Files in this item



application/pdfAndrey_Serov.pdf (5MB)
(no description provided)PDF


Title:Electrical and thermal transport in 2D materials: role of environment and imperfections
Author(s):Serov, Andrey
Director of Research:Pop, Eric
Doctoral Committee Chair(s):Pop, Eric
Doctoral Committee Member(s):Leburton, Jean-Pierre; Rosenbaum, Elyse; Ravaioli, Umberto
Department / Program:Electrical & Computer Eng
Discipline:Electrical & Computer Engr
Degree Granting Institution:University of Illinois at Urbana-Champaign
electrical transport
thermal transport
high-field transport
Abstract:Two-dimensional (2D) materials and devices such as graphene and MoS2 are of interest from a fundamental point of view, as well as for their practical applications in nanoelectronics. The role of interfaces and imperfections is an important topic for such nanoscale devices and materials due to their reduced dimensions and large surface-to-volume ratios. In this dissertation we address some of the factors limiting the electrical and thermal transport in 2D materials such as graphene and MoS2. First we investigate the influence of grain boundaries (GBs), line defects (LDs), and chirality on thermal transport in graphene using the nonequilibrium Green’s functions method. At room temperature the ballistic thermal conductance is ∼4.2 GWm−2K−1 and single GBs or LDs yield transmission of 50% to 80% of this value. We find that LDs with carbon atom octagon defects have lower thermal transmission than GBs with pentagon and heptagon defects. We apply our findings to study the thermal conductivity of polycrystalline graphene for practical applications, and find that the type and size of GBs play an important role when grain sizes are smaller than a few hundred nanometers. Then we investigate electrical transport in graphene supported on various dielectrics (SiO2, BN, Al2O3, HfO2) through a hydrodynamic model that includes self-heating and thermal coupling to the substrate, scattering with ionized impurities, graphene phonons, and dynamically screened interfacial plasmon-phonon (IPP) modes. We discover that while low-field transport is largely determined by impurity scattering, high-field transport is defined by scattering with IPP modes, and by a smaller contribution of graphene intrinsic phonons. We also find that lattice heating can lead to negative differential drift velocity (with respect to the electric field), which can be controlled by changing the underlying dielectric thermal properties or thickness. Graphene on BN exhibits the largest high-field drift velocity, while graphene on HfO2 has the lowest one due to strong influence of IPP modes. Moving from 2D material to device analysis, we present a simulation framework for graphene transistors, which includes quantum capacitance, generalized diffusion, carrier density dependent saturation velocity, and device electrostatics. We investigate how these graphene-specific effects change the results of conventional drift-diffusion simulation both in low and high drain bias regimes. Using our simulation methodology we also inspect the electronhole asymmetry in current vs. gate voltage (I-VG) characteristics, which is often attributed to differences in electron and hole mobility. However, we find that we can quantitatively understand such experimental results by simply accounting for chemical doping under the contacts and capacitive graphene-contact coupling, without making artificial assumptions about electron and hole mobility. Finally, we theoretically investigate electron transport in transistors based on another 2D material (few-layer MoS2) and compare our results to experimental data. We show that both a two-valley conduction band structure (K and Q) and device self-heating should be taken into account to reproduce the negative differential conductance experimentally observed at lower temperatures (e.g., < 150 K). We also demonstrate that the transport involving two valleys is necessary to describe the strong temperature dependence of mobility. Calibrating both low- and high-field transport models to experimental data, we discover that the Q-valley of MoS2 is approximately 130 meV above the K-valley. These results shed important insights into electrical current and heat flow in novel 2D materials and devices, which are of relevance for all their future applications in nano- and opto-electronics. Overall, methods and results presented in this dissertation can be extended for characterization and analysis of other 2D materials beyond graphene and MoS2, which are only starting to find their way into research and development.
Issue Date:2015-01-21
Rights Information:Copyright 2014 Andrey Serov
Date Available in IDEALS:2015-01-21
Date Deposited:2014-12

This item appears in the following Collection(s)

Item Statistics