Analysis of the atom probe microscopy (APM) data:
APM experiments determine the constituent atoms and their position in materials by using time-of-flight and mass-spectrometer measurements. We analyze a data set for a sample that is fabricated to be used as an STM tip by CINT user Prof. Krishna Rajan’s group at Iowa State University. The sample is intended to have a conical volume of Si atoms with an outer shell of Cu atoms. One of the goals of the analysis is to quantify the purity of the sample. By visualizing the structure and focusing on the identification of the impurity atoms and their distribution in the sample, we find that the tip (see Fig. 1b) and the boundary between the Si and Cu (see Fig. 1c and 1d) have many more impurities than intended for.
Fig.1 Cu (blue), Si (turquoise), O (green), CuO (yellow), SiO (red) are the constituents of the STM tip which is used in the atom probe microscopy experiment. The sample is intended to have Cu and Si atoms only. The size of the sample is 35.2 nm x 35.2 nm x 87.1 nm. The number of recorded atoms is about 1.7 million or roughly 50% of atoms are actually collected. a) Reconstructed 3D image of the STM tip. b) Atoms at the tip (Z=0). Atoms at the boundary c) on a plane at x=0 nm, and d) on a plane at z=60 nm. The color coding of the atomic species readily shows the impurity of the sample.
Using the advanced visualization tool (ParaView) we also quantify the distribution of the atoms in the sample. The result is shown in Fig. 2 (animation). Here the top-left figure shows the reconstructed 3D image of the STM tip. We take a box cut of width 2 nm along the Z-axis. The region is shown with a black bounding box. We show the constituent atoms in this region in the left-middle figure. The distribution of the atoms (number of atoms of a particular type) in this region is shown in the histogram of the atom distribution in the top-right panel. We can see that the number of Cu atoms dominates the number of all other atoms. So to amplify the distribution of the impurity in this region, we next filter out the Cu and Si atoms and display only the impurity species in the bottom-left panel. The corresponding histogram is shown in the bottom-right panel. By changing the position of the box cut we show all the atoms, impurity atoms and the corresponding histogram of the atom distribution within the box via the animation in Fig. 2.
(Put the animation here.)
Fig 2. The animation is created by letting the box cut to move along the Z-axis in steps of 0.5 nm. As the box sweeps along the Z-axis, the position of the box, all-atoms distribution, impurity distribution, and the histogram change to update the distribution of the atoms at the new Z location. In the top-right panel histogram, the number of Cu, Si, O, CuO and SiO is given by the bars from the left to the right respectively. In the bottom-right panel histogram, the number of O, CuO and SiO is given by the bars from the left to the right respectively. The animation is saved as either an image sequence (every time the box moves) or as a movie file (.avi). We created this animation by using only the Graphical User Interface of of the visualization tool ParaView.
To further quantify the distribution of the atoms in the system we calculate the density of atoms. We take a box cut of 1nm x 1nm x 1nm volume with the center of the box at the position of one of the atoms of the system. The number of atoms lying within this specified volume is then calculated. This gives the number density of the atoms (density). Now the position of the center of the box is moved to the position of the next atom in the sample. The result is shown in Fig. (3). We can clearly see that there is an inhomogeneous distribution of atoms in the sample.
Fig. 3. The number density of atoms at the position of the constituent atoms is shown. The sample has an inhomogeneous nanoscale distribution of atoms.