Projets de recherche au laboratoire

Presentation :

Ce projet vise à développer une nouvelle génération d'outils de simulation pour le transport quantique, en traitant les problèmes émergeant à l'échelle atomique; condition capitale pour la compréhension des dispositifs semi-conducteurs de faibles dimensions.

Sur la base d'un formalisme de fonction de Green éprouvé, les expertises complémentaires des partenaires, allant d'approches ab initio les plus pointues, à des modèles sophistiqués de liaisons fortes et de masse effective, seront combinées afin de développer des codes multi-échelles 3D de simulation de dispositifs. Ils permettront une analyse détaillée des propriétés de transport quantique sur des modèles réalistes de transistors: les MOSFETs ultimes obtenus par approches top-down et les nanofils semi-conducteurs fabriqués par croissance CVD (bottom-up).

Via une stratégie graduelle des schémas ab initio vers ceux de la masse effective, l’impact des mécanismes intrinsèques de diffusion sur les caractéristiques de courant sera étudié par un traitement auto-cohérent rigoureux de l’électrostatique. Simultanément, les effets quantiques cohérents à basse température, en particulier les longueurs de cohérence de spin, seront explorés dans la perspective d'une spintronique à base de nanofils semi-conducteurs.

Le projet QUANTAMONDE offre une occasion pertinente de fusionner des domaines d'expertises de différents groupes de recherche qui conduira à des avancées majeures dans la compréhension des nano-transistors.

This proposal focuses on the development of a new generation of quantum transport simulation tools tackling with atomic scale issues which become of paramount importance for understanding low dimensional semiconductor devices.

On the basis of a common Green’s function formalism, the complementary expertises of partners, ranging from state-of-the-art ab initio approach, to sophisticated tight-binding and effective mass models will be combined to develop multiscale 3D device simulation codes. It will allow an in-depth analysis of quantum transport properties and field effect transistor characteristics in realistic models of both top-down ultimate MOSFETs and bottom-up CVD-grown semiconductor nanowires.

Using an upscaling strategy from ab initio to effective mass schemes, the impact of intrinsic scattering mechanisms on current characteristics will be investigated, based on a rigorous self-consistent treatment of electrostatics. Simultaneously, the quantum coherent effects in the low temperature regimes will be explored, with a focus on the spin coherence lengths in the perspective of semiconductor nanowire-based spintronics.

QUANTAMONDE project offers a natural means for fusing the expertise of different research groups and leading to breakthroughs in the understanding of nano-transistors.

Partners :

Partner 1 - Coordinator et scientific responsible of the project
Marc Bescond, IM2NP Dispositifs Ultimes sur Silicium
IM2NP UMR CNRS 6242 • Bât. IRPHE • 49, rue Joliot-Curie • BP 146 • Technopôle de Château-Gombert – 13384 Marseille Cedex 13 • France

www.im2np.fr

Tel : +33 (0)4 96 13 97 31 • Fax : +33 (0)4 96 13 97 09
Mail : marc.bescond@im2np.fr

Partner 2
Yann-Michel Niquet
INAC CEA Grenoble • 17 rue des Martyrs • Grenoble Cedex 9 • France
Tel : +33 (0)4 38 78 43 22 • Fax : 04 38 78 51 97
Mail : yann-michel.niquet@cea.fr

www-drfmc.cea.fr

Partner 3
Christophe Delerue, IEMN, Groupe de Physique Théorique
IEMN UMR CNRS 8520 • Département ISEN • 41 Boulevard Vauban • 59046 Lille Cedex • France
Tel : +33 (0)3 20 30 40 53 • Fax : +33 (0)3 20 30 40 51
Mail : Christophe.delerue@isen.fr

www.iemn.univ-lille1.fr

Partner 4
Marco Pala, IMEP-LAHC
IMEP-LAHC UMR CNRS 5130 • MINATEC-INPG • 3 Parvis Louis Néel • BP 257 • 38016 Grenoble • France
Tel : +33 (0)4 56 52 95 49 • Fax : +33 (0)4 56 52 95 01
Mail : pala@minatec.inpg.fr

www.imep.enserg.fr

Partner 5
Xavier Blase, Institut Néel
Institut Néel UMR 2940 • CNRS 25 rue des Martyrs • BP 166 • 38042 Grenoble • France
Tel : +33 (0)4 76 88 74 68 •
Mail : xavier.blase@grenoble.cnrs.fr

neel.cnrs.fr

Partner 6
Clément Tavernier, STMicroelectronics
STMicroelectronics SA • 850 rue Jean Monnet • 38926 Crolles Cedex • France
Tel : +33 (0)4 76 92 64 89 • Fax : +33 (0)4 76 92 57 32
Mail : clement.tavernier@st.com

www.st.com

Scientific description of the project :

TASK 1 – Mechanical strain and electron-phonon coupling (Leader: C. Delerue, IEMN)

Transport properties in long channel transistors are mainly determined by charge carrier mobilities. Most of the works on nanotransistors use the drift diffusion equations which are only valid for long channel MOSFETs or fully ballistic calculations based on the Schrödinger equation. A detailed understanding of scattering mechanisms is crucial for determining the on-current of nanotransistors. This task aims to investigate fundamental questions on the physical mechanisms determining nanodevice mobilities from ab initio methods and to apply them to transport calculations of realistic structures. Two sub-tasks are proposed addressing the role of strain and that of electron-phonon scattering. Indeed, phonon scattering and strain are correlated since a model which correctly describes the strain should also treat efficiently the electron-phonon coupling which represents a displacement of the atoms around their equilibrium positions and are therefore addressed in the same task.

Task 1.1: Mechanical strain

Task 1.2: Electron-phonon coupling

Task 1.3: Implementation for quantum transport in nano-transistors

TASK 2 – Sources of intrinsic fluctuations and electrostatics effects (Leader: Y.M. Niquet, CEA-INAC)

Understanding the effects of the dielectric environment on the transport properties of nanodevices has become a fundamental issue. The dielectric environment of these devices indeed becomes more and more inhomogeneous, as the typical feature shrinks smaller and new materials are introduced. For instance, the channel can be surrounded by low-k oxides, high-k dielectrics, or (semi-)metallic contacts and gates. Most of the physics of the nanodevices is actually driven by electrostatic and electrodynamic effects. At variance with quantum confinement, which is mostly a “local” issue, these effects are long-ranged. The dielectric environment can actually alter the electrical behavior of the devices in many ways:

l It directly influences the gate efficiency.

l It is responsible for self-energy corrections to the quasiparticle band structure (a.k.a. image charge corrections).

l It affects the potential and binding energies of dopant impurities.

l It might alter the band offset between two materials, that has a long-range electrostatic contribution due to the transfer of charge at the interfaces.

These electrostatic effects might put severe constraints on the design of the devices, but can also be seen as an opportunity to tune their functionalities. Even in MOS devices the effect of the dielectric mismatch between Si and SiO₂ on the electron self-energies (image charges) seems little understood.

Task 2.1: Dielectric confinement: self-energies and impurity potentials

Task 2.2: The band offset problem

Task 2.3: Implementation in NEGF solvers and applications

TASK 3 – Coherent transport and Spin-dependent transport (Leader: M. Pala, IMEP-LAHC)

Quantum transport under the influence of an external magnetic field can furnish important information on the transport properties of nanodevices. It turns out that at large magnetic field magneto-conductivity obeys the semiclassical Drude formula from which it is possible to extract the charge carrier mobilities of the system. By performing full 3D magneto-transport simulation based on a recursive Green’s functions code the mobility due to the surface roughness or other spatial fluctuations can be obtained even in the subthreshold regime where the absence of significant charge in the channel prevents direct mobility evaluation from I-V curves.

On the other hand, at low magnetic field quantum corrections due to electron-phase coherence like weak or anti-weak localization phenomena are important and can be can be related to fundamental physical transport properties of the system such as the elastic and inelastic mean free paths, the quantum coherent length, and the spin-relaxation length, in situations where spin-dependent transport becomes relevant.

Task 3.1: Magneto- Transport length scales

Task 3.2: Spin-Transport

TASK 4 – Metal / Semiconductor contacts (Leader: F. Triozon and S. Roche, CEA-INAC)

In this task, the problem of metal / semiconductor contacts will be treated by three complementary approaches. Contrary to metal/nanotube contacts, the bulk limit (planar interface) exists and is well understood for some particular metal disilicides. These different approaches will be exploited to better analyze and understand the changes from planar interfaces to nanowire hetero-junctions.

Task 4.1: Compact model

Task 4.2: Ab initio validation

Task 4.3: NEGF transport calculations