Phillip J. Oldiges

Specific contact resistivity of n-type Si and Ge M-S and M-I-S contacts

We have theoretically investigated the specific contact resistivity of n-type Si and Ge metal-insulator-semiconductor contacts with various insulating oxides. We have found a significant reduction of the contact resistivity for both Si and Ge with an insertion of insulators at low and moderate donor doping levels. However, at the higher doping levels (>1020 cmu-3), the reduction of the contact resistivity is negligible and the contact resistivity increases as the insulator thickness increase. Thus, we have shown that the lowest possible contact resistivity can be achieved with the metal-semiconductor contact with highest possible activated doping density.

Technology modeling for emerging SOI devices

New physical models, algorithms, and parameters are needed to accurately model emerging silicon-on-insulator (SOI) devices. We discuss key modeling tools and methodologies used to support research, development, and manufacturing of emerging SOI device technology. Although commercial TCAD tools are available, new physical models, parameters, and algorithms are needed to study these novel device structures.

Critical analysis of 14nm device options

Modeling challenges and solutions for silicon based high performance device options at the 14nm node are presented. A variety of devices are being considered, using a variety of methods to analyze the devices objectively. Partially depleted silicon on insulator (PDSOI) devices are compared against extremely thin (ETSOI) and FinFET devices.

Impact of substrate bias on GIDL for thin-BOX ETSOI devices

We present a detailed analysis of substrate bias (Vbb) impact on gate induced drain leakage (GIDL) for thin-BOX extremely thin silicon-on-insulator (ETSOI) with BOX thickness (TBOX) ranging from 10 to 50 nm and inversion layer thicknesses (TINV) ranging from 1.1 to 1.3 nm. The GIDL behavior for thin-BOX under various substrate biases (Vbb) and partially depleted SOI (PDSOI) devices with different body doping are compared.

On the optimal shape and location of silicided source and drain contacts

A detailed simulation and analysis of the source/drain resistance is performed. It is shown that the placement and depth of silicide regions can have a strong influence on the total source/drain resistance. Simulations further show that moving the silicided regions closer to the channel of a device will not necessarily decrease source/drain resistance, and may actually cause the resistance to increase. Lumped contact resistance, distributed resistance, Schottky contact models, and a new local distributed resistance model are compared.