Plasma synthetic jet actuators are investigated experimentally, in which the geometrical design of single dielectric barrier discharge (SDBD) plasma actuators is modified to produce zero-mass flux jets similar to those created by mechanical devices. The SDBD plasma actuator consists of two rectangular electrodes oriented asymmetrically and separated by a layer of dielectric material. Under an input of high voltage, high frequency AC or pulsed DC, a region of plasma is created in the interfacial air gap on account of electrical breakdown of the ambient air. A coupling between the electric field in the plasma and the neutral air near the actuator is introduced, such that the latter experiences a net force which results in a horizontal wall jet. This effect of the actuator has been demonstrated to be useful in mitigating boundary layer separation in aerodynamic flows. To increase the impact that a plasma actuator may have on the flow field, this research investigates the development and characterization of a novel flow control device, the plasma synthetic jet actuator, which tailors the residual air in the form of a vertical jet resembling conventional continuous and synthetic jets. This jet can be either three dimensional using annular electrode arrays, or nearly two dimensional using two rectangular strip exposed electrodes and one embedded electrode. Detailed measurements on the isolated plasma synthetic jet reveal that pulsed operation of the actuator results in the formation of multiple counterrotating vortical structures in the flow field. The output jet velocity and momentum are found to be higher for unsteady pulsing as compared to steady operation. In the case of flow over a flat plate, the actuator is observed to create a localized interaction region within which the baseline flow direction and boundary layer characteristics are modified. The efficiency of the actuator in coupling momentum to the neutral air is found to be related to the plasma morphology, pulsing frequency, actuator dimension, and input power. An analytical scaling model is proposed to describe the effects of varying the above variables on the output jet characteristics and actuator efficiency, and the experimental data is used for model validation.

Atherosclerosis and related cardiovascular diseases (CVDs) are amongst the largest causes of morbidity and mortality in the developed world, causing considerable monetary pressure on public health systems worldwide. Atherosclerosis is characterised by the build up of vascular plaque in medium and large arteries and is a direct precursor to acute vascular syndromes such a myocardial infarction, stroke or peripheral arterial diseases. The causative factors leading to CVD still remain relatively poorly understood, but are becoming increasingly identifiable as a dysfunction of the endothelial cells that line the arterial wall. It is well known that the endothelium responds to the prevailing fluid mechanic (i.e. haemodynamic) environment, which plays a crucial role in the localised occurrence of atherosclerosis near vessel bends and bifurcations. In these areas, disturbed haemodynamics lead to flow separation and very low wall shear stress (WSS), which directly affects the functionality of the endothelium and impedes the transport of important blood borne agonists and antagonists. Detailed full field measurements assessing complex haemodynamics are sparse and consequently this thesis aims to address some of the important questions related to arterial haemodynamics and CVD by performing in-vitro flow measurements in physiologically relevant conditions. In particular, this research develops and uses state-of-the-art Particle Image Velocimetry (PIV) techniques to measure three-dimensional velocity and WSS fields in scaled models of the human carotid artery. For this purpose, the necessary theoretical and experimental concepts are developed and in-depth analyses of the underlying factors affecting the local haemodynamics and their relation to CVD are carried out. In the first part, a methodology for the construct of transparent hydraulic flow phantoms from medical imaging data is developed. The arterial geometries are reproduced in optically clear silicone and the flowing blood is modelled with a refractive index matched blood analogue. Subsequently, planar and Stereo-PIV techniques are developed and verified. A novel interfacial PIV (iPIV) technique is introduced to directly measure WSS by inferring the velocity gradient from the recorded particle images. The new technique offers a maximal achievable resolution of 1 pixel and therefore removes the resolution limit near the wall usually associated with PIV. Furthermore, the iPIV performance is assessed on a number of numerical and experimental test cases and iPIV offers a significantly improved measurement accuracy compared to more traditional techniques. Subsequently, the developed methodologies are applied in three studies to characterise the velocity and WSS fields in the human carotid artery under a number of physiological and experimental conditions. The first study focuses on idealised vessel geometries with and without disease and establishes a general understanding of the haemodynamic environment. Secondly, a physiological accurate vessel geometry under pulsatile flow conditions is investigated to provide a more realistic representation of the true in-vivo flow conditions. The prevailing flow structure in both cases is characterised by flow separation, strong secondary flows and large spatial and temporal variations in WSS. Large spatial and temporal differences exist between the different geometries and flow conditions; spatial variations appear to be more significant than transient events. Thirdly, the three-dimensional flow structure in the physiological carotid artery model is investigated by means of stereoscopic and tomographic PIV, permitting for the first time the measurement of the full 3D-3C velocity field and shear stress tensor in such geometries. The flow field within the model is complex and three-dimensional and inherently determined by the vessel geometry and the build up of an adverse pressure gradient. The main features include strong heliocoidal flow motions and large spatial variations in WSS. Lastly, the physiological implications of the current results are discussed in detail and reference to previous work is given. In summary, the present research develops a novel and versatile PIV methodology for haemodynamic in vitro studies and the functionality and accuracy is demonstrated through a number of physiological relevant flow measurements.

Inclined negatively buoyant jets are commonly used to dispose brine effluent produced by desalination plants. Desalination and associated research has expanded in recent years due to the continued depletion and degradation of natural potable water sources. Desalination plants are the preferred option for meeting water demand deficits in many countries around the world. Inclined negatively buoyant jets are produced when the brine is discharged at an upward inclined angle via an offshore pipeline and diffuser system. Previous experimental studies have focused on the rapid mixing and dilution achieved by these discharges, as well as geometric parameters. Dilution measurements between these experimental studies vary significantly, which is possibly due to variations in the location of a lower boundary on observed flow behaviour. In the present study, velocity field information is experimentally measured for inclined negatively buoyant jets and compared to integral model predictions. Experiments are conducted with and without a lower boundary influencing observed flow behaviour, thus allowing the effects of a lower boundary to be determined. The particle tracking velocimetry experimental technique is employed to measure near field velocities of these discharges. Firstly, discharges with source angles between 15\degree and 75\degree are investigated without boundary influence in stationary ambient conditions. The source was a minimum of 655 mm above the bottom of the experimental tank to ensure there was no lower boundary influence on observed behaviour. Time-averaged and fluctuating data are extracted along the trajectory of discharges. All non-dimensionalised geometric and centreline velocity parameters are found to collapse. Empirical coefficients are compared to previous experimental studies and integral model predictions. A new detrainment model is developed to predict the behaviour of inclined negatively buoyant jets without boundary influence. The model further develops recent attempts to allow for buoyancy flux reduction along the flow path. The reduction in buoyancy flux is dependent on the local parameters of the flow and simulates experimentally observed detrainment. Dilution, geometric, and velocity predictions are found to be improved over previous models when compared to experimental data. Finally, a raised platform was placed inside the experimental tank to determine the influence of a lower boundary on inclined negatively buoyant jets. Source angles of 30\degree, 45\degree, and 60\degree are investigated at three different non-dimensional source heights. The lower boundary is horizontal and ambient conditions are again stationary. Discharges impinge the lower boundary before forming a radially spreading layer along the boundary. Geometric and velocity data are compared to the first set of experiments in this study to determine the influence of the lower boundary on observed flow behaviour. Empirical coefficients at maximum height are similar with and without the influence of the boundary, whereas coefficients are substantially influenced at the return point when the boundary is present.

Tsunami are a very dangerous natural hazard, as highlighted in recent years by the Indian Ocean Tsunami of 2004 and the Japan Tsunami of 2011. In the last decade, tsunami have claimed hundreds of thousands of lives, and caused billions of dollars in damage around the world. The hazard posed to coastal communities by tsunami is expected to increase in the future, due to population growth, intensification of coastal development and sea level rise due to climate change. Tsunami may be generated by a number of different source mechanisms. One such source mechanism is a submarine landslide, which can occur in a number of marine environments containing significant sediment accumulation on a sloping seafloor. The high amplitudes and rapid celerities of landslide-generated tsunami make them very dangerous to communities in the vicinity of the landslide, although these waves do not possess the potential for transoceanic devastation. The objectives of this research project are to carry out a series of two-dimensional physical experiments investigating the waves generated by a rigid block landslide moving along a horizontal boundary. The use of a horizontal boundary has the advantage that waves propagating in the offshore and onshore directions may be measured (unlike previous studies using sloping boundaries). The landslide motion is provided by a mechanical system, allowing testing of a broad range of motion, and isolation of the wavemaking properties of different phases of landslide motion. Experiments are carried out in a 14.66 m long flume, with width 0.25 m and working depth 0.50 m. A false floor installed in the flume provides the sliding surface for the landslide motion, and houses the mechanical system. A series of preliminary particle tracking velocimetry experiments confirm the ability of the mechanical system to achieve its velocity targets to within 5% or better, depending on the parameters of the landslide motion. Full spatial and temporal resolution of the wave field is achieved using a laser-induced fluorescence technique to identify the air-water interface to sub-pixel accuracy. The measurements obtained using laser-induced fluorescence are validated against measurements from a resistance wave gauge, with sub-millimetre agreement. In an additional experiment, the particle tracking velocimetry technique provides measurements of the subsurface velocity field. The landslide motion during all experiments consists of an initial period of constant acceleration, followed by a period of constant velocity, followed by a deceleration to rest (at the same rate as the initial acceleration). The landslide acceleration generates two dispersive packets of waves, travelling in the offshore and onshore directions. The offshore-propagating wave packet contains a leading crest and the onshore-propagating wave packet contains a leading trough, with both waves approaching the shallow water limit. A free surface depression forms above the landslide during its constant-velocity motion, and its amplitude may be predicted to within approximately 20% using standard hydraulic theory (considering a frame of reference moving with the landslide). The offshore-propagating waves passing over the landslide cause the amplitude of this depression to fluctuate over time. The deceleration of the landslide generates two additional packets of waves with the opposite polarity to the waves generated by the landslide acceleration. The full spatial and temporal resolution of the generated wave field allows the calculation of the potential energy within the wave field. Additionally, the energy (and mass) within the onshore- and offshore-propagating wave packets may be estimated by calculating these quantities within the onshore and offshore regions of the experimental domain. The wave packets generated by the initial landslide acceleration transport positive mass in the offshore direction, and negative mass in the onshore direction. This mass transport is balanced by the waves generated during the deceleration of the landslide. The nondimensional landslide acceleration, landslide Froude number and submergence depth are varied during the physical experiments. The landslide Froude number has the greatest effect on the behaviour of the generated wave field. At low Froude numbers, the wave field is dominated by the waves generated by the acceleration and deceleration of the landslide. As the Froude number increases, the onshore-propagating waves become negligible in amplitude compared to the offshore-propagating waves. Additionally, the free surface depression increases in amplitude and a group of short-wavelength waves become trapped behind the landslide. These waves exhibit highly nonlinear behaviour at landslide Froude numbers greater than 0.5. The simple experimental geometry allows comparison between the measured wave fields with the predictions of three mathematical models. Two inviscid-irrotational models, differing in their treatment of the bottom boundary condition, provide comparisons over the entire parameter space. These models under-predict the amplitudes of the generated waves, and fail to correctly predict the ongoing interaction between the landslide and the offshore-propagating waves. The inclusion of bottom boundary nonlinearity improves the predictions of the amplitude of the leading waves, and the potential energy within the wave field. However, both of the inviscid models do not predict the extent of wave trapping behaviour behind the landslide observed in the experiments. A viscous model, formulated in the DNS solver Gerris, improves the predictions of wave trapping (and amplitude in general) in one experiment. Although the model still under-predicts the amplitudes of the generated waves, it correctly predicts the amplification of the waves behind the landslide during its constant-velocity motion. The failure of the inviscid models to predict the amplitudes of these waves can be mostly attributed to the linearised free surface condition used by both models. The presence of the turbulent wake may also have a secondary effect on these predictions. An extension of the linear inviscid-irrotational model to three dimensions allows the effect of the landslide width on the amplitudes of the generated waves to be determined. As the width increases, the behaviour of the waves approaches the two-dimensional limiting case.

The focus of this thesis is the study of turbulent solid-liquid stirred suspensions, which are involved in many common unit operations in the chemical, pharmaceutical and food industries. The studies of two-phase flows present a big challenge to researchers due to the complexity of experiments; hence there is a lack of quantitative solid and liquid hydrodynamic measurements. Therefore, an investigation of turbulence modulation by dispersed particles on the surrounding fluid in stirred vessels has been carried out, via two-phase fluorescent Particle Image Velocimetry (FPIV) and micromixing experiments. The main property of interest has been the local dissipation rate, as well as root-mean-square (rms) velocities and turbulent kinetic energy (TKE) of the fluid. Initially a single-phase PIV study was conducted to investigate the flow field generated by a sawtooth (EkatoMizer) impeller. The purpose of this study was to gain insight into various PIV techniques before moving on to more complex two-phase flows. Subsequently stereo-, highspeed and angle-resolved measurements were obtained. The EkatoMizer formed a good case study as information regarding its hydrodynamics is not readily available in literature, hence knowledge has been extended in this area. An analysis of the mean flow field elucidated the general structure of fluid drawn into the impeller region axially and discharged radially; the latter characterised the impeller stream. The radial rms velocity was considered to represent best the system turbulence, even though the tangential rms velocity was greater close to the blade; however the radial component was more prevalent in the discharge stream. Due to differences in rms velocities, TKE estimates obtained from two and three velocity components deviated, being greater in the latter case. Integral (1-D and 2-D) length scales were overestimated by the quantity W / 2 in the impeller region. Ratios of longitudinal-to-lateral length scales also indicated flow anisotropy (as they deviated from 2:1). The anisotropy tensor showed that the flow was anisotropic close to the blade, and returned to isotropy further away from the impeller. Instantaneous vector plots revealed vortices in the discharge stream, but these were not associated with flow periodicity. Alternatively, the vortex structures were interpreted as low frequency phenomena between 0-200 Hz; macro-instabilities were found to have a high probability of occurrence in the discharge stream. Dissipation is the turbulent property of most interest as it directly influences micromixing processes, and its calculation is also the most difficult to achieve. Its direct determination from definition requires highly resolved data. Alternative methods have been proposed in the literature, namely dimensional analysis, large eddy simulation (LES) analogy and deduction from the TKE balance. All methods were employed using 2-D and 3-D approximations from stereo-PIV data. The LES analogy was deemed to provide the best estimate, since it accounts for three-dimensionality of the flow and models turbulence at the smallest scales using a subgrid scale model. (Continues...).

Fluorescence lifetime imaging microscopy (FLIM) is a powerful technique to non-invasively map the physical and chemical environment within microfluidic devices. In this work FLIM has been used in conjunction with a variety of other techniques to provide a greater insight into flow behaviour and fluid properties at the microscale. The pH-sensitive fluorescent dyes, fluorescein and C-SNARF 1, have been used to generate pH maps of microfluidic devices with a time-gated camera and a time-and-space-correlated single photon counting (TSCSPC) detector, respectively. Using time-gated detection and fluorescein, the fluorescence lifetime images allow for direct reading of the pH. The relative contribution to fluorescence of the acid and basic forms of C-SNARF 1 was spatially resolved on the basis of pre-exponential factors, giving quantitative mapping of the pH in the microfluidic device. Three dimensional maps of solvent composition have been generated using 2-photon excitation FLIM (2PE-FLIM) in order to observe the importance of gravitational effects in microfluidic devices. Two fluidic systems have been studied: glycerol concentration in the microfluidic device was measured using Kiton red; water concentration in a methanolic solution was measured using ANS. The density mismatch between two solutions of different composition induced a rotation of the interface between two streams travelling side by side in a microchannel. The experiment has provided evidence of non-negligible gravitational effects in microflows. 2PE-FLIM has superior capability than methods used previously to assess similar phenomena. FLIM and micro-particle imaging velocimetry (μ-PIV) have been implemented on a custom-built open frame microscope and used simultaneously for multimodal mapping of fluid properties and flow characteristics. It has been shown that viscosity mismatch between two streams induces a non-constant advective transport across the channel and results in a flow profile that deviates from the usual Poiseuille profile, characteristic of pressure driven flow in microfluidic devices.

This thesis project examined the interactive fluid dynamics between a blood pump and the univentricular Fontan circulation. 2-D particle image velocimetry (PIV) measurements were conducted on an idealized total cavopulmonary connection (TCPC) with an axial pump prototype in the inferior vena cava (IVC). Fluid velocity profiles were examined under various physiologic conditions for Fontan patients. The velocity profiles for all cases demonstrated the shunting of flow from the IVC toward the right pulmonary artery. A rotational component in the pump outflow was observed forcing flow to the periphery as compared to the flow profile without a pump present in the IVC. The inclusion of the pump provides a pressure rise of 3 to 9 mmHg. These results demonstrate the ability of the intravascular blood pump to support a Fontan circulation and support the continued optimization and development of the pump.

Le sang est une suspension concentrée (45 % en volume) de cellules déformables, les globules rouges, dans un liquide newtonien, le plasma. Dans la microcirculation, i.e. le sous-ensemble du système de circulation sanguine où s'effectuent les échanges de matière entre le sang et les tissus, les tailles de vaisseaux sont comparables à la taille d'un globule rouge (environ 10 µm). En conséquence, les effets dynamiques liés à la présence de ces cellules induisent des comportements rhéologiques complexes, qui jouent un rôle important dans le transport de l'oxygène vers les tissus. En particulier, aux bifurcations microvasculaires divergentes, les débits de globules rouges et de plasma peuvent se répartir de façon non proportionnelle entre les deux branches filles. La fraction volumique de globules rouges (hématocrite) dans l'une des branches filles est alors plus élevée que celle de la branche d'entrée, et la fraction volumique dans l'autre branche y est plus faible. Cet effet, connu sous le nom d'effet de séparation de phase, induit une très grande hétérogénéité de l'hématocrite d'un vaisseau à l'autre dans la microcirculation. Il induit également un couplage entre l'architecture du réseau microvasculaire et la dynamique de l'écoulement sanguin dans ce réseau. L'objectif de ce travail de thèse est d'étudier finement l'effet de séparation de phase in vitro, dans un régime représentatif des conditions physiologiques, au moyen de dispositifs microfluidiques modélisant les bifurcations microvasculaires et de suspensions de globules rouges. Dans ce but, un dispositif expérimental microfluidique a d'abord été élaboré. Puis, les aspects métrologiques spécifiques aux suspensions concentrées ont été abordés afin de quantifier les paramètres de l'écoulement. En particulier, la technique de dual-slit a été comprise et optimisée, assurant une mesure précise de profils de vitesse de globules rouges en microcanaux. Des métrologies spécifiques à nos conditions expérimentales ont également été mises en place pour déterminer l'hématocrite. Ces techniques ont été validées par vérification du principe de conservation de la masse entre les trois branches d'une bifurcation, et elles nous ont permis de caractériser les écoulements de globules rouges dans des micro-canaux de différentes tailles (10 à 100 µm), et ce pour de larges gammes de débits et de concentrations. Enfin, l'écoulement de suspensions de globules rouges a été étudié au niveau de micro-bifurcations, dans l'objectif de caractériser l'effet de séparation de phase pour des tailles de canaux et des gammes d'hématocrites qui n'ont pas été étudiés auparavant en conditions d'écoulement maîtrisées. / Blood is a concentrated suspension (45% by volume) of deformable red blood cells, flowing in a Newtonian fluid called plasma. The microcirculation is the part of the blood circulation system where the exchanges of material (e.g. nutrients, oxygen) between the blood and tissues take place. The microvessels are characterized by diameters less than 100 microns, which is similar in size to the size of a red blood cell ( 10 microns). As a result, the presence of these cells considerably influences the dynamics of microvascular flows and induces complex rheological behaviors. In particular, at diverging microvascular bifurcations, red blood cells and plasma may be nonproportionally distributed between two daughter vessels : one gets a higher red blood cell volume fraction (hematocrit) than the feeding vessel, while the other gets a lower one. This effect, known as the phase separation effect, causes a tremendous heterogeneity of the hematocrit among vessels in microvascular networks and induces a coupling between the microvascular architecture and the blood flow dynamics. The aim of this thesis is to investigate the phase separation effect in vitro, in physiological conditions, using red blood cell suspensions and microfluidic devices modeling microvascular bifurcations. For this purpose, a microfluidic experimental device was first developed. Then the metrological aspects specific to concentrated suspensions were addressed in order to quantify all the flow parameters. In particular, the dual-slit technique has been understood and optimized, ensuring accurate measurement of velocity profiles of red blood cells in microchannels. Measurement methods for our experimental conditions were also implemented to determine the hematocrit. All these techniques have been validated by verification of the principle of mass conservation between the three branches of a bifurcation. They allowed us to characterize the flow of red blood cells in microchannels of different sizes (10 to 100 microns) and for wide ranges of flow rates and concentrations. Finally, the flow of red blood cell suspensions was investigated at micro-bifurcations, with the aim of characterizing the phase separation effect for channel sizes and hematocrit ranges never studied in controlled flow conditions.

Abstract An experimental study is conducted to determine the velocity fields, from development to steady state, in a square enclosure due to movement of a constant velocity lid using Particle Image Velocitmetry (PIV). Experiments were conducted with water, seeded with hollow glass sphere particles 10 microns in diameter, at three different lid velocities leading to Reynolds numbers in the high laminar to transitional range. Driven Cavity Flow is a classic fluid dynamics case often used for benchmarking of computational codes. Previous work has primarily focused on improving computational codes, experimental work is lacking and focused on obtaining steady state readings. The test cavity is 1 inch (25.4mm) high by 1 inch (25.4 mm) wide leading to an aspect ratio of 1.0. The depth is taken to be 5 (127mm) inches to reduce the three dimensional effects. Readings are taken from development to steady state allowing for a full spectrum of flow characteristics. PIV technique is successful in capturing the development of driven cavity flow. Circulation is shown to increase strength with time and Reynolds number. PIV capture and processing settings are determined. Keywords: Driven Cavity Flow, Particle Image Velocimeter (PIV)

<p>An accurate understanding of the respiratory fluid dynamics is instrumental for medical applications, such as drug delivery system and treatment of diseases. Substantial research has been done to study such flow. However, a great number of these studies have the prevailing assumption of having a smooth wall, in despite the human trachea and bronchi is sustain by a series of cartilaginous rings, which creates height differences near the wall. To study the effect of including cartilaginous rings in the respiratory flow we developed two experiments, presenting a comparison between a smooth model and a model with cartilaginous rings. First, we present an experimental observation of a simplified Weibel-based model of the human trachea and bronchi with cartilaginous rings. The experiments were carried out with a flow rate comparable with a resting state (trachea-based Reynolds number of Re_D = 2650). In the second experiment, we developed a similar experiment but in a model with a tracheal stenosis (70% in the middle of the model) and no bronchi. In this case we increase the Reynolds number to Re_D= 3350, still a resting breathing state condition.</p> <p>For both experiments, we used transparent models and refractive index-matching methods were used to observe the flow, particularly near the wall. The flow was seeded with tracers to perform particle image velocimetry and particle tracking velocimetry to quantify the effect the rings have on the flow near the trachea and bronchi walls. From the results, we present a previously unknown phenomenon in the cavities between the cartilaginous rings: a small recirculation is observed in the upstream side of the cavities throughout the trachea. This recirculation is due to the adverse pressure gradient created by the expansion, which traps particles within the ring cavity. In addition, we found that the cartilaginous rings induce velocity fluctuations into the flow, which enhances the near-wall momentum of the flow reducing the separation after the stenosis. Size of the recirculation is reduced by 11% and the maximum upstream velocity is reduced by 38%, resulting in a much weaker recirculation because of the rings. Also noticed a delay in the separation from the trachea to bronchi bifurcation. </p> <p>The detection of recirculation zones in the cartilage ring cavities and the perturbation sheds light on the particle deposition mechanism and helps explain results from previous studies that have observed an enhancement of particle deposition in models with cartilage rings. The results highlight the importance to include the cartilaginous rings in respiratory flow studies. Finally, we compared the results from the stenotic case with Reynolds-averaged Navier-Stokes (RANS) models (k – ε, k – ε RNG, k – ω, k – ω SST, k – ω SST LRN and 4-equation Transition SST). In the results, indicate significant discrepancies between the experimental measurements and the simulations, mainly in the area with flow separation after the contraction. Therefore, RANS algorithms should not be considered reliable for research purposes in respiratory fluid dynamics without experimental validation. </p>