## Abstract

The velocity and flux of spherical glass beads with nominal diameters of 200, 80 and 40 $\mu m$ have been obtained by phase-Doppler anemometry in a round unconfined air jet over the first 28 diameters. The jet diameter was 15 mm and the exit velocity was 13 m s$^{-1}$ giving a Reynolds number of 13 000 and a timescale of 1.15 ms, which increased quadratically with axial distance: the bead inertial time constants were 298, 48 and 12 ms. The purposes of the experiments were to quantify the velocity and flux characteristics of the beads and of the gas phase in the presence of the beads as a function of bead diameter and of the mass loading in the jet nozzle. Due to the large inertia of the 200 $\mu m$ beads, the mean bead velocity downstream of the exit of the jet was constant and independent of mass loading up to 0.37 and the axial root mean square (r.m.s.) bead velocity decayed by about one-fifth: at the exit of the jet, the axial r.m.s. bead velocity was higher than that of the corresponding clean jet. The mean centreline velocity of the 80 $\mu m$ beads decayed to about one-half of the bead exit velocity by 28 diameters downstream and was independent of mass loading up to 0.86. The decay rate of the mean gas centreline velocity in the presence of the beads reduced as the loading increased because of momentum transfer from the discrete to the gaseous phase. The axial r.m.s. velocity of the beads was comparable to that of the gas phase and both decreased with increasing loading and the rate of spread of the half width of the jet increased with increasing loading. For the 40 $\mu m$ beads, the decay rate of the mean centreline velocity of the beads decreased with increasing loading and, in contrast to the 80 $\mu m$ beads, the rate of spread decreased with increasing loading up to 0.80. The axial r.m.s. velocity of the beads became largest at a position downstream of the nozzle exit, which moved downstream with increasing loading and was larger than the axial r.m.s. velocity of the clean jet, although the beads were not expected to be responsive to the frequencies of the energy-containing eddies. The bead axial r.m.s. velocity was more than twice as large as the radial r.m.s. velocity and the correlation coefficient of the cross correlation was larger than that of the clean jet. The large bead turbulence, anisotropy and strong correlation coefficient are explained by the superposition of bead trajectories from regions of different bead mean velocity and are not because of acquisition of axial turbulent motion from the gaseous phase.