D.J. Larson

On the many advantages of local-electrode atom probes.

Local extraction electrodes offer several crucial advantages for operation of atom probes. Because of the proximity of the local extraction electrode to the specimen, the electric field produced at the specimen apex by a given voltage is enhanced and the voltage required for field evaporation is reduced. In a voltage-pulsed atom probe, the absolute magnitude of the energy uncertainty is correspondingly reduced. High mass resolution (m/deltam > 1000) may therefore be obtained by accelerating the evaporated ions to a greater total potential after the local extraction electrode. The low extraction voltage may also be pulsed rapidly (100 ps rise time) and at high repetition rates (up to 10(5) pulses per second) using currently available solid-state pulsers. Furthermore, a local electrode and intermediate electrodes may be used as optical elements to control the image magnification. All of these benefits may be applied to any type of atom probe. Local-electrode atom probes (LEAP) should be especially advantageous for developing three-dimensional atom probes with high mass resolution and a large field of view. A sample has been developed that consists of many microtips formed on a planar sample using ion beam mask etching. Microtip samples are especially suited to LEAP. Analysis of electrically insulating samples may also be possible with microtip samples in a LEAP. This combination of features suggests flexible, high speed, high mass resolution atom probes that can work with either conventional needle-shaped specimens or the new style of planar microtip specimens.

Magnification and mass resolution in local-electrode atom probes

Abstract The mass resolution of a local-electrode atom probe can be improved by accelerating the evaporated ions to a higher energy in order to reduce the energy-deficit-related dispersion. A simple model of the instrument is developed and used to estimate the effects of this secondary acceleration on image magnification and mass resolution. Effects of non-instantaneous secondary acceleration, variations in the secondary acceleration field distribution and electrode length are evaluated using the model. The addition of an acceleration electrode after the extraction electrode is shown to improve the performance of local-electrode atom probes.

A method for reconstructing and locating atoms on the crystal lattice in three-dimensional atom probe data

Abstract The physical process of field evaporation introduces lateral aberrations in the ion trajectories toward an atom probe detector. In three-dimensional atom probes, these aberrations blur information describing the 3D atomic stacking in the material. This work reports progress that has been made using Fourier transform and pattern recognition techniques to reconstruct an original lattice structure from simulated atom probe data and to subsequently force atoms to the nearest lattice point. Usually Fourier transform techniques are used in image processing to separate image noise from periodic information not to actually shift features in the image. The present technique analyzes a 2D data set and determines the statistically best lattice parameters, lattice orientation, lattice position and site occupation with no free parameters in the analysis. A randomly oriented Gaussian blurring function is used to simulate trajectory aberrations. For 151 atoms originally on a square lattice, atom locating errors are less than 4% when the mean displacement is one quarter of the lattice parameter. The repositioning efficiency increases rapidly with increasing data set size and decreases rapidly with increasing aberration magnitude.

Fabrication of microtips on planar specimens

Abstract Microtips were formed on planar samples using 3 and 6 μm diamond particles as masks for ion beam sputtering at normal incidence. Samples of copper, 304 stainless steel, a metal-oxide-semiconductor structure and a BiSrCaCuO superconductor were studied. It was found that tips could be formed from all materials examined. The tips were many microns tall with a radius of curvature at the apex of less than 100 nm and shank angles down to ∼ 20°. The use of carbon contamination spikes grown in a scanning electron microscope as specific-location masks is considered.

Simulated electron beam trajectories toward a field ion microscopy specimen

Abstract This article explores the conditions under which a directed electron beam originating nearly normal to the specimen axis can be made to impact the near-apex region of a field ion microscopy specimen in a high electric field. Electron trajectories were calculated using a modified Runge-Kutta numerical method. The results indicate that an electron beam can be directed to a specimen under typical field ion microscopy conditions using two methods: by varying initial beam tilt (less than 60 mrad) or by translating the initial beam position relative to the specimen apex (less than 5 mm). The net focusing effect of the high electric field on the electron beam can be treated, to first order, as an astigmatism and may be correctable by a post-lens deflection system.

Optimal field pulsing for atom probes with counter electrodes

Abstract We have calculated energy shifts and energy deficits arising from voltage-pulsing in both conventional atom probe (CAP) (electrode at 5 mm) and local-electrode atom probe (LEAP) (electrode at 1 μm). The effects of voltage pulse duration and rise time for pulsing of both the specimen (positive) and the electrode (negative) have been considered. For the negatively pulsed CAP case there is an optimum in the pulse duration that minimizes the energy shift which corresponds to the time required for the ions to reach the electrode. For pulse durations of less than this value, the energy shift increases with increasing mass-to-charge ratios. For pulse durations of greater than this value, the energy shift increases with decreasing mass-to-charge ratios. The energy deficit remains relatively constant with pulse duration. For the negatively pulsed electrode case in the LEAP, the energy shift increases with pulse duration due to the deceleration of the ions. This suggests that fabrication of an electrode with sufficient thickness to create a small field-free region is required if negative pulsing is to be employed. This scenario was investigated as a function of pulse rise time. Energy deficits for this case show that the shortest possible rise time is beneficial when voltage pulsing in the LEAP.

Simulation of rapid thermal pulsing for field evaporation

Abstract Pulsed field evaporation should be achievable using an electron beam for thermal pulsing. This article explores the optimum specimen heating configurations resulting from this pulsing. The subsequent evaporation rate and the evaporation probability per pulse are then calculated. The heat transfer due to volume heating of the specimen is calculated using a one-dimensional Crank-Nicholson method. The evaporation rate is calculated using an Arrhenius expression along with previously determined experimental data. The heat transfer results indicate that, for the conditions simulated, a heated width of 240 nm can have the largest range of heated offsets and still produce the desired temporal heating characteristics. The evaporation rate results indicate that an evaporation probability of one ion per 20 pulses should be obtainable. The timer-limited mass resolution for time-of-flight measurements is predicted to exceed 250 at full width tenth maximum, although experimental values are expected to be lower.

Median-style filters for noise reduction in composition analyses

Abstract Several commonly applied techniques for atom probe data noise reduction are compared to signal processing techniques specifically designed to maintain interfacial integrity. A median filter and a double-window modified trimmed mean filter, along with several other filters, are applied to both a simulated atom probe composition profile and to real atom probe data. The various methods are compared using average profile compositions and variance measurements. The double-window modified trimmed mean filter is found to smooth the data in the slowly varying regions of the profile without degrading sharp interfaces.