Beam Tilt, Defocus, and Vibration Contribute to HRTEM Gate Dielectric Thickness Measurement Uncertainty

Published

October 16, 2008

Author(s)

David S. Bright, Jake H. Scott

Abstract

Using statistical design-of-experiment (DEX) methods, beam tilt, defocus, and vibration of the microscope were determined to have large effects on the absolute measurement error of gate dielectric thickness measurements by high-resolution transmission electron microscopy (HRTEM). Beam-tilt+defocus and defocus+vibration were also identified as important two-term interactions in the analysis and will need to be considered in any empirical model of the uncertainty budget. These five terms were found to be the most significant among a set formed from 8 direct variables and all of their two-term interactions. Absolute errors were determined by simulating HRTEM micrographs using multislice calculations and physical gate stack models with known true dielectric thicknesses, then processing this known synthetic dataset using the same procedures used for experimental HRTEM micrographs. The physical models used for the simulation consisted of a variable SiO2 film approximately 2 nm thick positioned between two pieces of crystalline Si. This approximation to a gate stack was built atom-by-atom using commercial molecular modeling software supplemented with custom Tcl scripts to assemble the gate structures from simpler primitives. Using established DEX methodologies, a screening experiment based on a 2_IV^(8-4) fractional factorial design was employed to determine which factors had the greatest impact on the absolute error of the thickness measurement.

Citation

Journal of Microscopy-Oxford

Pub Type

Journals

Keywords

absolute error, design of experiment, multislice, screening design, uncertainty budget

Citation

Bright, D. and Scott, J. (2008), Beam Tilt, Defocus, and Vibration Contribute to HRTEM Gate Dielectric Thickness Measurement Uncertainty, Journal of Microscopy-Oxford (Accessed April 25, 2024)