Biological Decontamination by Microplasma

1. Introduction

Microplasma technology proved to be a more economical and ecological alternative to classic technologies for flow control [1,2], air treatment [3,4], surface treatment of polymers and glass [5,6,7], or biomedical applications [8,9]. Among biomedical applications, nonthermal plasma is of interest for the sterilization of bacterial contaminations since traditional methods could encounter limitations due to acquired resistance to antibiotics [10,11,12]. Moreover, nonthermal plasma is a more feasible alternative than other methods that use chemicals or heat. Biological decontamination can be achieved using atmospheric pressure nonthermal plasma with various types of plasma sources that can be classified considering their electrode geometry or the frequency of excitation [13]. The electrode geometry furthermore determines if the decontamination process is carried out for flat surfaces, complex geometries or is a localized process [14,15].

Previous studies showed how nonthermal plasma can be used for the bacterial decontamination i.e. Escherichia coli [16,17,18,19], Pseudomonas aeruginosa [20,21,22], and Staphylococcus aureus [23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39]. To inactivate these bacterial strains with nonthermal plasma, air was used as discharge gas [16,17,18,19,20,21,22,23,24,27,28,29,30,31,32,34,35,36,37,38,39], nitrogen [22,25], oxygen [38], mixtures of argon with oxygen [23,33,38], argon with air or helium with air [26]. Nonthermal plasma in dielectric barrier discharge (DBD) configuration was mostly used [16,17,18,19,21,22,23,24,25,26,27,29,31,32,33,34,35,36,37,39]. Other configurations were direct current (DC) microjet and corona discharge [20,30,38]. Although nonthermal plasma systems are already industrial-type produced [16,18], the discharge voltage is still high (more than 10 kV). Studies were also carried out by researchers that reduced the discharge voltage required for nonthermal plasma generation below 5 kV [19,25,31,34]. As mentioned also by Juozaitienė et. al. [17], the majority of nonthermal plasma devices used for plasma medicine applications, are based on high-voltage devices that are expensive and difficult to operate.

This research describes the use of microplasma as a type of atmospheric nonthermal plasma. It is a type of surface DBD that is generated facing two grid electrodes having a micrometer size dielectric material in between. Due to the micrometer size thickness of the dielectric material, a high intensity electric field generates microplasma at relatively low discharge voltages of about 1 kV. Conventionally nonthermal plasma is generated at tens of kilovolts. Because low discharge voltages are necessary to generate microplasma, less electrical insulation is required along with small-size power supplies and reactors. Also, microplasma is generated at atmospheric pressure, thus, costly vacuum enclosures are not necessary. The system of electrodes presented in this paper has the high voltage electrode covered on the back side with a dielectric layer, while the microplasma is generated at the surface of the grounded electrode, making the system safe to use. Moreover, the electrode can be scaled to larger or smaller surfaces while the voltage required voltage to ignite the discharge remains the same. The electrode is also flexible thus it can be bended and folded to fit various shapes. A portable decontamination system using microplasma could be fabricated due to the small size of electrode system and power supply thus making microplasma a more economical, safer and accessible technology for the sterilization of bacteria. A DBD plasma system similar to the one in this research was reported by Gershman et. al. although the discharge voltages are higher, over 2 kV.

The particle kinetics shows that microplasma produces effectively reactive species and charged particles at gas temperatures near the room temperature. It is a highly non-equilibrium characteristic that recommends microplasma for various applications [40]. This combined with the portability makes microplasma the ideal technology for biological decontamination. In this research we have investigated the microplasma treatment of Escherichia coli, Pseudomonas aeruginosa and Staphylococcus aureus.

2. Materials and Methods

2.1. Experimental Setup

In Figure 1(a) shows the schematic view of the experimental setup developed for the bacteria treatment using microplasma. The microplasma electrode is placed on a support that has electrical connections and also provides through a gas inlet the gas flow to the electrode as shown in Figure 1(b). During experiments the electrode was placed at a distance of 2 mm above the surface of the agar containing bacteria.

A pump (Mitsumi, R14-1604), was used to flow the air. It has a miniature size and it can be powered with voltages in range of 4.5-12 V thus making it suitable for battery use and furthermore for integration in a portable system. Room air with relative humidity RH at 39-41% was used and the air flow rate was set at 0.6 L/min. An ozone monitor (NORM ADKS-1 O3 Ozon) was placed at 2 mm under the microplasma electrode to measure the ozone concentration. The discharge voltage was increased from -1.2 kV to -1.7 kV with 1 kV incremental steps and ozone concentration was measured at each voltage value and at an air flow rate of 0.6 L/min.

A Marx Generator (laboratory manufactured) with MOSFET switches was developed to be used as the pulse power supply to energize the microplasma electrodes. The output of Marx Generator is a negative pulse with voltage peak values up to -3 kV. In our experiments the discharge voltages were set at -1.5 kV and -1.7 kV. The frequency was 500 Hz. A pulse wave was chosen to energize the electrodes since it has the advantage of producing less ozone than an AC excitation [23], thus is less harmful to humans. The energy efficiency is higher than the sinusoidal microplasma for pulsed microplasma because of a longer microplasma time off period that results in a lower power consumption [41]. The discharge voltage and corresponding discharge current were measured using a high voltage probe (Tektronix, Tek P5100), and a current probe (Tektronix, A622), respectively. The probes were connected to a digital oscilloscope (Tektronix, MSO 4104).

2.2. Microplasma Electrode

The microplasma electrode was facing the target to be sterilized.

Microplasma was generated at atmospheric pressure by a surface DBD type electrode using a configuration having a dielectric material of 25 µm thickness between a grid-shaped wire with 1.2 mm width on the high voltage side and 0.2 mm width on the ground side as shown in Figure 2 (a) and (b). A back dielectric layer of 50 µm thickness covers the high voltage electrode. The electrode has a diameter of 30 mm and the grid-shape is arranged with an interval of 3.5 mm.

The electrode has 31 perforated holes, with diameter Ø 2 mm, to allow the gas to flow and furthermore bring the active species generated by microplasma to a specific target. Microplasma discharge appeared near the grid of the grounded electrode as shown in Figure 2 (b) and (c).

Previously we have reported using a similar type of electrode, sterilization results obtained using a gas flow rate of 2 L/min [23]. Staphylococcus aureus was inactivated using pulse and AC voltages to energize microplasma electrodes. Air and Ar/O₂ mixtures were used as discharge gases.

The output of the Marx generator energized the electrodes at a negative pulse voltage with pulse width of 16.9 µs and a pulse rising time of 548 ns as shown in Figure 3. The measured discharge current is the total current, I_tot,

$$I_{tot}=I_{disp}+I_{cond},$$

(1)

where I_disp is the displacement current, that occurs due to the charging of the capacitance formed by the microplasma electrodes, and I_cond is the conduction current, that flows through the plasma.

The displacement current, I_disp,

$$I_{disp}=C{\displaystyle \frac{dV}{dt}},$$

(2)

where V is the applied voltage. The capacitance of the microplasma electrode C, was calculated considering the current at the discharge voltage of -400 V where the measured current is the displacement current I_disp, and there is no microplasma discharge [42,43]. Thus, I_tot=I_disp, and for the microplasma electrode it was calculated C=37 pF. After knowing C, we have calculated:

$$I_{cond}=I_{tot}-C{\displaystyle \frac{dV}{dt}}.$$

(3)

The microplasma instantaneous power is obtained by multiplying the voltage V and conduction current I_cond waveforms. The microplasma discharge power was calculated by integrating the power waveform in time and divided with the period corresponding to the frequency of 500 Hz. The calculated discharge power considering the conduction current I_cond at the discharge voltages -1.5 kV and -1.7 kV was 2.52 W and 2.91 W respectively.

2.3. Ozone Generation

The concentrations of ozone generated by the microplasma increased with discharge voltage at a gas flow rate of 0.6 L/min, as shown in Figure 4. At the discharge voltage of -1.2 kV no ozone was measured. The values of ozone concentration at -1.5 kV and -1.7 kV were 12.8 ppm and 19.2 ppm respectively.

2.4. Microbiology Assay

Three bacterial strains were used to determine the inhibitory capacity of the microplasma: Escherichia coli (ATCC: 25922), Pseudomonas aeruginosa (ATCC: 27853), and Staphylococcus aureus (ATCC: 25923). The microbiological assays were conducted according to EUCAST protocols [44,45,46]. The 24 h samples were adjusted to 10^-5 dilution and spread with a sterile swab on 10 cm Ø Petri dishes containing Nutrient Agar. The volume capacity of the swab is of 200 µL, and we estimated a volume of ~50 µL to be dispersed on the surface of the Petri dish. Two experimental designs were approached. The dishes were left to rest for 30 minutes, then the microplasma device was applied for 5 minutes at -1.5 and -1.7 kV. The plates were then incubated for 24 h at 35°C. Based on the best inhibitory efficiency, the next experimental design was employed. Here, using the same bacterial density, the strains were spread on the agar plates and left to grow for 24 h, after which the plasma was applied at -1.7 kV, for 5 minutes. The plates were incubated for another 24 h post-treatment and the results were quantified.

The approximate D values in Joule, for ~50 µL of bacterial suspension, were calculated as:

$$D=\left({\displaystyle \frac{Treatment time t}{{\log }_{10}Initial Concentration-{\log }_{10}Concentration after treatment time t}}\right)\ast Discharge Power$$

(4)

Initial concentration stands for initial concentration of bacteria.

2.5. Statistical Analyses

One Way ANOVA and Student’s t test were performed using Origin 8 (Origin Lab Corporation, USA) to determine the statistical significance between control groups and treated samples, at p levels below 0.05.

3. Results

Three bacterial strains were used to assess the antimicrobial efficacy of microplasma under two experimental designs (Figure 5).

The bacteria were inoculated on agar plates and the microplasma was applied at -1.5 kV and -1.7 kV for 5 minutes. At -1.5 kV only two out of the three bacterial strains were inhibited. The experiments were carried in duplicate, and only for S. aureus both replicates had a clear inhibition zone. At -1.7 kV the developmental of bacterial colonies was inhibited for all three strains, with clear inhibition zones that could be measured and compared (Figure 6). The microplasma had the best effect against S. aureus in this case as well, followed by E. coli, and P. aeruginosa, which was the least susceptible from the tested strains. Where possible, the colony forming units (CFU) were counted (Table 1). For both E. coli and S. aureus a decrease in CFU was observed, however, the results were not statistically significant compared to the untreated controls.

4. Discussions

The microplasma discharge in air generates Reactive Oxygen Species (ROS) such as O, O₂(a¹Δ_g), and O₃, and Reactive Nitrogen Species (RNS) such as NO, NO₂, and ONOO^-, [23,47,48,49]. The discharge using room air benefits from the presence of water vapors thus H₂O₂, OH, HNO₃, and HNO₂ are also generated.

The mechanisms responsible for the generation of ROS and RNS by microplasma are initiated by the electron impact dissociation of O₂, N₂, and H₂O (humidity in room air). Some ROS and RNS are generated at this stage and also the reactive products that furthermore are necessary to form other ROS and RNS [50]:

e + O₂ → O + O + e

(5)

e + N₂ → N + N + e

(6)

e + H₂O → OH + H + e

(7)

e + H₂O → H + H + O + e

(8)

e + H₂O → H₂ + O + e.

(9)

The electrons are losing energy mainly through inelastic collisions that are responsible for producing excited states of O₂, N₂, and H₂O. These excited electronically, rotationally, or vibrationally states are O₂(r), O₂(v), O₂^∗, O₂(¹S), N₂(r), N₂(v), N₂^∗, or H₂O(v). The formation of NO consists in the excited electronically states of N₂ (N₂^∗) which are quenched dissociatively by O₂ to produce O atoms, and furthermore participate in reactions with O. The three-body reaction of O₂ and O produces Ozone (O₃) [51]:

O + O₂ + M → O₃ + M

(10)

where M is a third collision partner: O, O₂ or O₃.

The direct three-body recombination is the mechanism for production of O₂(a¹Δ_g) [52]:

O + O + M → O₂(a¹Δ_g) + M.

(11)

The third body M can be in the room air primarily O₂/N₂ but also can be a third oxygen atom.

ROS, RON and also UV light affect the cell envelope and internal components of bacteria. Due to these active species the cellular envelope collapses and the intracellular components (DNA) is damaged [25,53]. T. Zhu et al. pointed out the differences between the inactivation mechanism of E. coli and S. aureus. For E. coli, a gram-negative bacteria, the main reason for the inactivation process occurs due to the action of ROS and RNS on the cell envelope thus the cell envelope of E. coli is affected and leads to cellular swelling, rupture and cytoplasmic leakage. Furthermore, a denaturation of biomacromolecules occurs because ROS and RNS enters excessively into the intracellular milieu. For S. aureus, a gram-positive bacteria, the deactivation mechanism consists of the accumulation in excess of ROS and RNS on internal structures. S. aureus has a larger cellular volume and also the peptidoglycan structure is thicker, thus, when the cell envelope is compromised there are minimal changes in particle size and consequently the likelihood of cytoplasmic leakage is reduced. Microplasma treatment causes damage to the intracellular macromolecules due to accumulation in excess of intracellular ROS that furthermore causes cellular death. For P. aeruginosa, a gram-negative bacteria, the inactivation mechanism is similar with E. Coli. According to Mosaka et. al. [20], a thinner peptidoglycan layer of P. aeruginosa compared with gram-positive bacteria and the presence of an outer membrane that contains proteins and lipopolysac-charide, which are sensitive to ROS and RNS in P. aeruginosa, will cause after microplasma treatment, cell breakage.

In various research papers is investigated the use of argon in mixture with air or oxygen as discharge gas for the plasma treatment of bacteria [23,33,38]. The discharge voltage is lowered when argon is used thus the power supply and the electrical insulation is reduced but the argon is expensive compared with air. Moreover, the use of room air is a more economical and practical solution since is freely available thus a portable microplasma device needs only a pump to generate a gas flow. The value of air flow rate was set considering our previous research [23]. In this research the value of the gas flow rate was set at 0.6 L/min due to the effect observed at a flow rate of 2 L/min [23]. The gas flowing to the holes, as shown in Figure 2 (b), has to have and optimum velocity, thus transporting the active species to the target but not flowing directly to the target through hole without taking the active species from the microplasma generated area near the grid of the electrode. Although in the gas velocity was higher compared with this research, on the agar the area directly under the electrode had spots of grown bacteria after the treatment. These spots were facing the electrode holes thus we concluded that the flow was too high and the gas was flowing directly to the agar without taking active species. Thus, after trials we reached the value of 0.6 L/min for the gas flow that allowed the gas to gather the active species generated by microplasma and transport them to the agar surface. Besides the gas flow from the pump, the air is directed to the target also by the vortexes induced by microplasma electrode due to the electrohydrodynamic (EHD) phenomenon [54].

At a gas flow of 0.6 L/min the calculated time for the gas to pass through the microplasma electrode’s holes and cross the distance of 2 mm to reach the agar that contains bacteria, was calculated to be at least 5 ms. The lifetime of highly reactive neutral species O, OH, and NO is about 0.1 ms and about 1 ms for N, O₂(a¹Δ_g), and HO₂ in the afterglow region [55]. According to Sakiyama et. al. these reactive species could be delivered to a surface to be treated by having a gap of under 1 mm and an introduced convective flow of at least 10 m/s. Considering this we can conclude that the inactivation of bacteria was mostly due to the long-lived neutrals such as H₂O₂, N₂O, NO₂, HNO₂, O₃, NO₃, HNO₃ and N₂O₅, although the simulation results shown in [55], indicate that the interpretation of results can be affected due to a significant evolution in time of the reactive species density. To benefit also from the effect of the reactive neutral species with shorter lifetime the distance between microplasma and bacteria should be decreased, as Gershman et. al. reported results with bacteria inoculated glass coverslip in direct contact with plasma. This is possible with our microplasma electrode system since microplasma is generated on the grounded electrode side. These types of experiments will be carried out in our future research. The values of ozone concentration as shown in Figure 4 are 12.8 ppm and 19.2 ppm at -15 kV and -1.7 kV respectively. These values are lower than what we reported in with microplasma electrodes energized at AC voltage. Among long-lived neutrals ozone plays a major role in the sterilization process. The lower concentration obtained with pulse microplasma compared with AC driven microplasma recommend microplasma energized by pulse power as a safer alternative for the biological decontamination due to ozone hazard for human health at higher concentrations.

For the experimental design 1, the inhibition zone as shown in Figure 6, corresponding to the microplasma treatment at -1.7 kV, shows an area of bacterial inhibition larger than the microplasma electrode area of 3 cm diameter. Due to the gas flow the active species are spread thus for all bacterial strains an inhibition area with a diameter of at least 4.5 cm was measured. The colony forming units counts as shown in Table 1 represent all the bacteria grown on the Petri dish. Because the Petri dish diameter is larger than the microplasma electrode diameter, for a more correct assessment of the inhibition efficiency of microplasma, it should be considered the area below the electrode. In this case as shown in Figure 5, at -1.7 kV discharge voltage, the microplasma inhibition efficiency could be considered that was 100% for all bacterial strains.

The Experimental design 2 in which the bacteria was inoculated on agar plates and incubated for 24 h, treated with microplasma at -1.7 kV for 5 min, and after re-incubated for an additional 24 h, showed no visible effect on bacteria after microplasma treatment.

The bacteria was inhibited 100% on the surface of agar below the electrode, as shown in Figure 5. Thus, the CFU count was carried out for the bacteria that appeared mostly on the sides of the agar plates where the gas active species generated by microplasma did not reach. The discharge power of microplasma electrode is proportional with the surface area of the electrode. It can be considered that in the case of swabbing 1 mL of bacterial liquid the surface of the bacterial spread on agar will increase 20 times compared with 50 µL of bacterial liquid and consequently, to inactivate this area, an electrode 20 times larger will be necessary. Thus, discharge power will increase 20 times. According to this the D values calculated for 1 mL of bacterial liquid are the D values calculated for 50 µL multiplied with 20. The D values are shown in Table 2. For E. coli the initial concentration of 6.96×10³ CFU/mL was reduced to 5.58×10³ CFU/mL at -1.5 kV and 4.45×10³ CFU/mL at -1.7 kV after 5 minutes of treatment. The experiments were carried out using 50 µL of bacterial liquid, thus the D values were 157.5 kJ/mL at -1.5 kV and 89.9 kJ/mL at -1.7 kV. Considering only the inactivation of 50 µL of bacterial liquid the D values were 7.88 kJ and 4.49 kJ at -1.5. kV and -1.7 kV respectively.

The experiments carried out with S. aureus had the initial concentration of 4.1×10³ CFU/mL. After 5 minutes of treatment time the concentration was reduced to 0.71×10³ CFU/mL at -1.5 kV. This corresponds to a D value of 19.9 kJ/mL and a D value considering 50 µL of bacterial liquid of 0.99 kJ. The calculated D values in Joule based on D values reported by Zhu et. al. [16], after 50 seconds of treatment time were 18 kJ/mL and 40 kJ/mL for E. coli and S. aureus respectively, considering 1000 W input power and a 4 cm distance between plasma jet and bacterial sample. Liao et. al. reported a 4.95 log CFU/mL reduction of S. aureus after 45 seconds of treatment time at 60 W input power. This corresponds to a D value of 0.55 kJ/mL. In another study from Liao et. al. a 0.85 log CFU/mL reduction was reported for S. aureus and a 1.72 log CFU/mL reduction for E. Coli using a DBD system. The results were obtained at a power of 50 W after 30 seconds of treatment. These correspond to the calculated D values in Joule of 1.76 kJ/mL and 0.87 kJ/mL for S. aureus and E. coli respectively. For the values obtained with S. aureus at -1.7 kV as shown in Table 1 was difficult to calculate the D values since the average control values were higher than the treated values. This can be explained by the variations in the initial bacterial count due to the method of swabbing that we have used to spread bacteria on the Petri dish. Nonetheless the effect of inactivation is clear as shown in Figure 5 and Figure 6. Considering only a 45 mm diameter area below electrode, as shown in Figure 5, where inhibition was 100%, the D value for E. coli at -1.7 kV was 5.6 kJ/mL and 0.28 kJ for 50 µL. These values are comparable with the results reported in [16,24,37] although our results are obtained at relatively low discharge voltage of -1.7 kV compared with voltages of more than 10 kV. Gershman et. al. obtained a 5.8 log CFU/mL reduction of E. coli using a flexible DBD device similar with ours at a discharge voltage of 3 kV. The discharge power reported in was 0.5 W/cm² and the electrode surface area was 2 cm². The authors used bacteria inoculated and dried glass coverslips to measure the surviving bacterial load and 20 μL of the bacterial culture were placed onto each coverslip. Considering that for the inactivation of 1 mL of bacterial liquid the surface area of electrode will increase 50 times thus the calculated D value in Joule is 1.55 kJ/mL. In our experiments the bacteria to be inactivated was placed 2 mm from the microplasma electrode while the results from Gershman et. al. [19], were obtained with the bacteria inoculated glass coverslip directly in contact with the discharge. Microplasma reported in this paper was generated at only -1.7 kV with also a flexible electrode thus making it a safer and economical alternative than other types of plasma for biological decontamination.

The colony forming units of P. aeruginosa were difficult to count thus the calculations of D values were not carried out. Considering the higher initial concentration of CFU/mL and an inhibition of 100% on an area with 45 mm diameter, after the microplasma treatment of P. aeruginosa at -1.7 kV, the D values in Joule would be lower than what was calculated for E. coli and S. aureus.

5. Conclusions

The biological decontamination of Escherichia coli, Pseudomonas aeruginosa, and Staphylococcus aureus was carried out by microplasma energized at -1.5 kV and -1.7 kV negative pulse voltage. Inhibition zones with 100% inhibition efficiency with at least 45 mm diameter were measured on the agar plates for all bacterial strains after microplasma treatment at -1.7 kV discharge voltage. The inhibition efficiency increased with the increase of discharge voltage. The calculated D values considering the inhibition zones below the electrode were for E. coli 5.4 kJ/mL at -1.7 kV. No inhibition was observed after the microplasma treatment of already grown bacteria, indicating that the system is more suited as a prophylactic measure, rather than a complete decontamination of a surface. The process of bacteria inactivation is due to action of ROS and RON. The results obtained recommend microplasma as a safer and more economical alternative to classical sterilization technologies and also other plasma technologies due the low discharge voltage and reduced size that furthermore can translate in the portability of the microplasma system.