Bipolar corona assisted jet flow for fluidic application tài liệu, giáo án, bài giảng , luận văn, luận án, đồ án, bài tậ...
Author’s Accepted Manuscript Bipolar corona assisted jet flow for fluidic application Van Thanh Dau, Thien Xuan Dinh, Tung Thanh Bui, Tibor Terebessy www.elsevier.com/locate/flowmeasinst PII: DOI: Reference: S0955-5986(16)30080-2 http://dx.doi.org/10.1016/j.flowmeasinst.2016.07.005 JFMI1227 To appear in: Flow Measurement and Instrumentation Received date: 11 March 2016 Revised date: 20 June 2016 Accepted date: 12 July 2016 Cite this article as: Van Thanh Dau, Thien Xuan Dinh, Tung Thanh Bui and Tibor Terebessy, Bipolar corona assisted jet flow for fluidic application, Flow Measurement and Instrumentation, http://dx.doi.org/10.1016/j.flowmeasinst.2016.07.005 This is a PDF file of an unedited manuscript that has been accepted for publication As a service to our customers we are providing this early version of the manuscript The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain Bipolar corona assisted jet flow for fluidic application Van Thanh Daua,*, Thien Xuan Dinhb, Tung Thanh Buic and Tibor Terebessyd a Research Group (Environmental Health), Sumitomo Chemical Ltd, Hyogo 665-8555, Japan (dauthanhvan@gmail.com) Graduate School of Science and Engineering, Ritsumeikan University, Shiga 525-8577, Japan (thien@cfd.ritsumei.ac.jp) c University of Engineering and Technology, Vietnam National University, Hanoi, Vietnam (tungbt@vnu.edu.vn) d Atrium Innovation Ltd., Lupton Road, OX10 9BT, Wallingford, United Kingdom (tibor.terebessy@clearview-intelligence.com) b Abstract In this paper, we present a study on a jet flow, assisted by low net charge ion wind from bipolar corona discharge setup The ion wind is simultaneously generated from both positive and negative electrodes placed in parallel, adding momentum to the bulk flow directed alongside the electrodes and focused in the middle of interelectrode space The electrodes are connected to a single battery-operated power source in a symmetrical arrangement, where the electrode creating charged ions of one polarity also serves as the reference electrode to establish the electric field required for ion creation by the opposite electrode, and vice versa Multiphysics numerical simulation is carried out with programmable open source OpenFOAM, where the measured currentvoltage is applied as a boundary condition to simulate the electrohydrodynamics flow The jet flow inside the device is verified by hotwire anemometry using hotwires embedded within the device, with the measured values in good agreement with simulation The corona discharge helped to focus the jet and increased the flow peak velocity from 1.41 m/s to 2.42 m/s with only 27.1 mW of consumed discharge power The device is robust, ready-to-use and minimal in cost In addition, as the oppositely charged corona flows are self-neutralized, the generated air flow remains neutral and therefore does not attach to a particular target, which expands the application range These are important features, which can contribute to the development of multi-axis fluidic inertial sensors, fluidic amplifiers, micro blowers, gas mixing, coupling and analysis with space constraints and/or where neutralized discharge process is required, such as circulatory flow heat transfer or the formation of low charged aerosol for inhalation and charged particle deposition Keywords: ⃗ Vc Vb Ic q s A U Ub d Ra Electric field Corona discharge voltage Driving voltage of bulk flow generator Corona discharge current Charge Charge density Electric current density Distance between electrodes Effective area of electrode tip Permittivity of free space Mobility of charge Ion recombination rate Air density Air kinematic viscosity Flow velocity Initial velocity of bulk flow Distance from electrode tip to hotwire Output voltage on hotwire Heat current for hotwire Hotwire resistance Hotwire resistance at ambient condition Temperature coefficient of the resistance (TCR) Introduction In pumping technology, synthetic jet is generally created by the interactions of counter-rotating vortex pairs at the orifice induced by an oscillating diaphragm in a sealed cavity [1,2] The potential applications for the integration of micro devices for synthetic jets cover various research topics such as, microelectronics cooling [3], lab-on-a-chip [4], gas monitoring [5], drug delivery [6], health monitoring [7] and biology research [8] One of the most common air jet micro generators is based on the piezoelectric lead zirconate titanate (PZT) diaphragm, which has advantages over the conventional fan/blower due to the minimum required volume, reduced power consumption, longer lifetime, and high output power density [9–11] These features also make micro jet generators preferable in micro fuel cells, where they overcome the disadvantage of low power density and oxygen concentration on the cathode side [12,13] In inertial applications, the advantages of a self-contained valveless micropump with its Coriolis effect-based sensors are the absence of moving parts and minimal crosseffects from linear acceleration [14–17], thereby reducing the risk of damage or fragility of mechanical sensors On the other hand, PZT diaphragm device is usually audible, exhibits moderate power consumption, and low spray rate/ device volume ratio Another way to create a jet flow is by electrokinetic actuation, which has recently attracted more interest Under strong electric field, every charged particle is subjected to Coulomb force and while accelerated by the field, the charged particles collide with neutral fluid molecules transferring momentum, which results in fluid drift commonly called the ion wind [18] Among the wide range of applications, the enhancement of heat transfer (spot cooling) and ion wind generators are the most popular Typically the electric field is provided by corona discharge, which has distinguishable advantages due to the absence of moving parts, small size and reduced weight, low acoustic noise, simple design and easy operation For spot cooling, either the reference electrode or both electrodes are placed on the target to direct or to alternate the flow near the target [19,20] In case of ion wind blowers, both electrodes are placed inside the device and the flow is aimed at an external target For both approaches, numerical simulations and experiments have been conducted in both academia and industry, e.g by using wire-to-plate or wire-to-cyclinder in order to demonstrate various implementations, such as ion wind blower in a laptop [21], ion wind propulsion [22,23] or ionic blower for proton exchange membrane fuel cell [24] Obviously, corona discharge wind blowers have more general applications, however they have their own challenges To maintain corona discharge, the electric field must have a high degree of inhomogeneity, requiring smaller characteristic dimensions for the sharp corona discharge electrode as the electrode separation decreases In addition, maintaining stable corona is a particularly difficult task in confined space, since the dielectric walls of the device adversely affect the desired field lines, and the charged ion wind will tend to attach to the wall of a practical device One of the methods to overcome this issue is to use multi-electrode system to assist the corona discharge, where a pair of electrodes is placed close together to initiate the discharge process and a third elecotrode, usually placed downstream, is used to accelerate the ion wind This is demonstrated in wirecylinder-plate [25] or pin-plate-plate configuration [26] with decoupled ionization and acceleration processes Another concern is the discharge ion current and space charge, which need to be compensated for by electrons in the downstream space, in order to prevent charging of the device [27] To overcome this issue, a dual positive and negative plasma thruster was proposed to avoid the need for additional neutralization, which indeed requires a special electronic power source [28] Other problems may also exist, such as the application in inertial sensing, where the flow must be able to freely vibrate in three dimensional space under inertial force, which is however dominated by the electrostatic force in limited space [29–31] In addition to improving discharge stability, charge neutralization is also of critical importance in bio-applications, where the aerosol particles with highly reactive ionization products can be destructive for living cells, spores or viruses [32–34] An alternative approach is demonstrated by a hybrid air blower, where the corona is used to assist the flow without significant power consumption, while the efficacy of corona wind is also boosted under the existence of bulk flow by up to 7.5% [35] In this concept, the corona assists the focus and magnitude of jet flow and the jet flow, in turn, eliminates the possibility of direct discharge breakdown in-between the electrodes caused by manufacturing tolerances or impurity contaminants, which is an inherent result of charge accumulation due to initial low mobility of charged particles It is worth to mention that an important feature of the blower is full control over the flow even in the presence of an arbitrarily charged external target, which essentially requires a neutral flow A vital approach presented in this paper is to assist the jet flow using corona discharge with negligible net charge Such neutralized corona flow is realized by a unique bipolar discharge configuration using symmetrically arranged electrodes powered from a single power source [36] The experiment and simulation results showed that in this configuration the air movement could be optimized to be parallel with the electrode axes, and directed away from both electrodes The corona flow was used to assist the bulk flow from PZT diaphragm flow generator, which has several practical advantages The device is simple and robust since the PZT does not require to be driven at high voltage, which can dramatically reduce lifetime of the diaphragm Moreover the system is easy to scale down while maintaining high flow rates Unlike conventional point-to-plane/ring/grid corona configurations, where the electrodes (usually the reference electrode) impede the external bulk flow [37–39], in our symmetrical configuration the electrodes are in parallel with the flow and thus have minimum negative impact on it With both charges simultaneously released from the electrodes driven by an isolated power source, the amount of net charge released by the device is negligible Owing to the easy scalability of the configuration and the low net charge, the proposed system is also beneficial for applications with space constraints [36], and for applications where neutralized ion wind is required, such as fluidic amplifiers, fluidic oscillators or fluidic actuators [30,40– 42] In this paper, we present the study of the above mentioned bipolar corona assisted bulk flow for jet flow generation The description of design and working principle is followed by experimental and numerical setup The air flow is validated by the integrated thermal sensing elements (hotwires) positioned along the downstream channel, as well as by an external reference flow sensor The simulation is conducted in an open-source code environment, OpenFOAM Concept design and working principle The schematic view of working principle is shown in Fig (top) The bulk jet flow is created by a bulk flow generator and the corona discharge is generated by two parallel discharge electrodes placed on both sides of the flow and pointed downstream The discharge electrodes are cylindrical pins with very small tips to ignite corona discharge and minimize the impact of the electrode itself on the jet flow In general, an ion wind generator is designed with a pin/needle/wire as a corona electrode and a plane/ring/grid as the collector electrode Unipolar ion wind is generated at the corona electrode and yields high velocity near the surface of the counter electrodes, where the charge has to be neutralized, an example of which (point-toring) is shown in Fig.1 (bottom) This configuration is not suitable to assist externally generated jet flow due to the positioning of the collector electrode To avoid impeding the flow, the collector electrode needs to be placed off center, or in case of a ring type electrode shown, the electrode diameter needs to exceed the cross sectional dimension of the flow This however creates a defocusing effect, as the charged particles in the generated ion wind tend to move towards the collector electrode In addition, such configuration usually leaves residual charge in the air flow, therefore in some applications vortex stream is needed to help with focusing the flow [43,44] In our configuration however, the two electrodes of opposite polarity are placed in parallel, and generate oppositely charged flows from a single power source ensuring both minimized impediment and focusing of the external jet flow This parallel configuration using a single power source is principally different from multi actuator designs powered by multiple power sources, providing not only cost savings, but also enabling a charge-balanced design with simultaneous charge neutralization as explained below The ion wind is simultaneously generated by both pins The pin tip can be modeled as a protruding hemisphere with extremely high curvature attached to the pin body, which focuses the electric field outwards and nearly parallel to the pin axis Thus, the generated charge (ion clouds) at the pin tips will gain an initial momentum downstream and in parallel with the electrodes (inset in Fig 1) Under the influence of the electric field, the clouds of oppositely charged ions tend to impinge on each other at the middle of electrode interspace, preventing them from reaching the counter electrode Due to the high speed of the ion wind and its forwarding momentum, the bulk of ions moves forward, resulting in net flow This movement of ion wind from the discharge electrodes to the centerline of electrode interspace helps to boost the jet flow in magnitude while maintaining the jet flow characteristics of the velocity profile Experimental setup Figure 2a shows the schematics of our experimental setup The prototype device (Fig 2b) is made from transparent polypropylene with internal cross section of 15 × 20 mm2 (height × width) The pin electrodes are held, aligned and positioned at one end of the device All parts are designed for press-fit structures to ensure the tolerance of the mechanical assembly A small amount of conformal coating is applied at the electrode holder to ensure electric isolation The electrodes are stainless steel SUS304, each 8.0 mm long and 0.4 mm in diameter, and placed in parallel with each other The spherical radius of the pin tip is approximately 80 µm The distance between the pins is set to s = 7.0 mm The bulk flow is generated by a conventional generator assembled at the back of the device While the study of PZT diaphragm actuated micro blower has been reported in previous work [11], to ensure the simplicity of the experimental setup, a commercially available bulk flow generator (Murata) with similar operation principle is used [45] For the electronics part, a single battery operated direct current high-voltage generator, capable of generating 10.0 kV, is connected to the pins The discharge current is recorded at the negative electrode by a precision measuring circuit, which is integrated in the high voltage generator The voltage output is calibrated using a high voltage meter (Japan Finechem) The isolation of the generator ensures that the current measured at the negative polarity representing the creation of the negative charge, is the mirror image of the current at the positive polarity for the positive charge Therefore the current at both electrodes is equal in magnitude as dictated by Kirchhoff's current law For the flow generator, a power source is used to drive the membrane at its natural frequency [45] The generated flow is measured by an array of hotwires placed across the downstream channel starting from a distance of 12.5 mm from the electrode tips, aligned in the plane of the discharge electrodes The spacing between the hotwires is 2.5 mm The hotwires are made of gold and are bonded to the electric stands embedded in the device Each hotwire has a diameter of 25.0 μm and length of 24.0 mm, the resistance and temperature coefficient of resistance (TCR) are 1.35Ω and 3700 ppm/°C, respectively The hotwires have been calibrated using the external bulk flow with the aid of simulation, as explained in section 5.1 During the flow measurement the hotwires are heated in turns by constant current of 0.2 A, in order to prevent the cross effect of heat transfer The voltage on the hotwires is measured by a digital multimeter using 4-point probe method to eliminate the effect of contact resistance Data is also streamed to a computer using a LabVIEW DAQ6220 data acquisition system with a sampling rate of Hz Conversion from the hotwire voltage to average air velocity is calculated by a C-code routine In addition, a reference flow anemometer (Sibata) is mounted 10.0 mm away from the device output, centered and aligned with the hotwire plane The hotwire length of the reference sensor is 10.0 mm, which is shorter compared with the hotwires embedded within the device, therefore the measured velocity by the reference sensor represents only the central part of the velocity profile The net charge of the released ion wind is measured using an aerosol electrometer 3068 (TSI), which replaced the reference flow sensor during experiments with the corona discharge on The results were recorded at Hz and averaged over every 60 seconds All the measurements were carried out at 22 – 24 °C and 55% – 60% relative humidity at atmospheric pressure Figure 2c shows a picture of the prototype in operation, where the bipolar corona discharge is observable at both pin electrodes The observed corona glow reveals that the pin tips are similar to a sphere partly embedded into the pin body and only a small part at the top hemisphere is unembedded and thus has extremely high curvature, which focuses the electric field outwards and almost in parallel with the pin axis Simulation setup The corona drift region is governed by the Poisson and current continuity equations [46,47], which cannot be solved globally [48] In this part, to avoid the complications of modeling of the discharge itself, we deploy multi physics simulation to analyze the flow characteristics of our system by treating the corona as a boundary condition The electric field ⃗ is represented as the gradient of an electric potential V, ⃗ , and can be calculated by Gauss’ law: ⃗ , (1) where is the permittivity of free space and is the total charge from the positive and negative pins The charge drift creates a total electric current density , which, neglecting the ion diffusion, is the sum of the ⃗⃗⃗ ⃗⃗⃗ ⃗ ⃗ ( where is mobility of charge) Because positive and negative current density the total charge is conserved, the total current density has a zero divergence The continuity equation of the positive/negative current density is separately described by the ion recombination process, which can be expressed as (where and are ion recombination rate and electron charge) ⃗⃗⃗ ⃗ ⃗ (2-1) ⃗⃗⃗ (2-2) The flow is assumed to be incompressible Newtonian fluid and is considered at steady state The buoyancy force due to temperature variations is neglected The flow is then described by the Navier–Stokes equations, including conservations of momentum and of mass density The impact of the electric field to the momentum of the gas is described by the volume force ⃗ on the right-hand side of Eq 3, ⃗ (⃗ ⃗ ) ( ⃗) (3) ⃗ (4) The solutions of Eqs (1)-(4) are obtained by the development of a solver in the finite volume library OpenFOAM [49] For a typical corona discharge, the electric field magnitude ⃗ is of the order of 106 V·m-1 which yields the drift velocity ⃗ ≈ 100 m·s-1 This is much larger than the air velocity ⃗ , which is of the order of several m·s-1 Therefore, the term ⃗ in Eq (2-1) is neglected For stable simulation, an additional solver was developed to solve Eqs (1)-(2) to provide the initial electric field condition for the coupled Eqs (1)-(4) in the main solver The three dimensional simulation domain was modelled as shown in Fig Non-slip condition was set on the pin electrode, the channel and the inlet channel walls, while free condition was used for the outlet and the inlet Uniform velocity Ub, obtained from the hotwire measurements and simulation, was applied to the inlet For the electric field, voltage was applied to the boundary of the electrodes and the Neumann condition was applied at the edges of the domain At the electrode, we assumed that the corona discharge maintained a constant ion density ( ) (5) where A is the total area of the tip The electric field at the pin tip Ew is determined based on Peek’s law for air, expressed as: ( ) ( ) (6) where R is the radius of the pin tip in units of centimeters Finally, with air used as the media, the following constants close the modelling portion: , , ̇ , , and Results and Discussion 5.1 Calibration of flow measurement Before evaluating the performance of the system as a whole, the built-in hotwires have been calibrated by both experiment and simulation without the corona discharge The calibration process is briefly presented as follows The temperature of the hotwire Thw, heated by the current Ihw is determined from the equilibrium equation of heat transfer at steady state between the hotwire and air: ( ), (7) where Ahw, h, and Ta are surface area of the hotwire, heat transfer coefficient, and ambient temperature, respectively Rhw is the hotwire resistance calculated as: ( ) (8) with Ra and are the resistance at temperature Ta and the TCR of the hotwire material, respectively Without bulk flow, stationary air defines the initial state of measurement by natural convection When the bulk flow is activated, it cools the hotwire down by forced convection The heat transfer coefficient of forced convection [50] and natural convection [51] are respectively calculated as presented in Eq (9) and Eq (10) (9) (10) where Ra is the Rayleigh number, D is the effective diameter of the hotwire, and is the Reynolds number The output voltage on the hotwire, offset to the initial value measured with still air, is measured as Figure 4a shows the voltage measured on the embedded hotwires for different bulk flow speeds generated by the flow generator with voltage applied to its PZT diaphragm in the range of Vb = 5.0 V to Vb = 11.0 V In addition the flow speed measured by the external reference flow sensor positioned 10 mm from the device output is also shown The flow velocity Uo applied at the outlet of bulk flow generator was treated as input parameter, iterated within valid range during the computation For each input, the simulated velocity profile at the hotwire positions was exported to further calculate the heat transfer process of Eqs (7-9) by a Ccode routine, considering infinitesimal hotwire elements on which the flow velocity is constant The process was repeated until the output on hotwire1 was matched with its experimental value at a typical driving voltage, chosen to be Vb = 8.0 V Once the boundary condition for input flow velocity was fixed, velocity profiles and the other hotwire outputs were interpolated accordingly Inherently, the boundary condition for other driving voltages, such as Vb = 11.0 V, is extrapolated based on the result of Vb = 8.0 V and the available data sheet information of bulk flow generator [52] The extrapolation was found to be in good agreement (< 6.8%) with the experimentally obtained hotwire voltage values for the entire driving voltage range Fig 4b shows both the simulation result and the measured hotwire output for all hotwire positions and two different driving voltages, 8.0 V and 11.0 V respectively In order to enable straightforward direct comparison, the flow profiles obtained by simulation are presented in terms of hotwire voltage The calibrated hotwires were used for the evaluation of the whole system with corona discharge, presented in the next sections 5.2 Flow characteristic of corona assisted bulk flow In order to facilitate the discussion, a Cartesian coordinate system is designated with the origin located at the centre of electrode interspace The corona discharge is turned on after the bulk flow has been established Fig presents the simulated result of the flow field inside the device where the bulk flow is simulated simultaneously with corona discharge The bulk flow is generated with driving voltage of Vb = 8.0 V and Vb = 11.0 V, and the corona is generated at fixed discharge voltage of Vc = 5.0 kV and corresponding discharge current of Ic = 5.78 µA and 5.41 µA, respectively The ion clouds generated in the vicinity of the tips gain an initial momentum away from the pin tips and in parallel with the electrodes Under the interaction with the electric field between the two electrodes, the jets of oppositely charged ions tend to impinge on each other at the middle of the electrode interspace and mix with the moving bulk flow, resulting in further pressure drop, charge neutralization and further focus into jet flow The overall view of the generated ion wind demonstrates that the jet flow is maintained downstream far away from the pins Figure also shows that the ion wind has stronger effect on the weaker jet flow (Vb = 8.0 V), which will be discussed quantitatively in section 5.3 The simulation results showed stable flow, inherent based on the low Reynolds number, which is 900 for the maximum discharge voltage (Vc = 5.5 kV) and the maximum bulk flow velocity at the inlet (5.5 m/s) The charge neutralization was confirmed by the aerosol electrometer at the outlet of device, with the total charge of the ion wind outside the wind collector measured to be very low, typically around -12 fA to +30 fA This charge is comparable with the value of the background noise, which was measured without the corona discharge, and not affected by the existence of the bulk flow Since this net charge of ion wind is very small compared with the discharge current (which is several μAs), this confirms that the positive and negative charges are well balanced 5.3 Effect of corona discharge on bulk flow Figure shows the output on hotwire1 when the corona is discharged at different voltages Vc from 4.0 kV to 5.25 kV and the driving voltage Vb is set from 5.0 V to 11.0 V As it can be seen by the increase of the total voltage on the hotwire, the corona discharge itself has added a considerable momentum to the flow At higher bulk flow velocity, the added hotwire voltage decreases However it is worth to note that the decrease of voltage is also contributed from the nonlinearity of hotwire anemometry, where the hotwire is exposed to large flow velocity and is considerably cooled down by the bulk flow [53,54] This is confirmed by Fig 7, where the experimental results for hotwire and hotwire are expressed in terms of average flow velocity Despite the small increase on hotwire voltage for the case of Vb = 11.0 V, the velocity shows that the corona wind does not considerably depend on the bulk flow, indeed it is more stable at high bulk flow The hotwires placed further away from the electrodes have smaller output voltage, which reflects the decay of the jet flow The flow velocity profile, which cannot be measured by our experimental setup, is demonstrated from our simulation as shown below The profiles of the velocity at hotwire locations and are plotted in each half of Fig for generator drive voltages of 8.0 V and 11.0 V and with and without corona discharge From Fig 8, it is evident that the profiles are significantly enhanced by the existence of corona discharge The tails of these profiles have negative values, which show that there is a circulatory flow in the channel By adding corona discharge of 5.0 kV and 5.78 μA, representing power consumption of 28.9 mW by the discharge itself, the flow peak velocity was increased from 0.86 m/s to 1.93 m/s with the bulk flow generated at 8.0 V drive voltage Similarly, the bulk flow generated at 11.0 V was increased from 1.41 m/s to 2.42 m/s with added corona driven at 5.0 kV and 5.41 μA, corresponding to 27.1 mW discharge power consumption The peak of velocity profile is off center at hotwire position when the bulk flow is driven at lower voltage of 8.0 V, indicating that the additional flow generated by corona is even stronger than the bulk flow Advancing downstream, the velocity profiles plotted at different locations are shown in Fig 9, where the focusing effect of the corona is also clearly observed As the ion wind is moving in from the pin tips to the center of interelectrode space, the velocity profiles at the hotwire locations, d = 12.5 – 20.0 mm, are wide and not decay with increasing distance Further away with d = 30.0 – 50.0 mm, the velocity profiles are narrow and decay with distance, exhibiting typical jet-like characteristics This characteristic of the device allows us to further develop multi axis inertial units, or multi direction synthetic jets in the future Figure 10 compares the hotwire anemometry result elicited in the experiment and the above simulation It is also important to note that while the corona is discharged with current of several μAs, the current supplied to the hotwires is six orders of magnitude larger, so the effect of discharge current on the hotwire voltage is small In addition, because both charges are released, this error is further minimized and is in fact negligible Therefore, the voltage on hotwire can be calculated by considering only the cooling effect of the flow This is confirmed through the observation of zero output voltage when turning on the corona discharge but without heating the hotwire As it can be seen, the simulation matches well with experiment, although we note a slight disagreement particularly at the wire closest to the pin Because the experimental device was fabricated with limited resolution, the device wall surface, uneven at a submillimeter scale, was excluded from the simulation Also the assembly in pin alignment made the tolerance at the closest hotwire larger compared with the others Better agreement can be expected if a microfabrication process is involved 5.4 Effect of bulk flow on corona discharge In order to investigate the effect of external flow on the corona discharge, the current – voltage characteristics of corona have been measured for different bulk flow speeds The current – voltage characteristics is a fundamental discharge property, and changes to the discharge condition can be revealed by its measurement The different bulk flow speeds were generated by the flow generator in the driving voltage range of 7.0 V to 11.0 V The results in Fig 11 show that the discharge voltage has a linear relation with the square root of discharge current, √ , and this relation does not change with the existence of bulk flow in the tested range Although this relation is much less common in comparison with the Townsend relationship , nevertheless it is in agreement with the reported literature for some configurations, e.g point-toplane for the positive corona with electrode distance of 50.0 mm [55], and spherically symmetric unipolar corona [56] For completeness, it is noted that although the √ relation is unchanged with the added bulk flow, the discharge current for a fixed corona voltage will slightly decrease with increasing bulk flow speed This is expected, since the additional external flow between the electrodes will reduce the possibility that charged particles reach the counter electrode Moreover, additional external flow will also present additional cooling to the electrodes, which proportionally reduces the ion mobility, thus decreasing the discharge current [57] The power consumed by the corona discharge is small, approximately 27 mW and the total power consumption of the high voltage generator (including losses) is less than 70 mW for the experimental condition in Fig For reference, the bulk flow generator consumes more than 250 mW for typical performance [52] Conclusion We have presented a study on corona assisted bulk flow by numerical simulations and experimental validation The corona is based on the simultaneous generation of both positive and negative ions using two sharp electrodes placed in parallel The ion winds originating from both electrodes assist the bulk flow moving along the centre of the electrode interspace, which results in a boosted jet without charging the flow Based on the measured current – voltage characteristics of bipolar corona discharge, simulations of flow rate and charge distribution have been carried out with and without the presence of bulk flow The simulations agree well with the measured flow using four hotwires embedded inside the device The system is very robust and easy to build because all the components are commercially available In fact, the system is scalable and is expected to have increased efficiency with reduced size The current dimension of the system is limited by the geometrical constraints of the system setup The pins and electrical connections, which are chosen based on their availability, still have finite size and thus impede the airflow around the pins While the electrode separation is selected as mm in consideration with bulk flow generator dimensions, there is no actual restriction to its minimum length because it is not a governing parameter In addition, the choice of electrode separation has also been affected by the relative size of electrode body, which is 0.54 mm in diameter from its commercial availability It is believed that with a revised system setup, such as integrating discharge electrode within the design step of PZT micro blower, usage of pins with smaller diameters or utilization of a microfabrication process, the efficiency could further increase and ion wind generation of similar magnitude could be expected at lower voltage levels and reduced applied power Finally we mention that the round jet is preferably refocused in the electrode plane, therefore several pairs of sharp electrodes may be needed to improve the symmetry of the flow References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] B.L Smith, A Glezer, The formation and evolution of synthetic jets, Phys Fluids 10 (1998) 2281 doi:10.1063/1.869828 K Tanaka, V.T Dau, R Sakamoto, T.X Dinh, D.V Dao, S Sugiyama, Fabrication and Basic Characterization of a Piezoelectric Valveless Micro Jet Pump, Jpn J Appl Phys 47 (2008) 8615–8618 doi:10.1143/JJAP.47.8615 D Jang, K Lee, Flow characteristics of dual piezoelectric cooling jets for cooling applications in ultra-slim electronics, Int J Heat Mass Transf 79 (2014) 201–211 doi:10.1016/j.ijheatmasstransfer.2014.08.013 P Gravesen, J Branebjerg, O.S Jensen, Microfluidics-a review, J Micromechanics Microengineering (1993) 168–182 doi:10.1088/0960-1317/3/4/002 V.T Dau, T.X Dinh, T Katsuhiko, S Susumu, A cross-junction channel valveless-micropump with PZT actuation, Microsyst Technol 15 (2009) 1039–1044 doi:10.1007/s00542-009-0878-2 M.K Araz, A.M Tentori, A.E Herr, Microfluidic multiplexing in bioanalyses., J Lab Autom 18 (2013) 350–66 doi:10.1177/2211068213491408 T Thorsen, S.J Maerkl, S.R Quake, Microfluidic large-scale integration., Science 298 (2002) 580–4 doi:10.1126/science.1076996 L.Y Yeo, H.-C Chang, P.P.Y Chan, J.R Friend, Microfluidic Devices for Bioapplications, Small (2011) 12–48 doi:10.1002/smll.201000946 X.-M Tan, J.-Z Zhang, Flow and heat transfer characteristics under synthetic jets impingement driven by piezoelectric actuator, Exp Therm Fluid Sci 48 (2013) 134–146 doi:10.1016/j.expthermflusci.2013.02.016 H.K.K Ma, H.C.C Su, W.F.F Luo, Investigation of a piezoelectric fan cooling system with multiple magnetic fans, Sensors Actuators, A Phys 189 (2013) 356–363 doi:10.1016/j.sna.2012.09.009 V.T Dau, T.X Dinh, Numerical study and experimental validation of a valveless piezoelectric air blower for fluidic applications, Sensors Actuators B Chem 221 (2015) 1077–1083 doi:10.1016/j.snb.2015.07.041 Y.H Seo, H.J Kim, W.K Jang, B.H Kim, Development of active breathing micro PEM fuel cell, Int J Precis Eng Manuf Technol (2014) 101–106 doi:10.1007/s40684-014-0014-3 [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] X Yang, Z Zhou, H Cho, X Luo, Study on a PZT-actuated diaphragm pump for air supply for micro fuel cells, Sensors Actuators, A Phys 130-131 (2006) 531–536 doi:10.1016/j.sna.2005.12.021 V.T Dau, D.V Dao, T Shiozawa, H Kumagai, S Sugiyama, A Single-Axis Thermal Convective Gas Gyroscope, Sensors Mater 17 (2005) 453–463 H Chang, P Zhou, Z Xie, X Gong, Y Yang, W Yuan, Theoretical Modeling for a Six-DOF Vortex Inertial Sensor and Experimental Verification, J Microelectromechanical Syst 22 (2013) 1100–1108 doi:10.1109/JMEMS.2013.2271862 V.T Dau, D.V Dao, T Shiozawa, H Kumagai, S Sugiyama, Development of a dual-axis thermal convective gas gyroscope, J Micromechanics Microengineering 16 (2006) 1301–1306 doi:10.1088/0960-1317/16/7/026 V.T Dau, T.X Dinh, D.V Dao, S Sugiyama, Design and Simulation of a Novel 3-DOF MEMS Convective Gyroscope, IEEJ Trans Sensors Micromachines 128 (2008) 219–224 doi:10.1541/ieejsmas.128.219 Y.P Raizer, J.E Allen, V.I Kisin, Gas Discharge Physics, Springer, 2011 D.B Go, S V Garimella, T.S Fisher, R.K Mongia, Ionic winds for locally enhanced cooling, J Appl Phys 102 (2007) doi:10.1063/1.2776164 P Zhao, S Portugal, S Roy, Efficient needle plasma actuators for flow control and surface cooling, Appl Phys Lett 107 (2015) 033501 doi:10.1063/1.4927051 N.E Jewell-Larsen, H Ran, Y Zhang, M.K Schwiebert, K.A.H Tessera, A V Mamishev, Electrohydrodynamic (EHD) cooled laptop, in: 2009 25th Annu IEEE Semicond Therm Meas Manag Symp., IEEE, 2009: pp 261– 266 doi:10.1109/STHERM.2009.4810773 W Jack, P Hugh, T William, An Investigation of Ionic Wind Propulsion, NASA Rep NASA/TM (2009) 2009– 215822 L Pekker, M Young, Model of Ideal Electrohydrodynamic Thruster, J Propuls Power 27 (2011) 786–792 doi:10.2514/1.B34097 K Kwon, L Li, B.H Park, S.J Lee, D Kim, Air supply using an ionic wind generator in a proton exchange membrane fuel cell, J Power Sources 284 (2015) 504–510 doi:10.1016/j.jpowsour.2015.03.053 D.F Colas, A Ferret, D.Z Pai, D a Lacoste, C.O Laux, Ionic wind generation by a wire-cylinder-plate corona discharge in air at atmospheric pressure, J Appl Phys 108 (2010) 0–6 doi:10.1063/1.3514131 R Tirumala, D Go, Multi-electrode assisted corona discharge for electrohydrodynamic flow generation in narrow channels, IEEE Trans Dielectr Electr Insul 18 (2011) 1854–1863 doi:10.1109/TDEI.2011.6118623 K Nishiyama, H Kuninaka, Discussion on Performance History and Operations of Hayabusa Ion Engines, Trans Japan Soc Aeronaut Sp Sci Aerosp Technol Japan 10 (2012) Tb_1–Tb_8 doi:10.2322/tastj.10.Tb_1 D Rafalskyi, L Popelier, A Aanesland, Experimental validation of the dual positive and negative ion beam acceleration in the plasma propulsion with electronegative gases thruster, J Appl Phys 115 (2014) 053301 doi:10.1063/1.4863876 Y Fukatsu, E Nomura, K Matsu, Gas rate gyro, US4941353, 1990 Y Cai, Y Zhao, Ion discharge gyroscope, US8146423, 2012 R.B Schlesinger, M Lippmann, Particle deposition in the trachea: in vivo and in hollow casts., Thorax 31 (1976) 678–684 doi:10.1136/thx.31.6.678 V.N Morozov, Generation of biologically active nano-aerosol by an electrospray-neutralization method, J Aerosol Sci 42 (2011) 341–354 doi:10.1016/j.jaerosci.2011.02.008 J.C Almekinders, C Jones, Multiple jet electrohydrodynamic spraying and applications, J Aerosol Sci 30 (1999) 969–971 doi:10.1016/S0021-8502(98)00755-1 D Camelot, J.C.M Marijnissen, B Scarlett, Bipolar Coagulation Process for the Production of Powders, Ind Eng Chem Res 38 (1999) 631–638 doi:10.1021/ie980435j H Bondar, D Bastien, Effect of neutral fluid velocity on direct conversion from electricla to fluid kinetic energy in an electro-fluid-dynamics (EFD) device, J Phys D Appl Phys 19 (1986) 1657–1663 V.T Dau, T.T Bui, T.X Dinh, T Terebessy, Pressure sensor based on bipolar discharge corona configuration, Sensors Actuators A Phys 237 (2016) 81–90 doi:10.1016/j.sna.2015.11.024 G.F.L Ferreira, O.N Oliveira, J a Giacometti, Point-to-plane corona: Current-voltage characteristics for positive and negative polarity with evidence of an electronic component, J Appl Phys 59 (1986) 3045–3049 doi:10.1063/1.336926 K Yamada, An empirical formula for negative corona discharge current in point-grid electrode geometry, J Appl Phys 96 (2004) 2472–2475 doi:10.1063/1.1775301 A.M Drews, L Cademartiri, G.M Whitesides, K.J.M Bishop, Electric winds driven by time oscillating corona discharges, J Appl Phys 114 (2013) doi:10.1063/1.4824748 V.T Dau, D.V Dao, T Shiozawa, S Sugiyama, Simulation and Fabrication of a Convective Gyroscope, IEEE Sens J (2008) 1530–1538 doi:10.1109/JSEN.2008.925457 T.X Dinh, V.T Dau, S Sugiyama, P.H Pham, Fluidic device with pumping and sensing functions for precise flow control, Sensors Actuators, B Chem 150 (2010) 819–824 doi:10.1016/j.snb.2010.08.009 V.T Dau, T.X Dinh, T.T Bui, Jet flow generation in a circulatory miniaturized system, Sensors Actuators B Chem 223 (2015) 820–826 doi:10.1016/j.snb.2015.09.151 Y.N Kolomiets, V V Pervukhin, Atmospheric pressure ion focusing with a vortex stream, Talanta 85 (2011) 1792–1797 doi:10.1016/j.talanta.2011.07.005 Y.N Kolomiets, V V Pervukhin, Vortex focusing of ions produced in corona discharge, Talanta 110 (2013) 39– 45 doi:10.1016/j.talanta.2013.02.016 A Hirata, G Kamitani, H Wada, M Sugana, S Kanai, Murata Blower Patent No US 8678787B2, US 8678787B2, 2014 [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] J.S Townsend, Electricity in Gases, Oxford University Press, 1915 doi:10.1038/095611a0 J.J Thomson, G.P Thomson, Conduction of Electricity Through Gases, London CUP, 1933 K Adamiak, Numerical models in simulating wire-plate electrostatic precipitators: A review, J Electrostat 71 (2013) 673–680 doi:10.1016/j.elstat.2013.03.001 OpenFOAM®, OpenFOAM® | The OpenFOAM Foundation, (2016) http://openfoam.org/ M Mikheyev, Fundamentals of Heat Transfer, Peace Publisher, Moscow, 1968 I Mabuchi, T Tanaka, Experimental Study on Effect of Vibration on Natural Convective Heat Transfer from a Horizontal Fine Wire, Bull JSME 10 (1967) 808–816 doi:10.1299/jsme1958.10.808 Murata Ltd, Features and Technical Specification of Microblower | Murata Manufacturing Co., Ltd., (n.d.) http://www.murata.com/en-eu/products/mechatronics/fluid/feature G Comte-Bellot, Hot-wire anemometry, Annu Rev Fluid Mech (1976) 209–231 A Berson, P Blanc-Benon, G Comte-Bellot, A strategy to eliminate all nonlinear effects in constant-voltage hotwire anemometry, Rev Sci Instrum 80 (2009) doi:10.1063/1.3103948 A.F Kip, Onset studies of positive point-to-plane corona in air at atmospheric pressure, Phys Rev 55 (1939) 549– 556 doi:10.1103/PhysRev.55.549 R.S Sigmond, Simple approximate treatment of unipolar space-charge-dominated coronas: The Warburg law and the saturation current, J Appl Phys 53 (1982) 891–898 doi:10.1063/1.330557 M Tabrizchi, F Rouholahnejad, Comparing the effect of pressure and temperature on ion mobilities, J Phys D Appl Phys 38 (2005) 857–862 doi:10.1088/0022-3727/38/6/012 Author Biogarphies Van Thanh Dau received the B.S degree in aerospace engineering from Hochiminh City University of Technology, Vietnam, in 2002, and the M.S and Ph.D degrees in micro-mechatronics from Ritsumeikan University, Japan, in 2004 and 2007, respectively From 2007 to 2009, he was a Postdoctoral Fellow with Japan Society for the Promotion of Science (JSPS) at Micro Nano Integrated Devices Laboratory, Ritsumeikan University Since 2010 he has been with Research Group, Sumitomo Chemical Co., Ltd where he works on integrated micro electrospray and atomization methods His research subjects are micro fluidics, electro hydrodynamics, microsensors and microactuators He is the author and co-author of 80 scientific articles and 20 inventions Thien Xuan Dinh received the B.S degree in aerospace engineering fromHochiminh City University of Technology in 2002, Vietnam and the M.Sc and Ph.D degrees in mechanical engineering from Ritsumeikan University in 2004 and 2007, respectively He was recipient of Japan Government Scholarship (MEXT) for Outstanding Student to pursuits his M Sc and Ph D courses and Japan Society for the Promotion of Science postdoctoral fellowship from 2011 to 2013 His general research interest is computation of fluid flow The large parts of his research are turbulence modeling using Large Eddy Simulation, multiphase modeling using Volume of Fluid technique, and simulation of turbulence and dispersion Recently, he has focused on computation of fluid flow for developing microfluidic devices as electrohydrodynamics, microsensors, micropump, and micromixer for biochemical engineering Tung Thanh Bui received the B.S degree in electrical engineering from Vietnam National University, Hanoi (VNUH) in 2004, and the M.E and D.Eng degrees in Science and Engineering from Ritsumeikan University, Shiga, Japan, in 2008 and 2011, respectively From 2011 to 2015 he was a post-doctoral researcher with the 3D Integration System Group, Nanoelectronics Research Institute (NeRI), National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Japan Currently, he is an assistant professor at the Faculty of Electronics and Telecommunication (FET), University of Engineering and Technology (UET), Vietnam National University, Hanoi (VNUH) His current research interests are 3D system integration technology and MEMS based sensors, actuators and applications He is the author and co-author of more than 60 scientific articles and inventions Tibor Terebessy received his M.S degree with honour in plasma physics from Comenius University, Slovakia, in 1998 and his Ph.D degree in electronics engineering from Shizuoka University, Japan, in 2002 He was then awarded a Postdoctoral Fellowship by the Japan Society for the Promotion of Science (JSPS), continuing his research in large area microwave discharges and their industrial applications at Graduate School of Electronic Science and Technology, Shizuoka University, Japan His main areas of research interests include atmospheric pressure discharges, microwave plasmas, nanoparticle generation and electrohydrodynamics He is the author and co-author of more than 20 scientific articles and 17 inventions Figure Schematic description of working principle of corona assisted jet flow device The ion wind is simultaneously generated by two symmetric pin electrodes connected to a single power source (top), in comparison with a typical point-to-ring corona configuration (bottom) Figure Schematic view of measurement setup (a) Cross-sectional view of the fabricated device (b) and a picture showing device in operation with bipolar discharge (c), the inset shows the magnified view of corona glow at negative electrode with discharge current of µA 10 Figure Meshing and boundary conditions for numerical setup of device The inset shows the meshing at electrode tip vicinity The inlet condition Ub is iterated to match with the bulk flow generator Figure 4a Hotwire voltage and flow speed vs applied voltage on bulk flow generator The flow speed is measured by the reference sensor positioned 10.0 mm from the device output 11 Figure 4b Comparison of simulated results with the experimentally obtained hotwire voltage values for bulk flow driving voltage 8.0 V and 11.0 V Figure Flow field inside the device showing bulk flow from generator (1) in the middle and ionic jets from two electrode (2) and (2’) The bulk flow is generated with driving voltage of 8.0 V (top) and 11.0 V (bottom), and the corona is discharged at 5.0 kV and 5.78 μA and 5.41 μA, respectively 12 Figure Output on hotwire1 versus discharge voltage, corresponding to each applied voltage on bulk flow generator Figure Output on hotwires and versus discharge voltage, measured for V b = 8.0 V and Vb = 11.0 V The corona discharge ignition point is also indicated 13 Figure Velocity profiles at hotwire and hotwire 4, shown for driving voltage set to 8.0 V and 11.0 V, both with (5.0 kV discharge voltage) and without corona discharge The effect of corona is clearly demonstrated by the significant increase of peak velocity Figure Velocity profiles at different downstream positions shown for driving voltage set to 11.0 V with corona discharge 5.0 kV 14 Figure 10 Output voltage on hotwire for case in Fig The output voltage on hotwire represents the cooling effect of velocity profile in Fig Good agreement is obtained between simulation and experiment Figure 11 I-V characteristics of corona under the existence of external flow Highlights Jet flow refocusing by corona ion wind Corona discharge from parallel pin configuration Hotwire is integrated to measure flow inside device Simulation by OpenFOAM agrees well with experiment 15 .. .Bipolar corona assisted jet flow for fluidic application Van Thanh Daua,*, Thien Xuan Dinhb, Tung Thanh Buic and Tibor... above mentioned bipolar corona assisted bulk flow for jet flow generation The description of design and working principle is followed by experimental and numerical setup The air flow is validated... also beneficial for applications with space constraints [36], and for applications where neutralized ion wind is required, such as fluidic amplifiers, fluidic oscillators or fluidic actuators