"The importance of two-phase heat transfer for thermal management of aerospace avionic systems has become increasingly important as these systems have become miniaturized. Embedded active cooling systems are used to remove heat from processors and other electronic components and transferring this heat to radiators or other heat exchangers. As the characteristic dimension of flow channels for two-phase flow becomes comparable to bubble size, the mini-channels (< 3 mm) used to direct the cooling fluid can complicate nucleate boiling heat transfer. Bubbles can encounter other heated walls, rapidly expanding and greatly reducing heat transfer as well as causing pressure oscillations and flow instabilities. The use of eletrohydrodynamic (EHD) effects, through the introduction of non-uniform electric fields, can help mitigate this problem by altering the behavior of nucleating bubbles. A combined experimental and computational study was undertaken using HFE-7100, an engineered fluid used in heat transfer applications, to investigate the potential for enhancement of nucleate boiling using EHD effects induced by applying a non-uniform electric field. In the experimental study, a minichannel was constructed consisting of an upper and lower copper electrode and glass side walls to allow visualization. The channel height and width were 3mm and 4.76 mm respectively, representative of the minichannel regime. The upper electrode was grounded while the lower electrode was heated and biased to high voltage. Optical imaging combined with post-processing and statistical analysis was used to quantify the effect of EHD on the bubble behavior. Bubbles were found to form preferentially on nucleation sites resulting from imperfections in the heated copper surface over artificially created nucleation sites. When a high voltage is applied across the electrodes, the electric field enhancement along the rim of the nucleation site is believed to influence the force balance on the forming bubble and thereby influence the bubble departure size and frequency. EHD forces also act on the bubble surface as a result of the variation in permittivity between the liquid and vapor phases, altering its shape as has been previously reported in the literature. Test results are presented that demonstrate that the application of EHD increases the nucleation site density on the heated surface and increase the bubble departure frequency from individual sites. In addition, test results are presented to show that EHD forces alter the shape of bubbles during growth and the vertical position of the detached bubbles as they are carried along in the cross flow. To better understand the underlying phenomena affecting the bubble shape and departure frequency, a numerical simulation of the bubble growth and departure was performed using COMSOL multiphysics software customized to incorporate a user-defined body force based on the Maxwell Stress Tensor. Tracking of the bubble surface, including coalescence and breakup was incorporated using the phase field variable method in which the Navier-Stokes and heat transfer equations are solved for each phase of the fluid. Results from the simulations confirmed the sensitivity of the bubble elongation and neck formation to the nucleation site geometry, specifically the angle along the rim where field enhancement occurs. The enhanced constriction of the bubble neck resulted in early detachment of bubbles when compared to simulations in which EHD was not applied. This finding provides some insight into the higher bubble departure frequency and nucleation site density observed in the experiment. The results from the combined experimental and numerical study suggest that EHD enhancement may provide a mechanism for extending the use of nucleate heat transfer to minichannels, thereby enabling additional options for cooling in compact, embedded systems. "

Thesis--Ph. D., University of Hong Kong. / Mimeographed.

An alternating current electroosmotic micro pumping device has been designed, experimentally tested and theoretically analyzed using an electrohydrodynamic theoretical model applied to a computer simulation model. The device SP-1 is a microelectrode array which uses the principal of AC electroosmosis (EO), ions driven along microelectrode surfaces by coulomb forces produced by tangential electric fields. These ions, when driven, induce a net fluid motion due to viscous drag forces. Three submerged microelectrode wires were deposited on a substrate using microfabrication techniques such that a square spiral geometry was formed. Device SP-1 received asymmetrically applied AC signals creating a travelling wave of potential and resulted in a net fluid flow across the microelectrode array. Microsphere tracer particles were suspended in ethanol to measure the fluid velocity to determine pumping performance and the experimental operating frequency at which maximum fluid velocity is achieved. The experimental results were reviewed and at an AC signal frequency of 125 Hz, device SP-1 was capable of pumping ethanol at a fluid velocity of approximately 270 μm/s. The experimental results were in good agreement with the theoretical predictions produced using the computer simulation model. In addition, the computer simulation model predicted a similar flow profile to those previously predicted and experimentally observed. Overall, novel micropumping device SP-1 was found to produce a net flow comparable to previously tested devices and a computer simulation framework capable of analyzing future micropump design concepts was developed. / Thesis (Master, Mechanical and Materials Engineering) -- Queen's University, 2011-04-01 17:12:02.908

No description available.

No description available.

Electrohydrodynamic (EHD) micropumps have been developed and used in many diverse applications such as in microscale liquid cooling and various microfluidic systems. The objective of this research is to investigate different methods of enhancing the performance of ion drag EHD micropumps. In particular, the effect of electrode surface topology, applied electric field and doping agent in the dielectric liquid were investigated. The effect of 3D sharp features on the electrodes on charge injection in HFE 7100 as dielectric fluid was studied under an applied DC electric field. Micro and nano-scale features with high aspect ratio were developed on smooth copper electrodes by chemical etching or through electrophoretic deposition of single walled carbon nanotube (SWCNT). The spacing between the electrodes was kept at 250 µm. A reduction factor of 5 was achieved for SWCNT electrodes compared to the smooth case for the onset of charge injection. This study was then extended to determine its effects on the performance of ion drag EHD micropumps with 100 pairs of interdigitated electrodes. The emitter electrodes (20 µm) were half the width of the collector electrodes (40 µm), with one pump having an inter-electrode spacing of 120 µm and the other with 40 µm. Each micropump had a width of 5 mm and a height of 100 µm. SWCNT was deposited on the emitter electrodes of the micropump to generate a maximum static pressure of 4.7 kPa at 900 V, which is a 5 fold increase compared to the pump with smooth electrodes. Flow rate at no back pressure condition was improved by a factor of 3. The effect of Ferrocene as a doping agent in the working fluid HFE 7100 was studied under DC voltages. A maximum static pressure of 6.7 kPa was achieved at 700 V with 0.2% weight based doping agent, 11 times higher than when there was no doping agent at the same applied voltage. When there was no back pressure the pump generated a maximum flow rate of 0.47 mL/min at 700 V with 0.05% doping agent which is 9 times greater than with no doping agent. The effect of pulsed voltage on the performance of ion drag EHD micropump has been studied to exploit the displacement current at the sudden change of applied voltage magnitude. A range of pulse repetition rate and duty cycle were found to significantly enhance the pump performance. Static pressure generation was up to 75% and 88% greater at an optimal pulse repetition rate and duty cycle, respectively, compared to the average of the two DC levels. The effect of external flow on the discharge characteristics of an injection micropump was studied with DC volts. Higher discharge current and lower threshold voltage for the onset of charge injection in case of co-flow compared to the static case was observed. There was an optimum flow rate to generate maximum current for both co and counter-flow cases. / Thesis / Doctor of Philosophy (PhD)

In this investigation an electric field was applied to a phase change thermal storage system while it was discharging energy. The phase change material used was octadecane. Octadecane is a high purity dielectric material that has a melting temperature close to room temperature. The material was forced to solidify using a heat exchanger mount below the phase change material, cold water flowed through the heat exchanger to ensure it maintained a constant temperature below the melting temperature of the phase change material. By applying -8kV to 9 electrodes – positioned in the phase change material – and by using the heat exchanger as an electrical ground – an electric field was generated in the phase change material. The electric field caused unbalanced body forces in the fluid which generated electro-convection in the fluid. The system was designed such that electro-convection is the only source of convection in the system to isolate the effects of electro-convection, allowing for the underlying physics of electro-convection to be studied easier. To understand the effects of applying electro-convection, a case where there is no applied voltage on the electrodes was compared to a case where there was -8 kV applied to the electrodes. Experiments showed that the effect of applying electro-convection depends on the initial temperature; however, it was found that the improvement after two hours was less than 10%. For a wall temperature of 8.5℃ and an initial temperature of 50℃ - the melting temperate of octadecane is 28℃- then the maximum enhancement of the energy extracted is 50%, but two hours after the start of the test the enhancement approached zero. For a wall temperature of 8.5℃ and an initial temperature of 30℃, the maximum enhancement is 10% and similarly fall to zero after a few hours of application. A simple analytical model was developed. The experimental and numerical results showed that at the early stages of energy discharge the electro-convection case had a large improvement compared to a pure conduction case, however as time progresses this improvement decreases. The explanation for the trend is that adding convection only increases the rate that energy is taken out of the liquid, thus the maximum improvement is bounded by the amount of sensible energy in the liquid phase change material, once this sensible energy is removed applying electrohydrodynamics is no longer beneficial. / Thesis / Master of Applied Science (MASc)

The focus of this Ph.D. thesis is the theoretical, computational, and experimental analysis of electrohydrodynamics and ionization in the Array of Micromachined UltraSonic Electrospray (AMUSE) ion source. The AMUSE ion source, for mass spectrometry (MS), is a mechanically-driven, droplet-based ion source that can independently control charge separation and droplet formation, thereby conceptually differing from electrospray ionization (ESI). This aspect allows for low voltage soft ionization of a variety of analytes and flexibility in the choice of solvents, providing a multifunctional interface between liquid chromatography and mass spectrometry for bioanalysis. AMUSE is a versatile device that operates in an array format, enabling a wide range of configurations, including high-throughput and multiplexed modes of operation. This thesis establishes an in-depth understanding of the fundamental physics of analyte charging and electrokinetic charge separation in order to enhance droplet charging and ionization efficiency. A detailed electrohydrodynamic (EHD) computational model of charge transport during the droplet formation cycle in the AMUSE ion source is developed, coupling fluid dynamics, pressure and electric fields, and charge transport in multiphase flow. The developed EHD model presents a powerful tool for optimal design and operation of the AMUSE ion source, providing insight into the microscopic details of physicochemical phenomena, on the microsecond time scale. Analyte charging and electrohydrodynamics in AMUSE are characterized using dynamic charge generation measurements and high-spatial-resolution stroboscopic visualization of ejection phenomena. Specific regimes of charge transport, which control the final droplet charging, have been identified through experimental characterization and simulations. A scale analysis of the ejection phenomena provides a parametric regime map for AMUSE ejection modes in the presence of an external electric field. This analysis identifies the transition between inertia-dominated (mechanical) and electrically-dominated (electrospraying) ejection, where inertial and electric forces are comparable, producing coupled electromechanical atomization. The understanding of analyte charging and charge separation developed through complimentary theoretical and experimental investigations is utilized to improve signal abundance, sensitivity, and stability of the AMUSE-MS response. Finally, these tools and fundamental understanding provide a sound groundwork for the optimization of the AMUSE ion source and future MS investigations.

published_or_final_version / Electrical Engineering / Master / Master of Philosophy

Consideration is given to the conditions under which floating drops of water, here referred to as water boules, form, exist and coalesce. Particular emphasis is placed on the part played by electric charge and electric fields in these processes. The literature is reviewed in terms of both the phenomenon of floating drops and of the development of hydrostatics, hydrodynamics and electrohydrodynamics as applicable to the subject. . Experimental investigations to ascertain the boundary conditions to the influence of such electrical forces are described, together with observations of the connected electrical events. It is confirmed that boules will fail to form at all, i) under conditions of high humidity, and ii) in the presence of an electric field greater than a certain value. This is investigated experimentally, and shown to be approximately 34kV/m, this figure being about two-thirds that previously reported. Boules traversing a plane water surface are demonstrated to acquire additional charge in the process. In the case of drops dispensed from a grounded source, forming boules and crossing a bulk water surface some 15cm wide, the additional charge gathered is significant. Boules of 0.055g mass were found to have a mean charge of 1.6 x 10-12C on leaving a water surface, having arrived as drops with an average charge of 5.8 x 10-14C. Possible charging mechanisms are discussed. The origin of the initial drop charge is considered, and measurements of this are presented from (i), conventional Faraday cup determinations, and (ii), induction methods applied to free-falling drops. Experimental investigation of the time-dependent electrical records of the coalescence of a dispensed drop with a plane water surface shows the whole coalescence process to have a two-part form. This detail is commonly hidden within more conventional charge-transfer measurements. For the coalescences investigated experimentally an small initial event is shown to occur, involving a charge transfer in the range 1.2 – 4.8 x 10-12C. Oscillograms taken from a large number of coalescences show this preliminary event to be a general feature of the coalescence process, with a number of such traces being appended to the thesis. This initial event is followed by a larger one where the signs of the signals from the drop and the bulk surface are opposite to those of the initial event, and whose potential magnitude is broadly in agreement with that anticipated by double layer disruption. The interfacial potential difference necessary for the onset of instability and subsequent coalescence in the case of closely opposed drops is shown to be dependent on the relative humidity of the ambient air. Consideration is given to G I Taylor’s equation describing the critical potential for the onset of instability between closely spaced drops, and this is shown experimentally to require correction for different humidities. It is demonstrated that the critical potential, Vc, at a relative humidity of 100% is approximately 50% of that at 40% RH. Possible reasons for this are discussed, drawing attention to the problem of establishing an accurate DC relative permittivity value for vapour-laden air in small interfacial gaps. The rôle of evaporation in modifying the system geometry is considered experimentally and theoretically, and shown to be significant only for humidities < 50%. The complex nature of the interface in the case of very small air-gaps is discussed, together with the implications of these investigations for the interfacial stability of a floating drop or boule system. A theoretical model based on a consideration of the complex liquid-air-liquid interface as a capacitive system is developed, and shown to be in good agreement with practical observations. This model demonstrates that the parts played by electrical forces, together with environmental factors, are likely to be significant in terms of coalescence at stages prior to gap thinning to the point where London/van-der-Waals forces become dominant. Interfacial potentials are calculated in a boule system at a number of times between 0.1 and 10 seconds, and shown to be sufficient to promote instability and coalescence. Full data based on a number of values of instability potentials is appended to the thesis. Development of the model raises questions concerning the validity of currently accepted values both for interfacial stability in small gaps and for the relative permittivity of humid air in similar situations. Suggestions are made for future work in such areas, together with possible methodologies. The phenomenon of floating water drops is therefore shown to be compatible with the general coalescence process, the event time being modified by such diverse factors as the impact energy with the surface, the ambient humidity and the magnitude of the initial drop charge. The latter is shown to be the dominant factor in the case of drops arriving on a clean surface with low kinetic energies, with the small charge inherent on any water drop being sufficient to produce potentials adequate to promote eventual instability.