AVS 50th International Symposium
    Applied Surface Science Thursday Sessions
       Session AS-ThA

Invited Paper AS-ThA5
Photon, Electron, and Ion Spectroscopies Applied to Thin Strained Si Films

Thursday, November 6, 2003, 3:20 pm, Room 324/325

Session: Electron and Photon Spectroscopies
Presenter: S. Zollner, Motorola
Authors: S. Zollner, Motorola
R. Liu, Motorola
M. Canonico, Motorola
M. Kottke, Motorola
Q. Xie, Motorola
S. Lu, Motorola
M. Sadaka, Motorola
T. White, Motorola
A. Barr, Motorola
B.-Y. Nguyen, Motorola
S. Thomas, Motorola
C.S. Cook, Arizona State University
A. Volinsky, Motorola
Correspondent: Click to Email

Applied thin-film spectroscopies can be classified by the primary (incident) and secondary (scattered, specularly reflected, diffracted) particle (photon, electron, ion) and by the energy loss (elastic, inelastic) of the interaction. The term spectroscopy implies that the energy of the primary and/or secondary particle is well-known, measured, or varied. We describe applications of various thin-film spectroscopies to thin (15-20 nm) Si layers under tensile biaxial stress, grown pseudomorphically on thick relaxed Si@sub 1-x@Ge@sub x@ buffer layers. Such Si layers have higher electron and hole mobilities than regular bulk Si and are therefore considered for next-generation CMOS technologies. Particular emphasis is placed on the physical mechanism of each spectroscopy technique, the results obtained (thickness, composition, stress), and the potential shortcomings. Specifically, we use UV Raman spectroscopy (325 nm laser excitation) to determine the strain in the Si layer from the frequency shift of the Si-Si lattice vibration. The contributions of the tensile hydrostatic strain and the compressive shear strain to the E1 peak shift (near 3.4 eV) cancel almost exactly, therefore ellipsometry cannot be used to determine strain. However, the piezo-optical effects are significant enough to make ellipsometry unreliable for the determination of the Si thickness or the Ge content of the alloy buffer. Piezo-optical effects are not an issue for x-ray reflectivity, since the dielectric constant of most materials is very close to unity for Cu K@sub alpha@ radiation, but surface roughness can be a problem. We also discuss results from secondary ion mass spectrometry and Auger electron spectrometry to determine composition of the layers.