The current study explores ultrasound imaging velocimetry (UIV) as an alternative method for capturing flow fields in shear-thinning suspensions. In particular, UIV is used here together with pressure measurements to identify the reattachment point and to correlate the point's location with the pressure recovery downstream of an axisymmetric expansion. Four fluids are investigated: pure water, as well as three 1750ppm xanthum-gum-in-water solutions mixed with non-reactive mineral micro-particles at 0%, 15% and 30% volume fractions. Wall-pressure measurements were collected through taps located at 0 to 13 throat diameters downstream of the expansion with subsequent UIV measurements collected from 0.5 to 5 throat diameters downstream of the expansion. Pressure-tap measurements for the tested suspensions demonstrate 80% to 100% total pressure recovery achieved at approximately 4 to 5 throat diameters downstream from the throat. Furthermore, all four fluids exhibit total pressure recoveries within 10% of Borda-Carnot prediction. Despite similar pressure recoveries, UIV revealed different reattachment lengths for the tested fluids. Water exhibited a reattachment length of approximately four throat diameters, which is expected within the tested turbulent regime, while the xanthum-gum solution exhibited a delayed reattachment beyond the field-of-view when the effective Reynolds number is O(100). Pipe-wall velocity measurements suggest that the dense suspensions exhibit similar reattachment lengths to that of water despite expectations that the dense suspensions would behave similar to the pure xanthum-gum solution given: the similar Reynolds number between the pure xanthum-gum solution and the tested dense suspensions; and previous work done by Barnes et al. (2024) and others revealing shear-layer stabilization within dense suspensions. It is suspected that increased diffusion within the suspension mitigates the shear layer's ability to convect downstream, but improved measurements are required to state this conclusively.
Read full abstract