No abstract

In our earlier work ͓Itoh et al., Phys. Fluids 17, 075107 ͑2005͔͒, the additional maximum of the streamwise turbulence intensity near the center of the drag-reducing turbulent boundary layer was found in the homogeneous dilute aqueous surfactant solution which was a mixture of cetyltrimethyl ammonium chloride with sodium salicylate as counterion. In this work, we systematically investigated the influence of the drag-reducing surfactant on the velocity fields of the turbulent boundary layer at various Reynolds numbers Re from 301 to 1437 and the drag reduction ratio DR from 8% to 74% under different streamwise locations and concentration and temperature of solutions using a two-component laser-Doppler velocimetry ͑LDV͒ system. It was revealed that all data on DR versus the wall-shear rate obtained here were collapsed on a single curve. We verified the existence of the additional maximum of the streamwise turbulence intensity near the center of the boundary layer which appeared at relatively large drag reduction ratios and small Reynolds numbers. It was found that the additional maximum of streamwise turbulence intensity and its wall-normal location were independent of the streamwise location, wall-shear rate, Reynolds number, and drag reduction ratio. The additional maximum could be explained by the bilayered structure model proposed, in which the flow in the near-wall region is in shear-induced structure ͑SIS͒ and viscoelastic, whereas the flow in the region away from the wall is in non-SIS and nonviscoelastic. This model was based on measurements of the shear viscosity. We also performed particle image velocimetry measurements, which revealed that the fluctuating velocity vector fields showed two situations, with low and high activity. In low activity, the velocity fluctuations were attenuated largely across the turbulent boundary layer. In high activity, fluctuating velocity vectors were almost parallel to the wall and relatively large in both regions near the wall and the center of the boundary layer, which seemed to be a bilayered structure and supported the bilayered structure model.

No abstract

There are only a few studies on the drag-reducing effect of nonionic surfactant solutions which are nontoxic and biodegradable, while many investigations of cationic surfactant solutions have been performed so far. First, the drag-reducing effects of a nonionic surfactant (AROMOX), which mainly consisted of oleyldimethylamineoxide, was investigated by measuring the pressure drop in the pipe flow at solvent Reynolds numbers Re between 1000 and 60 000. Second, we investigated the drag-reducing effect of a nonionic surfactant on the turbulent boundary layer at momentum-thickness Reynolds numbers Reθ from 443 to 814 using two-component laser-Doppler velocimetry and particle image velocimetry systems. At the temperature of nonionic surfactant solutions, T=25 °C, the maximum drag reduction ratio for AROMOX 500 ppm was about 50%, in the boundary layer flow, although the drag reduction ratio was larger than 60% in pipe flow. Turbulence statistics and structures for AROMOX 500 ppm showed the behavior of typical drag-reducing flow such as suppression of turbulence and modification of near-wall vortices, but they were different from those of drag-reducing cationic surfactant solutions, in which bilayered structures of the fluctuating velocity vectors were observed in high activity.

The drag-reducing ability of the seal fur surface was tested in a rectangular channel flow using water and a glycerol-water mixture to measure the pressure drop along the channel in order to evaluate friction factors in a wide range of Reynolds number conditions, and the drag reduction effect was confirmed quantitatively. The maximum reduction ratio was evaluated to be 12% for the glycerol-water mixture. The effective range of the Reynolds number, where the drag reduction was remarkable, was wider for the seal fur surface compared to that of a riblet surface measured in this channel and in previous studies. It was also found that for the seal fur surface, unlike riblets, any drag increase due to the effect of surface roughness was not found up to the highest Reynolds number tested. Measurements of the seal fur surface using a 3D laser microscope revealed that there were riblet-like grooves, composed of arranged fibers, of which spacings were comparable to that of effective riblets and were distributed in various wavelengths. Using LDV measurements, it was found that the difference in the mean velocity scaled by the outer variable among the smooth, riblet, and seal fur surfaces did not appear at any spanwise locations. Streamwise turbulence intensity for the seal fur surface was found to be about 5% smaller than those for smooth and riblet surfaces.