Pitfalls measuring 1D inertial particle clustering

Abstract

We perform 1D Voronoï analysis on a time series from an optical probe detecting the passage of particles in a homogeneous. isotropic turbulent flow. The Voronoï analysis is unable to identify clustering in the particle locations along the measuring “line”, despite the flow being almost identical in terms of the Reynolds number based on the Taylor scale (Reλ), and Stokes (St) numbers to previous experiments in which 2D Voronoï analysis successfully detected and measured this phenomenon [8]. The optical probe accurately measured the particle average global concentration, and size distribution. This result stemmed from the sub-kolmogorov measuring volume of the probe, and seems to be in agreement with previously reported studies under totally different conditions [7] that referred to this issue as sub-poissonian events. If the instrument measurement window size is ‘large’ enough -but not too large to smooth out all correlations-, and the data satisfyies statistical convergence, 1D Voronoï diagrams effectively capture evidence of clustering, and constitute a reliable proof of preferential concentration within the flow.