This content has been downloaded from IOPscience. Please scroll down to see the full text.

Download details:

IP Address: 129.2.180.151

This content was downloaded on 04/05/2017 at 15:52

Please note that terms and conditions apply.

Model polymer etching and surface modification by a time modulated RF plasma jet: role of

atomic oxygen and water vapor

View the table of contents for this issue, or go to the journal homepage for more

2017 J. Phys. D: Appl. Phys. 50 03LT02

(http://iopscience.iop.org/0022-3727/50/3/03LT02)

Home Search Collections Journals About Contact us My IOPscience

You may also be interested in:

Inactivation of Bacillus atrophaeus by OH radicals

Ryo Ono, Kenta Yonetamari, Yusuke Tokumitsu et al.

Inactivation of virus in solution by cold atmospheric pressure plasma: identification of chemical

inactivation pathways

Hamada A Aboubakr, Urvashi Gangal, Mohammed M Youssef et al.

Influence of air diffusion on the OH radicals and atomic O distribution in an atmospheric Ar

(bio)plasma jet

A Nikiforov, L Li, N Britun et al.

Atmospheric pressure plasma jets: an overview of devices and new directions

J Winter, R Brandenburg and K-D Weltmann

Atmospheric pressure plasma treatment of lipopolysaccharide in a controlled environment

E A J Bartis, D B Graves, J Seog et al.

Optical diagnostics of reactive species in atmospheric-pressure nonthermal plasma

Ryo Ono

Atmospheric plasma generates oxygen atoms as oxidizing species in aqueous solutions

Mohamed Mokhtar Hefny, Cedric Pattyn, Petr Lukes et al.

Formation of reactive oxygen and nitrogen species by repetitive negatively pulsed helium

atmospheric pressure plasma jets propagating into humid air

Seth A Norberg, Eric Johnsen and Mark J Kushner

1 © 2016 IOP Publishing Ltd Printed in the UK

Journal of Physics D: Applied Physics

P Luan et al

Printed in the UK

03LT02

JPAPBE

© 2016 IOP Publishing Ltd

50

J. Phys. D: Appl. Phys.

JPD

1361-6463

10.1088/1361-6463/aa4e97

3

Journal of Physics D: Applied Physics

1. Introduction

Cold atmospheric pressure plasma jets (APPJs) are able to generate chemically reactive species desired for material processing [1, 2] and biomedical applications such as wound healing, disinfection and decontamination [3, 4]. While a great amount of attention has been devoted to characterizing these APPJ sources [1, 5–8] and applying them to cell culture

studies [9], in vitro experiments [10], clinical trials [11] and other applications, comparatively little is known about the interaction mechanisms between cold atmospheric pressure plasma and cells, biomolecules, polymers [12] or dielectrics [13]. Since both plasma source and material surfaces are constantly exposed to the surrounding ambient, the gaseous environment is coupled with the plasma and material surface and influences plasma-surface interactions (PSI). In particular

Model polymer etching and surface modification by a time modulated RF plasma jet: role of atomic oxygen and water vapor

P Luan1, A J Knoll1, H Wang1, V S S K Kondeti2, P J Bruggeman2 and G S Oehrlein1

1 Department of Materials Science and Engineering and the Institute for Research in Electronics and Applied Physics, University of Maryland, College Park, MD 20742, USA2 Department of Mechanical Engineering, University of Minnesota, Minneapolis, MN 55455, USA

E-mail: oehrlein@umd.edu

Received 2 September 2016, revised 27 October 2016Accepted for publication 18 November 2016Published 13 December 2016

AbstractThe surface interaction of a well-characterized time modulated radio frequency (RF) plasma jet with polystyrene, poly(methyl methacrylate) and poly(vinyl alcohol) as model polymers is investigated. The RF plasma jet shows fast polymer etching but mild chemical modification with a characteristic carbonate ester and NO formation on the etched surface. By varying the plasma treatment conditions including feed gas composition, environment gaseous composition, and treatment distance, we find that short lived species, especially atomic O for Ar/1% O2 and 1% air plasma and OH for Ar/1% H2O plasma, play an essential role for polymer etching. For O2 containing plasma, we find that atomic O initiates polymer etching and the etching depth mirrors the measured decay of O atoms in the gas phase as the nozzle-surface distance increases. The etching reaction probability of an O atom ranging from 10−4 to 10−3 is consistent with low pressure plasma research. We also find that adding O2 and H2O simultaneously into Ar feed gas quenches polymer etching compared to adding them separately which suggests the reduction of O and OH density in Ar/O2/H2O plasma.

Keywords: atmospheric pressure plasma, plasma jet, surface modification, etching, XPS, atomic O, hydroxyl radicall

(Some figures may appear in colour only in the online journal)

Letter

IOP

2017

1361-6463/17/03LT02+7$33.00

doi:10.1088/1361-6463/aa4e97J. Phys. D: Appl. Phys. 50 (2017) 03LT02 (7pp)

2

for plasma medicine applications, plasma sources are operated in a humid air environment which contains water vapor that might play an important role in both gas phase and surface characteristics since the dissociation of H2O can initiate a wide variety of plasma and surface chemical processes [14]. Previously we reported plasma processing of model polymers and biomolecules by kHz double ring-APPJ [15–17], kHz pin-APPJ [18] and surface micro-discharge [19, 20] sources using well-controlled environments. Surface chemical modification by these plasma sources was generically observed, whereas etching was highly source dependent. The lack of the quantified reactive species measurement for these plasma sources impedes the interpretation of the material surface response with plasma treatment. In the present work, the surface interaction of a well-characterized time modulated radio frequency (RF) plasma jet with model polymers is investigated using both controlled and uncontrolled environments. This plasma source has been previously described [21] and studied in detail including flow dynamics [22], gas temperature [21], electron density and temperature [21], biochemically reactive species density such as NO [23], O and O3 [24]. The numerical simulations of the NO and the O density profile in the plume region have also been performed which predict significant quenching of O and NO density with additional water impurities in the feed gas [25]. Thus the available information provides the opportunity to improve our understanding of PSI under atmospheric pressure. In this work we have investigated the interaction of the RF jet with several model polymers namely polystyrene, poly(methyl methacrylate) (PMMA) and poly(vinyl alcohol) (PVA). We find that the RF jet can induce fast polymer etching but mild chemical modification, both of which reflect plasma treatment conditions such as feed gas composition, composition of the gaseous environment and treatment distance. Based on measured atomic O density data of this source [25], we provide evidence of direct relation

between polymer etching rate and incident atomic O flux onto material surface. By introducing H2O from either the feed gas or the environment to Ar/O2 plasma, we also demonstrate the quenching of atomic O by water vapor (H2O) predicted by the simulation [25].

2. Experiments and methods

A schematic diagram and optical image of the time modulated RF jet are shown in figure 1. The jet consists of a 1 mm diam-eter RF driven tungsten needle surrounded by a quartz tube (3 mm outer, 1.5 mm inner diameter) around which is wrapped by a grounded copper ring electrode. A total flow of 1.5 stan-dard liters per minute (slm), either of pure Ar or Ar plus 1% molecular gas admixture, is fed through the quartz tube. The molecular gas admixtures that we have studied include air, O2, H2O and O2/H2O mixtures. The plasma is generated using a 20 kHz time modulated 13.3 MHz RF signal with 20% duty cycle (10 µs on, 40 µs off). The average dissipated power by the plasma is measured using the method described elsewhere [21]. In order to avoid direct coupling of a charged species with material surfaces [26], the average dissipated power applied to the jet is maintained as 0.83 W for pure Ar and 3.8 W for Ar plus 1% molecular gas admixture. These values are close to the maximum power without direct coupling to the surface at the closest treatment distance studied (4 mm) while maintaining a comparable plume length for all treat-ment conditions.

Since the plasma plume has a small cross sectional area (0.018 cm2) due to the small inner diameter (1.5 mm) of the quartz tube, the RF jet is mounted on a 2D scanning stage to produce uniformly treated material surfaces that can be more easily characterized. As shown in figure  1(a), the jet scans over the polymer surfaces with a speed of 2.4 mm s−1 and the

Figure 1. (a) Schematic diagram and (b) optical image of the time-modulated RF jet and its interaction with a polymer surface. The RF jet scans over the surface at a speed of 2.4 mm s−1 with the trajectory shown in (a). Distance d from the end of the quartz tube nozzle to the surface can be varied from 4 mm to 8 mm. The RF jet is tilted at θ = 30° to facilitate in situ ellipsometry characterization—the laser optical path of which is also illustrated in (b).

J. Phys. D: Appl. Phys. 50 (2017) 03LT02

3

distance between scanning lines is 0.8 mm which is about half of the ID of the quartz tube. These scanning parameters are chosen to help improve treatment uniformity and avoid heat accumulation. During scanning, the RF jet is tilted at θ = 30°. Three treatment distances of d = 4, 6 and 8 mm from the end of the nozzle to the top of the surface are investigated. To study the composition effect of the gaseous environment on PSI, the material is processed in both uncontrolled (humid room air) and controlled environments using a sealed 50 L chamber that is evacuated to 8 × 10−5 Torr and then refilled to atmospheric pressure with desired N2/O2/H2O mixture. Water vapor is generated by a custom-built delivery system utilizing a MKS M330AH mass flow controller (MFC). The inlet of the MFC is connected to a sealed cylinder in which liquid water is con-tained, while the outlet of the MFC is connected to the dry feed gas line of the plasma jet. The temperature of the MFC (135 °C), water cylinder (110 °C) and connecting pipelines (110 °C) is controlled by proportional–integral–derivative (PID) controllers. The film thickness and surface chemical composi-tion change of model polymers are investigated to show the interaction of RF jet with surfaces. Polymers are character-ized by both in situ and ex situ ellipsometry [16] using 1.5 mW HeNe laser at 632.8 nm as shown in figure 1(b). An opti-cal model is applied to the raw ellipsometric data to extract polymer thickness and refractive index. To avoid contamina-tion and removal of plasma induced weakly bound species by the exposure of surfaces to ambient air [15], the processing chamber is immediately evacuated and polymer samples are transferred through vacuum to a x-ray photoelectron spectr-oscopy (XPS) system (Vacuum Generators ESCALAB MK II) for surface chemical composition analysis. The treated sur-face area is 1 × 1 cm2 for ellipsometry and 2 × 2 cm2 for XPS.

3. Results and discussion

The first observation concerns the significant polymer etch-ing by the RF jet. In humid room air environment, polymers processed using pure Ar or Ar + 1% admixture show thick-ness reduction at all three (4, 6 and 8 mm) treatment distances. Figure 2 shows the dependence of polymer etching depth on

feed gas chemistry and treatment distance using in situ ellip-sometry. The total amount of removed C atoms is estimated by the etching depth and the density of each model polymer (1.04 g cm−3, 1.18 g cm−3 and 1.19 g cm−3 for polystyrene, PMMA and PVA, respectively). Ex situ thickness change evaluation, i.e. comparing film thickness between pre- and post-treatment, agrees well with the in situ data. When com-paring different feed gas chemistries, we find that pure Ar plasma is generally less effective than Ar plus 1% molecular feed gas admixtures if experiments are performed at constant plume length and with the required differences in dissipated power. For these conditions, Ar with 1% molecular gas admix-ture plasma generates more reactive species responsible for polymer etching than pure Ar whose effect is mainly from the VUV [18] and ambient gas entrainment [27]. At the same treatment distance Ar + 1% O2 plasma shows significantly more etching than other feed gases. The maximum instanta-neous etch rate at 4 mm can be 100–500 nm min−1 depending on the polymer treated.

Former reports of the same plasma jet have shown that Ar + 1% O2 plasma has lower electron temperature and density [28], and produces less NO but more atomic O [29] than Ar + 1% air plasma. Along with results shown in figure 1, we infer that atomic O plays an essential role for polymer etching in the case of Ar, Ar + 1% O2 and Ar + 1% air plasma. When comparing the impact of treatment distance for each feed gas, we find that the polymer etching depth decreases linearly with the distance away from the nozzle. This is consistent with the observed linear decrease of the atomic O density with distance in the gas phase reported for the same jet [25]. The numerical simulations indicate that this linear decrease of atomic O density with the distance results from the destruction of O atoms by the impurities from the environment [25]. Besides, the reduction of virucidal activity of feline calicivirus (FCV) also decreases with increasing treatment distance by the same RF plasma jet as reported in literature [30]. It is worth mentioning that due to the strong VUV photon absorption of O2, the VUV effect from Ar excimers is minimal in Ar + 1% admixture plasma [18].

To evaluate the relation between polymer etching and esti-mated incident atomic O flux, the C flux leaving the surface

Figure 2. (a) Polystyrene, (b) poly(methyl methacrylate) (PMMA) and (3) poly(vinyl alcohol) (PVA) etching depth comparison of pure Ar, Ar + 1% O2, Ar +1% air and Ar + 1% H2O plasma with 4, 6 and 8 mm treatment distance. The area density of removed C atom is calculated from the etching depth. The inserts are molecular structures of the corresponding polymer. The environment condition is not controlled and mostly humid ambient air mixed with Ar feed gas.

J. Phys. D: Appl. Phys. 50 (2017) 03LT02

4

versus the O flux bombarding the surface is plotted in figure 3 for polystyrene, PMMA and PVA treated by Ar + 1% air plasma. The average O flux is estimated from the measured and simulated gas temperature and spatial O density profiles of Ar + 2% air plasma [25] by using equation  (1) from the kinetic theory of ideal gas:

πΦ = =nv n

RT

M

1

4

1

4

8 . (1)

Here Φ is the average gas flux, n is the gas density, v is the average speed, R is the gas constant, T is the gas temperature, M is the atomic weight of O. For our plasma conditions, the O density and gas temperature used at 4 mm, 6 mm and 8 mm are 8 × 1015 cm−3 420 K, 5 × 1015 cm−3 390 K and 2 × 1015 cm−3 375 K, respectively. For additional information about the RF jet, the reader is referred to more detailed descriptions by Bruggeman et al on the plasma properties, i.e. ne, Te and Tg [21], flow dynamics [22] and production of NO [23], O, O3 [24]. The C flux is calculated from the measured etching depth of the three polymers treated by Ar + 1% air plasma. The experiments show little difference in O density between 1 and 2% molecular admixtures into the Ar feed gas when the power is in excess of 3 W and gas flow rate is between 1 and 5 slm [31]. Figure 3 indicates that the removed C flux is proportional to the estimated incident O flux, and a similar etching reaction probability of O atoms (slope of the curves in figure 3) is seen for all three polymers which is roughly 1.8 × 10−4–3.3 × 10−4. It is worth noting that although the value of the O/C ratio depends on the accuracy of measured O density spatial profile, the order of magnitude we esti-mated here agrees with that seen in other work [32]. However Benedikt et  al estimated the etching reaction probability to be ~10−2 by matching the profile of He/O2 micro-plasma treated hydrogenated amorphous carbon (a-C:H) films to the simulated O flux profile at the surface [33]. In regard to the difference between the model systems used in these reports, the exact O/C ratio value might need further investigation.

Furthermore, compared to the O density of air admixtures in Ar feed gas, O2 admixtures yield a 2.5 times higher O density at similar plasma power [24, 34]. This value is in good agree-ment with the ~2 times higher etching rate for Ar + 1% O2 compared to Ar +1% air. For Ar + 1% H2O plasma we expect more OH species which may impact polymer etching and will be discussed in more detail below.

Surface chemical composition of plasma processed poly-styrene films in N2 environment with 4 mm treatment distance is shown in figure 4. Since C and H-based pristine polysty-rene contains no O and N, this makes it ideal for studying O and N uptake on surface as a result of plasma treatment. The most distinctive feature of the RF jet etched polystyrene sur-face is oxidation, especially the formation of carbonate ester (O–CO–O) groups. As shown in figure 4(c), the O 1s spectrum of polystyrene treated by both Ar + 1% O2 and Ar + 1% H2O plasma in N2 environment shows a peak (533.9 eV) which can be assigned to carbonate ester. Correspondingly, the C 1s spectrum of treated polystyrene also shows a carbonate ester peak at 290.2 eV (figure 4(a)). Other surface changes include the damage to the polymer carbon backbone (C–C/H at 285 eV) and phenyl ring (π–π* shake-up at 291.7 eV) along with the formation of C–O (286.5 eV), O–C–O/C=O (288 eV) and O–C=O (289 eV) groups. As to surface N, the N 1s spec-trum (figure 4(b)) only shows a weak NO signal, despite the fact that the gas phase NO density measured by laser-induced fluorescence (LIF) and molecular beam mass spectrometry (MBMS) [35] is very high (between 1020 and 1021 m−3). The surface chemical composition of polystyrene treated by other plasma conditions, such as different feed gases or environment gaseous compositions, were also studied. They all exhibit the same functional groups as shown in figure 4, although relative amount differs. Therefore, we show elemental compositions obtained from the XPS raw data in the rest of this report.

To explore the effect of H2O on polymer etching by the RF jet, Ar + 1% O2/H2O mixtures with various O2/H2O ratios were used as feed gas to treat polystyrene surfaces. Pure N2 was used as the processing environment to exclude interaction of environmental O2 with the plasma plume or polystyrene surfaces. Therefore all observed surface oxygen originates from the feed gas. In figure 5 polystyrene etching rates and surface chemical compositions are compared. On the left Y-axis of figure 5(a), etching depth of polystyrene is plotted against the O2/H2O ratio of the feed gas. For both Ar + 1% O2 (O2:H2O = 1:0) and Ar + 1% H2O (O2:H2O = 0:1) sig-nificant polymer etching can be seen. The total etching depth drops strongly if Ar + 1% O2/H2O mixtures are used instead. This indicates that: (1) the species responsible for etching in Ar + 1% O2 plasma and Ar + 1% H2O plasma are different, and (2) the reactive species generated in Ar/O2 plasma and Ar/H2O plasma interact and might be mutually exclusive with each other. In the surface reaction model proposed by Dorai and Kushner [32], O and OH both initiate the attack of poly-mers by H abstraction. Compared to other species available with sufficient densities at the polymer surface, e.g. O2(a1Δg), O3, N, NO, H and HO2, the reactivity of O and OH is several orders of magnitudes higher for hydrogen abstraction from either the carbon backbone or the aromatic ring and these

Figure 3. Relation between estimated incident O flux onto polymer surface and calculated C flux out of polymer surface for Ar + 1% air plasma in room air environment.

J. Phys. D: Appl. Phys. 50 (2017) 03LT02

5

reactions of O and OH initiate polymer ablation and further surface oxidation [32]. For non-water containing plasma the above results, especially figure 3, show great consistency with the notion of atomic O initiating the etching process. In agree-ment with the literature [32], it is also reasonable to assume that OH species are the controlling factor for high rate poly-styrene etching by Ar + 1% H2O plasma as seen here. The reduction of etching depth by mixing a small amount of H2O into Ar/O2 plasma observed in figure 5(a) can be explained by the processes proposed by Gaens et al who has reported the reduction of atomic O density when adding H2O into Ar + 2% air plasma of the same RF jet [25]. Water addition to the plasma leads to the generation of H-containing species, such as H, OH and HO2, which cause the loss of O atoms through reactions (2)–(4) with rate coefficient k (m3 s−1) at 400 K [36]:

+ + = × −kOH O H O ; 2.7 10217→   (2)

+ + + = × −kH O M HO M; 5.6 102 219→   (3)

+ + = × −kHO O OH O ; 4.8 10 .2 217→   (4)

In fact these processes are a cycle in which the H atoms, after being generated from H2O, are converted into HO2 radicals (reaction (3)) which are subsequently reduced to OH (reaction (4)). The cycle ends with a further reduction that creates H atoms again (reaction (2)). These reactions give rise to a net production of one O2 molecule from a net loss of two O atoms [36]. Our finding on the reduction of etching depth by mixing a small amount of O2 into Ar/H2O plasma further suggests that feed gas O2 and its related reactive oxygen species (ROS) could also quench OH radicals through destruction pathways such as reaction (2). We plan to further investigate this mechanism using direct measurement of OH densities. On the right Y-axis of figure 5(a), surface elemental composition of treated polystyrene is plotted against the feed gas O2/H2O ratio. As shown by the dotted lines, the surface C and O composition are proportional to the feed gas O2 concentration rather than the etching depth (black solid line). This indicates that for the RF jet the surface chemical modification is not the

Figure 4. High resolution XPS (a) C 1s, (b) N 1s and (c) O 1s spectra of polystyrene treated by the RF jet with Ar + 1% O2 and Ar + 1% H2O plasma in N2 environment. Pristine polystyrene is also shown for comparison. The RF jet treated surface shows characteristic carbonate ester (O–CO–O) and NO formation.

Figure 5. (a) Feed gas H2O and O2 mixture effect on the etching depth and surface elemental composition of polystyrene treated in N2 environment. (b) Environment O2 effect on the etching depth and surface elemental composition of Ar + 1% H2O plasma treated polystyrene.

J. Phys. D: Appl. Phys. 50 (2017) 03LT02

6

rate limiting step for the etching process. The two processes—etching and surface modification—are separately controlled by different group of plasma species. One possible surface reaction mechanism could be that the O and OH flux to the surface controls the total etching depth but the density of other ROS near the surface, such as O2(a1Δg), O2 and O3, controls the oxidation level by reacting with the active alkyl radical sites on the etched polymer surface to form various oxygen containing functional groups.

The interaction of species generated in Ar/O2 and Ar/H2O plasma is further demonstrated by studying the plasma treated polystyrene surfaces in controlled environment with various O2 concentration. Figure  5(b) shows the relation between the environmental O2 concentration and the polystyrene surface response treated by Ar + 1% H2O plasma. With the increase amount of environmental O2, the polymer etching depth decreases whereas the surface O elemental composition increases. The cause of etching and surface modification can be explained by the following. Since the amount of O2 entrainment into Ar + 1% H2O plasma effluent is proportional to the environmental O2 concentration, the response of the etching depth with ambient O2 should behave similar to adding O2 directly into the Ar + H2O feed gas as discussed in figure 5(a): the loss of OH is caused by O2 related reactive species such as O, O3, NO and HO2 [37]. Alternatively, with more O2 and consequently more ROS in the surrounding environment of the material surface, the reactive radical sites on etched polymer surface may quickly combine with these species and thus cause the increase of surface elemental O composition as shown in figure  5(b). In addition, we also studied the environmental H2O effect on Ar + 1% O2 plasma by adding 1% and 2% humidity into N2 ambient. The result is similar to directly mixing H2O into Ar/O2 plasma: compared to dry N2 ambient the etching depth of 1% and 2% humidity drops from 33.6 nm to 29.3 nm and 22.5 nm, respectively.

The relatively fast polymer etching and carbonate ester rich surface after etching is very characteristic of this RF jet. We have investigated other atmospheric pressure plasma sources using similar approaches, e.g. a surface micro-discharge (SMD) for which we see no etching [12, 19, 20]. A key differ-ence is the higher OH and O densities generated in the RF jet effluent due to the higher power density and less O2 near the surface compared to the kHz driven SMD.

4. Conclusion

In summary, we have demonstrated that a time modulated RF jet in close proximity to model polymer surfaces induces relatively fast etching and characteristic carbonate-ester for-mation along with minor NO uptake on the polymer sur-faces. For these treatment conditions, short lived species, especially O atoms and OH radicals, play an essential role in the polymer etching process. The etching reaction probabil-ity of atomic O with different model polymers based on the ratio of removed C flux to incident O flux is estimated to be 1.8 × 10−4–3.3 × 10−4. Other longer lived ROS surrounding the etched surface determine the final surface oxidation level.

Whereas the separate addition of the O2 or H2O admixture to Ar induces polymer etching at fairly high rates, the simultane-ous adding O2 and H2O to Ar reduces polymer etching, which indicates the reduction of O and OH density in Ar/O2/H2O plasma.

Acknowledgment

The authors gratefully acknowledge financial support by the National Science Foundation (PHY-1415353) and the US Department of Energy (DE-SC0001939). We also thank C Anderson and D B Graves of UC Berkeley for helpful discus-sions on this collaborative project. We are grateful to E A J Bartis, D Metzler, C Li and A Pranda for helpful discussions and collaborations.

References

[1] Tendero C, Tixier C, Tristant P, Desmaison J and Leprince P 2006 Spectrochim. Acta B 61 2–30

[2] Vandencasteele N and Reniers F 2010 J. Electron. Spectrosc. Relat. Phenom. 178 394–408

[3] Kong M G, Kroesen G, Morfill G, Nosenko T, Shimizu T, van Dijk J and Zimmermann J L 2009 New J. Phys. 11 115012

[4] Samukawa S et al 2012 J. Phys. D: Appl. Phys. 45 253001 [5] Bruggeman P and Brandenburg R 2013 J. Phys. D: Appl.

Phys. 46 464001 [6] Walsh J L, Iza F, Janson N B, Law V J and Kong M G 2010

J. Phys. D: Appl. Phys. 43 075201 [7] Sretenović G B, Krstić I B, Kovačević V V, Obradović B M

and Kuraica M M 2011 Appl. Phys. Lett. 99 161502 [8] Jun-Seok O, Yolanda A-G and James W B 2011 J. Phys. D:

Appl. Phys. 44 365202 [9] Arndt S, Wacker E, Li Y F, Shimizu T, Thomas H M,

Morfill G E, Karrer S, Zimmermann J L and Bosserhoff A K 2013 Exp. Dermatology 22 284–9

[10] Fridman G, Friedman G, Gutsol A, Shekhter A B, Vasilets V N and Fridman A 2008 Plasma Process. Polym. 5 503–33

[11] Isbary G et al 2010 Br. J. Dermatol. 163 78–82 [12] Bartis E A J, Knoll A J, Luan P, Seog J and Oehrlein G S 2016

Plasma Chem. Plasma Process. 36 121–49 [13] Wild R, Gerling T, Bussiahn R, Weltmann K D and

Stollenwerk L 2014 J. Phys. D: Appl. Phys. 47 042001 [14] Bruggeman P and Schram D C 2010 Plasma Sources Sci.

Technol. 19 045025 [15] Bartis E A J, Graves D B, Seog J and Oehrlein G S 2013

J. Phys. D: Appl. Phys. 46 312002 [16] Knoll A J, Luan P, Bartis E A J, Hart C, Raitses Y and

Oehrlein G S 2014 Appl. Phys. Lett. 105 171601 [17] Bartis E A J, Luan P, Knoll A J, Hart C, Seog J and

Oehrlein G S 2015 Biointerphases 10 029512 [18] Knoll A J, Luan P, Bartis E A J, Kondeti V S S K, Bruggeman P J

and Oehrlein G S 2016 Plasma Processes Polym. 13 1069–79 [19] Bartis A J E, Luan P, Knoll J A, Graves B D, Seog J and

Oehrlein S G 2016 Eur. Phys. J. D 70 1–19 [20] Bartis E A J, Luan P, Knoll A J, Graves D B, Seog J and

Oehrlein G S 2016 Plasma Processes Polym. 13 410–8 [21] Hofmann S, van Gessel A F H, Verreycken T and Bruggeman P

2011 Plasma Sources Sci. Technol. 20 065010 [22] Zhang S, Sobota A, Veldhuizen E M V and Bruggeman P J

2015 J. Phys. D: Appl. Phys. 48 015203 [23] Iseni S, Zhang S, Gessel A F H V, Hofmann S, Ham B T J V,

Reuter S, Weltmann K D and Bruggeman P J 2014 New J. Phys. 16 123011

J. Phys. D: Appl. Phys. 50 (2017) 03LT02

7

[24] Zhang S, Sobota A, van Veldhuizen E M and Bruggeman P J 2015 Plasma Sources Sci. Technol. 24 045015

[25] Gaens W V, Bruggeman P J and Bogaerts A 2014 New J. Phys. 16 063054

[26] Hofmann S, van Gils K, van der Linden S, Iseni S and Bruggeman P 2014 Eur. Phys. J. D 68 1–14

[27] Reuter S, Winter J, Schmidt-Bleker A, Tresp H, Hammer M U and Weltmann K D 2012 IEEE Trans. Plasma Sci. 40 2788–94

[28] van Gessel B, Brandenburg R and Bruggeman P 2013 Appl. Phys. Lett. 103 064103

[29] van Gessel A F H, Alards K M J and Bruggeman P J 2013 J. Phys. D: Appl. Phys. 46 265202

[30] Aboubakr H A, Williams P, Gangal U, Youssef M M, El-Sohaimy S A A, Bruggeman P J and Goyal S M 2015 Appl. Environ. Microbiol. 81 3612–22

[31] Zhang S 2015 PhD Thesis Eindhoven University of Technology, Eindhoven

[32] Dorai R and Kushner M J 2003 J. Phys. D: Appl. Phys. 36 666–85

[33] Schneider S, Lackmann J W, Narberhaus F, Bandow J E, Denis B and Benedikt J 2011 J. Phys. D: Appl. Phys. 44 295201

[34] Zhang S, Gessel A F H V, Grootel S C V and Bruggeman P J 2014 Plasma Sources Sci. Technol. 23 025012

[35] Ham B T J V, Hofmann S, Brandenburg R and Bruggeman P J 2014 J. Phys. D: Appl. Phys. 47 224013

[36] Verreycken T and Bruggeman P J 2014 Plasma Chem. Plasma Process. 34 605–19

[37] Herron J T and Green D S 2001 Plasma Chem. Plasma Process. 21 459–81

J. Phys. D: Appl. Phys. 50 (2017) 03LT02