Numerical study of flow boiling in micro/mini channels

University dissertation from KTH Royal Institute of Technology

Abstract: Boiling phenomena in micro scale has emerged as an interesting topic due to its complexity and increasing usage in micro electronic and mechanical systems (MEMS). Experimental visualization has discovered five main flow regimes: nucleate boiling, isolated bubbles, confine bubbly flow, elongated bubbly (or slug) flow, and annular flow. Two of these patterns (confine bubbles and slug flow) are rarely found in macro channels and are believed to have very different heat transfer mechanisms to that of nucleate boiling.The development of a phenomenological model demands a deep understanding of each flow regime as well as the transition process between them. While studies in every individual flow pattern are available in literature, the mechanisms of transition processes between them remain mysterious. More specifically, how the isolated bubbles evolve into a confined bubbly flow, and how this further evolves into elongated bubbles and finally an annular flow. The effects of boundary conditions such as wall heat flux, surface tension, and interfacial velocity are unclear, too.The aims of this thesis are to develop and validate a new numerical algorithm, perform a comprehensive numerical study on these transition processes, uncover the transition mechanisms and investigate effects of boundary and operating conditions.Firstly, a sophisticated and robust numerical model is developed by combining a coupled level set method (CLSVOF) and a non-equilibrium phase change model, which enables an accurate capture of the two-phase interface, as well as the interface temperature.Secondly, several flow regime transitions are studied in this thesis: nucleate bubbles to confined bubbly flow, multi confined bubbles moving consecutively in a micro channel, and slug to annular flow transition. Effects of surface tension, heat flux, mass flux, and fluid properties are examined. All these regimes are studied separately, which means an appropriate initial condition is needed for each regime. The author developed a simplified model based on energy balance to set the initial and boundary conditions.