The stability characteristics of polymeric flows are often dramatically different from those of Newtonian flows. In some cases, polymers can stabilize Newtonian flows or significantly alter the nature of the instability, in other cases viscoelastic effects lead to new instabilities.
A dramatic instance where polymers lead to stabilization is the reduction of drag in wall-bounded turbulent flows such as pipe flow. Even a small amount of polymer can drastically reduce turbulence. This phenomenon, known since the 1940s, has yet to be fully explained. Studies of near wall turbulence have established that the cycle of streak formation and bursting contributes in a major way to the production of drag. Roughly speaking, a streamwise vortex leads to a speeding up of the flow where fluid is advected towards the wall and a slowing down where it is pushed away from the wall. This leads to the formation of streaks. The resulting distortion of the mean flow eventually causes an instability known as bursting. There is a consensus that polymers somehow interfere with this process. The conventional wisdom is that the elongational resistance of polymers is the crucial factor and either inhibits bursting or perhaps suppresses streamwise vortices. Doubts remain, however, if the elongation rates found in drag reducing flows are large enough to cause a major elongational response. In Renardy's paper, a different explanation is proposed, related to second normal stress differences. The effect of a second normal stress difference would be a negative feedback such that formation of streaks suppresses the streamwise vortex which is responsible for creating the streaks.