A new plasma source design that merges the main characteristics of capacitive dielectric barrier discharge (DBD) and cold atmospheric plasma jet (CAPJ) is discussed. The DBD system contains a flexible, porous matrix consisting of silica aerogel, which is comprised between two biased electrodes. The helium flow supply subjected to a sinusoidal voltage of around 5 kV in amplitude and 15 kHz in frequency provides a set of plasma jets that propagates more than 1 cm beyond the active DBD region. The studied plasma multi-jet system consists of an array of three aligned jets that flow in the laminar regime, and it is intended for treating the surfaces of 3D objects and large areas. CAPJ performance is discussed as a hypothetical morphing source in flat and bent configurations. Electrical characterization and optical emission spectroscopy diagnostics have provided current–voltage waveforms and the composition of the CAPJ through the aerogel layer, respectively. This novel source is promising for biomedical applications that require full adaptation of plasma parameters to delicate samples, such as wound healing and treatment of surgical margins in plasma-based cancer surgery.

A major challenge in plasma medicine applications consists in developing new plasma sources adaptable to soft surfaces that exhibit non-trivial topologies, such as asperities, sharp edges, cavities, and other places of complex accessibility.1,2 Two main approaches in plasma sources for biomedicine have been considered so far: cold atmospheric plasma jet source with all integrated components (CAPJ concept) and floating-electrode dielectric barrier discharge (DBD) plasma source (FE-DBD concept).3–6 CAPJ devices provide a well-defined plasma effluent with tunable chemistry and are very convenient for local plasma treatment since they are restricted to small areas (∼0.01 to 1 cm2).4,7–10 Such a feature is disadvantageous when homogeneous irradiation of large areas is needed. Therefore, several CAPJ arrays and other sophisticated architectures have been developed to increase the surface area and homogeneity of the interacting beam, especially aiming at decontamination, sterilization, and disinfection applications. The reported prototypes involved the fabrication of plasma jet arrays11–15 and the design of flexible DBD sources in the case of conformal treatment of the substrate.16–18 However, near-surface or close contact scanning routines are usually required to improve plasma uniformity over extended areas. For the treatment of large areas, it is thus convenient to design a thin, flexible, and adaptable plasma source that combines the advantages of the controllable parameters of multi-jet plasma and the flexibility of FE-DBD devices.

In the present article, we report on the characterization of a flexible, extended CAPJ source able to mimic surfaces with non-flat topography for plasma treatments in open air. The properties of silica aerogel make it a suitable material for its performance as a CAPJ dielectric matrix.19 Such a device concept is new and necessary for the above reasons, and it constitutes the first step toward the fabrication of a multi-jet CAP device with morphing capabilities, i.e., able to adapt to sample surfaces with arbitrary topologies. Electrical measurements of the flexible source have provided current–voltage characteristics and the power consumed by the plasma multi-jet. On the other hand, the gas composition is explored by optical emission spectroscopy (OES). A multi-jet flow regime has been examined at different helium rates, whereas the source performance is assessed in concave and convex bending configurations.

The basic elements of the electrical unit powering the multi-jet are described below. A similar circuit was used in a previous study.20 Briefly, the constant voltage from an SPS-606 Instek DC power supply is modulated by means of the AC signal from a 33220A Function/Arbitrary Waveform Generator from Agilent [Fig. 1(a)]. The frequency range between 14 and 16 kHz yielded stable CAPJs. The resulting waveform is then transformed into a high-voltage sinusoidal signal of around 5 kV in amplitude, which is supplied to the CAPJ electrodes. Discharge electrical parameters of the free jets were measured with a Murata current monitor Model 56050C and a P6015A voltage probe from Tektronix, both connected to a TPS 2024B Four Channel Digital Storage Oscilloscope from Tektronix. The sample holder containing the aerogel matrix consists of two planar polymer spacers of 5 mm thickness each, fabricated via 3D printing [Fig. 1(a)]. The inlet gas tube was directly connected into a circular aperture with an inner diameter of 8 mm at the side of the biased electrode [Fig. 1(b)]. The outlet nozzle, which has a diameter of 10 mm, is surrounded by a circular electrode connected to ground [Fig. 1(c)]. The two electrodes, axially separated by about 2 cm, were covered with Kapton tape. Helium (purity: 99.995%) was supplied at a total flow rate of between 1 and 4 slm.

FIG. 1.

(a) Cross section sketch of the flexible plasma source together with a diagram of the electrical powering circuit. In downstream mode, the upper electrode is biased with a high AC voltage and the lower electrode is grounded. (b) Rear view (biased electrode) with the gas inlet and (c) front view (grounded electrode) of the setup with three CAPJs (3 slm He, 10.5 kV peak to peak, 15 kHz). The nozzle is a circular aperture of 10 mm in diameter. The OES probe is located on the right and the ground contact on the left.

FIG. 1.

(a) Cross section sketch of the flexible plasma source together with a diagram of the electrical powering circuit. In downstream mode, the upper electrode is biased with a high AC voltage and the lower electrode is grounded. (b) Rear view (biased electrode) with the gas inlet and (c) front view (grounded electrode) of the setup with three CAPJs (3 slm He, 10.5 kV peak to peak, 15 kHz). The nozzle is a circular aperture of 10 mm in diameter. The OES probe is located on the right and the ground contact on the left.

Close modal

Figure 1(c) shows the CAP multi-jet system with three active, free jets in upstream mode. The length of the jet plumes is of the order of 1 cm. The slight divergence of the jet columns is basically attributed to the electrostatic interaction between the charged streamers.21 This issue is discussed in Sec. III. Gas flow expansion from the nozzle constitutes another possible source of jet divergence. In particular, the source is fully operational in all flat, concave, and convex states. In this study, we focus on an array of three aligned plasma jets.

A flexible and porous silica layer from Kudosale (4 mm thick with an accuracy of 0.5 mm) [Figs. 2(a) and 2(b)], sandwiched between two planar-parallel spacers with electrodes on the outer surfaces, is connected to a He gas supply system to provide multiple plasma jets with sufficient flow rate and energy [Fig. 1(a)]. Silica aerogel was selected as the dielectric barrier material because of its flexibility, high electrical resistivity, and refractory material properties. The layered structure of an aerogel sample is discernible by optical microscopy. The cross section image in Fig. 2(b) shows elementary layers of aerogel piled up in stacks of 4 mm of total thickness. These elementary layers are weakly adhered, which facilitates their detachment to obtain thinner samples. The aerogel CAPJ matrix exhibits a mesoporous structure reinforced with microscopic fiberglass, as verified via scanning electron microscopy (SEM). The aerogel porosity is estimated to be around 90% via microbalance measurements and comparing the measured density with the density of ceramic silica. However, this material exhibits reduced gas permeability because of the very small size of its pores, on the order of a nanometer.22 Indeed, ceramic barriers with at least sub-millimeter pores were necessary for adequate permeation of the gas flow in the formation of CAPJ.23 In the present study, the increase in permeability for multiple jet production was enabled by perforating a pattern of pinholes (≈1 mm in diameter) through the aerogel barrier [Fig. 2(a)]. CAPJ systems comprising from one to five jets have been tested. Here, we have focused on a three-jet linear arrangement. The holes were produced with an approximate separation of 2 mm. A DBD plasma can be then generated between the electrodes, before the level of pinholes, and projected as a CAP multi-jet along the He flow stream.

FIG. 2.

(a) Top view image of the multi-jet nozzle with perforated silica aerogel (hole diameter 1 mm). The nozzle is a circular aperture of 10 mm in diameter. (b) Cross section image of the stacks of silica layers of the aerogel (taken with an optical microscope). (c) and (d) SEM top view images of the silica aerogel sample, showing the zone scanned by EDS (area bounded by a dashed square).

FIG. 2.

(a) Top view image of the multi-jet nozzle with perforated silica aerogel (hole diameter 1 mm). The nozzle is a circular aperture of 10 mm in diameter. (b) Cross section image of the stacks of silica layers of the aerogel (taken with an optical microscope). (c) and (d) SEM top view images of the silica aerogel sample, showing the zone scanned by EDS (area bounded by a dashed square).

Close modal

Morphological characterization of the aerogel was performed by means of a Tescan GAIA FEG SEM with an accelerating voltage of 5 kV. Previously, a conductive thin film of Au was sputtered on top of the aerogel surface. The top view SEM micrographs in Figs. 2(c) and 2(d) show a landscape full of fibers covering a substrate with very small pores, which explains the very low gas permeability of the material. On the other hand, the highly cross-linked fiberglass network confers mechanical stability to the aerogel structure. The chemical composition on the region marked in Fig. 2(d) was assessed by energy dispersive x-ray spectroscopy (EDS). As expected for silica, the main elements are Si and O with the approximated stoichiometry 1:2 (32 at. % Si, 57 at. % O). Carbon impurities were also detected (≈10 at. % C).

Figure 3 shows images of the three-jet system operating at flow rates of 1–4 slm at 15 kHz and 12 kVpp (peak-to-peak voltage). The central jet systematically displays a lower luminosity attributed essentially to a weakening of the intensity of the electric field in the axis of the nozzle. In fact, the equipotential surfaces created by the charged electrodes are not perfectly parallel at the level of the aerogel. Uniformity issues associated with the electric field distribution have been addressed by Finite Element Method Magnetics (FEMM) simulation in the supplementary material. The average jet length increases monotonically with the He rate. The observed jet divergence is associated with electrostatic repulsion between adjacent plumes, and it is reduced at higher flow rates mainly because of the increase of momentum transfer between plasma species along the jet direction.21 It should be noted that the CAPJ system exhibited fully laminar flow, as reported in systems operating at similar flow rates and orifice diameters.24,25 At higher flow rates, the plasma jet performance will likely degrade because of the expected transition from a laminar pattern to a turbulent regime.

FIG. 3.

Images of the CAP multi-jet system operating at 15 kHz, 12 kVpp, at He flow rates of 1–4 slm are shown. The circular nozzle, surrounded by the grounded electrode, has a diameter of 10 mm.

FIG. 3.

Images of the CAP multi-jet system operating at 15 kHz, 12 kVpp, at He flow rates of 1–4 slm are shown. The circular nozzle, surrounded by the grounded electrode, has a diameter of 10 mm.

Close modal

Jet lengths of between 4 and 16 mm were measured from the projected jet images in Fig. 3 by correcting the parallax angle of observation. The evolution of the jet length with respect to the gas velocity per hole is represented in Fig. 4. Values of the linear velocity through each hole, v, are obtained from the expression for the total He rate, ϕ = 3 × v × π × (D/2)2, where D is the hole diameter (≈1 mm). The average He rate per hole is approximated to one third of the total flow rate, although the middle hole is generating a slightly longer jet at 3 and 4 slm (Fig. 3). A finer approach to partial gas rates requires the calculation of gas flow velocity profiles by numerical fluid dynamics simulations, which is out of the scope of this article. Linear fit analysis on the data from jet length, L, and velocity, v, with the model L = v × τ provides a characteristic time of τ = 0.6 ms. This parameter can be understood as the approximated lifetime of the plasma species in the DBD afterglow, and the fitted value of 0.6 ms is consistent with the timescale of plasma chemical reactions (neutrals).26 

FIG. 4.

Plot of the length of middle jet, L, vs average linear velocity, v, at different He flow rates. The dashed line is calculated from a linear fit of the experimental data using the indicated model. The fitted slope defines a characteristic time, τ = 0.6 ms. The upper scale shows the range of the associated Reynolds number, which is proportional to the velocity values.

FIG. 4.

Plot of the length of middle jet, L, vs average linear velocity, v, at different He flow rates. The dashed line is calculated from a linear fit of the experimental data using the indicated model. The fitted slope defines a characteristic time, τ = 0.6 ms. The upper scale shows the range of the associated Reynolds number, which is proportional to the velocity values.

Close modal

The multi-jet flow dynamics has been approached by estimating the Reynolds number associated with a single jet (middle jet), Re = ρ ×v × D/μ, where ρ is the He gas density (0.179 kg/m3), v is the outlet velocity calculated above, D is the characteristic jet size (1 mm in diameter), and μ is the He viscosity constant (1.96 × 10−5 Pa s). The linear correlation between the jet length and Reynolds number (see Fig. 4), together with the relatively low Re values reached here (Re < 300), confirms the hypothesis that CAPJ develops within the laminar regime. A completely turbulent pattern would be expected at much higher gas velocities (Re > 650).25 

Figure 5 shows the voltage and current waveforms corresponding to the three-jet system ignited at 15 kHz with variations in the flow rate and maximum discharge voltage. The acquired oscilloscopic data have been averaged to better visualize the basic features of the waveforms. The discharge current is obtained by subtracting the displacement current measured without plasma (He flow is cut) from the total current measured when the plasma multi-jet is switched on (He flow is established) at a constant voltage amplitude. The consumed power is calculated in each case by averaging the product of voltage and current curves over one AC period. In general, the discharge current profile shows an asymmetric AC component together with current peaks periodically occurring at both positive and negative half-cycles. The current peak in phase with the rising slope is significantly stronger than the peak occurring during the voltage reversal. The observed asymmetries are related to the different anode and cathode geometries.

FIG. 5.

Typical waveforms of the discharge current and voltage corresponding to a multi-jet nozzle with three aligned apertures operating at 15 kHz with He flow rates of 1, 2, 3, and 4 slm.

FIG. 5.

Typical waveforms of the discharge current and voltage corresponding to a multi-jet nozzle with three aligned apertures operating at 15 kHz with He flow rates of 1, 2, 3, and 4 slm.

Close modal

The plasma plume is the load of the transformer, and the conductivity of the plasma determines the discharge current when the discharge voltage is a constant. Apparently, a higher electron density leads to a higher conductivity. Therefore, when the He flow rate is rising from 1 to 2 slm, the ionization rate increases. This leads to a higher current as shown in the 1 and 2 slm cases in Fig. 5. When the flow rate reached 3 slm, the streamer propagation effect was more significant, leading to a peak of electron density boost during the propagation. This means a boost of conductivity and, thus, a peak of discharge current during the propagation, which usually lasts about 1–4 µs.27 When the flow rate reached 4 slm, the gas flow was approaching to a laminar–turbulence transition state. Therefore, a small amount of extra He–air mixing occurs, which causes random losses of electron density, such as the dissociative attachment e + O2 => O + O, and other sinks of energy on the extra degrees of freedom of diatomic molecules. This generates the random loss of the current peaks as shown in the case of 4 slm, while the peaks constantly appear in the 3 slm case. On the other hand, as a capacitively coupled plasma, a low electron density means that the scenario is closer to an ideal capacitor. The current of an ideal capacitor should have a 90°-phase ahead of the voltage. As mentioned earlier, the plasma conductivity scales with the increase in electron density. In this case, the plasma resembles a resistor, which has an in-phase current. Note that the electrodes are also connected to the coil of a transformer. Thus, overall, the circuit is becoming an RL (resistive-inductive) circuit from an RLC (resistive-inductive-capacitive) one when the electron density is increasing because of the mixing of a higher He flux in air. This mechanism is, thus, causing a current phase shift from the left to the right along the time axis.

Although similar structures are observed in all waveforms, some differences are identified by varying the voltage and He rate. In this article, the influences of flow rate and voltage amplitude on the current–voltage waveforms are studied:

  • He flow rate: Variations from 1 to 4 slm at peak-to-peak voltage values of 12 kVpp have been considered. The current reached 2 mA (1 mA at 1 slm) during the positive half-cycle of the oscillation, whereas −4 mA (−2 mA at 1 slm) was reached during the negative half-cycle. The discharge current in the CAPJ maintained at 1 slm or less, and 12 kVpp, is limited by the low He rate, so that one can speak of a flow-limited regime. The coincidence in I–V maxima and minima measured at 1 slm suggests a resistor-like behavior typical of purely dissipative media. The consumed power is ∼3 W at 1 slm. The discharges operated at rates greater than 1 slm do not undergo such a flow limitation, as observed in Fig. 5. The consumed power is around 5 W within the 2–4 slm interval. Additionally, reactant components become significant in the plasma multi-jet system by the formation of a phase shift in the I–V waveforms. The equivalent circuit of the multi-jet platform will be modeled in a future article.

  • Voltage amplitude: In the 4 slm case, the initial peak-to-peak voltage of 12 kVpp was set to a lower value of 10.3 kVpp to assess the dependence of current–voltage waveforms on the supplied power. As a result, the positive current still reaches 2 mA, but the negative part only reaches −3 mA instead of the −4 mA amplitude registered at 12 kVpp. The average power is around 3 W, similar to the previous case of the jet with a limited flow rate of 1 slm. Therefore, the three-jet system operates in the power-limited regime at flow rates greater than 2 slm He. As observed in the He rate variation study, the phase shift in the I–V waveforms evidences the participation of reactant elements in the generated plasma.

See the supplementary material to check the waveforms corresponding to the range 14–16 kHz. In the present study, discharges at frequencies below 14 kHz were not explored because they are near the lower frequency tolerances of both the transformer and current monitor. The CAPJ aerogel matrix remained at room temperature during operation as measured by a non-contact infrared thermometer. The non-thermal nature of this plasma source allows its application for treatment of delicate and temperature-sensitive materials.

The optical characterization of the three-jet plume system was performed by OES measurements, which provided the optical emission spectrum displayed in Fig. 6. The UV-visible spectrum of the He plasma was obtained by means of a StellarNet spectrometer operating in the spectral range of 191.0–851.5 nm with a spectral resolution of 0.5 nm. The OES probe was aligned in front of the middle jet, as shown in Figs. 2(a) and 6 (inset), so that the light emitted by the three jets was captured. The spectrum was collected at an axial distance of 7 mm (middle length of the plume) and a radial distance of 5 mm with an integration time of 2 s. It is important to quantify the weight of each plasma column in the measured signal for real-time control of the multi-jet. In this specific geometry, the individual contribution of each lateral jet to the total intensity measured by OES is around 90% of the contribution from the central plasma column (see details in the  Appendix).

FIG. 6.

Optical emission spectrum of a three-jet helium plasma at half length (distances: 7 mm axial, 5 mm radial) through the aerogel matrix. The optically active plasma species are identified. The jet was operated at 15 kHz and 12 kVpp with a He flow rate of 4 slm. Inset: lateral view of the analyzed three-jet system with the OES probe on the right side.

FIG. 6.

Optical emission spectrum of a three-jet helium plasma at half length (distances: 7 mm axial, 5 mm radial) through the aerogel matrix. The optically active plasma species are identified. The jet was operated at 15 kHz and 12 kVpp with a He flow rate of 4 slm. Inset: lateral view of the analyzed three-jet system with the OES probe on the right side.

Close modal

A strong N2 emission is observed in the OES spectrum (Fig. 6), as reported for He CAPJ afterglow elsewhere.28 The emission lines have been attributed to the following transitions: 315.93 nm N2 (C3Πu → B3Πg, ν′ = 1, ν″ = 0), which, if blue-shifted, may overlap with the less intense but close 308.9 nm OH (A2Σ+X2Π, ν′ = ν″ = 0); 337.13 nm N2 (C3Πu → B3Πg, ν′ = ν″ = 0); 353.67 nm N2 (C3Πu → B3Πg, ν′ = 1, ν″ = 2); 357.69 nm N2 (C3Πu → B3Πg, ν′ = 0, ν″ = 1); 375.54 nm N2 (C3Πu → B3Πg, ν′ = 1, ν″ = 3); 380.49 nm N2 (C3Πu → B3Πg, ν′ = 0, ν″ = 2); 399.84 nm N2 (C3Πu → B3Πg, ν′ = 1, ν″ = 4); 405.94 nm N2 (C3Πu → B3Πg, ν′ = 0, ν″ = 3); 391.4 nm N2+ (B2Σu+ → X2Σg+, ν′ = 0, ν″ = 0); 427.81 nm N2+ (B2Σu+ → X2Σg+, ν′ = 0, ν″ = 1); 667.8 nm He I (31D → 21P); and 706.52 nm He I (33S → 23P). The typical contribution of CAPJ at 777.41 nm O (35P → 35S) is below the detection threshold.28,29

A spatially resolved optical diagnostics along each single jet and the determination of plasma parameters, such as the average temperature and density of electrons by Rayleigh scattering,30 are planned in future work. For gas temperature evaluation, the use of a spectrometer with higher spectral resolution (much better than 0.5 nm employed here) is needed, for instance, to estimate the Doppler broadening from line shape analysis.31 

The basic morphing capabilities of the CAPJ system have been tested by shaping the aerogel outlet, which constitutes the nozzle region, with concave and convex geometries (Fig. 7). To this end, hard silica spheres of different diameters have been used to shape conveniently a 4 mm-thick aerogel layer with three aligned apertures. After removing the sphere, the flexible dielectric resulted in a static bent position toward the rear part (concave) or front part (convex) of the sample holder, showing respective curvature radii of roughly 5 and 7 mm. Figure 7(a) shows that the three plasma jets emerged from the convex nozzle with a significant divergence, even stronger than in the flat nozzle jet scenario. Here, the natural effect on jet bending due to nozzle curvature is added to the electrostatic repulsion mentioned above between adjacent plasma columns and the hydrodynamic effect caused by gas expansion. In contrast, the CAP multi-jet system tended to converge at the nozzle axis in the concave configuration, as shown in Fig. 7(b). In fact, an important step is to control the directionality of all the jets to allow uniform treatments on substrates. The non-uniformity of the jets could be corrected by improving the distribution of the gas inlet and by using segmented electrodes polarized at different voltages, each voltage being adapted to control the directionality of the multi-jet in any nozzle configuration. Movies of the multi-jet discharges are available in the supplementary material.

FIG. 7.

Images of a CAP multi-jet generated in two states of nozzle bending: (a) convex bending (7 mm in curvature radius) and (b) concave bending (5 mm in curvature radius). The multi-jet was operated at 15 kHz and 9.5 kVpp-1 W (convex) and 12 kVpp-5 W (concave), with a He flow rate of 4 slm. The lower voltage amplitude used in the convex state is intended to avoid short-circuits between the lateral jets and the grounded electrode.

FIG. 7.

Images of a CAP multi-jet generated in two states of nozzle bending: (a) convex bending (7 mm in curvature radius) and (b) concave bending (5 mm in curvature radius). The multi-jet was operated at 15 kHz and 9.5 kVpp-1 W (convex) and 12 kVpp-5 W (concave), with a He flow rate of 4 slm. The lower voltage amplitude used in the convex state is intended to avoid short-circuits between the lateral jets and the grounded electrode.

Close modal

The changes in aerogel concavity did not significantly affect the current and voltage waveforms (not shown here) compared to the original curves shown in Fig. 5. Small variations in the geometry of the DBD system have no significant impact on the electrical parameters, a fact that proves the stability of the generated multi-jet plasma. Figure 8 shows that OES profiles are roughly similar in the different bending states. The OES probe is in the same position as in the flat aerogel measurements. Similarly, the light from each single jet contributes to the measured OES profile. However, in the case of convex bending, significant enhancements in N2+ and He emissions compared with flat configuration are detected. In addition, the helium and atomic oxygen species, which respectively cause the 587.6 nm He I (33D → 23P) and 777.41 nm O (35P → 35S) transitions, are now observable in the convex state. These substantial increases in the intensities of the N2+, He, and O lines are attributed to stronger dissociation and ionization events within the CAPJ. Indeed, the convex modification of the aerogel is expected to bring the active DBD region, defined as the source of the plasma multi-jet, closer to the position of the OES probe.

FIG. 8.

Optical emission spectra of the three-jet helium plasma through the aerogel matrix in concave and convex states. The OES probe is placed in the same position (distances: 7 mm axial, 5 mm radial) as for the analysis in flat configuration. The optically active plasma species are identified. The jet was operated at 15 kHz with bias at 12 kVpp (5 W) in the concave state and bias at 9.5 kVpp (1 W) in the convex state with a He flow rate of 4 slm.

FIG. 8.

Optical emission spectra of the three-jet helium plasma through the aerogel matrix in concave and convex states. The OES probe is placed in the same position (distances: 7 mm axial, 5 mm radial) as for the analysis in flat configuration. The optically active plasma species are identified. The jet was operated at 15 kHz with bias at 12 kVpp (5 W) in the concave state and bias at 9.5 kVpp (1 W) in the convex state with a He flow rate of 4 slm.

Close modal

Here, we have reported the basic characteristics of a cold plasma multi-jet generated in a flexible matrix. The platform is operational with a stable plasma, which exhibits reproducible characteristics. However, the system mentioned corresponds to a small-scale system, which can constitute the elementary cell, the building block of a larger device. Indeed, the CAPJ can be extended from a 1D array of jets to a 2D matrix of jets, and the resulting 2D distribution of jets can be upscaled into a larger morphing source formed by a cluster of many identical cells. The total flow rate diverted to the integrated source must be proportional to the number of jet outlets. Individual or collective control of several elementary cells will be enabled by using segmented electrodes. Hence, the electrode system can be controlled using a set of independent power supplies.

Alternative jet arrangements using an aerogel matrix have already been tested. Images of jet distributions with different hole patterns (up to five holes) perforated on the aerogel layer are presented in a preliminary study.32 Therefore, ignition of a plasma multi-jet from a 2D matrix source is feasible, and a detailed approach to the performance of 2D jet distributions emanating from flexible sources will be discussed in another paper.

One can wonder about the maximum bending tolerance by such an aerogel-based CAPJ. This point will be explored by constructing an upgraded sample holder with an adjustable bending angle, so that CAPJ morphing performance can be examined from small tilt angles to maximum compliance. The critical bending angles and sharpness of the folded surface will be obviously limited by the size of the plasma device and the thickness of the aerogel matrix. Moreover, the ratio between the gas velocity and discharge voltage should be adjusted for each specific plasma treatment conditions. For example, flow-limited operation (shorter jets) will be preferred for near-surface treatments, while power-limited operation (longer jets) should be suitable for treating samples located at a greater distance. The operation mode should be selected according to the gap between the sample and the plasma device.

In summary, the dielectric barrier performance of a flexible aerogel layer in a CAP multi-jet source has been demonstrated, thereby suggesting the construction of a morphing prototype based on this working principle. A study on the CAPJ directionality control (collimation) by exploring alternative electrode configurations and gas inlet distributions is envisaged to address non-uniformity issues in the substrate treatments. In a next step, focused on unraveling the plasma chemistry associated with large-area CAPJ operation, we will learn the main reaction kinetics and chemical agents required to design smart plasma treatments for biomedical applications, such as wound healing and treatment of surgical margins in cancer therapy. In addition, the basic interactions at the CAPJ–target interface must also be considered to provide a complete picture of the physicochemical mechanisms involved in such treatments.33 The scope of applications in plasma medicine will be broadened by upgrading the conventional, “rigid” plasma sources into flexible platforms adaptable to complex topographies.

See the supplementary material for (1) the distribution of instantaneous electric field strength at the sample holder simulated by FEMM, (2) current and voltage waveforms in discharges developed in the frequency interval 14–16 kHz, and (3) movies showing the multi-jet configuration operating in concave and convex modes.

This work was supported by the National Science Foundation through Award No. 1919019. The authors acknowledge the assistance by Dr. Jiancun Rao from AIM Laboratory at the Maryland Nano Center.

The data that support the findings of this study are available from the corresponding author upon reasonable request.

This appendix is intended to estimate the individual contribution from each jet to the optical spectra measurements. Figure 9 sketches the relative positions of the three formed jets, separated with a distance d = 2 mm, and the OES probe, which is located at a distance R = 5 mm from the middle jet. Assuming that each plasma column emits the same light intensity, I0, and that the intensity decays with distance as 1/r, it is straightforward to derive the following expressions for the individual contributions in the OES probe by the middle jet, Imid, and each lateral jet, Ilat,

Imid=I0×r0/R,
(A1)
Ilat=I0×r0/(d2+R2)1/2,
(A2)

where the constant r0 has length units. After dividing Eq. (A2) by Eq. (A1), and substituting the numerical values of R and d, it is found that Ilat = 0.93 × Imid. Hence, the total intensity at the OES probe is defined as Itot = Imid + 2 × Ilat, which is equivalent to Itot = 2.86 × Imid. In conclusion, the measured spectra collect the partial contributions Imid/Itot = 0.35 for the central jet and Ilat/Itot = 0.325 for each lateral jet. More accurate estimations require considering differences in jet intensities and possible coupling effects between the jets.

FIG. 9.

Schematic top view of the plasma multi-jet generated at the aerogel matrix (10 mm in diameter) together with the OES probe. The indicated distances are R = 5 mm and d = 2 mm.

FIG. 9.

Schematic top view of the plasma multi-jet generated at the aerogel matrix (10 mm in diameter) together with the OES probe. The indicated distances are R = 5 mm and d = 2 mm.

Close modal
1.
D.
Yan
,
L.
Lin
,
E.
Gjika
,
C.
Corbella
,
A.
Malyavko
,
I. I.
Beilis
,
J. H.
Sherman
, and
M.
Keidar
, “
Current understanding of mechanisms in plasma cancer therapy and recent advances in technology
,” in
Plasma Cancer Therapy
, edited by
M.
Keidar
(
Springer
,
Heidelberg
,
2020
), pp.
271
287
.
2.
G.
Busco
,
E.
Robert
,
N.
Chettouh-Hammas
,
J.-M.
Pouvesle
, and
C.
Grillon
, “
The emerging potential of cold atmospheric plasma in skin biology
,”
Free Radicals Biol. Med.
161
,
290
304
(
2020
).
3.
G.
Fridman
,
G.
Friedman
,
A.
Gutsol
,
A. B.
Shekhter
,
V. N.
Vasilets
, and
A.
Fridman
, “
Applied plasma medicine
,”
Plasma Processes Polym.
5
,
503
533
(
2008
).
4.
J.
Golda
,
J.
Held
,
B.
Redeker
,
M.
Konkowski
,
P.
Beijer
,
A.
Sobota
,
G.
Kroesen
,
N. S. J.
Braithwaite
,
S.
Reuter
,
M. M.
Turner
,
T.
Gans
,
D.
O’Connell
, and
V.
Schulz-von der Gathen
, “
Concepts and characteristics of the ‘COST reference microplasma jet
,’”
J. Phys. D: Appl. Phys.
49
,
084003
(
2016
).
5.
K.-D.
Weltmann
and
T.
Von Woedtke
, “
Plasma medicine—Current state of research and medical application
,”
Plasma Phys. Controlled Fusion
59
,
014031
(
2017
).
6.
M.
Laroussi
, “
Plasma medicine: A brief introduction
,”
Plasma
1
,
47
60
(
2018
).
7.
T.
Kawasaki
,
W.
Eto
,
M.
Hamada
,
Y.
Wakabayashi
,
Y.
Abe
, and
K.
Kihara
, “
Detection of reactive oxygen species supplied into the water bottom by atmospheric non-thermal plasma jet using iodine-starch reaction
,”
Jpn. J. Appl. Phys., Part 1
54
,
086201
(
2015
).
8.
O.
Volotskova
,
L.
Dubrovsky
,
M.
Keidar
, and
M.
Bukrinsky
, “
Cold atmospheric plasma inhibits HIV-1 replication in macrophages by targeting both the virus and the cells
,”
PLoS One
11
,
e0165322
(
2016
).
9.
A. V.
Omran
,
G.
Busco
,
L.
Ridou
,
S.
Dozias
,
C.
Grillon
,
J.-M.
Pouvesle
, and
E.
Robert
, “
Cold atmospheric single plasma jet for RONS delivery on large biological surfaces
,”
Plasma Sources Sci. Technol.
29
,
105002
(
2020
).
10.
L.
Lin
and
M.
Keidar
, “
A map of control for cold atmospheric plasma jets: From physical mechanisms to optimizations
,”
Appl. Phys. Rev.
8
,
011306
(
2021
).
11.
Q.
Nie
,
Z.
Cao
,
C. S.
Ren
,
D. Z.
Wang
, and
M. G.
Kong
, “
A two-dimensional cold atmospheric plasma jet array for uniform treatment of large-area surfaces for plasma medicine
,”
New J. Phys.
11
,
115015
(
2009
).
12.
P. P.
Sun
,
E. M.
Araud
,
C.
Huang
,
Y.
Shen
,
G. L.
Monroy
,
S.
Zhong
,
Z.
Tong
,
S. A.
Boppart
,
J. G.
Eden
, and
T. H.
Nguyen
, “
Disintegration of simulated drinking water biofilms with arrays of microchannel plasma jets
,”
npj Biofilms Microbiomes
4
,
24
(
2018
).
13.
T.
Maho
,
X.
Damany
,
S.
Dozias
,
J.-M.
Pouvesle
, and
E.
Robert
, “
Atmospheric pressure multijet plasma sources for cancer treatments
,”
Clin. Plasma Med.
9
(
Suppl.
),
3
4
(
2018
).
14.
S.
Bekeschus
,
P.
Favia
,
E.
Robert
, and
T.
von Woedtke
, “
White paper on plasma for medicine and hygiene: Future in plasma health sciences
,”
Plasma Processes Polym.
16
,
1800033
(
2019
).
15.
Y.
Lv
,
L.
Nie
,
J.
Duan
,
Z.
Li
, and
X.
Lu
, “
Cold atmospheric plasma jet array for transdermal drug delivery
,”
Plasma Processes Polym.
18
,
2000180
(
2021
).
16.
J.
Kim
,
K.-H.
Choi
,
Y.
Kim
,
B. J.
Park
, and
G.
Cho
, “
Wearable plasma pads for biomedical applications
,”
Appl. Sci.
7
,
1308
(
2017
).
17.
H.
Jung
,
J. A.
Seo
, and
S.
Choi
, “
Wearable atmospheric pressure plasma fabrics produced by knitting flexible wire electrodes for the decontamination of chemical warfare agents
,”
Sci. Rep.
7
,
40746
(
2017
).
18.
J.
Xie
,
Q.
Chen
,
P.
Suresh
,
S.
Roy
,
J. F.
White
, and
A. D.
Mazzeo
, “
Paper-based plasma sanitizers
,”
Proc. Natl. Acad. Sci. U. S. A.
114
,
5119
5124
(
2017
).
19.
C. J.
Brinker
and
G. W.
Scherer
,
Sol-Gel Science: The Physics and Chemistry of Sol-Gel Processing
(
Elsevier
,
Boston
,
1990
).
20.
L.
Lin
,
Y.
Lyu
,
B.
Trink
,
J.
Canady
, and
M.
Keidar
, “
Cold atmospheric helium plasma jet in humid air environment
,”
J. Appl. Phys.
125
,
153301
(
2019
).
21.
M.
Ghasemi
,
P.
Olszewski
,
J. W.
Bradley
, and
J. L.
Walsh
, “
Interaction of multiple plasma plumes in an atmospheric pressure plasma jet array
,”
J. Phys. D: Appl. Phys.
46
,
052001
(
2013
).
22.
J.
Phalippou
,
T.
Woignier
,
R.
Sempere
, and
P.
Dieudonne
, “
Highly porous aerogels of very low permeability
,”
Mater. Sci.
20
,
29
42
(
2002
).
23.
S.
Ma
,
K.
Kim
,
S.
Lee
,
S.
Moon
, and
Y.
Hong
, “
Effects of a porous dielectric in atmospheric-pressure plasma jets submerged in water
,”
Phys. Plasmas
25
,
083519
(
2018
).
24.
Q.
Li
,
X.-M.
Zhu
,
J.-T.
Li
, and
Y.-K.
Pu
, “
Role of metastable atoms in the propagation of atmospheric pressure dielectric barrier discharge jets
,”
J. Appl. Phys.
107
,
043304
(
2010
).
25.
R.
Xiong
,
Q.
Xiong
,
A. Y.
Nikiforov
,
P.
Vanraes
, and
C.
Leys
, “
Influence of helium mole fraction distribution on the properties of cold atmospheric pressure helium plasma jets
,”
J. Appl. Phys.
112
,
033305
(
2012
).
26.
S.
Park
,
W.
Choe
,
S. Y.
Moon
, and
S. J.
Yoo
, “
Electron characterization in weakly ionized collisional plasmas: From principles to techniques
,”
Adv. Phys.: X
4
,
1526114
(
2019
).
27.
A.
Shashurin
,
M. N.
Shneider
,
A.
Dogariu
,
R. B.
Miles
, and
M.
Keidar
, “
Temporary-resolved measurement of electron density in small atmospheric plasmas
,”
Appl. Phys. Lett.
96
,
171502
(
2010
).
28.
M.
Thiyagarajan
,
A.
Sarani
, and
C.
Nicula
, “
Optical emission spectroscopic diagnostics of a non-thermal atmospheric pressure helium-oxygen plasma jet for biomedical applications
,”
J. Appl. Phys.
113
,
233302
(
2013
).
29.
R. W. B.
Pearse
and
A. G.
Gaydon
,
The Identification of Molecular Spectra
(
Chapman & Hall, Ltd.
,
London
,
1941
).
30.
L.
Lin
,
Y.
Lyu
,
M. N.
Shneider
, and
M.
Keidar
, “
Average electron temperature estimation of streamer discharge in ambient air
,”
Rev. Sci. Instrum.
89
,
113502
(
2018
).
31.
U.
Fantz
, “
Basics of plasma spectroscopy
,”
Plasma Sources Sci. Technol.
15
,
S137
S147
(
2006
).
32.
C.
Corbella
,
S.
Portal
,
L.
Lin
, and
M.
Keidar
, “
Towards the fabrication of a morphing plasma source for biomedical applications
,” arXiv:2102.02937 [physics.plasm-ph] (
2021
).
33.
L.
Martinez
,
A.
Dhruv
,
L.
Lin
,
E.
Balaras
, and
M.
Keidar
, “
Interaction between a helium atmospheric plasma jet and targets and dynamics of the interface
,”
Plasma Sources Sci. Technol.
28
,
115002
(
2019
).

Supplementary Material