Nitrogen plasma functionalization of low density polyethylene

1 Nitrogen plasmas functionalization of low density polyethylene Carmen López-Santos1, Francisco Yubero1, José Cotrino1,2, Agustín R. González-Elipe1 1.-Instituto de Ciencia de Materiales de Sevilla (CSIC-Univ. Sevilla). Avda. Américo Vespucio 49. 41092 Sevilla. Spain Phone nº: +0034 954489500 Fax nº: +0034 954460665 2.-Física Teórica, Departamento de Física Atómica, Molecular y Nuclear, Facultad de Física, (Univ. de Sevilla), Avda. Reina Mercedes, 41012, Sevilla. Spain Phone nº: +0034 954489500 Fax nº: +0034 954460665 Abstract Low density polyethylene (LDPE) plates have been treated with different nitrogen containing plasmas with the purpose of incorporating nitrogen functional groups on its surface and analyzing the changes experienced in their surface tension. Effects of a dielectric barrier discharge (DBD) at atmospheric pressure and a microwave discharge (MW) at reduced pressure are compared with those obtained by using an atom source supplied with N2 and mixtures Ar+NH3 as plasma gas. X-ray Photoelectron Spectroscopy (XPS) analysis has provided information about the chemical surface changes whereas the surface topography of the treated samples has been examined by Atomic Force Microscopy (AFM). Non-destructive depth profiles of oxygen and carbon have been obtained for the treated and one month aged samples by means of the nondestructive Tougaard´s method of XPS background analysis. A DBD plasma with a mixture of Ar+NH3 was the most efficient treatment for nitrogen and amine group functionalization of this polymer as determined by chemical derivatization. It is argued that the high concentration of NH* species in this plasma is the most important factor in enhancing the nitrogen functionalization of this polymer. It has been also found that the observed increase in hydrophilicity and surface tension cannot be attributed to the anchored nitrogen functional groups formed on plasma treated LDPE. Keywords: nitrogen plasma, LDPE surface, amine groups, wetting, AFM 2 1. Introduction Most applications of polymeric materials require the modification of their surface properties. Plasma techniques constitute an efficient technology of surface polymer activation, aiming at the control of their wetting properties and the improvement of adhesion [1-6]. In a plasma process, gas molecules are dissociated into ions, electrons, free radicals and neutral species. The interaction of these species with the polymer surface causes a series of chemical and/or physical changes which usually extend trough a surface layer of up to 10 nm thickness. [3,6,7] Plasmas of nitrogen are usually employed to incorporate nitrogen functional groups like -NH2 on the surface of the treated polymers [7-12], aiming at modifying their bioactive or biocide activity [13-15]. Due to their extraordinary bulk properties (low density, flexibility, and high chemical resistance), Polyethylene (PE) is much used for industrial and medical applications, although its low surface energy and the low or null concentration of surface functional groups sometimes result in incomplete adhesion and poor chemical reactivity [16]. Trying to get around these drawbacks, plasma activation of PE, particularly with regard to its nitrogen functionalization, has been an active area of research during the last years [16-18]. These previous works have shown that low pressure ammonia plasmas are efficient to incorporate surface amine groups, while simultaneously inducing other side effects like an increased hydrophilicity and changes in roughness. Improving the efficiency of nitrogen plasma functionalization and describing the aging effects after plasma treatment constitute a first motivation of the present work. Moreover, trying to understand the effect of the different species of the plasma in these surface functionalization processes, we propose a comparative study of the effect of different types of plasmas and plasma gases on the surface functionalization of LDPE. Our study focuses on the incorporation of nitrogen functional groups, particularly amine groups, and on the determination of the changes in the chemical depth distribution through the first surface layers of the treated specimen. A related problem is to determine to which extent the chemical changes induced by the nitrogen plasmas are the cause of the modification of the surface tension and wettability of the plasma treated polymers. This point is critical for nitrogen functionalized polymers where the question of whether amine or other nitrogen functional groups induce an enhancement of the surface hydrophilicity is still controversial and of great importance 3 for the development of biomaterials [19, 20]. This work is the continuation of previous investigations of our group where we have intended a similar approach with polyethylene terephthalate (PET) and DLC thin films [21, 22]. Differences with respect to these two materials are highlighted here, emphasizing the importance of chemical etching processes in the control of the functionalization processes in LDPE. As analysis tools we have used X-ray photoemission spectroscopy (XPS), water and iodomethane contact angle (WCA) measurements for estimation of the surface tension [23, 24] and Atomic Force Microscopy (AFM) for the characterization of the surface topography. These characterization studies have been carried out just after the plasma treatments or after the storage of the samples for one month (i.e, to verify their aging behaviour). A very important issue is the assessment of the in-depth atom distribution after different plasma treatments. With this aim, we have used the XPS peak shape analysis developed by Tougaard [25, 26]. This analysis has provided a semiquantitative description of the oxygen/carbon depth distributions for both the “astreated” and “aged” samples. This method has been previously used by our group to determine the oxygen in-depth distribution in PET and LDPE exposed to different oxygen plasmas [27] and in PET treated with nitrogen plasmas [21]. In agreement with these previous works we have found here that the changes in atom composition affect not only to the upmost surface layers but extend to deeper zones up to some tenths of angstroms. 2. Experimental: Materials Commercial plates of LDPE (0.5 mm thickness) from GOODFELLOW (Goodfellow Ltd, Cambridge) are exposed to the different plasma treatments. Previously, the plates were cleaned with ethanol to remove the possible surface contamination. In the text, these samples will be designed as “original” samples. Plasma treatments To get a deep insight into the role played by neutral, ion and atom species in the plasma activation of LDPE we have used three different methods of surface activation. These procedures include low pressure MW plasma, atmospheric pressure DBD plasma 4 and a beam of neutral species containing nitrogen. A full account of the experimental details can be found in our previous publications [21, 22, 27]. In the low pressure MW plasma reactor a surfatron [28] launcher at 60W of microwave power generates a plasma discharge with a typical density of ionic species of about 1011 cm-3 [29]. N2 (17 sscm) or mixtures Ar+NH3 with Ar as majority component (40 sscm) and NH3 (1.2 sscm) as minority component are used as plasma gases, working at a pressure of 0.3 Torr. Before the plasma activation treatments, the discharges of the different gas mixtures were maintained ignited for 30 min to ensure that no traces of air remain in the reaction chamber. By opening a shutter, the polymer surfaces were exposed to the plasma discharge for 1 min, a time that was sufficient to achieve steady state conditions. The DBD reactor consists of two active electrodes made of stainless steel. LDPE samples were placed on the bottom electrode, where it acted as a dielectric barrier. The distance between the upper metallic electrode and the polymer was fixed at one millimetre. The discharge was generated by an AC high voltage power supply working at a typical voltage of 8 kV and a frequency of 1.5 kHz. Formation of small filaments was observed with an oscilloscope connected in parallel between the two electrodes [27]. N2 (47 sscm) or mixtures of Ar (76 sscm) and NH3 (2 sscm) were employed as plasma gases. The purity of the plasma gas within the reactor was checked by analysis of the outlet gases by mass spectrometry. LDPE surfaces were exposed to the plasma discharge during 1 min. It is expected that the density of ionic species in this system has an approximate value of 109 cm-3 [30]. Plasma treated samples were stored in a diseccator in air at room temperature until their surface characterization. Treatments with neutral species These treatments were made “in situ” in the same XPS device used for the analysis of the samples treated “ex-situ” by DBD and MW plasmas. The beam of neutral atoms/species is generated with an Oxford OSPrey source (Oxford, UK) supplied with pure nitrogen or ammonia at a pressure of 5x10-5Torr. Details of the main operating parameters of this source can be found in our previous publications [21, 22, 27]. From these previous analysis the dose of species coming from the source can be estimated in 5x1013 atoms cm-2 s-1. In the case of ammonia, it was found by mass spectrometry that species of hydrogen, nitrogen and NHx neutral species (all of them 5 designed as N-Hx in the text) impinge on the surface of the polymer. The base pressure in this pre-treatment chamber was 2x10-8 Torr. Long exposure times of up to 15 min were used in this experiment to ensure that the surface of samples reach the steady state. Analysis methods The surface tension and wetting properties of the treated polymer were determined by measuring the wetting contact angles (WCA) of water and iodomethane [23, 24]. The roughness was assessed by AFM and the chemical and in-depth compositions by XPS before and after the polymer was exposed to the different treatments or after several aging times. Here, only a brief account about these experimental details will be provided, a more thorough description can be found in ref. [21, 22, 27]. XPS spectra were obtained with a PHOIBOS-100 5MCD (SPECS) electron spectrometer with an un-monochromatized Mg Kα radiation as excitation source. The spectra were acquired at normal take-off angles in the constant pass energy mode fixed at a value of 20 eV and the pressure was about 5x10-9 mbar. For calibration of the binding energy (BE) scale, a value of 284.5 eV was assigned to the C 1s signal component corresponding to C-H and C-C bonds. Average surface composition was obtained from the area behind the C1s, O1s and N1s peaks corrected with their corresponding sensitivity factors. The concentration of amino groups incorporated onto the treated surfaces was estimated by vapour phase chemical derivatization according to the method of Chevallier et al. [31]. Details can be also found in our previous works [21,22,27]. The following specific reaction involving amino groups occurs at the surface: Cl-C6H4 -CHO + NH2 -[R]  Cl-C6H4 -CH=N-[R] + H2O (1) and the chlorine concentration can be correlated with the amount of amino groups that have reacted according to (1) using the expression [32] : 27 [Cl] [NH ] [C] [Cl]  (2) XPS peak shape analysis using the QUASES software was applied to analyze the in-depth distribution of C and O atoms within the upmost surface regions of the samples before and after plasma treatments [33-35]. Wide scans of C1s and O1s peaks 6 including their inelastic background were studied. However, the relatively low intensity of the nitrogen signal does not make possible to carry out a similar analysis with this peak. Static WCA measurements of the contact angle of deionized and bidistilled water and iodomethane droplets (5l) were made by the Young method with a CAM100 instrument (KSV Instruments Ltd, Findland). The mean value of a minimum of seven measurements on each examined surface is taken as its WCA value. Polar and non-polar contributions to surface tension were estimated according to the Owens-Wendt-Kaelble approximation [36]. Changes on the treated surface topography were studied with a Nanotec AFM microscope and a Dulcinea electronics in tapping mode. AFM images were taken with high frequency levers and analyzed by the WSxM software. The roughness is expressed through the RMS coefficient. 3.- Results 3.1 Standard XPS analysis of functional groups after plasma and beam treatments Chemical changes on the treated LDPE surface are observed by XPS analysis. Figs. 1-3 present the C1s, O1s and N1s spectra of samples just after the different treatments and after their storage for one month. Meanwhile, Table 1 presents the surface composition determined with this technique for LDPE subjected to the different plasma and beam treatments. The C1s spectra of the original LDPE consists of a peak associated to the -C-H / -C-C bonds at 284.5eV of binding energy and a small shoulder at higher binding energies [3, 37]. Original LDPE presents some oxygen and nitrogen impurities on its surface that must be attributed to the surface contamination during the polymer fabrication and manipulation [8]. This contamination generates the observed shoulder in the C1s spectra. This shoulder increased slightly in intensity after the treatment of the original samples with the different plasmas. Such an increase in the intensity shoulder has been usually attributed to the surface incorporation of oxygen and/or nitrogen functional groups characterized by different binding energies: C-N / C-O (285.9eV), CN (286.7eV), N-C=O / C=O (287.6eV) y N-C(=O)-N / O-C=O (288.9eV) [38-42]. Because of the superposition in this spectral region of the possible contribution of quite 7 different functional groups of carbon bonded to oxygen and/or nitrogen, no fitting analysis will be intended here. The observed increase in the C1s shoulder intensity is accompanied by a significant incorporation of oxygen at the surface as revealed by the O1s spectra in Figure 2 and the surface atomic compositions gathered in Table 1. The incorporation of oxygen functional groups during the nitrogen plasma treatment of LDPE must be attributed to the effect of excited species of oxygen always present in conventional plasma reactors. These species have been characterized by optical emission spectroscopy (OES) analysis in a previous work [22]. The data in Fig. 2 and Table 1 reveal that DBD and MW / N2 plasmas incorporate greater amounts of oxygen than the other treatments. The obtained O1s signal is characterized by a binding energy of 532.5eV, usually attributed to C-OH groups [38, 42]. Less oxygen becomes incorporated after the DBD and MW/ Ar + NH3 plasma treatments. In this case, the peak is also broader and appears slightly displaced to a binding energy of 531.7 eV, thus suggesting the additional incorporation of species such as OC = O or NC = O. [42] Incorporation of nitrogen onto the surface during these treatments is evidenced by the development of a XPS signal with a BE comprised between 399 eV and 400 eV (Figure 3). From the evaluation of the area of these peaks, the atomic percentages of nitrogen summarized in Table 1 are obtained. The width of the recorded signals lies between 2.4 and 2.5 eV (in contrast with the C1s peak at 284.5 eV characterized by a width of ca. 1.6eV) and suggests that the obtained spectra are the contribution of several types of functional groups. According to literature, nitrogen species with BEs around this energy zone can be attributed to C-N / C-NH2 groups at around 399.2eV and C = N /N-C=O species at around 400.1eV [40, 42, 43]. Since no clear shoulders/bands can be identified in the experimental spectra no fitting analysis has been intended. Nonetheless, a qualitative assessment of the functional groups is possible by following the evolution of the centroid of the experimental band [9]. XPS signal with a BE around 399.2 eV suggests a special formation of amine-like species. On the other hand, bands appearing at higher BEs would indicate the formation of imide, amide and other related nitrogen species. The N1s spectra in Fig. 3 evidence a certain shift towards 399.2 eV after the treatments with MW Ar+NH3, DBD N2 plasmas and the exposure to the beam of neutral species. It is thus likely that in these samples the relative concentration of amine groups is higher. In addition, on the DBD N2 plasma treated sample the broadening of the peak towards higher BEs suggests the formation of –NO like species characterized by a BE 8 of 401.5eV [40, 42]. According to the spectra in Fig. 3 and the data in Table 1, DBD Ar+NH3 plasma is the most efficient treatment for nitrogen incorporation in LDPE, where a N / C ratio of 15% is reached. Owing to the wide spam of the obtained signal, it is likely that in this sample there are various chemical forms of nitrogen. According to Figure 3 and Table 1, after aging, the treated LDPE surfaces undergo important losses of nitrogen, especially for the samples treated with nitrogen plasmas. The oxygen concentration in these samples also decreased after one month of storage. By contrast, it is noticeable that the nitrogen concentration at the surface is little affected in the aged samples treated with the Ar+NH3 plasmas. A similar behaviour can be signalled for the samples treated with the neutral beams of N or N-Hx. 3.1.- Formation of amino groups determined by surface derivatization The incorporation of amino surface groups can be assess specifically by means of the exposure of the plasma treated LDPE to vapours of chlorobenzaldehyde. According to Chevallier et al. [31] the surface concentration of chlorine after reaction (1) enables an estimation of the surface concentration of amino groups before reaction. In general, per each reacted amino group seven carbon atoms and a chlorine atom of the chlorobenzaldehyde radical become incorporated onto the LDPE surface. This fact was verified as the surface concentration of nitrogen atoms decreased and that of C increased after derivatization. Percentages of chlorine atoms shown in parenthesis in Table 1 indicate that the plasma treatments at atmospheric pressure, in particular the DBD Ar+NH3, produce the maximum concentration of –NH2 surface groups, followed by the MW plasma treatments and the exposure to the beam of nitrogen/hydrogen neutral species. By contrast, no chlorine atoms could be detected on the LDPE sample exposed to the beam of nitrogen atoms despite the relatively high concentration of incorporated nitrogen. Figure 4 shows the obtained the –NH2 concentrations formed on each sample: The reported values have been obtained according to expression (2). A maximum amine concentration, closed to 2% of all surface atoms, is reached after the DBD Ar+NH3 plasma treatment. Meanwhile, the MW N2 plasma treatment and the exposure to the beam of nitrogen/hydrogen neutral species produced –NH2 concentrations higher than 1%. By contrast, the amine concentration was smaller after the treatment with the DBD N2 and MW Ar+NH3 plasmas. Since no amine could be detected on the sample 9 exposed to the beam of nitrogen atoms, we must conclude that this treatment produces another type of nitrogen species on the LDPE surface and/or amine groups which are not exposed at the external surface. Except for the ammonia plasma treated samples that retained an amine concentration higher than 0.5%, after storage for one month the activated LDPE surface lost all the amine groups. 3.3 Oxygen and carbon in-depth distributions from XPS peak shape analysis According to the Tougaard’s concepts it is possible to examine the atomic indepth concentration profiles by XPS peak shape analysis to estimate atoms depth distributions [33, 34]. This technique has been previously used by us to study oxygen depth profiles in PET and LDPE exposed to different plasmas of oxygen [27] or nitrogen [21]. In agreement with other studies in the literature [23], the obtained results showed that oxygen can penetrate up to several nanometers into the outmost surface layers of the plasma activated polymers. We have applied the XPS peak shape analysis to the oxygen and carbon photoemission signals to estimate their in-depth distribution in the activated LDPE. We also tried to apply this method to determine nitrogen depth profiles, but the poor intensity of the N1s signal and the absence of significant differences in the backgrounds behind the N1s peak precluded the use of this technique to assess depth distribution profiles of this element. The method is based on the simulation of the inelastic losses induced by the transport of photoelectrons through a given material [33,34]. To do that, it is considered that a primary spectrum F(E) obtained just after excitation by the X-rays transforms into the measured spectrum J(E) because of the energy losses underwent by the primary electrons during their transport through the analysed material. For the original LDPE sample, the F(E) carbon and oxygen functions are assumed to result from a homogeneous in-depth distribution of these elements and the probability for electronic energy losses leading to the experimental J(E) is represented by the polymer inelastic scattering cross sections collected in the QUASES software package [35]. As an example, Fig. 5 shows the differences in the shape of the backgrounds and the derived depth profiles wide scan spectra corresponding to the O1s signal of LDPE before and after being exposed to the DBD Ar+NH3 plasma. The spectra have been normalized to the intensity of the elastic O 1s peaks to clearly see the changes in the inelastic backgrounds. The simulated spectrum of the plasma treated LDPE and the depth profiles obtained with the QUASES software are also represented in this figure. 16 [12] C.-A. Dai, Y.-H. Lee, A.-C. Chiu, T.-A. Tsui, K.-J. Lin, K.-L. Chen, M.-W. Liu, Polymer 47 (2006) 8583. [13] N.A. Bullet, D.P. Bullet, F.-E. Truica-Marasescu, S. Lerouge, F. Mwale, M.R. Wertheimer, Appl. Surf. Sci. 235 (2004) 395. [14] A.A. Meyer-Plath, K. Schröder, B. Finke, A. Ohl, Vaccum 71 (2003) 391 [15] A.A. Meyer-Plath, B. Finke, K. Schröder, A. Ohl, Surf. Coat. Technol. 174-175 (2003) 877. [16] H. Drnovska, L. Lapcık Jr, V. Bursıkova, J. Zemek, A.M. Barros-Timmons, Colloid Polym. Sci. 281 (2003) 1025. [17] F. Bretagnol, M. Tatoulian, F. Arefi-Khonsari, G. Lorang, J. Amouroux, React. Funct. Polym. 61 (2004) 221. [18] M. Kral, A. Ogino, M. Nagatsu, J. Phys. D: Appl. Phys. 41 (2008) 105213. [19] R. Morent, N. De Geyter, L. Gengembre, C. Leys, E. Payen, S. Van Vlierberghe, E. Schacht, Eur. Phys. J. Appl. Phys. 43 (2008) 289. [20] R. Li, J.R. Chen, J. Plasma Sci. Technol. 8 (2006) 325. [21] C. Lopez-Santos, F. Yubero, J. Cotrino, A.R. Gonzalez-Elipe, ACS Appl. Mater. Interfaces 2 (2010) 980. [22] C. Lopez-Santos, F. Yubero, J. Cotrino, L. Contreras, A. Barranco, A.R. Gonzalez- Elipe, Plasma Process. Polym. 6 (2009) 555. [23] C. van Oss, J. Interfacial Forces in Aqueous Media, Nueva York (EEUU): Marcel Decker (1994). [24] B. Jańczuk, M.L. Kerkeb, T. Bialopiotrowicz, F. González-Caballero, J. Colloid Interface Sci. 151 (1992) 333. [25] S. Tougaard, J. Vac. Sci. Technol. A 14 (1996) 1415. [26] S. Tougaard, Surf. Interface Anal. 26 (1998) 249. [27] C. López-Santos, F. Yubero, J. Cotrino, A. Barranco, A.R. González-Elipe, J. Phys. D: Appl. Phys. 41 (2008) 225209. [28] M. Moisan, Z. Zakrzewski, R. Pantel, J. Phys. D: Appl. Phys. 12 (1979) 219. [29] Y.A. Lebedev, I.L. Epstein, A.V. Tatarinov, V.A. Shakhatov, J. Phys: Conference Series 44 (2006) 30. [30] J.H. Kim, Y.H. Choi, Y.S. Hwang, Phys. Plasma 13 (2006) 093501. [31] P. Chevallier, M. Castonguay, S. Turgeon, N. Dubrulle, D. Mantovani, P.H. McBreen, J.-C. Wittmann, G. Laroche, J. Phys. Chem. B 105 (2001) 12490. 17 [32] A. Choukourov, H. Biederman, I. Kholodkov, D. Slavinska, M. Trchova, A. Hollander, J. Appl. Polym. Sci. 92 (2004) 979. [33] T.K. Markkula, J.A. Hunt, F.R. Pu, R.L. Williams, Surf. Interf. Anal. 34 (2002) 583. [34] P. Favia, M.V. Stendardo, R. d’Agostino, Plasmas Polym. 1 (1996) 91. [35] S. Tougaard, QUASES, Software package for quantitative XPS/AES of surface nanostructures by inelastic peak shape analysis; see www.QUASES.com. [36] D.K. Owens, R.C. Wendt, J. Appl. Polym. Sci. 13 (1969) 1714. [37] V. Kotal, V. Svorcik, P. Slepicka, P. Sajdl, O. Blahova, P. Sutta, V. Hnatowicz, Plasma Process. Polym. 4 (2007) 69. [38] J. Grimblot, B. Mutel, V. Moineau, T. Colson, O. Dessaux, P. Goudmand, Surf. Interface Anal. 30 (2000) 415 [39] J. Larrieu, F. Clément, B. Held, N. Soulem, F. Luthon, C. Guimon, H. Martinez, Surf. Interface Anal. 37 (2005) 544. [40] A. Vesel, I. Junkar, U. Cvelbar, J. Kovac, M. Mozetic, Surf. Interface Anal. M 40 (2008) 1444. [41] K. Narushima, N. Yamashita, Y. Isono, M.R. Islam, M. Takeuchi, Jap. J. Appl. Phys. 47 (2008) 3603. [42] S. Guimond, I. Radu, G. Czeremuszkin, D.J. Carlsson, M.R. Wertheimer, Plasmas Polym. 7 (2002) 71. [43] R.W. Paynter, Surf. Interface Anal. 26 (1998) 674. [44] M. Strobel, V. Jones, Ch.S. Lyons, M. Ulsh, M.J. Kushner, R. Dorai, M. Branch Plasmas Polym. 8 (2003) 61. [45] A.N. Bhoj, M.J. Kushner, J. Phys. D: Appl. Phys. 40 (2007) 6953. [46] A.N. Bhoj, M.J. Kushner, J. Phys. D: Appl. Phys. 39 (2006) 1594. [47] A.N. Bhoj, M.J. Kushner, Plasma Sources Sci. Technol. 17 (2008) 035025. [48] C.P. Klages, A. Grishin, Plasma Process. Polym. 5 (2008) 368. 18 List of tables: Table 1.- Atomic percentages in LDPE subjected to the indicated treatments. In parenthesis, percentage of chlorine determined after derivatization Sample / treatment 1hour (%) (  0.5%) 1month (%) (  0.5%) C O N C O N LDPE original 93.4 4.5 2.1 MW N2 65.3 30.1 4.7(0.7) 70.1 27.7 2.2 DBD N 2 62.8 32.1 5.1(0.5) 73.4 25.3 1.3 N x species 86.9 7.7 5.4(0.0) 89.1 6.3 4.6 MW Ar+NH 3 81.7 14.3 4.0(0.5) 83.7 12.7 3.6 DBD Ar+NH3 66.7 23.2 10.1(1.5) 67.3 24.3 8.4 NH x species 85.5 9.9 4.6(0.8) 88.6 8.5 2.9 Table 2.- RMS values of LDPE subjected to the indicated treatments Sample/ treatment RMS /nm (1mx1m) LDPE original 2.1 MW N2 3.9 DBD N2 15.4 N species 3.3 MW Ar+NH3 1.6 DBD Ar+NH3 5.0 NHx species 2.1 19 Figure caption Figure 1.- C1s photoemission spectra of LDPE subjected to the indicated surface activation treatments by plasma and a beam of neutral species (black line) and for these films after storage for one month (gray dashed line). Figure 2.- O1s photoemission spectra of LDPE subjected to the indicated surface activation treatments by plasma and a beam of neutral species (black line) and for these films after storage for one month (gray dashed line). 20 Figure 3.- N1s photoemission spectra of LDPE subjected to the indicated surface activation treatments by plasma and a beam of neutral species (black line) and for these films after storage for one month (gray dashed line) Figure 4.- amine concentration of LDPE surface subjected to the indicated surface activation treatments by plasma and a beam of neutral species (black) and for these films after storage for one month (gray). original MW N2 DBD N2 Nx especies MW Ar+NH3 DBD Ar+NH3 NHx especies 0 1 2 3 NH 2 (%) - NH2 concentration 21 Figure 5.- O1s wide scan spectra of the original LDPE and of this sample treated with the Ar-NH3/DBD plasma. The simulated spectrum that best fits this latter is also included in the figure (dash line). The two insets show the oxygen depth profiles determined from background analysis of the spectra of the original and plasma treated sample. 22 Figure 6.- Calculated depth profiles of oxygen and carbon for LDPE subjected to the different plasma and beam treatments. Continuous full gray and black color profiles correspond to the original and plasma and beam treated samples, respectively, while the dashed profiles correspond to the samples aged for one month. 23 Figure 7.- AFM images of the LDPE surface subjected to the indicated treatments. Figure 8.- Bearing plots of the AFM images of the LDPE surface subjected to the indicated treatments. 150100500 100 80 60 40 20 0 X[nm] original DBD N2 MW N2 Nx species DBD Ar+NH3 MW Ar+NH3 NHx species 150100500 100 80 60 40 20 0 X[nm] 150100500 100 80 60 40 20 0 X[nm] original DBD N2 MW N2 Nx species DBD Ar+NH3 MW Ar+NH3 NHx species 24 Figure 9.- (Top) Water contact angle of LDPE subjected to the indicated plasma and beam treatments and then aged for the indicated period of time. (Bottom) surface tension of LDPE just after its surface activation by plasma and beam treatments and after storage for one month. 0 100 200 600 800 -1.0 -0.8 -0.6 -0.4 -0.2 -0.0  A  i time / min MW Ar+NH3 MW N2 DBD Ar+NH3 DBD N2 NHx species Nx species 0 100 200 600 800 -1.0 -0.8 -0.6 -0.4 -0.2 -0.0  A  i time / min MW Ar+NH3 MW N2 DBD Ar+NH3 DBD N2 NHx species Nx species