Artem Fediai

Towards a multiscale modeling framework for metal-CNT interfaces

This paper gives a short overview on our recent investigations towards a multiscale modeling and simulation framework for metal-CNT interfaces. We employ three simulation approaches with well defined interfaces. For the simulation at device level we make use of a recently developed wave-function based effective-mass Schrödinger-Poisson solver which employs a hetero-junction like contact model to capture the physics in the contact region where the CNT is embedded into metal. The required model parameters are adjusted to TB and DFT simulation results. A comparison with experimental data for a short channel device shows the applicability of the proposed approach.

In Situ Electron Driven Carbon Nanopillar-Fullerene Transformation through Cr Atom Mediation

The promise of sp2 nanomaterials remains immense, and ways to strategically combine and manipulate these nanostructures will further enhance their potential as well as advance nanotechnology as a whole. The scale of these structures requires precision at the atomic scale. In this sense electron microscopes are attractive as they offer both atomic imaging and a means to structurally modify structures. Here we show how Cr atoms can be used as physical linkers to connect carbon nanotubes and fullerenes to graphene. Crucially, while under electron irradiation, the Cr atoms can drive transformations such as catalytic healing of a hole in graphene with simultaneous transformation of a single wall carbon nanotube into a fullerene. The atomic resolution of the electron microscopy along with density functional theory based total energy calculations provide insight into the dynamic transformations of Cr atom linkers. The work augments the potential of transmission electron microscopes as nanolaboratories.

Impact of incomplete metal coverage on the electrical properties of metal-CNT contacts: A large-scale ab initio study

Using a dedicated combination of the non-equilibrium Green function formalism and large-scale density functional theory calculations, we investigated how incomplete metal coverage influences two of the most important electrical properties of carbon nanotube (CNT)-based transistors: contact resistance and its scaling with contact length, and maximum current. These quantities have been derived from parameter-free simulations of atomic systems that are as close as possible to experimental geometries. Physical mechanisms that govern these dependences have been identified for various metals, representing different CNT-metal interaction strengths from chemisorption to physisorption. Our results pave the way for an application-oriented design of CNT-metal contacts.

The modular approach enables a fully ab initio simulation of the contacts between 3D and 2D materials

Up to now, the electrical properties of the contacts between 3D metals and 2D materials have never been computed at a fully ab initio level due to the huge number of atomic orbitals involved in a current path from an electrode to a pristine 2D material. As a result, there are still numerous open questions and controversial theories on the electrical properties of systems with 3D/2D interfaces-for example, the current path and the contact length scalability. Our work provides a first-principles solution to this long-standing problem with the use of the modular approach, a method which rigorously combines a Green function formalism with the density functional theory (DFT) for this particular contact type. The modular approach is a general approach valid for any 3D/2D contact. As an example, we apply it to the most investigated among 3D/2D contacts-metal/graphene contacts-and show its abilities and consistency by comparison with existing experimental data. As it is applicable to any 3D/2D interface, the modular approach allows the engineering of 3D/2D contacts with the pre-defined electrical properties.

Multi-scale modeling of metal-CNT interfaces

The authors studied the impact of contact materials on CNTFET behavior using multiscale modeling and simulation framework. A strong correlation between metal-CNT coupling strength, contact length and contact resistance was found. The atomistic simulation was used to adjust the contact model used within the transport studies at the device level.

Electrical characteristics of the carbon nanotube field-effect transistors with extended contacts obtained within ab-initio based model

In our previous work the model of the electron transport in CNTFETs with extended contacts was developed for equilibrium case using combination of the density-functional theory and equilibrium Green functions formalism. Here we extrapolate it to the case of the arbitrary biased CNTFET using simplest thinkable assumptions, which allows us to calculate drain current as a function of the gate and drain potentials. For Al and Pd electrodes we show qualitative agreement of our model with the existing experimental results both in terms of polarity of the device (p- or n-type FET) and dependence on the contact length. Most of the prediction justified by experimental data were done at ab-initio level for the first time.

Contact properties of ultrasmall carbon nanotube transistors from Ab-initio

Based on large-scale ab-initio calculations contact properties of carbon nanotubes field effect transistors with 9-nm channel were analyzed. We have found that the internal part of the nanotube is metalized, while the uncovered CNT portion is p-doped due to the influence of the metal. Our findings can explain unique scaling of the CNT-FETs based on first-principles.