Impurity diffusion coefficients D of five kinds of solute elements in liquid Sn were measured using the Foton shear cell with stable density layering on the ground. This experiment involved diffusion from a thin alloy layer of 3-mm thickness into pure Sn. The Sn alloys contained Ag, Bi, In, Pb, or Sb at 5at%. The diffusion couple was set vertically so that the side with higher density was on the bottom, which is called stable density layering. Four identical parallel experiments were performed simultaneously for each condition. The diffusion temperature was 573 K, and the diffusion time was 8 h. Each obtained profile agreed well with the theoretical equation for the thick layer solution of the diffusion equation (coefficient of determination r2>0.999). The reproducibility of the diffusion coefficients among the four parallel experiments was very good with a standard deviation less than 2.5%. The obtained diffusion coefficients DBi and DIn agreed well with μg-reference data. Therefore, the buoyancy convection was assumed to be suppressed by the stable density layering during the diffusion experiments in this study, including the experiments of SnPb, SnAg, and SnSb with high density gradient. The impurity diffusion coefficients in liquid Sn at 573 K can be expressed as a proportional relationship to the product of the ratio of atomic radii rs/ri of the two substances and thermodynamic factor Φ　is, with a gradient equivalent to the value of the self-diffusion coefficient of Sn.

View full abstract

The X-ray fluorescence spectra of a liquid metal were analyzed using an X-ray fluorescence analysis, and the effects of the container materials on the peak intensities of a liquid metal using an X-ray fluorescence analysis were experimentally clarified. The spectrum of a liquid metal (Sn57Bi43) at 443 K with a container material (graphite or quartz) was measured using an X-ray fluorescence analysis. As a result of the experiments, sharp peaks of Sn and Bi were observed with each container material, and the Compton scattering peak from the X-ray tube was detected using graphite. In addition, by combining the Lambert-Beer law with the model used in the experiments, a theoretical formula was derived. Because the derived formula agreed well with a determination coefficient of 0.979 or higher, the validity of this formula was confirmed. The variation coefficient, which indicates the reproducibility, of an X-ray fluorescence analysis using the container materials was 2.3% for quartz (thickness of 0.3 mm) and 2.1% for graphite (thickness of 2.0 mm) at a detection time of 180 s. Moreover, to obtain the reproducibility, the detection time should be set to longer than 300 s (graphite) and 480 s (quartz), which are comparable to the times required by inductively coupled plasma-optical emission spectrometry.

View full abstract

Diffusion is a common phenomenon for the all state of materials. We have developed a new technique for the observation of diffusion in liquid metals by using a fluorescent X-rays analysis and X-rays transmission device. This apparatus makes available the in-situ observation and real-time analysis of diffusion profile in liquid meals. In this study, we applied this apparatus to the measurement of diffusion between liquid Ag and liquid Cu. The diffusion coefficient obtained by this method is consistent with the orders of magnitude of previous results. Technical issues of this method emerged from this study.

View full abstract

A high power laser, such as CO2 or LD, is necessary for the sample heating in the container less processing. The high power laser is expensive generally, therefore that interrupt the spread of the container less processing of high temperature condition. Recently, a glass tube CO2 laser is provided, which is used as a heat source of computer controlled laser cutter. The glass tube CO2 laser is quite inexpensive, however we can hardly find the information of how to use it. Then, I actually bought the CO2 laser whose maximum output is 100 watt and try to set up it by myself with trial and error for testing the performance. I describe a way to the setup and the result of the first emission of this laser.

View full abstract