85a Downstream Development of Drag Reduction by Pipe Wall Injection of Concentrated Polyox into Water at Reynolds Numbers from 55000 to 140000

Preetinder S. Virk, Fernando Funakoshi, Nikhil Pradhan, and Jason Whittaker. Department of Chemical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA 02139

The downstream development of drag reduction following wall injection of concentrated polyethyleneoxide solutions into a turbulent water flow was studied by simultaneous pressure drop and flow visualization measurements at Reynolds numbers from 55000 to 140000 in a transparent acrylic test pipe of 15.7 mm ID and L/D = 430. The experiments partially modeled drag reduction in the Trans-Alaska Pipeline System, by roughly matching (i) the pipe friction velocity uô ~ 0.15 m/s, (ii) the backbone chain links Nbb ~ 0.5x106 and downstream concentrations Cline ~ 5 wppm of the polymer additive and (iii) the wall injection of concentrated polymer through equi-spaced circumferential ports at friction-normalized injection velocities Vinj+ ~ 1 << mean velocity V+ ~ 25. Polymer dissolution into the flow was assessed two ways, (i) from the initial increase in fractional drag reduction DR towards the eventual downstream DRline, which provided a characteristic development distance (L/D)dr and (ii) from visual observation and flash photography to reveal a characteristic distance (L/D)pv when red-dyed strands of the injected solution disappeared. Specific results are: (i) Friction factors for solvent, deionized water, adhered to the Prandtl-Karman law for fully-developed turbulent flow in smooth pipes, and injection of red-dyed water “blanks” gave (L/D)pv ~ 50, with corresponding eddy diffusivity to kinematic viscosity ratio ~ 150. (ii) Concentrated polymer solutions with Cinj from 500 to 5000 wppm provided downstream polymer concentrations Cline from 0.1 to 100 wppm that exhibited DRline from 0.15 to 0.80, the latter close to the asymptotic maximum drag reduction possible at the present Re. (iii) Drag reduction development distances (L/D)dr ~ 100 were generally much lower than the dyed polymer strand disappearance distances (L/D)pv ~ 150 to > 400, at all Cline and Cinj. The contrast between the insensitivity of (L/D)dr to Cline on the one hand, and the sensitivity of (L/D)pv upon Cline on the other, is discussed using modern notions of Type B turbulent drag reduction physics.