Experimental determination of cooling and spray characteristics of the water electrospray

https://doi.org/10.1016/j.icheatmasstransfer.2020.105046Get rights and content

Abstract

Over the last two decades, the dimensions of electromechanical systems have decreased to micro, even nano scale, as well as the capacity of these systems has increased, leading to the gradual increase of heat flux density. Hence, the conventional cooling methods cannot completely remove heat from the surface. Electrospray cooling is a novel method used in high heat flux systems with small surface area. In literature, studies investigating the relationship between electrospray cooling and spray characteristics are limited. In the present work, electrospray and cooling characteristics were determined for 3different nozzle diameters, 2 different nozzle – plate separation distances, 4 voltages and 3volumetric flow rates. The relationship between mean flow velocity, mean particle diameter and electrospray modes also were determined by digital image processing. The cooling enhancement ratio (CER) values were calculated for varying voltages, mean flow velocities and mean particle diameters. As a result, it was observed that with the increased voltage, the CER and mean flow velocity increased and the mean particle diameter decreased. According to the experimental results, the maximum value of CER was determined to be 1.9, where the mean velocity of the deionised water and the mean particle diameter were 9 m/s and 102 μm, respectively.

Introduction

Increased joint temperature during operation of electromechanical systems decreases operating efficiency and lifespan, causing parts to be deformed over time due to thermal expansion and fail to fulfill functions. Various cooling methods are used in order to keep the operating temperatures of the system at optimum working range due to the heat generated during the operation of these systems. With the current rapid development of the technology, systems have been reduced from milimetric dimensions to nano dimensions and increased data processing capacities result in increased joint temperatures and conventional cooling methods which cannot achieve the desired optimum temperatures. The reason is that surface effects due to the reduction in the size of the electronics become more effective in MEMS. This downsizing in the surface area to volume ratio of MEMS, affects the mass, momentum and energy transfer significantly along the surface. Eventually, the small length scale of micro devices completely invalidates the continuum approach. Mass, momentum and energy conservation laws should be re-evaluated with concentration, viscous diffusion, compressibility, intermolecular forces and other unknown effects [1]. Therefore, the techniques used to cool high power systems should be reevaluated, because cooling is a key point for the further development of high-tech electronics products. A number of studies have been performed to remove the heat from the surface quickly with high efficiency, due to the small surface area of the MEMS, the inefficiency of conventional cooling methods and the low heat transfer capacity. The existing studies of the high heat flux cooling methods have concentrated on spray cooling, impinging jet, liquid jet cooling, microchannel cooling, pool boiling, and micro pump cooling [2,3]. Gao and Li summarized the variation of cooling techniques by comparing the achievable range of heat flux and heat transfer coefficient [4]. Fig. 1 shows that the cooling performance of water spray cooling is better than other methods. Therefore, spray cooling is one of the best cooling options for achieving the high – heat fluxes necessary to cool very small devices.

The atomization of liquid in spray applications is very important. There are many methods in the literature to atomize liquids, where Ang et al. compared the flow trends obtained with different types of atomization in the literature [9]. According to this comparison, having approximately the same volumetric flow rate range, surface acoustic atomization, piezoelectric actuator, electrospray techniques and having a higher volumetric flow rate range, piezoelectric pump techniques were compared and the lowest droplet diameters were obtained by electrospray atomization [[10], [11], [12], [13]]. As the droplets get smaller, their thermal resistance decreases and heat transfer coefficient increases [14]. Besides, droplets evaporate faster, and therefore lead to increased critical heat flux (CHF) and cooling efficiency [10]. Different single nozzle electrospray cooling studies vary in their heat removal capacities and CER values are summarized in Table 1. (See Table 2.)

Electrospray is the best method of providing homogeneous flow and proper particle size distribution when used in low flow rate systems, thus providing high heat transfer and heat transfer homogeneity in cooling. This method is based on the ionization of dielectric fluids in the high electric field. When exposed to an electric field, the liquid which swells on its surface tries to avoid this effect through surface tension. As the electric field increases, the swell becomes a conical structure. If sufficient electric field force is generated by the system, then the electrostatic forces dominate the surface tension and this is called electrospray ionization [19,20] when it is observed that the conical structure is separated from the top into droplets. The electro hydrodynamic (EHD) strengthening mechanism was first summarized in a study by Allen et al. [21]. The necessary research studies were also discussed in order to clarify the question of multiplying and to encourage its use in practical systems. Sheu et al. investigated heat transfer characteristics of oscillatory electrohydrodynamic with applied voltage ranging from 4 to 9.5 kV on a flat plate [22]. The waveforms of input voltage were constant or gradual and the frequency varied from 0.5 to 2 Hz. The distance between the electrode - test plate was arranged as 5 and 15 mm, respectively. At the same applied voltage, the heat transfer performance, subject to oscillating electrohydrodynamic, was found to be lower than continuous electrohydrodynamic for all frequencies. Brewer et al. patented a system for electrostatically moving air to cool the computer. The electrostatically charged system includes a heat sink assembly in the computer and an ionization strip for filtering at high voltage. The blown air flows towards the heat sink to form a flow of cooling air [23]. Ramadhan et al. designed a innovative electronic cooling method using optimized electrohydrodynamic blowers integrated with a plate-fin heat sink. Numerical results of the recommended electrohydrodynamic cooling method show effective cooling performance; besides, operating temperature was safe for typical operating ranges [24]. The application of electrospraying in practice is rare, due to device miniaturization, reliability, high voltage generation and electromagnetic interference. Larsen et al. integrated an electrohydrodynamic cooling system into a laptop computer and the fundamental principles of electrospray cooling were investigated theoretically and experimentally. Although they did not optimize the system, they obtained values close to fan cooling [25].

Boroomandpour et al. investigated the thermal conductivity (knf) of a trio hybrid nanofluid containing Multi Walled Carbon Nanotubes (MWCNTs) and zinc oxide / titania / Ethylene Glycol (EG) – water (20:80) in φ: 0.1–0.4% and T: 25–50 °C. The results showed that the highest increment in knf containing MWCNTs/ EG – Water was observed at a φ: 0.4% and T: 50 °C [26]. Yan et al. performed on the rheological properties of non-newtonian hybrid nanofluid (MWCNTs–ZnO / EG– water (20:80)) in φ: 0.075–1.2% and T: 25–50 °C to develop a new model. They found the effect of changes is very clear as the φ increases [27]. Soltani et al. measured the knf of WO3-MWCNTs / hybrid nano-engine oil, and a new interest was established. They obtained two mathematical models for assessing knf [28]. He et al. proposed an algorithm for the best neuron number in the Artificial Neural Network (ANN), after creating experimental datum points of Ag – ZnO (50:50) / water nanofluid, and calculated the correlation coefficient and performance for ANN. They found that all ANN outputs, the maximum absolute error is 0.0095, and the performance of train is 1.6684e-05 [29]. Zheng et al. used aqueous solution of calcium dichloride / silicon dioxide nanoparticles as a desiccant in a gas–liquid Hollow Fiber Membrane (HFM) contactor system for air dehumidification. They examined effects of vary parameters on heat and mass transfer system characteristics. According to their results, it was found that the effect of particles on the rate of moisture dissipation is greatest for the highest particle and highest temperature conditions [30]. Rostami et al. developed ANNs to predict the knf of MWCNTs – copper(II) oxide / water nanofluid. An algorithm is suggested to find the optimum ANN concerning the best performance. The knf increased more than 30.38% as against the ambient temperature in φ: 0.6% and T: 50 °C [31]. Rostami et al. proposed an algorithm to enhance the best neuron number in the hidden layer ANN method, to find the best construction and then to predict the knf of silicon dioxide / EG - water (50:50) nanofluid. They found the correlation coefficient for all outputs of ANN and eight neuron numbers to be 0.993861 [32]. Afshari et al. presented experimental investigation into the effects of φ (0.0625–1%) and temperature (25–50 °C) on dynamic viscosity of the MWCNTs hybrid nanofluid and Al2O3in a mixture of EG – water (20:80). Their results revealed that viscosity has a direct relationship with φ of the nanofluid and the apparent viscosity usually increases as the φ increases [33]. Shahsavar et al. investigated the thermal conductivity variation and viscosity of liquid paraffin – aluminum oxide nanofluid containing C18H34O2 surfactant versus temperature, nanoparticle mass and surfactant concentration. The results revealed increasing the nanoparticle concentration reasons an increase in the thermal conductivity and viscosity, while increasing the temperature results in a decrease in the viscosity and an increase in the thermal conductivity [34]. Ruhani et al. proposed a model for calculating viscosity of the nanofluid is based on present data. The results revealed the dynamic viscosity reduces as temperature increases and augments as φ of nanoparticles increases [35]. Shahsavar et al. investigated the entropy generation and heat transfer characteristics of hybrid water based nanofluid in natural convection flow inside a horizontal concentric annulus. The entropy generation studies show that thermal entropy generation rates and frictional increase with increasing iron (II,III) oxide and concentrations of CNT at several Rayleigh numbers [36]. Ruhani et al. developed new model for rheological behavior of water / silica – EG (70:30) newtonian hybrid nanofluid. They found a relationship between shear stress and shear rate is linear [37]. Aghahadi et al. carried out an experimental study on the rheological behavior of hybrid tungsten trioxide MWCNTs / hybrid newtonian nano-engine oil. They proposed a mathematical model to predict viscosity of nanofluids. Model and experimental results are shown to be in a good agreement [38].

In literature, few electrospray cooling studies have included spray characteristics, and these studies attempts to determine the relationship of spray characteristics with heat transfer are insufficient. In addition, in studies where spray characteristics were determined, the capacities of the cameras used are limited in determining spray characteristics. Electrospray cooling depends on a number of independent parameters, and the prediction possibilities are limited. Therefore, more experiments are needed to better understand the cooling mechanism. The inability to accurately measure the size of the micro-sized, and even nano-sized particles causes the spray characteristics to be incompletely determined. In addition, the sprayer needs to have both high speed and high resolution cameras that can be used to examine the formation in real time. In the present study, effective parameters of the electrospray cooling process, one of the indispensable cooling technologies for the increasingly increasingly applied Micro Electro-Mechanical Systems (MEMS), was determined through the use of a high speed camera (HSC).

There are very few studies on electrospray cooling in the literature and the results obtained are limited. In addition, although the spray mode is very important in electrospray, the effect of spray modes has been ignored in the studies. In the present study, the effects of electrospray modes on CER, particle diameter and velocity were investigated for the first time.

Section snippets

Experimental setups and procedure

An experimental system for the spray and cooling characteristics of deionised water electrospray was established. The schematic representation of the experimental system is seen in Fig. 2.

The first part of the study aimed to determine the cooling performance of the electrospray used in local cooling applications. It was aimed to increase the cooling performance by reducing mean diameter of the spray with flow rates of the dielectric fluid by changing the nozzle properties and applied voltage

Calculations

At the steady state condition, the heat generated (Qgen) is transferred by conduction heat transfer to the cooling plate and is then transferred by convection (Qconv), vaporization (Qvap), thermal radiation (Qrad) and heat loss (Qloss) from the heated surface [16]. For this reason, the heat removed from the cooling plate (Qcp) was calculated and expressed asQcp=Qconv+Qvap=QgenQradQloss

The cooling plate was made of 6063 series aluminum; its surface was fully cleaned and polished and working

Electrospray modes

In the study, in order to provide evaporative cooling conditions, low volumetric flow rates of 0.10, 0.15 and 0.20 ml/min were studied. Under the same conditions, it is clear that the spray mode for the various volumetic flow rates in Fig. 4 does not change. Therefore, the images obtained for 0.15 ml / min are given below in Fig. 5 and Fig. 6.

Fig. 5 highlights the effects that voltage and nozzle diameter have on the electrospray modes for a constant heat flux at 0.15 ml/min volumetric flow

Conclusion

In the present study, the effect of electrospray modes created by changing effective parameters in electrospray on spray and cooling characteristics was determined. The mean flow velocity and CER values were determined with high precision, while the mean particle diameters were measured with an uncertainty of 15%. The reason is that particles below 30 μm could not be visualized because they evaporate in the air and the camera is not enough. The accuracy of the data should be determined by

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The research was supported by the Scientific and Technological Research Council of Turkey (2214A) and funded by West Virginia University.

References (56)

  • Y. Kim et al.

    Heat transfer performance of water-based electrospray cooling

    Int. Commun. Heat Mass Transf.

    (2020)
  • P. Allen et al.

    Electrohydrodynamic enhancement of heat transfer and fluid flow

    Heat Recov. Syst. CHP

    (1995)
  • W.-J. Sheu et al.

    Effect of oscillatory EHD on the heat transfer performance of a flat plate

    Int. J. Heat Mass Transf.

    (2013)
  • A.A. Ramadhan et al.

    Numerical development of EHD cooling systems for laptop applications

    Appl. Therm. Eng.

    (2018)
  • A. Boroomandpour et al.

    A comprehensive experimental investigation of thermal conductivity of a ternary hybrid nanofluid containing MWCNTs-titania-zinc oxide/water-ethylene glycol (80, 20) as well as binary and mono nanofluids

    Synth. Met.

    (2020)
  • S.-R. Yan et al.

    The rheological behavior of MWCNTs–ZnO/Water–Ethylene glycol hybrid non-Newtonian nanofluid by using of an experimental investigation

    J. Mater. Res. Technol.

    (2020)
  • F. Soltani et al.

    Experimental measurements of thermal conductivity of engine oil-based hybrid and mono nanofluids with tungsten oxide (WO3) and MWCNTs inclusions

    Powder Technol.

    (30 June 2020)
  • W. He et al.

    Using of artificial neural networks (ANNs) to predict the thermal conductivity of zinc oxide–silver (50%–50%)/water hybrid Newtonian nanofluid

    Int. Commun. Heat Mass Transf.

    (2020)
  • Y. Zheng et al.

    Experimental investigation of heat and moisture transfer performance of CaCl2/H2O-SiO2 nanofluid in a gas–liquid microporous hollow fiber membrane contactor

    Int. Commun. Heat Mass Transf.

    (2020)
  • A. Shahsavar et al.

    Experimental investigation and develop ANNs by introducing the suitable architectures and training algorithms supported by sensitivity analysis: measure thermal conductivity and viscosity for liquid paraffin based nanofluid containing Al2O3 nanoparticles

    J. Mol. Liq.

    (2019)
  • B. Ruhani et al.

    Statistical investigation for developing a new model for rheological behavior of ZnO–Ag (50%–50%)/water hybrid Newtonian nanofluid using experimental data

    Phys. A Stat. Mech. Appl.

    (2019)
  • B. Ruhani et al.

    Statistical investigation for developing a new model for rheological behavior of silica–ethylene glycol/water hybrid Newtonian nanofluid using experimental data

    Phys. A Stat. Mech. Appl.

    (2019)
  • M.H. Aghahadi et al.

    An experimental study on the rheological behavior of hybrid tungsten oxide (WO3)-MWCNTs/engine oil Newtonian nanofluids

    J. Mol. Struct.

    (2019)
  • G.E. Amidon et al.

    Particle, powder, and compact characterization

  • X. Zhang

    Dynamics of growth and breakup of viscous pendant drops into air

    J. Colloid Interface Sci.

    (1999)
  • A. Jaworek et al.

    Classification of the modes of EHD spraying

    J. Aerosol Sci.

    (1999)
  • E. Moreau et al.

    Enhancing the mechanical efficiency of electric wind in corona discharges

    J. Electrost.

    (2008)
  • Z. Wang et al.

    Natural periodicity of electrohydrodynamic spraying in ethanol

    J. Aerosol Sci.

    (2018)
  • Cited by (24)

    View all citing articles on Scopus
    View full text