Grid Turbulence in Dilute High-Polymer Solutions 



Singh (2), and the intention was to obtain a comparison of theory and experiment. 

 Due to the nature of the results, this comparison cannot be made yet, and no 

 theoretical discussion is needed in this work beyond that given in the later dis- 

 cussion of the experimental results. 



Several high-molecular-weight industrial polymers which are known to be 

 effective friction reducers were used as additives: 



Polymer Supplier 



Poly(ethylene oxide) (Polyox WSR-301) Union Carbide Chem. Co. 



Poly(ethylene oxide) (Polyox Coagulant) Union Carbide Chem. Co. 



Guar gum (J2FP) Western Company 



Polyacrylamide (Separan AP-30) Dow Chemical Company 



Polyacrylamide (JlOO) Dowell Division, 



Dow Chemical Company. 



The polymers were used as received, without drying, screening, or frac- 

 tionation. Most of the work was done with a 50-pound barrel of Polyox WSR-301, 

 within several months of being shipped "from new stock." This polymer will be 

 indicated by P301B. The last three additives were used only briefly, since the 

 results indicated no advantage in their use. Because of the nature of the results 

 no detailed comparison of the additive molecular properties is needed in this 

 work. 



It is clear that the test solutions would have produced significant friction 

 reduction in turbulent pipe flow at typical wall shear stresses, say of the order 

 of 10^ dynes/cm^. In two cases, friction reduction tests with solutions made 

 from Polyox WSR-301 and from JlOO showed the expected high degree of effec- 

 tiveness (e.g., 70% pressure-drop reduction) at low concentrations (e.g., at 20 

 ppm —parts per million by weight). Maximum concentrations in the towing-tank 

 tests were about 140 ppm. All the towing-tank solutions are "dilute" (3) in that 

 the dimensionless concentration c' [17] , where c' is the polymer concentration 

 in g/100 cc or g/dl and [r]] is the intrinsic viscosity in dl/g, is at most about 

 0.3. In general, local tap water was used as the solvent. As discussed later, 

 there was no evidence that the results were dependent on the solvent impurities. 



PROCEDURE 



A Turbulence Research Towing Tank, 48 feet long with a cross section of 

 about 4 by 4 feet, was constructed for this work. The tank is lined with a 

 smooth vinyl film which was hosed clean between tests. A grid-turbulence ex- 

 periment is shown in progress in Fig. 1. The grid carriage, carrying the verti- 

 cal grid, is pulled along by cables from constant-speed winches. The sensor 

 carriage, generally towed by a cable from the grid carriage, follows at a fixed 

 distance x from the grid (i.e., at a constant value of x m, where M is the grid 

 mesh width). The duration of useful run time is about 25 sec for the standard 

 towing speed of 36 cm/sec. 



41 



