Surface dielectric barrier discharge in closed volume air

A. Dascalu1, A. Demeter1, F. Samoila1, V. Anita1, K Shimizu2, L. Sirghi1.*

1
Iasi Plasma Advanced Research Center (IPARC), Faculty of Physics, “Alexandru Ioan Cuza”
University, Iasi, 700506, Romania
2
Organization for Innovation and Social Collaboration, Shizuoka University, 3-5-1, Naka-ku, Johoku,
Hamamatsu, Shizuoka 432-8561, Japan
e-mail: lsirghi@uaic.ro

The efficiency of reactive species produced by a surface dielectric barrier discharge (SDBD)

working in closed volume air for decomposition of organic molecules in aqueous solutions

was investigated. A SDBD with surface of 6 cm2 was operated in a closed volume (0.5 L) of

air at atmospheric pressure. The voltage across the plasma gap, the charge transferred through

the plasma and the dissipated power were computed from the charge-voltage diagram to

values of 3900 V, 0.2 C and 6 W, respectively. It is shown that the reactive species

produced by the SDBD in closed volume of air are active for decomposition of methylene

blue molecules in aqueous solution for a relatively long period of time after the discharge was

cut off. This determined a life time of species active for decomposition of organic molecules

to about 11 minutes. However, SDBD operation in closed volume air shortened the lifetime

of polyimide dielectric of the discharge microelectrode system due to the increased humidity

and reactivity of the active species generated by plasma. Measurement of relative humidity

revealed that the discharge enhanced water evaporation, which determined a fast increase of

the humidity in air. Atomic force microscopy investigations of dielectric surface in the

discharge region reveled important modifications of surface morphology during the

discharge.

*Corresponding author, e-mail: lsirghi@uaic.ro

In the recent years, medical applications of atmospheric pressure plasmas are intensively

studied1. Among various atmospheric pressure plasmas, the surface dielectric barrier

discharge (SDBD) plasma proved to be an important candidate for plasma treatments for

wound healing2, drug administration3, sterilization4, 5 and removal of contaminants from air6

or water7. Most of the SDBD studies have been devoted to discharges working in open air or

in closed vessels through which were flown air or rare gases. Then, the effluents generated by

plasma are transported and used for treatments. There are very few studies of SDBD plasma

treatments performed in closed volume air7. The present study is devoted to SDBD operated

in closed volume air at atmospheric pressure for treatment of contaminant organic molecules

in aqueous solution. Oehmigen et al 7 studied the kinetic of chemical reactions taking place

in liquid as result of active species produced in the discharge plasma and their diffusion into

liquid. In their experimental arrangement, the plasma faced the liquid surface at a small

distance (5 mm) and the volume of air available for the discharge was very small (14 mL). In

the present study the plasma do not face the liquid surface and the volume of air enclosed in

vessel is much larger (0.5 L). Therefore, in the present study the plasma treatment can be

considered totally indirect, the treatment being determined the plasma generated active

species in the gas phase and their free diffusion from the gas phase in water. The plasma

active species involved in treatment of water or aqueous solutions are reactive oxygen species

(ROS) and reactive nitrogen species (RNS)8. Among these species, H2O2, NO2–, NO3– and O2

are the most important active species with important role in treatments for removal of organic

contaminants from water, inactivation of bacteria, or for inducing biologic reactions in living

tissue9. Oh et al10 showed that these species are generated immediately in water once the

atmospheric-pressure plasma is ignited and, if not consumed in reactions with organic

molecules, they survive for long time. In the present study, kinetic of active species generated

2
by SDBD operated in closed volume air is investigated indirectly by the analysis of the

decomposition rate of methylenne blue (MB) molecules in a small amount of water exposed

to plasma effluents. It is shown that the plasma generates species have a long lifetime (more

than 10 minutes), they being available for reactions in liquid phase for relatively long time

(half an hour) after the discharge has been cut off. Thus, as compared with operations in open

air or gas flow, the operation of SDBD in closed volume air increases the efficiency of

treatment. However, operation of the SDBD in closed volume air has also drawbacks. It is

shown that plasma active species promote water evaporation, which results in a fast increase

of water vapor in the air during the discharge. The high concentration of active species in gas

phase and the increase of humidity determine erosion of the dielectric polymer of the SDBD

electrode system due to numerous reactions of the reactive plasma species with the dielectric

material.

2 Experimental setup

The SDBD was generated in atmospheric air into a glass vessel with the volume of 0.5

liter. Figure 1 presents schematically the experimental setup. The SDBD microelectrodes

used in present study had a sandwich structure (area of 2.3  3 cm2) formed by two copper

electrodes (18 m in thickness) deposited on the two sides of a polyimide foil (Kapton® with

14 m in thickness and relative dielectric permittivity of 3.4). The upper (hot) electrode was

structured as 23 equidistant and parallel stripes (0.2 mm in width and 30 mm in length) with a

gap width of 0.8 mm. In the glass vessel, side by side with the SDBD electrodes, there was

placed a Petri dish (40 mm in diameter) with a small volume (5 mL or 3 mL) of a solution of

400 ppm methylene blue (MB) in deionized water. The SDBD was powered by a sinusoidal

wave high voltage amplifier. The voltage and intensity of discharge current signals were

collected by specialized voltage probes and monitored by a digital oscilloscope (DPO 2024

from Tektronix, USA). The electric characterization of the discharge is provided in the next

3
section. Decomposition of MB in solution was determined by measurements of optical

absorbance in the region 400-800 nm (with a peak in absorption at 664 nm) by a UV-VIS

spectrophotometer (Evolution 300 from Thermo Scientific). The humidity in the glass vessel

has been monitored by a precision hygrometer (HM34C, Vaisala Oyj, Finland). The

topography images and roughness of the dielectric before and after operation of SDBD in

open and closed volume air have been obtained by atomic force microscopy (AFM)

measurements performed by a commercial AFM apparatus (XE 100 from Park, South Korea)

working in non-contact mode. The measurements were performed with a silicon AFM probe

(HQ:NSG 35 from NT-MDT, Russia) with a sharpened tip (nominal curvature radius of 8

nm) and a stiff cantilever (nominal resonant frequency and force constant of 150 kHz and 5.4

N/m, respectively).

Fig. 1 Schematic representation of the experimental setup.

4
3 Electric characterization of discharge

Figure 2a shows the typical waveforms of the applied voltage and intensity of the

electric current through the SDBD during one oscillation period. The peak values of the

applied voltage and current intensity are around 1900V and 35 mA, respectively. As shown in

detailed figure 2b, the current intensity presents the typical pattern of a multi filamentary

discharge11 with spikes with the amplitude around 5 mA and life time around 20 ns. The

amplitude and number of current intensity spikes are larger during the positive discharge than

during the negative one. Important electrical parameters of the discharge as the energy

dissipated during one oscillation period, discharge power, electric charge transferred to

dielectric surface during one cycle the discharge, Qdis, and the voltage across the gas gap, Ub,

can be determined from charge-voltage Lissajous diagrams. Figure 3 shows a sketch of the

measuring circuit of SDBD that indicates also the capacitances involved in the discharge

mechanism. The capacitance C0 accounts for the capacitor formed by the dielectric between

the top and bottom electrodes and does not change during the discharge. The capacitance C1

account for the dielectric capacitor formed between the bottom electrode and bare surface of

the dielectric. The gap capacitance accounts for the air capacitor formed between the top

electrode and bare surface of the dielectric. During the multi filamentary discharges, the air

gap is partially shortcut by the low impedance of air plasma. If f is the fraction of bare

dielectric area covered by plasma, the impedance of SDBD circuit during positive and

negative discharge periods is approximately:

Cd  C0  f  C1 (1)

While the discharge is off, the capacitance of the SDBD electrode system is

C1  Cg
C 'd  C0  (2)
C1  Cg

The value of C0 has been determined to about 309 pF from plane capacitor formula:

5
  0 R  S
C0  , (3)
d

where S is the area of the upper electrode (S = 140mm2), d the thickness of the dielectric (d =

14 m), r the relative electric permittivity of the dielectric (r = 3.4) and 0 the electric

permittivity of vacuum (0 = 8.85 10-12 F/m). Same formula computed a value of 1324 pF for

C1.

Fig. 2 a) Typical waveforms of voltage and current intensity during a period of SDBD. b)

Detailed image showing current intensity variation at a smaller scale of time (1 s).

Fig. 3 Sketch of the SDBD circuit. The current intensity, I, signal is collected on a 12 Ohm

resistor and the discharge voltage, U, is collected by a voltage probe connected to the top

electrode. The bottom electrode has been grounded. Capacitance C0, C1 and Cg are shown

schematically.

6
 This change of circuit capacitance from Cd’ to Cd during the discharge determines a

change of the slope in Q-U diagram, which have typically a parallelogram shape. Figure 4

presents the Q-U diagram of the SDBD used in the present work. The regions AB and CD

show a smaller slope, which corresponds to capacitance Cd’, when the discharge is off, while

the regions BC and DA show a larger slope, which corresponds to capacitance Cd, when the

discharge is on. From the Q-U diagram Cd‘ = 260 pF and Cd = 390 pF. The values of Cd and

C1 show that only 10% of bare dielectric surface is covered by conductive plasma. The value

of Cd’ determined from the Q-U diagram is close to the value of C0 computed by plane

capacitor formula. To determine the relevant parameters of the discharge, the Q-V diagram of

the discharge12 has been computed by extracting the contribution of the displacement current

Id = Cd’dU/dt flowing through the dielectric (capacitance C’d). The Q-V diagram of the

discharge allows for determination of voltage across the plasma, Ub, and the charge Qdis

transferred through the plasma to values of 3900 V and 0.2 C, respectively. These values

determine the energy dissipated in one cycle to about 0.5 mJ and the discharge power to

about 6 W.

7
 Fig. 4 Q-U diagram of SDBD showing typical parallelogram shape (ABCD) before (blue)

and after (black) extraction of contribution of displacement current through the capacitance

Cd’. The slopes of quasi linear parts of the diagram compute the capacitances Cd and Cd’.

4 Results and discussion

4.1 Degradation of methylene blue solution

Figure 5 shows the variation of MB concentration in the solution exposed to the action of

reactive species produced by SDBD plasma. Due to dielectric failure, the SDBD electrodes

could not be operated more than 10 minutes in closed volume air (red plot in figure 5). Then,

the discharge has been operated only for 4 minutes, and the time evolution of MB

concentration in the solution exposed to active species generated by SDBD plasma has been

investigated during the discharge and after the discharge was cut off. The decay of MB

concentration in the solution shows that the reactive species generated by SDBD plasma have

a long life time and are decomposing MB molecules for about half hour after the SDBD was

cut off. After this time there was no further decomposition of MB. The time variations of MB

concentration show that for small amounts of solution (less than 3 mL) the amount of the

decomposed MB molecules is independent on the volume of the solution used in the

experiments. This indicates that the decomposition process is limited by the flux of active

species coming from air and dissolved in solution. Therefore, the time evolution of the

decomposition rate of MB molecules in aqueous solution is an indicator of the time variation

of the concentration of active species in the gaseous phase. Fitting of the time variation of

MB concentration with an exponential decay (Fig. 5) determines a decay time for the active

species in the closed volume air of about 11 minutes. This decay time may depend on the

volume and surface of the glass vessel that contain the discharge gas. Also, the air humidity

may affect the production and consumption of reactive species. Therefore, we monitored the

8
relative humidity (RH) of air during the discharge and found that the SDBD enhances

evaporation of water from the MB solution, probably due to contribution of radical species

generated by plasma and the ultra sounds produced by the SDBD dielectric.

Fig. 5 Time variation of concentration of MB in water during SDBD plasma operation and

after SDBD has been cut off.

Figure 6 shows comparatively the increase of RH in the closed volume air without and

with SDBD. Evaporation of water in a closed volume air raises the RH from the initial value

of ambient air towards saturation. This process is relatively slow in absence of SDBD plasma

and faster in its presence. Therefore SDBD enhanced the evaporation of the liquid water

exposed to plasma active species. We believe that the presence of reactive species and the

ultrasound produced by the discharge promote water evaporation. When SDBD is operated

for relatively long time (10 minutes), this may contribute to the electric failure of the

dielectric material because water is absorbed on the dielectric surface and modify its electric

properties. Also, it is known that the content of water vapor in the discharge gas have a great

impact on the discharge current and ultimately to the discharge power13.

9
Fig. 6 Variation of relative humidity in closed volume air during SDBD operation (squares)

and without SDBD (dots).

4.2 Degradation of dielectric material

The dielectric materials used in surface dielectric barrier discharge (SDB) devices are

inevitably exposed to plasma of the discharge filaments forming at their surface.

These discharges determine erosion and degradation of dielectric material because

action of plasma active high-energy species (electrons, ions, UV photons, radical and

metastable molecules), and finally can cause dielectric failure. Rise of the air

humidity during SDBD may also contribute to degradation of the dielectric material.

In the SDBD, plasma filaments form on the bare dielectric surface at the edge of the

hot electrode. Therefore we have investigated the evolution of dielectric surface

topography at this particular position by atomic force microscopy (AFM). It is

expected that operation of SDBD in open air to degrade less the dielectric because in

this case the active species formed in plasma are diluted and transported in the

environmental air. Therefore, we have investigated comparatively the changes of the

dielectric surface morphology during the SDBD operation in open and closed volume

10
 (0.5 L) air. Figure 7 shows comparatively the AFM topography images of dielectric

surface obtained before operation (a) and after 4minutes of SDBD operation in open

(b) and closed volume air (c). The AFM topography image of the dielectric surface

before operation shows a relatively smooth surface (root mean square roughness of

about 20 nm). The AFM topography images of the dielectric surface after 4 minutes

of operation revealed drastic changes consisting in surface buckling at nanoscale, the

RMS roughness of surfaces being 73 nm for operation in open air and 95 nm for

operation in closed volume air. Figure 8 shows the evolution of RMS roughness of

dielectric surface as result of SDBD operation. The difference between RMS

roughness values observed in the two cases is not very large for operation times lower

than 7 minutes. However, at larger operation time values the difference increases

significantly, probably due to the effect of increased humidity in the closed-volume

air.

Fig. 7 Atomic force microscopy images of the dielectric surface before (a) and after 4

minutes of operation of SDBD in open air (b) and closed volume air (c).

11
Fig. 8 Changes of root mean square (RMS) roughness of the dielectric surface during SDBD

operation in open and closed volume air.

5 Conclusion

In summary, this work investigated the efficiency of reactive species produced by

SDBD plasma in closed volume air at atmospheric pressure for decomposition of methylene

blue molecules in aqueous solutions. The SDBD electrode system consisted by a dielectric

foil of polyimide covered on one face with a planar copper electrode (grounded) and on the

other face with a striped cooper electrode (connected to an high voltage amplifier). The

voltage across the plasma gap, the charge transferred through the plasma and the dissipated

power were determined from the charge-voltage Lissajous diagrams. After operation for 4

minutes, the discharge was turned off. However, the decomposition of methilene blue

molecules continued for about half hour, which indicated that the reactive species produced

by SDBD were active for a relatively long period of time. The decomposition rate of

methilene blue molecules in solution after the discharge was cut off determined the life time

of active species to about 11 minutes. The operation of SDBD in closed volume air in

presence of water shortened the lifetime of polyimide dielectric of the discharge

microelectrode system due to increased humidity and reactivity of the active species

generated by plasma. Atomic force microscopy was used to observe the changes in the

surface morphology of the dielectric material during the discharge operation. The AFM

topography images of the dielectric surface revealed drastic changes consisting in surface

buckling at nanoscale with formation of grains after only four minutes of operation. These

surface morphology transformations were indicated by the surface root mean square

roughness, which increased from around 20 nm for the dielectric surface before operation to

about 90 nm after 4 minutes discharge time. Operation of SDBD in closed volume air

resulted in larger modifications, especially for long operation times, as compared with

12
operation in open air. This is probably an effect of the increased humidity observed during

SDBD operation in closed volume air.

Acknowledgement

This work was supported by CNCSIS, IDEI Research Program of Romanian Research,

Development and Integration National Plan II, Grant no. 267/2011.

References

1. Fridman G, Friedman G, Gutsol A, Shekhter A B, Vasilets V N, Fridman A, Applied

plasma medicine. Plasma Process. Polym. 2008; 5:503–533.

2. Babaeva N Y, Kushner M J, Reactive fluxes delivered by dielectric barrier discharge

filaments to slightly wounded skin, J. Phys. D: Appl. Phys. 2013;46:025401.

3. Shimizu K, Hayashida K, Blajan M, Novel method to improve transdermal drug delivery

by atmospheric microplasma irradiation. Biointerphases. 2015; 10:029517.

4. Shimizu K, Blajan M, Tatematsu S. Basic study of remote disinfection and sterilization

effect by using atmospheric microplasma. IEEE Trans. Indust. Appl. 2012; 48:1182–1188.

5. Oehmigen K, Winter J, Hahnel M, Wilke Ch, Brandenburg R, , Weltmann K-D, von

Woedtke Th, Estimation of Possible Mechanisms of Escherichia coli Inactivation by Plasma

Treated Sodium Chloride Solution, Plasma Process. Polym. 2011; 8(10):904-913.

6. Shimizu K, Blajan M and Kuwabara T. Removal of indoor air contaminant by atmospheric

microplasma. IEEE Trans. Ind. 2011;47:2351–2358.

7. Oehmigen K, Hoder T, Wilke Ch, Brandenburg R, Hähnel M, Weltmann K-D, von

Woedtke Th, Volume effects of atmospheric-pressure plasma in liquids IEEE Trans. Plasma

Sci. 2011; 39(11): 2646.

8. Kong M G, Kroesen G, Morfill G, Nosenko T, Shimizu T, Van Dijk J, Zimmermann J L.

Plasma medicine: an introductory review. New J Phys. 2009;11:115012.

13
9. Oehmigen K, Hoder T, Brandenburg R, Wilke Ch, Weltmann K-D, von Woedtke Th, The

role of acidification for antimicrobial activity of atmospheric pressure plasma in liquids,

Plasma Process. Polym. 2010; 7:250–257.

10. Oh J-S, Szili E J, Ito S, Hong S-H, Gaur N, Furuta H, Short R D, Hatta A, Slow

molecular transport of plasma-generated reactive oxygen and nitrogen species and O2 through

Agarose as a surrogate for tissue, Plasma Medicine 2015; 5(2–4): 125–143.

11 Ozkan A, Dufour T, Bogaerts A, Reniers F, How do the barrier thickness and dielectric

material influence the filamentary mode and CO2 conversion in a flowing DBD? Plasma

Sources Sci. Technol. 2016; 25:045016.

12 Peeters F J J, van de Sanden M C M. The influence of partial surface discharging on the

electrical characterization of DBDs, Plasma Sources Sci. Technol. 2015; 24:015016.

13. El Shaer M, Mobasher M, Zaki A, Effect of Flowing Mist Relative Humidity on the

Electric Characteristics of Helium Dielectric Barrier Discharge, Plasma Medicine 2015; 5(2–

4): 249–256.

14