Characteristics and origins of viscoelastic turbulence in a 3-stream contraction-expansion micro-channel

Abstract

At low Reynolds number (Re < 1), the flow of viscous liquids e.g. water, is laminar. An aqueous solution e.g. water becomes viscoelastic when a small amount of polymer additives (< 1 wt%) is added to it; its flow behavior can become drastically different and turbulent i.e. viscoelastic turbulence. This phenomenon has gained increasing attention because it violates the conventional school of thought i.e. high-Re criteria, for creating chaos and disorder in a fluid dynamics system. As the polymer molecules are invisible, the indirect deduction of molecular behavior via motion analyses of tracer additives has been adopted as the main investigative approach in viscoelastic turbulent flows. Based on this approach, reported works attribute viscoelastic turbulence to the release of elastic energy by the polymer molecules, which had been extended due to strong velocity gradients in the flow field. The release of energy occurs over a range of time scales which is dependent on the characteristic time scales of the molecules and bulk viscoelastic liquid. Although the reported characteristic time scales vary significantly, their effects on the structure of the viscoelastic turbulent flow field have not been investigated. Despite the significant number of investigations based on the conventional approach, the underlying mechanisms in viscoelastic turbulence remains elusive; several outstanding questions on viscoelastic flows remain unresolved e.g. the high Weissenberg number problem, and the drag reduction theory debate. "How do the polymer molecules change the flow field so drastically when they are only present in minute amounts?". This fundamental question has yet to be clearly answered. Although fluorescent-tagged DNA molecules are commercially available, because I would like to express my heartfelt gratitude to my supervisor, Professor Lam Yee Cheong. His advice and guidance has been invaluable throughout my Ph.D. study. In addition, I would like to thank NTU for the funding and support, which made this research work possible. The excellent research environment has greatly facilitated the experimental work required in this study. I would also like to thank all staffs of Materials Laboratory 1, Thermal and Fluids Laboratory and Precision Engineering Laboratory for their support and help in the management and usage of facilities. Most importantly, I wish to thank my parents and friends for their unwavering support. Blank