Simultaneous Deposition of Submicron Aerosols onto Both Surfaces of a Plate Substrate by Electrostatic Forces

We demonstrate one-step deposition of submicrometer-sized particles suspended in the gas-phase onto a plate type substrate using an electrostatic-assisted spray system. The spray nozzle was set perpendicularly to the substrates (facing the front surface). The particles were deposited on plate-type metallic surfaces, on both front and rear sides of the substrate. This “both-side” deposition can be ascribed to deflection of charged particles in the front side, and then drifting of the particles around the rear side. A numerical simulation also showed that the deposition mechanism was found to be dependent on the center and the edge of the substrate. The electrostatic effect is more effective on both the center and the edge than the diffusion effect. [DOI: 10.1380/ejssnt.2014.238]


I. INTRODUCTION
Charged particles trace a specific path under a suitable electric field.For instance, the particles go to a frontfacing surface and around a rear-facing surface.This "both-side" deposition is known as "wrap-around" effect, and has been utilized for agricultural spraying [1], and powder coating [2].Up to now, the investigated size ranges are larger than a few tens µm.
Submicron particles are of keen importance as functional materials and atmospheric particles [3,4].These particles contribute to transboundary air pollution as well as indoor system [5,6].A theoretical study has proposed the importance of electrostatic effect for deposition of atmospheric particles on surfaces in nature [7].Although the role of inertial force of submicron particle is inadequate [8], electrostatic effect can provide an additional force, resulting in a high deposition rate [9,10].However, no experimental study has been reported on the electrostatic deposition of submicron particles.The present study demonstrates the electrostatic deposition of submicron particles on a plate type substrate, also in order to provide experimental evidence for the "both-side" deposition of submicron particles.

II. EXPERIMENTAL
An electrostatic spray system is designed to generate positively charged submicron particles in the gas phase (Fig. 1).Stable spray was confirmed by optical microscopic observation and by measuring spray current derived from charged droplets.The particle size distributions in the liquid phase were measured by a dynamic light scattering (DLS) method (HPPS, Malvern Instruments, Worcestershire).The size distributions of dry particles in * Corresponding author: wuled@cc.tuat.ac.jp the gas phase were measured by an online (real-time) particle size/mobility analyzer system (SMPS, Model 3034, TSI, St. Paul).Two combustion-made nanopowders, which consists of 99.9% of elemental carbon, were used as representative material of atmospheric particles, denoted as sample A (Tokablack #5500, Tokai Carbon Co. Ltd., Tokyo) and sample B (Mikuni Color Ltd., Himeji).Suspensions with concentrations of 0.01-0.03wt% were prepared as following procedures.The nanopowders were mixed with ultrapure water sample, and mechanically dispersed using an ultrasonic-force based homogenizer (19.5 kHz, 300 W, US-300T, Nihonseiki, Osaka).
To mimic the shape of the natural surface (e.g.leaf), square plates with 10×10×0.02mm 3 were used as model substrates.If an insulator or semiconductor substrate is used, the surface-potential effect of the substrate on the mobility of particles (i.e.selective deposition) can be involved [11].For this reason, an aluminum substrate was used and grounded to create uniform surface potential.A spray nozzle was positioned perpendicularly to the substrate (front side) facing to the nozzle.The dis- tance between the substrate and the nozzle tip was set to be 100 mm.The liquid flow rate and the applied voltage were set to 0.2 mL/h and between 2-3 kV, respectively.A deposition experiment of charged particles was conducted for 60 min.The morphology of the deposited particles was observed by a field-emission scanning electron microscope (FE-SEM; JSM-6330FS, JEOL, Tokyo).The flow regimes around the substrate were calculated using COMSOL Multiphysics 3.5a (Comsol AB, Stockholm).The flow rate of the carrier gas (CO 2 ) around the tip of the nozzle was 3 L/min.The calculated flow regime (Fig. 2(a) and (b)) provides the minimum velocity (0.0221 m/s) at the nearest of substrate.According to a theoretical study, this velocity may not give an effect of wind on the particle deposition [7].Furthermore, electric fields around a conductive plate (Fig. 2(c) and (d)) were also numerically calculated, based on space charge of aerosols.The particle charges were calculated using a value of the spray current.

III. RESULTS AND DISCUSSION
The particle concentration in a liquid sample in the order of 0.01 wt% was found to be the necessary condition to form submicron (100-200 nm) aerosols.The size measurements (Fig. 3) reveal that individual carbonaceous particles (samples A, B) form aggregates (100-200 nm) in the liquid-and gas-phase.In the gas-phase, sample B show sharp size distribution and high concentration compared to sample A, suggesting that the aerosols of sample A could have a broad size distribution including size exceeding 500 nm, which is out of the range for the present instrument (SMPS).Most runs in this study then use sample B to obtain sharper particle size distribution in the gas-phase.The deposition mechanism of charged particles on four regions in a substrate (center and edge in front or rear side) is discussed.Despite the angle of the spray nozzle perpendicularly to the front side, the particles were found on both front and rear sides (Fig. 4).The number of particles and the sizes were calculated using an image analyzer (WinRoof, Mitani Corporation, Fukui/Tokyo).The number and the size distributions on the positions are described inset Figs. 4 and 5, respectively.The particle number concentrations on the edge (of both sides) are higher than those of the center.The sizes of deposited particles (Fig. 5) are ∼100 nm in all regions moderately consistent with the gas-phase measurements (Fig. 3(d)).
Particles deposited on the center and the edge are anhttp://www.sssj.org/ejssnt(J-Stage: http://www.jstage.jst.go.jp/browse/ejssnt/)Fig. 4 Front alyzed in terms of deposition velocities by diffusion and electrostatic effect.First, the diffusion deposition velocity is calculated using the following equation [7]: where k and T are the Boltzmann constant (1.38×10 −23 J/K) and the absolute temperature (293.15K) respectively; B and D are the mechanical mobility and the characteristic length respectively.Re and Sc are the Reynolds number and the Schmidt number respectively as described: where U a and ν are the flow velocity of air near a deposition substrate (0.0221 m/s) and the kinematic viscosity of air (1.6×10 −5 m 2 /s) respectively.When ReSc > 0.2 is satisfied, the validity of Eq. ( 1) is ensured.Since a spherical particle with 100 nm gives B = 1.72×10 11 m/(s N) and size of the substrate is used as the characteristic length (D = 0.01 m), the U D is calculated to be 4.6 × 10 −6 m/s.The electrostatic deposition velocity, U e is described as the following form [12]: where E and q p are the electric field and the particle charge respectively.The calculation results for the electrostatic deposition velocities are 6.7×10 −4 and 2.3×10 −2 m/s on the center and the edge, when the electric field and the particle charge are given E = 2300 and 80000 V/m (Fig. 2(d)) and q p = 1.7 × 10 −18 C, respectively.The values of E and q p were estimated by the numerical simulation and the spray current respectively.The diffusion deposition velocity is not affected by the electric field and hence the comparison of both deposition velocities provides the orders of U D < U e on the center and U D ≪ U e on the edge.In the present study, the electrostatic effect is dominant around both the center and the edge of the substrate compared to the diffusion effect.The electrostatic-assisted spray method is used to generate unipolar charged particles [13], which exert mutual repulsion.The particles approaching to the front side of the substrate are deflected to the edge because of the deposition velocity difference between the center and the edge.In the same time, some of the particles go around the rear side of the substrate.Furthermore, the image force [14] and Coulomb force assist the particles to be deposited on the rear side.
In a theoretical study [15] for agricultural application, "both-side deposition" is characterized in term of the charge-to-mass ratio, q m of the particles having size between 5 and 30 µm.Zhao et al. [15] also concluded that the electrostatic effect can enhance the both-side deposition with high number of the charge-to-mass ratio.In an experimental study using 100-µm charged particles [16], the both-side deposition has shown that the deposition rate of the approaching particles on the front side of a substrate decreased with increasing the charge-to-mass ratio.Number or weight concentration per unit area on the rear side is close to that of the front side (i.e.frontto-rear side deposition ratio is 1:1) [16] as q m increases up to 5 mC/kg.However, our experimental data did not follow the trend; the total number concentrations of the deposited particles on the front and rear sides are estimated to be 5.4 and 2.7 numbers/µm 2 , respectively, resulting in the ratio of 2:1.This inconsistency might be attributed to the different particle charge between large (> 1 µm) and small (< 1 µm) particles and the difference in the experimental setup.For instance, the particle charge ratio of 100-µm particle (q m = 5 mC/kg) to 0.1-µm particle (q m = 1.8 C/kg) is estimated to be 2.8×10 6 .Higher particle charge increases the image force and Coulomb force affecting the rear-side deposition.In addition to the charge-to-mass ratio, the particle charge should be considered for both-side deposition.Electro-static effect can play an important role in the deposition mechanism for submicron particles.In that sense, it is important to consider the electrostatic effect in the study of particles suspended as well as their applications in the field of atmospheric-environment-related studies.

IV. CONCLUSIONS
Both-side deposition using a plate type substrate was demonstrated for the first time using submicro-meter sized particles.The deposition mechanism was found to be dependent on the regions of the substrate (center and edge).The electrostatic effect is dominant on both the center and the edge compared to the diffusion effect.The charged particles are deflected by the deposition velocity difference between the center and the edge and its mutual repulsion, on the facing surface, and thereafter drift around the rear side.

FIG. 1 :
FIG.1:A schematic diagram of electrostatic-assisted spray and deposition system.

FIG. 2 :
FIG. 2: Contour lines and direction arrows of velocity fields from outside a spray system to around a substrate (a) and in the vicinity of the substrate (b).(c) A distribution of electric fields around substrate and (d) plots of electric field versus substrate length on front side: substrate length = 0 is the center and substrate length = 5 and −5 are the edge.

FIG. 3 :
FIG.3: Size distributions of particles of two samples A and B measured in the liquid (a, c) and gas phases (b, d).

FIG. 4 :FIG. 5 :
FIG. 4: FE-SEM images of the particles (sample B) deposited on four positions of a grounded aluminum substrate: edge/front (a), center/front (b), edge/rear (c), center/rear (d).Values inserted in these images represent numbers of the deposited particles in the corresponding area (approximately 400 µm 2 ).