A total-electron-yield technique is described in which near-surface extended x-ray-absorption 6ne-structure (EXAFS) data are obtained from direct measurements of specimen current. Experiments with several model systemsamorphous germanium and crystalline germanium, nickel, and cobalt; and arsenic ion implanted into silicondemonstrate that this technique can reproduce EXAFS X(k) functions obtained from transmission and fluorescence measurements. Experiments also reveal that EXAFS amplitudes from total-electron-yield data can be 5-10% smaller than those from transmission measurements for samples where the very-near-surface structure, at depths of tens to hundreds of angstroms, divers from the bulk structure. Measurements with buried layers con6rm that the sampling depth for this total-electron-yield technique is determined primarily by the penetration ranges of Auger electrons emitted from the absorbing atoms. For the model systems listed above, LMM Auger electrons have ranges of hundreds of angstroms and ELL Auger electrons have ranges of thousands of' angstroms. Expressions are derived for the sampling depth for total-electron-yield EXAFS experiments. The total-electron-yield technique described here is particularly useful for studying impurities within a few thousand angstroms of the surface of single crystals, where Bragg difFraction complicates the use of fluorescence measurements.sult of contributions of inelastically scattered photoelectrons. Guo and denBoer' and Lytle et al. ' used a gas-Qow electron detector' to measure electron total yield EXAFS signals for Ni, Cu, Fe, and Cr K edges. They observed considerable differences for Ni, Cu"and Fe between the amplitude values obtained from transmission measurements and the corresponding values from a gasflow detector. The range of subsurface sensitivity for total-electron-yield EXAFS has been estimated to be 1000 A from measurements for Cu (E edge), less than 390 A for A1203 (Al K edge), and 700-1000 A for GaAs (E edge). '6The present paper describes a method for totalelectron-yield measurements" ' mhich is particularly mell suited for near-surface EXAFS from single-crystal samples, where Bragg difFraction often makes fluorescence measurements diScult or impossible. The method reported here detects the total yield of electrons from the sample by measuring the sample current. Following descriptions of the technique and its applications, comparisons of total-electron-yield and transmission or fluorescence EXAFS measurements are given, to demonstrate the reliability of total-electron-yield measurements, particularly in light of previously reported di5culties with EXAFS amplitudes from total-electronyield measurements. ' ' ' Measurements and calculations are then given concerning the range of subsurface sensitivity for total-electron-yield EXAFS measurements.Results from the present study are also compared with those of previous investigations. 37 24SO

We report on the implementation of crystal ion slicing in lithium niobate (LiNbO3). Deep-ion implantation is used to create a buried sacrificial layer in single-crystal c-cut poled wafers of LiNbO3, inducing a large etch selectivity between the sacrificial layer and the rest of the sample. 9-μm-thick films of excellent quality are separated from the bulk and bonded to silicon and gallium arsenide substrates. These single-crystal films have the same room-temperature dielectric and pyroelectric characteristics, and ferroelectric transition temperature as single-crystal bulk. A stronger high-temperature pyroelectric response is found in the films.

No abstract

Electromigration-induced stress distributions in 200 μm long, 10 μm wide aluminum conductor lines in 1.5 μm SiO2 passivation layers have been investigated in real time using synchrotron-based white-beam x-ray microdiffraction. The results show that a steady-state linear stress gradient along the length of the line developed within the first few hours of electromigration and that the stress gradient could be manipulated by controlling the magnitude and the direction of the current flow. From the current density dependence of the steady-state stress gradient, the effective valence Z* was determined to be 1.6 at 260 °C. From the time dependence of the transient-state stress gradient, the effective grain boundary diffusion coefficient Deff was estimated to be 8.2×10−11 cm2/s at 260 °C using Korhonen’s stress evolution model [M. A. Korhonen, P. Bo/rgesen, K. N. Tu, and C.-Y. Li, J. Appl. Phys. 73, 3790 (1993)]. Both Z* and Deff values are in good agreement with the previously reported values.

No abstract