3.2 Spatially resolved spectroscopy – mixing of effluent Ar plasma with air The spatially resolved spectra were obtained by moving the optical fibre in a plane 2 cm in front of the discharge effluent plasma. The minimum lateral distance between two measured points was 1 mm in both directions. Measuring the points closer to each other would be of no benefit due to the angle from which the optical fibre collects the light and which corresponds to a 2 mm diameter spot. The results are plotted as a colour map of spectral feature intensity (peak height). The position [0,0] corresponds to the middle of the nozzle. The scale of intensity is non-linear in order to achieve the best visualization of results in both extremes of the intensity range. A photograph of the discharge flame flowing out of the tube (Figs. 3a and 4a) is added for easier readability and interpretation of the results. In Fig. 3, the results for the two strongest bands originating in the airborne admixtures are shown. As can be seen, the emission intensity distribution is very similar for both spectral areas – the OH radical at 310 nm (Fig. 3b) and NH at 336 nm (Fig. 3c). They have a common intensity maximum inside the flame at a vertical position 3-4 mm from the nozzle. Beyond this point (getting further from the nozzle and approaching the afterglow region) the intensity sharply drops. This trend is almost the same for OH and NH molecules. However, they differ closer to the nozzle and inside the discharge tube. Here, the OH intensity decreases more rapidly than the intensity of NH emissions at 336 nm. Fig. 4 shows the results for other two spectral areas of interest – the 1st negative system of N2+ around 391 nm (Fig. 4b) and the Hα line at 656 nm (Fig. 4c). The spatial distribution of their emissions exhibits a different tendency than OH and NH molecules. The strongest N2+ and Hα emission intensity was detected inside the tube and then steadily decreased. The results imply that the processes governing the effluent surfatron plasma jet are similar to those described for the TIA device by [22]. The N2+ molecular ion produced
Unauthenticated Download Date | 11/24/15 8:51 AM
544 Lucia Potočňáková et al.
by charge transfer collisions with argon ions plays an important role in the discharge kinetics. Dissociative recombination of this molecular ion may results in a large amount of N atoms, which then participate in various reactions leading to the formation of NH. The emission intensity evolution at 391 nm and 336 nm supports this hypothesis. Close to the nozzle, the number of N2+ ions is the highest, but as the dissociation and consequent reactions start to take place, the amount of N2+ decreases in favour of an increasing amount of NH molecules. Similarly, the formation of the NH molecule can cause continuous losses of hydrogen, evidenced by Fig. 4c. Another process responsible for the lack of hydrogen atoms farther in the effluent plasma flame could be the association reactions of hydrogen atoms with oxygen atoms, forming the OH molecules, which were found in large numbers especially in the middle of the plasma flame. The presence of OH molecules can however also be attributed to the dissociation of water traces present in the surrounding air.
3.3 Temperature profile from OES
It is rather difficult to directly determine the gas temperature in the discharge. However, thanks to the balance between the rotational and translational degrees of freedom established in atmospheric pressure plasmas, the rotational temperature, easily calculated from intensities of molecular bands, can be used instead. In this study, we used the Boltzmann plot technique [23] to determine the rotational temperature of the OH radical using the intensities of five OH lines coloured red in Fig. 5a. This particular selection of OH lines was used in many other studies [24-27]. The measurements were carried out for the same positions as relative OH intensity measurements (Fig. 3b). It can be seen from the results in Fig. 5b that the temperature inside the discharge tube is lower than the temperature in the outer region, where the amount of molecular admixture is higher. This can be caused by a gradual cascade transfer of energy from the electric field driven electrons to translations
Figure 2: Overview optical emission spectra of surfatron argon plasma (a) inside the discharge tube and (b) outside the discharge tube. For the meaning of position values please compare with Fig. 3 or 4.
Figure 3: To-scale photograph and colour maps of relative intensity of optical emissions at wavelengths of 310 nm and 336 nm.
Unauthenticated Download Date | 11/24/15 8:51 AM
Spatially resolved spectroscopy of microwave plasma jet 545
via electronically excited Ar states and vibrations and rotations of molecular admixtures. It should be noted that the actual temperature on the sample surface during plasma treatment is significantly reduced due to cooling by the compressed air [12].
3.4 Surface treatment
For the treatment of many materials, the impinging plasma temperature is a strongly limiting factor. We have previously published studies of treatment of heat-sensitive materials like wood and polyamide [13,28], when this limit was overcome by placing the sample only at the very end of the plasma flame and setting a very high flow of cooling
air, which helped to preserve the sample. However, the higher energy carried by plasma and the significantly higher amount of active species in the middle of the flame indicate the possibility of more effective treatment in this area. In this study, we have treated materials that are sufficiently heat resistant to endure treatment inside the flame without any damage – stainless steel, aluminium and silicon wafer – and compared the results for 1 s treatment at the end of the flame (vertical position 8-9 mm) and in the middle of the flame (vertical position 3-4 mm). The results are summarized in Fig. 6 and in Table 1, where the contact angles for untreated samples are shown in red, contact angles for sufficiently treated samples in green and orange colour indicates a plasma treated sample with insufficient results. As can be seen,
Figure 4: To-scale photograph and colour maps of relative intensity of optical emissions at wavelengths of 391 nm and 656 nm.
Figure 5: (a) Resolved spectrum of OH radical emissions around 309 nm wavelength. Red lines were used for temperature calculation by the Boltzmann plot technique. (b) Colour map of OH rotational temperature.
Unauthenticated Download Date | 11/24/15 8:51 AM
546 Lucia Potočňáková et al.
the treatment at the end of the flame is perfectly suitable for silicon wafer, as the contact angles achieved in both regions are the same and very low. There is some small difference in contact angles for aluminium, but compared to the contact angle of the reference sample, the difference is not relevant. However, the treatment of stainless steel was significantly improved by putting the samples deeper inside the flame, as the treatment of steel at the end of flame resulted only in minor effects. To investigate this more thoroughly, we performed more experiments using only the stainless steel samples. The results of treatment of stainless steel can be seen in Fig. 7. First, we have performed experiments with more vertical positions of the samples. From Fig. 7b it can be seen that treatment near the nozzle leads to better results, although at some point it has no further effect. The “red position” where nitrogen and hydrogen intensities are the highest and blue position, where hydroxyl radical is the most intensive, both give superior results. Fig. 7b represents the dependency of the treatment efficiency on treatment time. Here, the samples were placed closest
to the nozzle (red position) to obtain the best results. Prolonging the exposure to plasma led to smaller contact angles but the trend is not linear. While doubling the treatment time from 0.5 s to 1 s improved the activation significantly, longer (t > 1s) treatment times did not result in adequately enhanced treatment.