Interferometry of plasma bursts in helium atmospheric-pressure plasma jets Free

Rapid progress has been achieved in our understanding of the physical phenomena related to atmospheric-pressure plasma jets (APPJs) even though these types of discharges have been investigated for only two decades.^1,2 A number of plasma sources have been developed with different geometries of dielectric barrier-coated electrodes through which a gas flows, or with the application of different voltage waveforms and frequencies, to launch an APPJ into the open atmosphere.^1,3–5 In these sources, plasma produced in a region confined by the electrodes is conveyed outward beyond the source, thereby generating the APPJ. One of the critical processes responsible for plasma propagation is Penning ionization,^6,7 which occurs when the flowing gas comprises a majority of atoms or molecules that have the high first excitation state, e.g., He (19.8 eV), with an admixture of gaseous atoms or molecules with ionization energies lower than the first excitation state, e.g., N₂ (15.58 eV). If the amount of N₂ gas constitutes a relatively high fraction of total flow, the electrons produced during discharge lose their energy in inelastic collisions with N₂ and thus fail to reach the energy level necessary for ionization. In order to sustain plasma propagation, the APPJ follows the trajectory of the He flow in the open atmosphere. Furthermore, the plasma discharge affects the properties of gas flow, including the transition from a laminar to turbulent flow.^8,9 This can have tremendous implications, especially for APPJ applications in which the goal is to target a specific region of treated samples. The objective of this study was to use interferometry as a complementary method of APPJ diagnostics to investigate the effect of plasma discharge on gas flow as an example of plasma phenomena. This effect can have a potential drastic impact on APPJ delivery in medical applications, and thus, it is still under examination by many research groups. The interference patterns were compared with the images obtained from schlieren photography. The possible thermal effects responsible for the gas flow transition were also considered using an optical emission technique.

An APPJ source in this study consisted of a fused silica tube with inner and outer diameters of 5 and 6 mm, respectively, as well as two 50-mm brass electrodes separated by 30 mm and located on the outside of the tube. A detailed description of the entire APPJ source assembly was reported previously.^10,11 In the continuous pulsing mode, direct current (DC) pulses of up to 12 kV were applied with a repetition frequency between 1.0 and 1.5 kHz to the electrode that was closer to the tube outlet; the other electrode was grounded. In the burst mode, 200 pulses with a repetition rate of 2 kHz were applied in one burst, which had a duration of 100 ms separated from the next burst by 400 ms. Ultra-high-purity He (less than 5 ppm of N₂) flowed through the tube at a rate within the range of 1–10 standard liters per minute (slm). A plasma discharge ignited in the tube and propagated into the open atmosphere, forming a so-called “plasma jet” with a length of up to 10 cm beyond the tube. This continuous plasma jet consists of fast-moving plasma plumes, which were detected using intensified charge-coupled device (ICCD) cameras.^6,12 The plumes propagate at speeds as high as 60 km/s,¹³ which is much faster than the gas flow rate (up to 10 m/s) or even the thermal speed of plasma atoms and ions. Thus, in this case, rather than fast-moving plasma species, the plume consists of a fast-moving ionization front caused by a combination of both physical events, Penning photoionization and electron acceleration toward the positively charged plasma species, which occur in the direction opposite to that of the ionization front.^6,14,15

Two experimental setups were used for schlieren photography and interferometry as presented in Figs. 1(a) and 1(b), respectively.

Figure 2 shows a composite photograph taken using schlieren photography, a technique based on the refraction of light focused by a parabolic mirror and passing through transparent media with different refractive indices^8,9,16–18—in this case, He and air. In addition to schlieren photographs, this figure consists of a long-exposure digital photograph of the APPJ launched at a flow rate of 1 slm. As mentioned above, the plasma jet follows the He flow, which tends to be pulled upstream. Moreover, a comparison of the schlieren photographs with and without plasma ignition showed differences in apparent He flow rates. In Fig. 2, the photograph of the nominal He flow rate of 1 slm with plasma ignition resembles that for higher rates without plasma ignition (videos of He flow at different flow rates with and without plasma discharge are presented in the supplementary material),²⁵ implying that ignition of the plasma increased the flow rate. One possible explanation of this effect may be gas heating; however, to achieve this increase in flow, the gas temperature must be significantly increased.

The gas temperature was extracted from the analysis of plasma emission using rotational lines of the (A-X) system of OH at 310 nm and the first negative system of N₂⁺ at 390 nm. The optical emission spectra (OES) were recorded using a HORIBA Scientific iHR550 spectrometer equipped with a Stanford Computer Optics 4 Picos ICCD camera. No significant changes were observed in the rotational spectra at the beginning and the end of the pulse or between pulses in either the continuous or burst mode, indicating that the gas temperature did not increase substantially (see supplementary material). We compared the OES spectra recorded and those generated using the LIFBASE program^19,20 and obtained the best match with a rotational temperature of 300 K. We estimated the accuracy of our method to be 10 K. In addition, simplified calculations of the power pumped into the plasma (i.e., the integral of the product of the voltage applied and the current measured on the electrode) indicated that the gas temperature is likely to increase by only a few degrees. We also varied the pulse frequency from 0.5 to 2 kHz, which resulted in undetectable changes in the He flow, even though the average power pumped into the plasma source was quadrupled. Taking all the aforementioned findings into account, we concluded that the phenomenon observed could not be explained by an increased gas flow rate due to plasma heating. Recent investigations of plasma effects on gas flow were reported for both AC and nanosecond pulse-driven plasmas,^8,9,16,18,21 offering a range of possible explanations for changes in the flow rate and type, including an increase in gas temperature and local pressure, different transport properties of gas, and different momentum transfers between ions and neutral species; however, no clear consensus has been reached concerning the underlying mechanism.^15,22,23

To further support the observation of the transition from laminar to turbulent gas flow when plasma is ignited, we used the interferometry experimental setup presented in Fig. 1(b). If the composition of the gas media differs between the two branches, then the optical path lengths, which govern the interference of light, also vary. Therefore, it is possible to determine the He fraction in the air using interferometry, as the interference pattern is determined by the refractivity of the gas.²⁴ We used this approach here to study gas fluctuations of APPJs and compared the results with those from schlieren photography. Comparative images of the horizontal and vertical APPJs for 1 slm He flow are shown in Figs. 3 and 4, respectively. Although both schlieren photography and interference images showed similar trends in He flow, the inner structure of the flow and the He fraction gradient are visible only in the interference images.

In addition to qualitative analysis of the He flow, we also calculated the actual fraction of He in air using the Lorentz–Lorenz equation,²⁴ by applying the corresponding refractive indices of gases and taking into account the fact that the optical path is actually integrated across cylindrically symmetrical beams of He (i.e., Abel inversion of the calculated phase must be applied)

where $ϕi$ represents the fractions of helium $(i=1)$ and air $(i=2)$⁠, $ni$ is the refractive index of He $(i=1)$ and air $(i=2)$⁠, and $n12$ is the refractive index of the mixture. Figure 5 shows the He fraction for a He flow of 1 slm for the vertical position of the APPJ source (without plasma discharge), as calculated from the interference image in Fig. 4.

To investigate the speed of turbulence propagation in the He flow of 1 slm caused by plasma ignition, we used the burst mode. As shown in Fig. 6, the flow gradually became turbulent during the first 100 ms after plasma ignition. Then, the turbulence was reduced, resuming a laminar flow within 100 ms after plasma was quenched. At a flow rate of 1 slm, the speed of He through the tube was approximately 1 m/s; thus, the time required for He to travel from the region between the electrodes, where the plasma was ignited approximately 10 cm from the tube outlet, is similar to the time required to switch from laminar to turbulent flow.

In summary, in addition to the schlieren photography, we used an interferometry technique to characterize the flow of He in the open atmosphere and investigated gas fluctuation in the APPJ. Both techniques showed that although the APPJ follows the He flow strictly, the flow itself is affected by the plasma discharge and switches from laminar to turbulent movement; however, thermal heating can be excluded as the primary reason for the transition because the OES measurements indicated no detectable changes. Moreover, the interferometry of the plasma bursts showed that the perturbation was not instantaneous but rather propagated at a speed comparable to that of the flow. Although these results indicate that the plasma discharge affects the He flow, an exact explanation of the flow transition requires further investigation.

This material is based upon the work supported by the U.S. Department of Energy Office of Science, Office of Basic Energy Sciences under Award No. DE-FC02-04ER15533. The contribution number from the Notre Dame Radiation Laboratory is NDRL 5125.