Click to return to the VINSE homepage

February 21, 2001


Dr. Stefan Zollner
Section Manager, Wireless Technology AnalysisProcess and Materials Characterization Laboratory
Digital DNA Laboratories
Motorolo SPS
"Optical, Vibrational, and Structural Properties of High-k Transition Metal Oxides"

Abstract. According to the International Technology Roadmap for Semiconductors (ITRS), the present structure of CMOS devices using SiO2 gate oxides (maybe lightly nitrided) will change around 2005, when the SiO2 thickness reaches 15–20 Å and the gate length 0.1 mm. Beyond this technology node, the physical thickness of the gate oxide can no longer shrink, since the leakage currents due to quantum-mechanical tunneling would render the devices useless. Instead of reducing the physical thickness, scaling the gate length is accompanied by increasing the low-frequency dielectric constant e0 of the gate material, thus increasing the areal capacitance. Therefore, the industry is searching for novel high-e0 (or high-k) materials.  

These materials need to fulfill a number of requirements: (1) A large band gap with reasonable band offsets compared to the valence and conduction bands of Si. (2) They should be amorphous to prevent diffusion of dopants (from the poly-Si gate to the channel in the Si substrate) along grain boundaries. (3) They need to be thermodynamically stable (particularly towards reaction with Si or O) to allow subsequent CMOS processing. (A simple phase diagram would help.) (4) A large lattice polarizability (e0–e¥–1), which is usually associated with a soft phonon driving a transition between two crystalline phases. No simple requirements! 

Optical spectroscopies (spectroscopic ellipsometry from 300 mm to 9 eV, FTIR transmission and reflection, deep-UV Raman scattering) and x-ray structural analysis are ideally suited for a rapid screening of a large class of materials for potential high-k applications, since they do not require processing for electrical measurements. Ab initio band structure calculations can assist in the interpretation of experimental results. This talk will describe recent work on SrTiO3, various transition metal oxides (TiO2, ZrO2, and HfO2), as well as other classes of emerging materials, both in bulk and thin-film form. Of particular importance is the detection of an amorphous interfacial layer that may form during deposition on Si.

Vanderbilt University