Macro-to-Microchannel Transition in Two-Phase Flow and Evaporation

Continuous development of highly efficient and compact cooling elements are driven by extensive developments of Micro-Electro-Mechanical Systems (MEMS) devices such as microprocessors or micro high-powered laser systems. As microprocessors continue to head in the direction of miniaturization, it was quickly realized that the major obstacle to overcome was "heat". This justified the need for the development of a new generation of liquid cooling technologies that involves phase-change processes. However, significant differences in two-phase transport phenomena have been reported in the microscale sized channels as compared to conventional macroscale channels. The classification of macroscale, mesoscale and microscale channels with respect to two-phase processes is still an open question. This research project focuses on investigating the macro-to-microscale transition during flow boiling in small scale channels of various size and refrigerants to investigate the effects of channel confinement on flow boiling heat transfer, two-phase flow patterns, pressure drop, critical heat flux and liquid film stratification in a single circular horizontal channel. In this project, we systematically investigated the trends observed in the heat transfer data, pressure drop, critical heat flux and film stratifications for wide range of conditions as a function of flow pattern, creating a uniquely comprehensive experimental database for R134a (high pressure), R236fa (medium pressure) and R245fa (low pressure) in the 1.03, 2.20 and 3.04 mm channels. From a heat transfer viewpoint, the prediction of the heat transfer requires first the prediction of the two-phase flow regimes in these small scale channels. So, in that respect an improved flow pattern map has been proposed and by determining the flow patterns transitions existing under different conditions, the effects of the flow regimes and channel dimensions on flow boiling heat transfer is thus explained. A flow pattern based heat transfer prediction model, i.e. the three-zone model showed good agreement in the prediction of heat transfer coefficients corresponding to the isolated bubble (IB) and coalescing bubble (CB) flows. As for the two-phase pressure drops, various statistical comparisons have been made by comparing the pressure drop database with various macroscale and microscale prediction methods, finding that three macroscale and two microscale methods predicted the pressure drop database with good accuracy. From this research, a distinctive peak in the CHF has been identified when comparing the experimental CHF database for 0.50, 0.80, 1.03, 2.20 and 3.04 mm channels (the first two from previous study). A new CHF prediction method was developed that is applicable for the CHF prediction of single, square multi-microchannels and split flow multi-channels. Finally, the processing of the high speed flow visualization images provided the interpretation of the lower threshold of the macro-microscale transition. As an approximate rule, the lower threshold of macro-scale flow is Co= 0.3-0.4 while the upper threshold of the symmetric microscale flow is Co≈1.

State of the Art of High Heat Flux Cooling Technologies

Bruno Agostini, Bruno Michel, Jung Eung Park, John Richard Thome, Leszek Wojtan

The purpose of this literature review is to compare different cooling technologies currently in development in research laboratories that are competing to solve the challenge of cooling the next generation of high heat flux computer chips. Today, most development efforts are focused on three technologies: liquid cooling in copper or silicon micro-geometry heat dissipation elements, impingement of liquid jets directly on the silicon surface of the chip, and two-phase flow boiling in copper heat dissipation elements or plates with numerous microchannels. The principal challenge is to dissipate the high heat fluxes (current objective is 300 W/cm2) while maintaining the chip temperature below the targeted temperature of 85°C, while of second importance is how to predict the heat transfer coefficients and pressure drops of the cooling process. In this study, the state of the art of these three technologies from recent experimental articles (since 2003) is analyzed and a comparison of the respective merits and drawbacks of each technology is presented. The conclusion is that two-phase flow boiling in microchannels is the most promising approach; impingement cooling also has good prospects but single-phase liquid cooling is probably only a short-term solution. As an example of the state of the first technology, the Heat and Mass Transfer Laboratory at Ecole Polytechnique Fédérale de Lausanne has already achieved 200 W/cm2 of cooling in a first prototype, with a low pumping power, good temperature uniformity, and at the required maximal operating temperature.

2007

Heat Transfer and Pressure Drop in the Dryout Region of Intube Evaporation with Refrigerant/Lubricant Mixtures.

Daniel Favrat, John Richard Thome, Olivier Zürcher

A summary of the objectives and conclusions reached in this project are: •A litterature search on post dryout heat transfer for pure refrigerants and for refrigerant-oil mixtures was performed and a comprehensive review written (Chapters 9, 10, and 11); •The correct thermodynamic definition of the heat transfer coefficient for evaporation of refrigerant-oil mixtures was presented and then used in the project, i.e. defining the heat transfer coefficient using the local bubble point temperature Tbub of the refrigerant-oil mixture rather than with the saturation temperature of the pure refrigerant Tsat incorrectly used in prior studies (Chapter 3); •A new generalized approach, that is simple to implement, was developed for predicting bubble point temperatures of refrigerant-oil mixtures for miscible oils that can be applied to any refrigerant-oil combination and is accurate over the local oil concentration range from 0-50 wt.% oil range confronted in direct- expansion evaporators with 0-5 wt.% oil in their refrigerant charge (Chapter 4); •A new thermodynamic method for preparation of temperature-enthalpy- vapor quality (T-h-x) curves for refrigerant-oil mixtures for general application was developed and validated against test data (Chapters 5 and 6), including a simple expression for calculating local oil concentrations as a function of vapor quality; •New, accurate thermodynamic methods for preparation of temperature-enthalpy-vapor (T-h-x) curves for the refrigerant blend R-407C and R-407C/oil mixtures were developed for thermal design of evaporators and for the present experimental test, and was validated against measured test data (Chapter 12); •Recommendations were made on how to include oil effects and refrigerant-oil T-h-x curves into evaporator design methods (Chapter 7); •The effects of oil on set-point parameters for control devices were analyzed and some recommendations made (Chapter 8); •An online oil concentration measurement system was perfected using a high accuracy density flowmeter that allowed oil concentrations to be measured accurately and continuously during experiments, such that oil holdup in the heat transfer test sections at high vapor quality could be identified and the correct inlet oil concentrations be measured, and thus utilized to reduce experimental data; a simpler industrial method for applying the measurement system was also developed and presented (Chapter 15); •Experimental results (local heat transfer coefficients, two-phase pressure drops and some two-phase flow patterns) were obtained for R-134a and R-134a/oil mixtures evaporating in plain and microfinned tubes over a wide range of test conditions (Chapters 16 and 17); •Experimental results (local heat transfer coefficients, two-phase pressure drops and some two-phase flow patterns ) were obtained for R-407C and R- 407C/oil mixtures evaporating in plain and microfinned tubes over a wide range of test conditions (Chapters 18 and 19); •The recently proposed flow boiling model and flow pattern map of Kattan-Thome-Favrat, that importantly models heat transfer coefficients based on local flow pattern and predicts the onset of dryout in horizontal tubes and local heat transfer coefficients in partial dryout regimes, was described in detail; the flow pattern map is also useful to designers wishing to design in specific flow pattern regimes and avoid inefficient heat transfer regimes, such as mist flow and stratified flow (Chapters 20 and 21); •The stratified-wavy flow regime model was modified based on the new test data at high vapor quality for pure R-134a and R-407C, which increased the accuracy of the method significantly in this flow regime, now referred to as the “modified” Kattan- Thome-Favrat flow boiling model (Chapter 22); •The modified Kattan- Thome-Favrat flow boiling model and flow pattern map were shown to accurately predict R-407C heat transfer and flow pattern data and hence is recommended for industrial use for designing direct-expansion evaporators with refrigerant blends [the original version was also verified for R-402 and R-404A blends, see Chapter 21] (Chapter 22); •The effect of local refrigerant-oil viscosity on liquid-phase convection in flow boiling heat transfer and on flow patterns was successfully incorporated into the modified Kattan-Thome-Favrat model, establishing the first flow boiling design method that predicts oil effects on heat transfer (with quite reasonable accuracy too), and included a simple method for calculating the values of µref-oil for use in the model (Chapter 23); •The oil effects on microfin heat transfer coefficients were analyzed and heat transfer augmentation ratios were determined for R-134a/oil and R-407C/oil mixtures (Chapter 25); •Plain tube, two-phase pressure drop gradients for R-134a and R-407C at three mass velocities were compared to the Friedel frictional two-phase pressure drop correlation, attaining accurate results and also microfin pressure drop augmentation ratios were determined (Chapter 25); •The ratio of [µoil/µref]mw, derived from the Arrhenius logrithmic viscosity mixing rule and local oil concentration w, was shown to accurately predict the effect of oil on two-phase pressure drop gradients at high vapor qualities, with an empirical exponent m = 0.18355 found for R-134a/oil mixtures (no foaming) and an empirical expression for m for R-407C/oil mixtures (with foaming). This ratio incorporated into the Friedel correlation accurately predicts two-phase pressure drops of evaporating refrigerant-oil flows (Chapter 25).

1996

Heat Transfer and Pressure Drop in the Dryout Region of Intube Evaporation with Refrigerant/Lubricant Mixtures.

Daniel Favrat, John Richard Thome, Olivier Zürcher

1996