Experimental investigation of electrosprayed droplets behaviour of water and KCl aqueous solutions in silicone oil

Experimental Thermal and Fluid Science 36 (2012) 249–255

Contents lists available at SciVerse ScienceDirect

Experimental Thermal and Fluid Science

journal homepage: www.elsevier .com/locate /et fs

Experimental investigation of electrosprayed droplets behaviour of waterand KCl aqueous solutions in silicone oil

Behnam Sadri, Babak Vajdi Hokmabad, Esmaeil Esmaeilzadeh ⇑, Reza GharraeiHeat and Fluid Flow Research Laboratory, Department of Mechanical Engineering, University of Tabriz, Tabriz, Iran

a r t i c l e i n f o

Article history:Received 18 May 2011Received in revised form 29 August 2011Accepted 26 September 2011Available online 4 October 2011

Keywords:ElectrohydrodynimicDielectric liquid mediaDropletsElectrical conductivity

0894-1777/$ - see front matter � 2011 Elsevier Inc. Adoi:10.1016/j.expthermflusci.2011.09.021

⇑ Corresponding author. Tel.: +98 411 3393054; faxE-mail address: [email protected] (E. Esmae

a b s t r a c t

Electrostatic atomization of liquids is now considered as a well established technique for creating finelymono-dispersed liquid droplets. One of the techniques in producing stable emulsion is to make finerdroplets of dispersed liquids using electrohydrodynamic actuators. This work is to investigate the hydro-dynamics of dispersed droplets of water and aqueous solutions of KCl in a continuum dielectric fluid ofsilicone 100 cSt. The effects of dispersed liquid flow rate, applied voltage and electrical conductivity ofliquids on droplet motion and its diameter were considered. By means of a nozzle with inner diameterof 0.8 mm, water and KCl solution droplets were generated and forwarded into the dielectric siliconeoil as a base of quiescence medium. In order to produce fine droplets, EHD was applied using high voltagenozzle and grounded ring. By this method, stable water and KCl solution droplets were produced in oilwithout any surfactants. It was observed that increasing dispersed liquid flow rate reduces the effectof EHD leading to droplet size increment. The conductivity of dispersed media has significant effectson droplet size and stability of them. For KCl solution with more conductivity than water, the directeffects of this property on droplet size and their stability were studied.

� 2011 Elsevier Inc. All rights reserved.

1. Introduction

Emulsions are systems consisting of two or more liquid immis-cible phases [1]. Emulsions are widely used in many branches ofindustry: power engineering, combustion engines, chemical andpetrochemical, pharmaceutical, food and nutrition. The stabilityof produced emulsion is important in all of the previous applica-tions. Many methods for this aim are used. Some of the methodsare relied on application of high shear stress to liquid mixtures.For a given droplet, increasing of the drift velocity between dropletand surrounding liquid normally cause more deformation. So, thedroplets in emulsions are owned higher shear stresses which causemore stability. A simple method of achieving this phenomenon isto mix two immiscible liquids in a baffled vessel by means of anagitator. In mass production of emulsions the homogeniser equip-ments are commonly used for producing smaller droplets [2].Homogenisation or homogenisation is any of several processesused to make a chemical mixture the same throughout. Intensivemixing of mutually insoluble substance or groups of substance iscarried out during this procedure to obtain a soluble suspension.

In customary production of emulsions, the stabilising agents arerequired. If two pure insoluble liquids as water and oil are mixed,the droplets forming the disperse phase would quickly coalesce

ll rights reserved.

: +98 411 3354153.ilzadeh).

and settle either to the top or the bottom of the continuous phaseas a completely separate layer concerning their densities. In orderto prevent or reduce this phenomenon, surfactants are added tothe emulsions [3].

A new and successful method to produce emulsions is the elec-trohydrodynamic atomization (EHDA). In most cases for producinga water-in-oil emulsion, atomization of water is carried out byusing a fine metallic nozzle, which is connected to a high voltagesource. The disperse phase is charged in the metallic nozzle andthen is drawn out to the gas-dielectric liquid interface influencedby electric field forces. In this method there is no necessity tosurfactants anymore.

The electrohydrodynamic inkjet functions by pressurising thefluid so that it forms a convex meniscus. By EHDA an electric fieldis applied that draws out this convex meniscus into a sharp cone.When the electric field strength is high enough to overcome thesurface tension of meniscus, the free jet can break. Depending uponthe strength of electric field, duration and amplitude of the ejectionpulse this technique can be used to produce a wide-angle spray, acontinuous stream and a discrete mono-disperse drops [4].

Advantages of making emulsion by electro spraying over exist-ing techniques may be listed as follows [2,5]:

– No necessity for high velocity liquid transfer: Reversely duringmechanical preparation of emulsions high revolution speedsfor homogeniser are required.

Nomenclature

E strength of electric field (V/m)fe body force of electric field (N)g gravity (m/s2)l length scale (m)r droplet radius (mm)T temperature (K)t settling time of droplet (s)t⁄ settling parameterT Maxwell stress

e permittivity (F/m)q fluid density (kg/m3)qc electric charge density (C/m3)ql oil density (kg/m3)qd dispersed (droplet) density (kg/m3)ll oil dynamic viscosity (kg/m s)ld dispersed (droplet) dynamic viscosity (kg/m s)M molaln normal direction

Fig. 1. Schematic illustration of main experimental setup.

250 B. Sadri et al. / Experimental Thermal and Fluid Science 36 (2012) 249–255

– The creation of mono-disperse droplets: The sizes of dropletsproduced in electrospraying method are almost the same, com-pared to wide range of produced droplets sizes by othermethods.

– The creation of an emulsion in a single step process: Mechanicalmanufacturing of emulsions, generally, is performed in morethan one step. Heating, homogenisation, mixing and coolingprocesses have to be carried out to prepare an emulsion butin EHDA in one step the emulsion is obtained.

– Low energy consumption required compared with other sys-tems: Each method of emulsion preparation requires thatenergy be put into the system in some form. The energy is sup-plied in a variety of ways: Trituration, homogenisation, agita-tion, heat and direct applied electric energy (electrosparying).Between all these methods, the lowest consumption is referredto electrosparying.

Limitations to the system may be regarded as the suitability ofthe liquids for electrostatic atomisation, and the necessity of highvoltages (up to 50 kV) which may cause problems on equipmentfrequently required in industrial flame-proof areas. An early at-tempt at producing emulsions by electrostatic means was reportedby Nawab and Mason [6] in 1958. A metal capillary nozzle forreleasing of the disperse phase liquid was immersed in the contin-uous phase liquid, and although this was successful at low concen-trations of disperse phase, it was reported that when theconcentration reached about 1% the technique was not successfuland the emulsion broke. The final system comprised a nozzle sus-pended above the interface of continuous phase liquid with theresulting aerosol collected on the surface and dispersed in the sec-ond liquid by stirring. With this system, particle size distributionwas quite narrow. Watanable et al. found that when an electricfield is applied to an oil/water interface, the interfacial tension isreduced almost to zero and spontaneous emulsification occurs,due to the interfacial fluctuation. Their experiments led to the nec-essary condition for emulsion production in their case: the dis-perse phase conductivity must be greater than that of thecontinuous phase [7].

The studies of emulsions atomization were also directed to im-prove the combustion performances of emulsions as fuels in fur-naces for local power generation [8–11].

Yusuke et al. [12] investigated the distribution of droplets andtheir size as a function of applied voltage and its frequency in orderto achieve a stable emulsion. Esmaielzadeh et al. [13] introduced anovel method for preparing metal powder by means of EHD. Theyfound that Due to the presence of electrostatic forces, the disinte-grated droplets and ligaments repel each other and therefore par-ticle coalescence and agglomeration is minimised in this method.Experiments of Broniarz et al. [14] showed that the changes inphysical properties of a liquid phase lead to the significant changesin the spray characteristics and the droplet sizes are dependent onliquid flow rates, geometry of nozzle and properties of liquid.

Jayasinghe and Edirisinghe [15] applied constant voltage for elec-trostatic atomisation of Water, glycerol and their mixtures withconstant flow rates. From their Results the effects of viscosity onmodes of atomisation and size distribution approved. Also risingviscosity increases the size and size distribution of obtained drop-lets. The measurements of Tang and Kebarle demonstrated that thetotal electrospray current increases with the conductivity of thesolution [16]. But in the literature, the influence of the conductivityon droplet sizes and stability of drops in dielectric medium rarelyhas been considered.

In this paper hydrodynamic behaviour of dispersed droplets ofwater and aqueous KCl solutions in continuum dielectric 100 cStsilicone oil has been experimentally studied. The effects of flowrate, applied voltage and electrical conductivity of dispersedliquid on droplet size and stability of drops in silicone oil arereported.

2. Experimental setup and methods

A schematic illustration of the experimental apparatus for pro-ducing droplets by means of electrostatic atomization method isshown in Fig. 1. A stainless steel needle, used as the nozzle capil-lary, with inner diameter of 0.8 mm was mounted vertically on astand. The dispersed liquid was supplied to the nozzle by meansof a syringe pump which obtained particular uniform flow rates.Below the nozzle a stainless steel ring with inner diameter of30 mm was used as ground electrode. The distance between thenozzle and the ground electrode was fixed at 25 mm. It is notablethat the distance between the nozzle and the oil/air interface was60 mm. 100 cSt silicone oil (AK-100WACKER) was used as contin-uous-phase liquid because of its excellent dielectric properties.Distilled water and aqueous solutions of KCl were used as dis-persed-phase liquids through the nozzle electrode. In order to pro-duce fine droplets, positive polarity high voltage DC was applied to

Table 1Electrical conductivity of dispersed solution.

Materials Distilledwater (0.00Molal)

KCl solution(0.01 Molal)

KCl solution(0.10 Molal)

KCl solution(1.00 Molal)

Conductivity(S/m)

0.042E-3 1.634E-3 11.82E-3 108.620E-3

Fig. 2. schematic illustration of setup for measurement of current.

Fig. 3. Mean droplet diameter of four high-voltages versus flow rate for distilledwater.

B. Sadri et al. / Experimental Thermal and Fluid Science 36 (2012) 249–255 251

the nozzle electrode. Droplets were injected vertically downstreamthrough the nozzle into the Pyrex vessel.

The nozzle capillary was fed with distilled water and KCl aque-ous solutions with various concentrations. The solution concentra-tion was changed to attain liquids with various conductivities.

Electrical conductivities of the dispersed liquids are noted inTable 1. With adding KCl the permittivity of water does not change,therefore during the experiments this property was assumed con-stant [17].

The conductivities of water and KCl solution were measured byconductivity-Meter (EZTECH AZ-8361).

In Fig. 2 the schematic illustration of the second apparatus isshown. This arrangement was used in order to measure the totalelectrospraying current (I). In the mentioned setup all of the dis-persed droplets hit the ring and plane electrodes and deliver allcharges to them. It was shown in earlier works that the positivecharge leaving the capillary tip is equal to the positive charge arriv-ing at the plane electrode [18]. Therefore this setup gives accurateelectrospray currents. This current to the plane and ring electrode,which were at ground potential, was measured by a micro-milliammeter.

The camera and the light source were fixed in front of the vesseland behind of it, respectively. After the electrosparying procedure,the imaging process begins. In order to investigate droplets behav-iour the camera was set to capture every 1 min, which is a suffi-cient time step due to very low droplet velocities. Images wereprocessed by means of image processor software. In the images,a scaling length, in focus area, was considered. By means of thisparticular length, precise length of one pixel was extracted exactly.Then all measured lengths, in pixels, were changed into actualmetric system.

Experiments were conducted in the room temperature andatmosphere pressure.

3. Results and discussion

The configuration for electrostatic atomization, illustrated inFig. 1, leads to smaller size of droplets in comparison with casesin which the atomization is carried out in the liquid media dueto secondary break up of droplets impacting the surface of liquidmedium [13]. Therefore in this study two simultaneous phenom-ena were observed, first the effect of EHD on atomization of the li-quid and the second is break up of droplets due to impacting theinterface.

As it is shown in Fig. 3 droplets become bigger when raising theflow rate. With an increase in flow rate the velocity of liquid growsand liquid has less time to get charged. Therefore the droplets haveless electric charge and an increase in droplet size is observed as aresult of decrease in amount of electric force which coincides withGanan-Calvo’s study [19]. In other words, in specific period of timethe quantity of water passing through the charging region of thenozzle increases so with increasing flow rate, droplets emergethe nozzle capillary with less charge density. Furthermore due togrowth of droplets velocity, electric field has less time to breakthem, therefore bigger droplets and lower quality of dispersion isobserved. It is noteworthy that in the same voltage, the hydrody-namic reason is the main factor for increasing droplet sizes. Whenflow rate increases, the hydrodynamics forces make the emanateddroplets bigger [20].

The figure illustrates that the maximum and minimum meandroplet diameter take place in 10 kV and 16 kV voltages, respec-tively. The reason is that The Maxwell stresses increase in highervoltages. Maxwell stress is defined as [21]:

T ¼ eðn � EÞ � E� e2

E2nþ ql

2dedql

E2n

where E is the electric field vector and E is the strength of electricfield.

Mean droplet diameter of different KCl solutions versus flowrate, in 10 kV high voltage, has been shown in Fig. 4. In order toinvestigate the dependence of droplet size on conductivity of dis-persed phase, three KCl solutions with different concentrations of0.01, 0.1 and 1 molal was prepared. The effect of various conductiv-ities of the dispersed phase on the droplet size is explained theo-retically in the following paragraphs.

As shown in Table 1 each concentration of KCl solutions hassensible conductivity difference with the others. Fig. 4 illustratesthat 1 molal KCl solution has the minimum mean diameter be-cause EHD force is stronger and mean droplet size of the 0.01 molalKCl solution is bigger than 0.1 molal KCl solution. Because EHDforce which is exerted to 0.01 molal is less than 0.1 molal KCl

Fig. 4. Mean droplet diameters of KCl solutions versus flow rate in 10 kV.

Fig. 5. Mean droplet diameter of four KCl solutions versus flow rate in 10 kVapplied voltage.

252 B. Sadri et al. / Experimental Thermal and Fluid Science 36 (2012) 249–255

solution due to the reason explained in the following. The EHDbody force consists of three terms which can be expressed as [21]:

fe ¼ qcE� 12

E2reþ 12r qE2 @e

@q

� �T

� �

The first term is coulomb force (electrophoretic) which exertson positive and negative free charges, The second term is dielec-trophoretic force, mainly depends on electric permittivity gradient,and the third term is electrostrictive force related only to com-pressible fluids.

When the conductivity of fluid increases the electrophoreticforce increases. Also with raising conductivity, the required timefor charging each droplet decreases, and in particular interval oftime, droplet can take more charge. In the present work the secondapparatus was conducted to prove this fact. From theoretical pointof view, for liquid with more charge the coulomb force is biggerand increase in amount of electric force results in smaller droplets.

Experiments on the second apparatus showed that current ofelectrospraying increases with higher concentration of KCl solu-tion. By raising concentration the conductivity of solution in-creases, therefore the electrospraying current increases [22]. Thegrowth rate at 12 kV is more sensible than at 10 kV whichillustrates that the charging rate of droplets increases with appliedvoltage. These results are in good agreement with Ganan-Calvo et al.’s study [19]. More charge in the droplets make strongerstress on capillary of the nozzle which leads to better atomizationand consequently smaller droplets [19]. Table 2 shows that thecurrent increases with higher concentration of KCl solution andsubsequently the charges in the droplet increases [23] which re-sults in more exerted Maxwell stresses.

Fig. 5 is presented to compare the size of distilled water dropswith the size of KCl solutions drops. It was seen that distilled waterand conductive solutions have a considerable difference in meandroplet diameter because of great differences in conductivityamounts.

The stability of droplets in continuum liquid was investigated,too. As depicted in Fig. 6, columns a–d present the settling proce-

Table 2Electrospraying current (l A) for different experimental conditions.

Materials

Voltage (kV) Distilled water(0.00 Molal)

KCl sol(0.01 M

8 3.94 3.9510 6.58 6.6512 10.50 11.50

dure of distilled water, 0.01, 0.1 and 1 molal KCl solutions dropletsin silicone oil. The settling procedure was defined as the time inwhich the droplets settled in the continuum phase. As time elapsesdistilled water droplets show the most stable behaviour in contin-uum phase. In other words the settlement duration of water drop-lets is the longest and the stability of drops lessens withconductivity. With increasing the conductivity of dispersed phasein spite of the fact that droplets diameters lessen, stability de-creases. In continuum phase, each droplet has electric surfacecharge. According to double layer theory, all surface charges in flu-ids are screened by a diffuse layer. This diffuse layer has the sameabsolute charge value, but with opposite sign from the surfacecharge. Due to these charges, the droplets affect each other bymeans of electrostatic Coulomb forces. There is one more electricforce, which is associated with deviation of the double layer fromspherical symmetry and surface conductivity due to the excessions in the diffuse layer. These forces are balanced with hydrody-namic drag force, which affects all bodies moving in viscous fluidswith low Reynolds number [24]. Because of existence of mentionedforces exerted on droplets, high quantity of droplets and unbal-anced distribution pattern of them in continuum liquid, irregularbehaviour and unpredictable motion of drops was observed.

Two effective properties in this procedure are droplet densityand conductivity. By dissolving KCl in distilled water the densityof the dispersed phase increases making droplets settle fasterwhich reduces the stability of droplets. The settling time decreaseswith increasing the conductivity of dispersed phase. More conduc-tive droplets become neutral earlier because they deliver theircharge faster to the dielectric medium. When the charges of drop-lets reduce, the irregular motion of droplets in the continuumphase diminishes and changes into distinct motion toward thewalls and bottom of the container gradually. As has been men-tioned, since the conductive drop loses its charge earlier, so with

utionolal)

KCl solution(0.10 Molal)

KCl solution(1.00 Molal)

3.95 4.006.90 7.00

13.67 16.23

t

m

t

m

t

m

t

m

t

m

m

t

m

t

m

t

m

t

m

t=0

min

t=10

min

t=20

min

t=30

min

t=40

min

t=50

min

t=60

min

t=70

min

t=80

min

t=90

min

n

0

n

0

n

0

n

0

n

n

0

n

0

n

0

n

0

n

(a) Distilled water (b) 0.01 M KCl Time sequences (c) 0.1 M KCl (d) 1.0 M KCl

Fig. 6. Settling time of different KCl solution droplets in silicone oil.

T= 0 second

T==1 minute

T=30 second

T=90 second

Fig. 7. The irregular motion of charged droplets. Vectors show the droplet path at mentioned moment.

B. Sadri et al. / Experimental Thermal and Fluid Science 36 (2012) 249–255 253

T=0 second T=1 minutes

T=2 minutes T=4 minutes

T=6 minutes T=8 minutes

T=10 minutes T=12 minutes

Fig. 8. the settling pattern of distilled water charged drops.

Fig. 9. Coarse droplets adjacent to the wall.

254 B. Sadri et al. / Experimental Thermal and Fluid Science 36 (2012) 249–255

increasing the conductivity of drop the irregular motion (Fig. 7) ofdrops disappear more quickly and they settle faster.

Settling pattern of droplets is illustrated in Fig. 7. In theseimages ten particular drops were selected and numbered. Motiondirection of each drop is determined by means of a vector so dropscould be followed. Unpredictable motion of drops is clear in the fig-ure. Same surface charges of these droplets make them repel eachother or move toward the vessel’s walls in the settling procedure.

In Fig. 8 the bulk motion and settling pattern of charged dis-tilled water drops are presented in 1 min time steps. At first, dropsare aggregated and close to each other due to spraying mode. Butas time elapses, same electric charge makes drops repel each otherso that they scatter in the continuum phase. This procedure carrieson until drops lose their charge. For neutral water droplets threephenomena were observed. Some drops travel toward the vessel’swalls and discharge their electric charges there. Some others settleat the bottom of the vessel. And small number of the drops coa-lesce each other making a bigger droplet which finally settles.

According to Fig. 9 some of seen droplets have bigger sizes incomparison with our analysis. These coarse droplets have been

gathered on the vessel’s walls which are not considered in thiscase. During the settling procedure, some drops approach to the

Table 3Relative time parameter, t⁄, for different experimental conditions.

Material Distilled water(0.00 Molal)

KCl solution(0.01 Molal)

KCl solution(0.10 Molal)

KCl solution(1.00 Molal)

t⁄ 2.454 0.1937 0.135 0.118

B. Sadri et al. / Experimental Thermal and Fluid Science 36 (2012) 249–255 255

walls of the vessel, discharge there and after adhering to the wallsact as grounded electrode which draws the other charged drops in-ward. Other charged drops move toward mentioned unchargeddroplets and after discharging coalesce into them. This coalescencemakes the drops, adjacent to the wall, bigger and bigger.

For good understanding the behaviour of droplets in the presentwork, a new non-dimensional parameter was defined as t⁄:

t� ¼ tactual

tstokeststokes ¼

9lll1þ2ll=ld1þll=ld

2ðqd � qlÞgr2

The t⁄ is a non-dimensional parameter expressing the behaviourof charged droplet with respect to an uncharged droplet. The mo-tion of charged droplets is influenced by gravity, hydrodynamicforces and repulsion of other charged drops while in the case of un-charged droplet the behaviour is affected by hydrodynamic forcesand gravity. In each case the motion of one particular charged dropwas tracked and settlement time was measured. In order to char-acterise the motion of an uncharged drop the Stokes regime wasconsidered and the settlement time of drop was calculated bymeans of mentioned formula. Table 3 depicts the considerable dif-ference between distilled water and KCl solution droplets. Gener-ally, any increase in conductivity lead to lower stability of dropswhich is obviously illustrated in Fig. 6.

4. Conclusion

The behaviour of droplets, produced by means of electrostaticatomization method, was investigated. The behaviour of waterand aqueous solutions of KCl, injected from the capillary nozzleinto silicone oil, was demonstrated and the diagrams of each sam-ple are depicted in the results section. The dispersion of water inpresence of electric field was studied and then atomization ofdroplets was enhanced, by increasing conductivity of dispersed li-quid, using KCl. In the first part of study the effect of altering volt-age and flow rate on the size of droplets was studied and it wasfound that rising the voltage and decreasing the flow rate leadsto decrease in the size of droplets. Then influence of altering con-ductivity was investigated by dispersing KCl aqueous solutionswith different concentrations. Increase in conductivity made

droplets finer and it was found that the droplets in dielectric liquidmedium are more unstable.

References

[1] Rajinder Pal, Rheology of simple and multiple emulsions, Curr. Opin. ColloidsInterface Sci. 16 (2011) 41–60.

[2] J.F. Hughes, I.D. Pavey, Electrostat. Emulsification, J. Electrostat. 10 (1981) 45–55.

[3] S. Kanazawa, Y. Takahashi, Y. Nomoto, Emulsification and demulsificationprocesses in liquid–liquid system by electrostatic atomization technique, IEEETrans. Indus. Appl. 44 (2008) 1084–1089.

[4] EricR Lee, Microdrop Generation, CRC PRESS, 2003 (23).[5] B. Vonnegut, R.L. Meubauer, Detection and measurement of aerosol particles

by the use of an electrically heated filament, J. Colloid Sci. (1952) 1000–1005.[6] M.A. Nawab, S.G. Mason, The preparation of uniform emulsions by electrical

dispersion, J. Colloid Sci. 10 (1958) 179.[7] Watanable et al., Studies on electrocapillary emulsification, J. Colloid Interface

Sci. 64 (1978) 279.[8] G.N. Laryea, S.Y. NO, Spray angle and breakup length of charge-injected

electrostatic pressure-swirl nozzle, J. Electrostat. 60 (2004) 37–47.[9] A.S. Ramadhas et al., Use of vegetable oils as I.C engine fuels: a review,

Renewable Energy 29 (2004) 727–742.[10] S. Slupek, The poly nuclear aromatic hydrocarbons formation during

combustion, in: Proceedings of the Conference on Low-Emission CombustionTechniques, Ustron-Zawodzie, Poland, 1996.

[11] G. Vaitilingom et al., Development of rape seed oil burners for drying andheating, Indus. Crops Prod. 7 (1998) 273–279.

[12] T. Yusuke et al., Electrostatic process in liquid–liquid system (I)-production ofW/O emulsion by electrostatic atomization, Seidenki Gakkai Koen Ronbunshu30 (2005) 135–140.

[13] E. Esmaielzadeh et al., Powder production via electrohydrodynamic-assistedmolten metal jet impingment in to a viscous medium, Powder Technol. 203(2010) 518–528.

[14] L. Broniarz-Press et al., The atomization of water–oil emulsions, Exp. ThermalFluid Sci. 33 (2009) 955–962.

[15] S.N. Jayasinghe, M.J. Edirisinghe, Effect of viscosity on the size of relicsproduced by electrostatic atomization, J. Aerosol Sci. 33 (2002) 1379–1388.

[16] Liang Tang, Effect of the conductivity of the electrosprayed solution on theelectrospray current. Factors determining analyte sensitivity in electrospraymass spectrometry, Anal. Chem 63 (1991) 2709–2715.

[17] F.H. Drake, G.W. Pierce, M.T. Dow, Physical review measurement of thedielectric constant and index of refraction of water and aqueous solutions ofKCl at high frequencies, Phys. Rev. (1930) 35.

[18] Ikonomou, M.0. Blades, A.T. Kebarle, Electrospray mass spectrometry ofmethanol and water solutions suppression of electric discharge with SF6 gas,Anal. Ct 63 (1991) 1989–1998.

[19] A.M. Ganan-Calvo, Current and droplet size in the electrospraying of liquidsscaling laws, Aerosol Sci. 28 (1997) 249–275.

[20] Bon Ki Ku, Sang Soo Kim, Yu Dong Kim, Sang Yong Lee, Direct measurementofelectrospray droplets in submicron diameter using a freezing method and aTEM image processing technique, Aerosol Sci. 32 (2001) 1459–1477.

[21] F. Chen, Y. Peng, Y.Z. Song, M. Chen, EHD behavior of nitrogen bubbles in DCelectric fields, Exp. Thermal Fluid Sci. 32 (2007) 174–181.

[22] Liang Tang, Paul Kebarle, Effect of the conductivity of the electrosprayedsolution on the electrospray current. Factors determining analyte sensitivity inelectrospray mass spectrometry, Anal. Chem 63 (1981) 2709–2715.

[23] A.M. Ganan-Calvo, J. Davila, Barrero, The emitted current and droplet size lawsin steady cone-jet electrosprays of polar and non-polar liquids, A. in: Proc. 4thInt. Aerosol Con., 1994.

[24] H.A. Pohl, Some effects of nonuniform fields on dielectrics, J. Appl. Phys. 29(1958) 1182–1188.