Quantitative Z-contrast imaging in the scanning transmission electron microscope with size-selected clusters

Abstract

This paper describes a new approach of quantification of annular-dark-field or Z-contrast image intensity as a function of inner acceptance angle of the detector in a scanning transmission electron microscope. By using size-selected nanoclusters of Pd (Z = 46) and Au (Z = 79), it is shown experimentally that the exponent in the power law I ∼ Z α varies strongly between 1.2 and 1.8 as the collection angle changes from 14 to 103 mrad. The result is discussed in line with existing theoretical models. Factors, such as cluster size, structure, and orientation as well as the detector geometry, are also discussed for potential use of the work. PACS number(s): 36.40.−c With today's developments in nanoscience and technology, there is an increasing demand to correlate, in a quantitative way, chemical composition, structure, and property of materials with reduced dimension and size. For example, Serpell et al. showed recently that the design, synthesis, and understanding of industrial catalysts with improved selectivity could benefit from knowing internal structures and composition of core-shell bimetallic nanoparticles. 1 Similarly, examples have been seen in nanoalloyed AuAg particles with potential use of tunable optical properties. 2 Here, Z-contrast imaging in scanning transmission electron microscopy (STEM) are applied in obtaining chemical composition within nanoparticles.The so-called Z-contrast imaging method originates from the strong dependence of STEM image intensity on atomic numbers of elements when the signals are collected by a highangle annular dark-field (HAADF) detector. [3][4][5] In this case, as proposed by Howie in 1979, the signals are predominantly thermal scattered electrons, and coherence is largely reduced. 6 Since then, there have been many studies on the capabilities and limitation of Z-contrast imaging. 7-14 While there are a number of specimen-related factors that contribute to the image contrast, including not only atomic number, but also the specimen thickness, the crystal structure, and the crystal orientation, the geometry of annular detector also plays an important role. The dependence of Z contrast on the detector geometry was investigated theoretically by Hartel et al., who pointed out that, for thin objects, an analytical expression for the Z dependence of image intensity can be approximated by an exponential function of the form I ∼ Z α , where α is smaller than 2 and in the range 1.6-1.9 for most cases. 11 This power law has been widely accepted by the materials and microscopy community, however, without being experimentally verified. For effective quantitative characterizations using Z-contrast imaging, it is advantageous to know the critical exponent α as a function of detector angles.Zhu et al. pointed out from their combination of simulation and experimental work on SrTiO 3 crystals (∼10-50 nm thickness) that the dependence of the imaging contrast on collection angle varies with sample thickness, 15 while Klenov and Stemmer show experimentally for a 40-nm-thick PdTiO 3 film on a...

Catalysis with Supported Size-selected Pt Clusters

“…Though there are plenty of theoretical calculations available for size-selected clusters in the gas phase [227,233,234,235] and even results on surfaces [236,11], experimental investigation (by means of local methods, e.g. STM or TEM) on the local structure have been (recently) reported [229,237,238,95,239,211] and are heavily dependent on cluster and support material as well the experimental conditions. 10 Further, with larger clusters support effects become more important along with an increasing number of structural isomers.…”

“…For other metals with high atomic number Z, i.e. Au, Pd [336,337,238,237], the dependency of the STEM image intensity on Z, so called Z-contrast imaging, in the ADF mode was used in order to discriminate between different sized clusters. The feasibility of this approach is shown, however the work suffers from poor size selection (±5%), the use of considerably large clusters and neglecting the influence of other parameters on the Z-contrast [290].…”

Quantum correlations among optical and vibrational quanta

“…where a laser photon absorption is accompanied by the generation of a phonon and an optical cavity photon, respectively, while ∆ 1 = ∆. A frequency shift occurring due to nonlinear effects existing in the dispersive limit applied here (see, also, [16][17][18][19]) is responsible for the resonance around ω and not exactly at this value. Furthermore, the environment temperature that explicitly influences the vibrational degrees of freedom, implicitly affects also the mean photon number in the optical resonator mode (compare solid and dashed curves) demonstrating, thus, the existence of correlations among them.…”

“…Consequently, the quantum dot variables can be eliminated from the whole quantum dynamics, an approximation valid for Ω ≫ γ ≫ κ a,b as well as Ω ≫ {g, λ}. Notice that this approach is widely used in other somehow related systems [16][17][18][19]. In the following, we write down the master equation (2) after tracing over quantum-dot degrees of freedom:…”

Molecule-by-Molecule Writing Using a Focused Electron Beam

“…In these conditions, the ADF signal is proportional to the membrane thickness, so that it is valid to use the ADF signal for quantification. 17 When imaging this membrane with the standard FEI imaging software, this generates 5.5 Â 10 3 counts, or I 1nmC = 176 counts/nm.…”