Nucleate boiling is ubiquitous from daily lives to engineering applications, and the increasing demand for efficient heat transfer in the industry requires further understanding of it, especially on small scales where the knowledge on conventional scales may no longer apply. This study investigated the effects of heater size, contact angle, surrounding vessel size, and surface structure on nucleate boiling heat transfer occurring on horizontal flat surfaces where the heater size is comparable to the two-dimensional ''most dangerous'' Taylor wavelength, aiming to comprehend the parametric effects on nucleate boiling heat flux and critical heat flux (CHF) and revisit the hydrodynamic theory near critical condition.

Saturated water at one-atmosphere pressure was boiled on horizontal flat copper discs of diameters 1.0, 1.5, and 2.0 cm, respectively. The contact angle was varied from about 10 to 80 deg by controlling thermal oxidation of the discs, while the surrounding vessel size was changed by placing glass tubes of different inner diameters around the discs. The surface structure in the form of microgrooves was fabricated by sanding the disc top surface. Boiling heat transfer data were obtained up to CHF. Boiling curves and CHFs measured under different experimental configurations were compared in terms of each parameter. Rohsenow's correlation was employed to assess the parametric effects on nucleate boiling heat flux quantitatively. Vapor removal patterns were photographed in nucleate boiling regime and near CHF. Vapor jet diameter and the dominant wavelength at water-steam interface were measured from the photographs for the well wetted discs and used to predict the corresponding CHF based on the hydrodynamic theory.

For well wetted surfaces, the boiling curve was insensitive to the heater size, but the CHF increased when the heater size was reduced from 2.0 to 1.0 cm. Improving the wettability delayed the onset of nucleate boiling and shifted the boiling curve to the right while enhancing the CHF substantially. Enlarging the liquid-holding vessel hardly affected the boiling curve at low heat fluxes but improved the CHF slightly. The structured surface featured higher nucleate boiling heat flux resulting from more active nucleation sites but showed no advantage in CHF over the plain surface of similar size and contact angle. In this study, the highest measured CHFs for plain and structured surfaces are close and both about 2.1 times Zuber's CHF prediction for infinite horizontal flat plates. They were obtained on well wetted 1.0-cm-diameter discs surrounded by large vessels.

Boiling heat transfer research spans several decades with extensive data accumulation from several experiments. Several mechanistic models and empirical correlations have also been put forth but their applicability is limited to the narrow range of parameters over which they have been developed. For the past few decades several numerical methods have also been developed and gained considerable momentum to study boiling process. The most popular of these numerical methods are volume of fluid method, level set method and front tracking methods, however some other computational methods like Lattice Boltzmann, Moving particle semi-implicit gridless and cellular automata SIMPLER methods have also been developed.

In the present study level set method is used to simulate the entire boiling curve in a temperature controlled mode spanning all the three regimes viz. nucleate, transition and film boiling with a unified numerical model supplemented with correlations specifying nucleation site density and bubble waiting time. In order to improve the performance of the code parallel computing has also been implemented. Both two-dimensional and three-dimensional simulations have been done for saturated water with different contact angles for a horizontal surface with uniform wall superheat applied to it. Temporal and spatial averaged wall heat flux and wall void fraction computed for a fixed wall superheat case are plotted and analyzed. For a specified contact angle and by incrementing the wall superheat as two independent input parameters, the entire boiling curve along with vapor removal patterns capturing its vital points like the maximum heat flux and minimum heat flux are shown.

The two-dimensional assumption yields opposite trend for the nucleate boiling regime heat flux variation with contact angle but this anomaly was not observed in the three-dimensional simulations which infers that the two-dimensional assumption is an incorrect representation to study the essential physics of the boiling process. The trend of critical heat flux with contact angle was found to be decreasing with increase in contact angle for both two-dimensional and three-dimensional case, with the trend being steeper for the former.

Wall void fraction was found to increase with increase in wall superheat as different regimes of boiling were traversed, and also with increase in contact angle at a given wall superheat. Mushroom type vapor bubbles are seen in the nucleate boiling regime with liquid macrolayer trapped underneath it while long column of sustained vapor is seen at the critical heat flux condition continuously being fed by nucleating cavities at the surface. Upon increasing the wall superheat beyond critical heat flux the negative slope of the boiling curve is captured characterized by the transition boiling regime with intermittent liquid solid contacts seen. Finally, the transition to film boiling is seen with entire surface covered with superheated vapor and wall void fraction reaching unity. Energy partitioning from wall into liquid, interface and microlayer has also been examined for the 3D cases. For the 3D coarse grid case it was found that, as the wall void fraction increases the percent energy going into liquid decreases from about 85% in lower nucleate boiling to 6% in film boiling while the microlayer contribution peaks around CHF to a value of about 45%. The energy partition from the fine grid cases were however inconclusive as they couldn`t be run long enough to obtain any meaningful results.

The experimental data of a simulation for a desalination system utilizing swirl flow phase separation is presented. Previously, the author investigated a desalination system that utilizes flash evaporation and swirl flow phase separation. To better understand the requirements for high separation efficiency, adiabatic experiments with air and water were conducted with a similar test section. In the experiments, air and water are premixed prior to entering the injection tubes and injection passages. The two-phase mixture tangentially injected into the test section experiences centrifugal force. Centrifugal force pushes heavier liquid toward the wall and air being lighter stays in the inner core. A retrieval tube that is placed in the center of the separator tube collects the air from the air core. Superficial air and water velocities were parametrically varied. Air core diameter, air core coverage length, and separation efficiency were measured. It was found that air core coverage length is a significant factor in determining the efficacy of phase separation. The diameter of the injector and the position of the retrieval tube inlet were also determined to be important parameters in the system.

Due to the complex nature of the subprocesses involved in nucleate boiling, it has not been possible to develop comprehensive models or correlations despite decades of accumulated data and analysis. Complications such as the presence of dissolved gas in the liquid further confound attempts at modeling nucleate boiling. Moreover, existing empirical correlations may not be suitable for new applications, especially with regards to varying gravity level. More recently, numerical simulations of the boiling process have proven to be capable of reliably predicting bubble dynamics and associated heat transfer by showing excellent agreement with experimental data.

However, most simulations decouple the solid substrate by assuming constant wall temperature. In the present study complete numerical simulations of the boiling process are performed--including conjugate transient conduction in the solid substrate and the effects of dissolved gas in the liquid at different levels of gravity. Finite difference schemes are used to discretize the governing equations in the liquid, vapor, and solid phases. The interface between liquid and vapor phases is tracked by a level set method. An iterative procedure is used at the interface between the solid and fluid phases.

Near the three-phase contact line, temperatures in the solid are observed to fluctuate significantly over short periods. The results show good agreement with the data available in the literature. The results also show that waiting and growth periods can be related directly to wall superheat. The functional relationship between waiting period and wall superheat is found to agree well with empirical correlations reported in the literature. For the case of a single bubble in subcooled nucleate boiling, the presence of dissolved gas in the liquid is found to cause noncondensables to accumulate at the top of the bubble where most condensation occurs. This results in reduced local saturation temperature and condensation rates. The numerical predictions show reasonable agreement with the results from experiments performed at microgravity. For nucleate boiling at microgravity the simulations predict a drastic change in vapor removal pattern when compared to Earth normal gravity. The predictions match well with experimental results. However, simulated heat transfer rates were significantly under-predicted.

The two-phase heat transfer technology has been a major thermal control hot-point for cooling or heat rejection during the last decade. Two-phase pumped cooling systems are applied when it is required to maintain a very stable temperature for heat dissipation in a system. A novel additively manufactured evaporator for two-phase thermal control has been developed at NASA Jet Propulsion Laboratory. The Two-Phase Mechanically Pumped Loop (2PMPL) permits latent heat transfer to be implemented with much wider breath of control authority compared to capillary-based systems while alleviating the system sensitivity to pressure drops. The flexibility of the 2PMPL stays in the capability to reclaim waste heat dissipation and to reject heat from discrete locations (multiple evaporators and condensers) and to control loop temperature with minimal control power. Another key aspect of this technology is the ability to integrate multi-phase heat transfer provisions directly with structural elements thanks to the conjunct use of the Additive Manufacturing Technology.

A collaboration between the NASA Jet Propulsion Laboratory, University of California Los Angeles and University of Michigan has been established to develop numerical models for the prediction of the thermal behavior and performance of the 2PMPL. The Boiling and Phase Change Heat Transfer Laboratory in the Department of Mechanical and Aerospace Engineering at University of California Los Angeles has focused on the understanding and capture of the micro-scale evaporation occurring in the porous structure of the evaporator.

The main goal of the research has been to develop a modelling tool that can be used to design and optimize the evaporator. An all-encompassing numerical simulation can be carried out to predict the operational thermal behaviour of the evaporator taking into account the effect of the liquid-vapor interface at the wick-to-vapor border. The numerical model allows to study the effect of different parameters, such as boundary conditions (inlet temperature, input heat loads, etc.), geometrical parameters and wick (effective thermal conductivity, permeability, pore size, etc.) and fluid properties.

An experimental setup has been built in order to characterize the heat transfer within an additively manufactured porous sample and in particular its evaporative heat load under certain heat inputs. The experimental campaign has served also as validation for the numerical results and for the characterization of the transient phenomena such as dry-out.

Knowledge of the physical mechanisms governing bubble dynamics and two-phase heat transfer is critical in order to accurately predict and scale the performance of two-phase systems, most importantly in low-g environments. To better understand flow boiling, especially under microgravity conditions, the dynamics of single and multiple bubbles under different levels of bulk liquid velocity, surface orientation, contact angle, and substrate materials are studied in this work. Microfabricated cavities at the center of a flat heating surface are used to generate bubbles. Silicon and aluminum are used as substrate materials, with contact angles of 56o and 19o, respectively, with water as test liquid. The investigated bulk liquid velocities ranged from 0 m/s to 0.25 m/s, while surface orientation varies from horizontal to vertical, through 30^o, 45^o and 60^o, and cavity spacing from 0.4 mm to 1.2 mm, in upflow conditions. Bulk liquid temperature was set close to saturation temperature, with bulk liquid subcooling less than 1 ^oC, and wall superheat was maintained between 5.0 ^oC and 6.0 ^oC.

Based on the experimental data, a simple force balance model was developed, and is used to develop a model to predict bubble lift off. These forces are the lift force (F_l), the buoyancy force (F_b), the surface tension force (F_s), the contact pressure force (F_cp), and the inertia of both the vapor and the liquid displaced by the growing bubble. It is showed that at the instant when bubble lift off is initiated, the sum of forces acting on the bubble is equal to zero (and then becomes positive in the direction normal to the heater). This force balance is used to develop an expression for bubble lift off diameter. It also is found that for single and merged bubbles, when lift off occurs, buoyancy and lift forces are the only forces acting on the bubble, regardless of orientation, contact angle and flow velocity, and that for all cases, the ratio (F_b + F_l) / A_l is constant and equal to 2.25 N/m², where A_l is the bubble surface area at lift off.