On the accuracy of the interdiffusion coefficient measurements of high-temperature binary mixtures under ISS conditions

Abstract

An accelerometric record from the IVIDIL experiment (ESA Columbus module) has exhaustively been studied. The analysis involved the determination of basic statistical properties as, for instance, the auto-correlation and the power spectrum (second-order statistical analyses). Also, and taking into account the shape of the associated histograms, we address another important question, the non-Gaussian nature of the time series using the bispectrum and the bicoherence of the signals. Extrapolating the above-mentioned results, a computational model of a high-temperature shear cell has been performed. A scalar indicator has been used to quantify the accuracy of the diffusion coefficient measurements in the case of binary mixtures involving photovoltaic silicon or liquid Al-Cu binary alloys. Three different initial arrangements have been considered, the so-called interdiffusion, centred thick layer and the lateral thick layer. Results allow us to conclude that, under the conditions of the present work, the diffusion coefficient is insensitive to the environmental conditions, that is to say, accelerometric disturbances and initial shear cell arrangement.

Dynamics of a binary mixture subjected to a temperature gradient and oscillatory forcing

“…Shaking the test cell creates an additional contribution at the micro-acceleration level. A recent analysis by Saez et al (2013) indicated that only in experiments with higher frequencies, i.e. f > 1 Hz, the g rms ( f ) peaks were above the requirement curve at a forcing frequency of f 0 and its third harmonic.…”