Design considerations and emerging challenges for nanotube-, nanowire-, and negative capacitor-field effect transistors

Abdul Wahab MD., Purdue University

Abstract

As the era of classical planar metal-oxide-semiconductor field-effect transistors (MOSFETs) comes to an end, the semiconductor industry is beginning to adopt 3D device architectures, such as FinFETs, starting at the 22 nm technology node. Since physical limits such as short channel effect (SCE) and self-heating may dominate, it may be difficult to scale Si FinFET below 10 nm. In this regard, transistors with different materials, geometries, or operating principles may help. For example, gate has excellent electrostatic control over 2D thin film channel with planar geometry, and 1D nanowire (NW) channel with gate-all-around (GAA) geometry to reduce SCE. High carrier mobility of single wall carbon nanotube (SWNT) or III-V channels may reduce VDD to reduce power consumption. Therefore, as channel of transistor, 2D thin film of array SWNTs and 1D III-V multi NWs are promising for sub 10 nm technology nodes. In this thesis, we analyze the potential of these transistors from process, performance, and reliability perspectives. For SWNT FETs, we discuss a set of challenges (such as how to (i) characterize diameter distribution, (ii) remove metallic (m)-SWNTs, and (iii) avoid electrostatic cross-talk among the neighboring SWNTs), and demonstrate solution strategies both theoretically and experimentally. Regarding self-heating in these new class of devices (SWNT FET and GAA NW FET including state-of-the-art FinFET), higher thermal resistance from poor thermal conducting oxides results significant temperature rise, and reduces the IC life-time. For GAA NW FETs, we discuss accurate self-heating evaluation with good spatial, temporal, and thermal resolutions. The introduction of negative capacitor (NC), as gate dielectric stack of transistor, allows sub 60 mV/dec operation to reduce power consumption significantly. Taken together, our work provides a comprehensive perspective regarding the challenges and opportunities of sub 10 nm technology nodes.

Degree

Ph.D.

Advisors

Alam, Purdue University.

Subject Area

Electrical engineering|Nanotechnology

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS