Skip to main content
eScholarship
Open Access Publications from the University of California

UC Berkeley

UC Berkeley Electronic Theses and Dissertations bannerUC Berkeley

Heat Pipe Performance Enhancement with Binary Mixture Fluids that Exhibit Strong Concentration Marangoni Effects

Abstract

This research investigates the impact of Marangoni phenomena, with low mixture concentrations of alcohol and water, to enhance thermal transport capability of gravity-assisted heat pipes. The use of binary mixture working fluids in gravity-assisted heat pipes are shown to improve the critical heat flux (CHF) and operating performance, more so than with pure fluids. The CHF is responsible for dryout when the pumping rate of a liquid flow structure is not sufficient to provide enough fluid to the evaporator section.

In the first study, heat pipe performance experiments were conducted for pure water and 2-propanol solutions with varying concentrations. Initial tests with pure water determined the optimal working fluid charge for the heat pipe; subsequent performance tests over a wide range of heat input levels were then conducted for each working fluid at this optimum value. The results indicated that some mixtures significantly enhance the heat transfer coefficient and heat flux capability of the heat pipe evaporator. For the best mixture tested, the maximum evaporator heat flux carried by the coolant without dryout was found to be 52% higher than the value for the same heat pipe using pure water as a coolant under comparable conditions. Peak evaporator heat flux values above 100 W/cm2 were achieved with some mixtures. Evaporator and condenser heat transfer coefficient data are presented and the trends are examined in the context of the expected effect of the Marangoni mechanisms on heat transfer.

Analytical modeling effort was also conducted investigating the impact of Marangoni phenomena for low concentrations of 2-propanol/water and methanol/water mixtures. In real systems the addition of small levels of surface-active contaminants can affect the surface tension of the liquid-vapor interface and thermodynamic conditions in this region. Analysis was performed for three widely accepted binary mixture correlations to predict heat flux and superheat values for subatmospheric experimental data using bulk fluid and film thermodynamic properties. Due to the non-ideal nature of these alcohol/water mixtures, this study employs an average pseudo single-component (PSC) coefficient in place of an ideal heat transfer coefficient (HTC) to improve the correlation predictions. This investigation evaluates the ability for these correlations to predict strong Marangoni effects of mixtures that have large surface tension variation with concentration under subatmospheric conditions. It is not always clear that evaluation of bulk fluid properties will satisfactorily account for Marangoni effects. Analysis is also performed to assess correlation predictions for interfacial film properties rather than that of the bulk fluid. The results indicate that the use of film properties along with the PSC coefficient improves heat flux model predictions of subatmospheric experimental data by as much as 59.3% for 0.015M 2-propanol and 49.1% for 0.04M methanol/water mixtures, where strong Marangoni effects are believed to be more evident.

A second experimental study was also performed of a 37° inclined, gravity-assisted, brass heat pipe with a 0.05M 2-Propanol/water binary mixture. The device design was developed from the first study by enlarging the evaporator and condenser surface areas. Strip heaters were also employed to provide larger input heat flux levels, for enhanced heat pipe performance testing. These experiments were carried out for varying liquid charge ratios between 30% and 70%, to determine an optimal value that would enhance heat transport performance by maximizing the critical heat flux (CHF) condition, while reducing the evaporator wall superheat. A 45% fill ratio was found to have the lowest overall superheat and highest thermal conductance by as much as 7.5W/K, as well as an enhanced CHF condition of 114.8W/cm2. A heat pipe analytical model, that characterizes binary mixture pool boiling is also presented, which was developed based on modeling efforts presented in studies 1 and 2. Model results with a 45% liquid charge ratio were found to provide good correspondence with the experimental data with an average rms evaporator vaporization heat flux deviation of 6.5%.

The final study of this investigation assesses the cooling of single and dual-junction solar cells with the inclined, gravity-assisted, brass heat pipe, with a 0.05M 2-propanol/water mixture. Thermal behavior of this heat pipe solar collector system was investigated theoretically and semi-empirically through experimentation of varying input heat loads from attached strip-heaters to simulate waste heat production of single-junction monocrystalline silicon (Si), and tandem multijunction GaInP/GaAs solar cells. It was also found that the 45% liquid charge was capable of achieving the lowest superheat levels and highest critical heat flux (CHF) condition of 114.8 W/cm2, at a predicted solar concentration of 162 suns. Solar cell semiconductor theory was employed to evaluate the effects of increasing temperature and solar concentration on solar cell performance. Results showed that a combined PV/heat pipe system had a 1.7% higher electrical efficiency, at a concentration ratio 132 suns higher than a stand-alone PV system. The dual-junction system also exhibited enhanced performance at elevated system temperatures with a 2.1% greater electrical efficiency, at an operational concentration level of 560 suns higher than a stand-alone PV system. Waste heat recovery analysis of the silicon solar cell, revealed respective thermal and system efficiencies as high as 56.3% and 66.3% as the incident solar radiation and corresponding condenser heat removal factor increased to 82 suns.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View