Viscosity flow cell experiments

The flow cell translates time into distance and the combination of the three and varying the flow rates gave a range of observations from 0 to 30 s. SHG measurements of the static aqueous/dodecane interface were made at each port before and after the flow experiment to calibrate the observations from each port For a laminar (non-turbulent) flow, the two flow rates should be in the inverse ratio of the fluid viscosities this ratio for dodecane on water is 0.65 at 25°C, very close to the observed flow rate ratio of 0.67. The bulk flow rates for each liquid were measured by collecting the volume of liquid flowing in a known time. Since the cell operates under non-turbulent conditions, the velocity of each layer at the interface must be the same, but the average velocities of the two layers are different. Ideally a model of the flow conditions inside the cell would be used to accurately determine the velocity of the interface. Since this was not... 

Interestingly, Rajamani and Milin found no need to iterate for feed concentrations up to 20% by weight where clean water velocity profiles were used and the separation results compared well with experiments (limestone in water, in a 75 mm hydrocyclone). This is due to the fast dilution process taking place in the flow as pointed out in section 1.4. Only for concentrations above 20% by weight the authors needed to iterate the computations the first iteration of particle trajectories is done with the values of water viscosity and density, from which particle concentrations (and slurry viscosities and densities) are computed at each point. For this the authors use computed particle trajectories and assume that the concentration is proportional to particle residence time in each computational cell. 

---