit{ab initio} quantum transport simulations to demonstrate that the presence of an oxide layer between a metallic contact and a MoS$_2$ monolayer, for example TiO$_2$ in case of titanium electrodes, favors an area-dependent process with a long transfer length, while a perfectly clean metal-semiconductor interface would lead to an edge process. These findings reconcile several theories that have been postulated about the physics of metal/MoS$_2$ interfaces and provide a framework to design future devices with lower contact resistances.

cond-mat.mes-hall↗

arXiv2019

First-principles approach to electric polarization and dielectric constant calculations using generalized Wannier functions

We describe a method to calculate the electronic properties of an insulator under an applied electric field. It is based on the minimization of an electric enthalpy functional with respect to the orbitals, which behave as Wannier functions under crystal translations, but are not necessarily orthogonal. This paper extends the approach of Nunes and Vanderbilt (NV) [Phys. Rev. Lett. 73, 712 (1994)], who demonstrated that a Wannier function representation can be used to study insulating crystals in the presence of a finite electric field. According to a study by Fernández et al. [Phys. Rev. B. 58, R7480 (1998)], first-principles implementations of the NV approach suffer from the impact of the localization constraint on the orthogonal wave functions, what affects the accuracy of the physical results. We show that because non-orthogonal generalized Wannier functions can be more localized than their orthogonal counterparts, the error due to localization constraints is reduced, thus improving the accuracy of the calculated physical quantities.

cond-mat.mtrl-sci↗

arXiv2019

One-dimensional edge contacts to a monolayer semiconductor

Integration of electrical contacts into van der Waals (vdW) heterostructures is critical for realizing electronic and optoelectronic functionalities. However, to date no scalable methodology for gaining electrical access to buried monolayer two-dimensional (2D) semiconductors exists. Here we report viable edge contact formation to hexagonal boron nitride (hBN) encapsulated monolayer MoS$_2$. By combining reactive ion etching, in situ Ar$^+$ sputtering and annealing, we achieve a relatively low edge contact resistance, high mobility (up to ~30 cm$^2$/Vs) and high on-current density (>50 uA/um at V$_{\rm DS}$ = 3V), comparable to top contacts. Furthermore, the atomically smooth hBN environment also preserves the intrinsic MoS$_2$ channel quality during fabrication, leading to a steep subthreshold swing of 116 mV/dec with a negligible hysteresis. Hence, edge contacts are highly promising for large-scale practical implementation of encapsulated heterostructure devices, especially those involving air sensitive materials, and can be arbitrarily narrow, which opens the door to further shrinkage of 2D device footprint.

cond-mat.mtrl-sci↗

arXiv2019

COUNTDOWN Slack: a Run-time Library to Reduce Energy Footprint in Large-scale MPI Applications

The power consumption of supercomputers is a major challenge for system owners, users, and society. It limits the capacity of system installations, it requires large cooling infrastructures, and it is the cause of a large carbon footprint. Reducing power during application execution without changing the application source code or increasing time-to-completion is highly desirable in real-life high-performance computing scenarios. The power management run-time frameworks proposed in the last decade are based on the assumption that the duration of communication and application phases in an MPI application can be predicted and used at run-time to trade-off communication slack with power consumption. In this manuscript, we first show that this assumption is too general and leads to mispredictions, slowing down applications, thereby jeopardizing the claimed benefits. We then propose a new approach based on (i) the separation of communication phases and slack during MPI calls and (ii) a timeout algorithm to cope with the hardware power management latency, which jointly makes it possible to achieve performance-neutral power saving in MPI applications without requiring labor-intensive and risky application source code modifications. We validate our approach in a tier-1 production environment with widely adopted scientific applications. Our approach has a time-to-completion overhead lower than 1%, while it successfully exploits slack in communication phases to achieve an average energy saving of 10%. If we focus on a large-scale application runs, the proposed approach achieves 22% energy saving with an overhead of only 0.4%. With respect to state-of-the-art approaches, COUNTDOWN Slack is the only that always leads to an energy saving with negligible overhead (<3%).

cs.DC↗

arXiv2019

Lithiation-delithiation cycles of amorphous Si nanowires investigated by molecular dynamics simulations

The atomistic mechanisms during lithiation and delithiation of amorphous Si nanowires ($a$-SiNW) have been investigated over cycles by molecular dynamics simulations. First, the Modified Embedded Atom Method (MEAM) potential from Cui et al. [J. Power Sources. 2012, (207) 150] has been further optimized on static (Li$_x$Si alloy phases and point defect energies) and on dynamic properties (Li diffusion) to reproduce the lithiation of small crystalline Si nanowires calculated at the {\it ab initio} level. The lithiation of $a$-SiNW reveals a two-phase process of lithiation with a larger diffusion interface compared to crystalline Si lithiation. Compressive axial stresses are observed in the amorphous Si$_x$Li alloy outer shell. They are easily released thanks to the soft glassy behavior of the amorphous alloy. Conversely, in crystalline SiNW, the larger stress in the narrow crystalline lithiated interface is hardly released and requires a phase transformation to amorphous to operate, which delays the lithiation. The history of the charge-discharge cycles as well as the temperature appear as driving forces for phase transformation from amorphous Li$_x$Si alloy to the more stable crystalline phase counterpart. Our work suggest that a full delithiation could heal the SiNWs to improve the life cycles of Li-ion batteries with Si anode.

cond-mat.mtrl-sci↗

arXiv2019

Light from van der Waals quantum tunneling devices

The understanding of and control over light emission from quantum tunneling has challenged researchers for more than four decades due to the intricate interplay of electrical and optical properties in atomic scale volumes. Here we introduce a device architecture that allows for the disentanglement of electronic and photonic pathways - van der Waals quantum tunneling devices. The electronic properties are defined by a stack of two-dimensional atomic crystals whereas the optical properties are controlled via an external photonic architecture. In van der Waals heterostructure made of gold, hexagonal boron nitride and graphene we find that inelastic tunneling results in the emission of photons and surface plasmon polaritons. By coupling these heterostructures to optical nanocube antennas we achieve resonant enhancement of the photon emission rate in narrow frequency bands by four orders of magnitude. Our results lead the way towards a new generation of nanophotonic devices that are driven by quantum tunneling.

cond-mat.mes-hall↗

arXiv2018

Single Atom Plasmonic Switch

The atom sets an ultimate scaling limit to Moores law in the electronics industry. And while electronics research already explores atomic scales devices, photonics research still deals with devices at the micrometer scale. Here we demonstrate that photonic scaling-similar to electronics-is only limited by the atom. More precisely, we introduce an electrically controlled single atom plasmonic switch. The switch allows for fast and reproducible switching by means of the relocation of an individual or at most -- a few atoms in a plasmonic cavity. Depending on the location of the atom either of two distinct plasmonic cavity resonance states are supported. Experimental results show reversible digital optical switching with an extinction ration of 10 dB and operation at room temperature with femtojoule (fJ) power consumption for a single switch operation. This demonstration of a CMOS compatible, integrated quantum device allowing to control photons at the single-atom level opens intriguing perspectives for a fully integrated and highly scalable chip platform -- a platform where optics, electronics and memory may be controlled at the single-atom level.

physics.optics↗

arXiv2018

Ab-initio quantum transport simulation of self-heating in single-layer 2-D materials

Through advanced quantum mechanical simulations combining electron and phonon transport from first-principles self-heating effects are investigated in n-type transistors with a single-layer MoS2, WS2, and black phosphorus as channel materials. The selected 2-D crystals all exhibit different phonon-limited mobility values, as well as electron and phonon properties, which has a direct influence on the increase of their lattice temperature and on the power dissipated inside their channel as a function of the applied gate voltage and electrical current magnitude. This computational study reveals (i) that self-heating plays a much more important role in 2-D materials than in Si nanowires, (ii) that it could severely limit the performance of 2-D devices at high current densities, and (iii) that black phosphorus appears less sensitive to this phenomenon than transition metal dichalcogenides.

cond-mat.mes-hall↗

arXiv2018

Pushing Back the Limit of Ab-initio Quantum Transport Simulations on Hybrid Supercomputers

The capabilities of CP2K, a density-functional theory package and OMEN, a nano-device simulator, are combined to study transport phenomena from first-principles in unprecedentedly large nanostructures. Based on the Hamiltonian and overlap matrices generated by CP2K for a given system, OMEN solves the Schroedinger equation with open boundary conditions (OBCs) for all possible electron momenta and energies. To accelerate this core operation a robust algorithm called SplitSolve has been developed. It allows to simultaneously treat the OBCs on CPUs and the Schroedinger equation on GPUs, taking advantage of hybrid nodes. Our key achievements on the Cray-XK7 Titan are (i) a reduction in time-to-solution by more than one order of magnitude as compared to standard methods, enabling the simulation of structures with more than 50000 atoms, (ii) a parallel efficiency of 97% when scaling from 756 up to 18564 nodes, and (iii) a sustained performance of 15 DP-PFlop/s.

physics.comp-ph↗

arXiv2018

Thermal Properties of Disordered LixMoS2 - An Ab-Initio Study

An atomistic study of the thermal properties of lithiated molybdenum disulfide (MoS2) is presented and an explanation for the experimentally determined anisotropic behavior of the in- and through-plane thermal conductivity is proposed. Configurations with different levels of lithium concentration are simulated using Density Functional Theory (DFT) and their structural, electronic, and thermal transport properties are evaluated as a function of the degree of lithiation. Comparisons between regular and disordered lithium distributions as well as to the experimental data reveal that the measured ratio of the in-plane to through-plane thermal conductivity can only be reproduced if the disorder of the lithium ions is taken into account. These results suggest that it is not only possible to modify the thermal properties of MoS2 by changing the degree of lithiation, but also by controlling its level of disorder.

cond-mat.mtrl-sci↗

arXiv2018

Electronic Properties of Lithiated SnO-based Anode Materials

In this paper we use an ab-initio quantum transport approach to study the electron current flowing through lithiated SnO anodes for potential applications in Li-ion batteries. By investigating a set of lithiated structures with varying lithium concentrations, it is revealed that LixSnO can be a good conductor, with values comparable to bulk $β$-Sn and Li. A deeper insight into the current distribution indicates that electrons preferably follow specific trajectories, which offer superior conducting properties than others. These channels have been identified and it is shown here how they can enhance or deteriorate the current flow in lithiated anode materials.

cond-mat.mtrl-sci↗

arXiv2017

Ab-initio Modeling of CBRAM Cells: from Ballistic Transport Properties to Electro-Thermal Effects

We present atomistic simulations of conductive bridging random access memory (CBRAM) cells from first-principles combining density-functional theory and the Non-equilibrium Green's Function formalism. Realistic device structures with an atomic-scale filament connecting two metallic contacts have been constructed. Their transport properties have been studied in the ballistic limit and in the presence of electron-phonon scattering, showing good agreement with experimental data. It has been found that the relocation of few atoms is sufficient to change the resistance of the CBRAM by 6 orders of magnitude, that the electron trajectories strongly depend on the filament morphology, and that self-heating does not affect the device performance at currents below 1 $μ$A.

cond-mat.mes-hall↗

arXiv2016

Influence of the Surface of a Nanocrystal on its Electronic and Phononic Properties

Over the past thirty years, it has been consistently observed that surface engineering of colloidal nanocrystals (NC) is key to their performance parameters. In the case of lead chalcogenide NCs, for example, replacing thiols with halide anion surface termination has been shown to increase power conversion efficiency in NC-based solar cells. To gain insight into the origins of these improvements, we perform ab initio molecular dynamics (AIMD) on experimentally-relevant sized lead sulfide (PbS) NCs constructed with thiol or Cl, Br, and I anion surfaces. The surface of both the thiol- and halide-terminated NCs exhibit low and high-energy phonon modes with large thermal displacements not present in bulk PbS; however, halide anion surface termination reduces the overlap of the electronic wavefunctions with these vibration modes. These findings suggest that electron-phonon interactions will be reduced in the halide terminated NCs, a conclusion that is supported by analyzing the time-dependent evolution of the electronic energies and wavefunctions extracted from the AIMD. This work explains why electron-phonon interactions are crucial to charge carrier dynamics in NCs and how surface engineering can be applied to systematically control their electronic and phononic properties. Furthermore, we propose that the computationally efficient approach of gauging electron-phonon interaction implemented here can be used to guide the design of application-specific surface terminations for arbitrary nanomaterials.

cond-mat.mtrl-sci↗

arXiv2015

Diffusion Mechanisms in Li$_{0.5}$CoO$_2$ -- A Computational Study

An atomistic study of the order-effect occurring in Li$_{x}$CoO$_{2}$ at $x=0.5$ is presented and an explanation for the computationally and experimentally observed dip in the Li diffusivity is proposed. Configurations where a single half-filled Li layer arranged in either a linear or a zig-zag pattern are simulated. It is found that the lowest energy phase is the zig-zag pattern rather than the linear arrangement that currently is considered to be of lowest energy. Atomic interactions are modeled at the DFT level of accuracy and energy barriers for Li-ion diffusion are determined from searches for first order saddle points on the resulting potential energy surface. The determined saddle points reveal that the barriers for diffusion parallel and perpendicular to the zig-zag phase differ significantly and explain the observed dip in diffusivity.

cond-mat.mtrl-sci↗

arXiv2015

Compositional bowing of band energies and their deformation potentials in strained InGaAs ternary alloys: a first-principles study

Using first-principles calculations, we show that the conduction and valence band energies and their deformation potentials exhibit a non-negligible compositional bowing in strained ternary semiconductor alloys such as InGaAs. The electronic structure of these compounds has been calculated within the framework of local density approximation and hybrid functional approach for large cubic supercells and special quasi-random structures, which represent two kinds of model structures for random alloys. We find that the predicted bowing effect for the band energy deformation potentials is rather insensitive to the choice of the functional and alloy structural model. The direction of bowing is determined by In cations that give a stronger contribution to the formation of the In$_{x}$Ga$_{1-x}$As valence band states with $x\gtrsim 0.5$, compared to Ga cations.

cond-mat.mtrl-sci↗

arXiv2015

The reversible lithiation of SnO: a three-phase process

A high reversible capacity is a key feature for any rechargeable battery. In the lithium-ion battery technology, tin-oxide anodes do fulfill this requirement, but a fast loss of capacity hinders a full commercialization. Using first-principles calculations, we propose a microscopic model that sheds light on the reversible lithiation/delithiation of SnO and reveals that a sintering of Sn causes a strong degradation of SnO-based anodes. When the initial irreversible transformation ends, active anode grains consist of Li-oxide layers separated by Sn bilayers. During the following reversible lithiation, the Li-oxide undergoes two phase transformations that give rise to a Li-enrichment of the oxide and the formation of a layered SnLi composite. We find that the model-predicted anode volume expansion and voltage profile agree well with experiment, and a layered anode grain is highly-conductive and has a theoretical reversible capacity of 4.5 Li atoms per a SnO host unit. The model suggests that the grain structure has to remain layered to sustain its reversible capacity and a thin-film design of battery anodes could be a remedy for the capacity loss.

cond-mat.mtrl-sci↗