Polymer Interface - American

Department of Chemistry, Science Laboratories, UniVersity of Durham, Durham, DH1 3LE England, U.K.. ReceiVed: December 20, 1995; In Final Form: May 21...
12 downloads 0 Views 502KB Size
14062

J. Phys. Chem. 1996, 100, 14062-14066

Synergistic Oxidation at the Plasma/Polymer Interface J. Hopkins, S. H. Wheale, and J. P. S. Badyal* Department of Chemistry, Science Laboratories, UniVersity of Durham, Durham, DH1 3LE England, U.K. ReceiVed: December 20, 1995; In Final Form: May 21, 1996X

Non-equilibrium O2 plasma modification of polyethylene and polystyrene is compared with photooxidation using O2, Ar, Kr, and Xe electrical discharge vacuum-UV irradiation. The differing levels of surface oxygenation can be accounted for on the basis of direct and radiative energy transfer processes taking place at the polymer surface.

Introduction Low-pressure plasma modification of polymeric substrates is becoming increasingly popular in industry.1,2 Applications include improved hydrophobicity, printability, biocompatibility, adhesion, and gas barrier. Such non-equilibrium glow discharges consist of electromagnetic radiation, ions, atoms, metastables, and electrons. Typically, ion densities lie in the range of 1081010 cm-3 with energies of 0-100 eV.3-6 Mean electron energies 〈〉 span 0-20 eV with a high-energy tail reaching out to 100 eV due to reflections at sheath boundaries.7 The emitted electromagnetic photon flux spans infrared to soft X-rays.8 The interaction of non-isothermal inert gas plasmas with organic substrates is fairly well understood on the basis of a direct energy transfer contribution arising from ions and metastables down to ∼10 Å, combined with a vacuum ultraviolet (VUV) radiative transfer component which can penetrate up to ∼10 µm below the polymer surface.9 However, there currently exists a poor understanding at the molecular level of the specific reaction pathways taking place at non-noble gas plasma/polymer interfaces due to additional chemical factors.10 For instance, in the case of O2 plasmas, it is not clear exactly what role is played by atomic oxygen species contained within the glow discharge during surface oxidation. This article describes a series of experiments carried out to differentiate between direct and radiative chemical processes occurring at the substrate surface during the plasma oxidation of a straight-chain hydrocarbon polymer (polyethylene) and its phenyl-substituted counterpart (polystyrene). Plasma oxidation CH CH2

CH2

CH2 n

Polyethylene

n Polystyrene

of these structurally related polymers is compared with VUV photooxidation under an O2 atmosphere using a range of emission lines (105-147 nm) produced by a variety of lowpressure gas discharges (O2, Ar, Kr, and Xe). Experimental Section Low-density polyethylene (ICI, MW ) 250 000) and polystyrene (BP, MW ) 100 000) films were carefully washed in a * To whom correspondence should be addressed. X Abstract published in AdVance ACS Abstracts, July 15, 1996.

S0022-3654(95)03786-5 CCC: $12.00

Figure 1. Schematic of apparatus used for plasma and VUV irradiation experiments.

1:1 2-propanol/cyclohexane mixture for 30 s and then dried in air; this was found to give a contaminant-free surface (as measured by XPS). High-purity oxygen, argon, krypton, and xenon (BOC, 99.6, 99.999, 99.995, and 99.999%, respectively) gases were used for the various types of glow discharge/VUV treatment. A cylindrical glass reactor (internal diameter 5 cm, volume 490 cm3) was inductively coupled to a 13.56 MHz rf generator via an externally wound copper coil (0.5 cm diameter, 10 turns, spanning 8-16 cm from the gas inlet) and an LC matching unit (Figure 1). This was enclosed in a Faraday cage and continuously pumped by a 33 dm3 h-1 Edwards E2M2 mechanical rotary pump via a liquid nitrogen cold trap yielding a base pressure of 2 × 10-3 mbar and a leak rate of better than 2.3 × 10-12 kg s-1 (calculated assuming ideal gas behavior11). For each O2 plasma modification experiment, a clean piece of polymer film was positioned in the center of the copper coils, and the reactor was purged with 0.2 mbar of oxygen for 2 min prior to igniting the glow discharge at a power of 30 W for a total duration of 5 min (this was found to be sufficient, since longer periods resulted in no further detectable chemical changes at the polymer surfaces). Upon termination, the reactor was vented to O2 gas. An additional glass chamber was attached to the aforementioned plasma reactor for the VUV photooxidation experiments, with the two vessels being connected via a lithium fluoride window (cutoff wavelength below 105 nm 8) (Figure 1); this allowed only the VUV component of the glow discharge to reach the polymer substrate. Both sides of the apparatus were evacuated by separate two-stage rotary pumps to a base pressure of better than 2 × 10-3 mbar. A strip of polymer was placed at a distance of 1 cm facing the LiF port under 0.2 mbar of O2 and irradiated with VUV (produced by igniting a 0.2 mbar glow discharge on the other side of the LiF window). VUV exposures at 30 W glow discharge power for a duration of 30 min were found to produce a limiting level of oxidation at the polymer © 1996 American Chemical Society

Synergistic Oxidation at the Plasma/Polymer Interface

J. Phys. Chem., Vol. 100, No. 33, 1996 14063

TABLE 1: Summary of Oxidation Treatments for Polyethylene % functionalities treatment

CxHy

-C-CO2

-C-O

〉CdO

OdCdO

-OdC(O)dO

O2 plasma O2-VUV/O2 Ar-VUV/O2 Kr-VUV/O2 Xe-VUV/O2

72.0 ( 0.6 93.6 ( 0.4 69.6 ( 0.1 69.8 ( 0.3 62.4 ( 0.3

4.7 ( 0.4 0.0 ( 0.0 7.6 ( 0.2 6.5 ( 0.1 8.1 ( 0.1

8.5 ( 2.3 4.3 ( 0.2 10.1 ( 0.6 10.3 ( 0.4 11.7 ( 0.1

6.1( 0.5 2.1 ( 0.2 5.1 ( 0.4 5.6 ( 0.4 7.8 ( 0.1

4.7 ( 0.4 0.0 ( 0.0 7.6 ( 0.0 6.5 ( 0.1 8.1 ( 0.1

3.7 ( 0.3 0.0 ( 0.0 0.0 ( 0.0 1.3 ( 0.4 1.8 ( 0.1

Figure 2. Compilation of C(1s) XPS spectra for the various oxidation treatments of polyethylene: (a) clean; (b) O2 plasma treatment; (c) O2VUV/O2; (d) Ar-VUV/O2; (e) Kr-VUV/O2; and (f) Xe-VUV/O2.

surface. These treatment times were much longer than those required for the aforementioned O2 plasma experiments, since in the latter case, the substrate is located in closer proximity to the reactive plasma species. X-ray photoelectron spectra were acquired on a Kratos ES300 surface analysis instrument equipped with a Mg (KR1,2 ) 1253.6 eV) X-ray source and a concentric hemispherical analyser (CHA) operating in fixed retarding ratio mode (FRR ) 22:1). XPS measurements were taken with an electron take-off angle of 30° from the surface normal. No evidence was obtained for radiation damage to the samples during the typical time scale involved in these experiments. A PC computer was used for data accumulation and component peak analysis based on a Marquardt minimization algorithm which assumed Gaussian peak shapes with fixed relative full width at half-maximum12 (except for the π-π* shake-up transition in polystyrene13). All binding energies are referenced to the hydrocarbon component at 285.0 eV.14 Under these experimental conditions, the instrumentally determined sensitivity factors for unit stoichiometry were taken as C(1s):O(1s) ∼ 0.55. Results (a) Polyethylene. XPS spectra measured for clean polyethylene displayed a single C(1s) peak at 285.0 eV arising from the -CH2- polymer backbone, (Figure 2). Detailed chemical information about the oxidized polymer surfaces was obtained by peak fitting the C(1s) XPS spectra to a range of carbon functionalities:15 carbon adjacent to a carboxylate group (CCO2 ∼ 285.7 eV), carbon singly bonded to one oxygen atom (C-O ∼ 286.6 eV), carbon singly bound to two oxygen atoms or carbon doubly bonded to one oxygen atom (O-C-O or CdO ∼ 287.9 eV), carboxylate groups (O-CdO ∼ 289.0 eV), and carbonate carbons (O-CO-O ∼ 290.4 eV) (Figure 3 and Table 1). In all cases, C-O groups were found to be the most prominent oxidized carbon species. Surprisingly, very little

Figure 3. Typical C(1s) peak fit for Xe-VUV/O2 treatment of polyethylene.

Figure 4. Variation in O:C ratio with surface treatment of polyethylene.

oxygen incorporation into the polyethylene surface was found during O2-VUV/O2 exposure, while Xe-VUV/O2 gave rise to a level of surface oxidation which was comparable to O2 plasma treatment (Figure 4). (b) Polystyrene. Polystyrene consists of an alkyl chain backbone, to which phenyl rings are attached. Two peaks are observed in the C(1s) region of the XPS spectrum for the clean starting material: a hydrocarbon component (285.0 eV, 94% of total C(1s) signal), and a distinctive satellite structure at ∼291.6 eV (6% of total C(1s) signal) which is associated with low-energy π-π* shake-up transitions that accompany core level ionization13 (Figure 5). Oxygen plasma treatment of polystyrene produced a range of oxygenated surface functionalities, and the extent of oxidation was found to be much greater than that observed for polyethylene (Figure 5 and Table 2). Oxygen uptake by the polystyrene surface was noted to be very small during O2-VUV/O2 exposure, while noble gas-VUV/ O2 treatments led to a significant level of surface oxidation, with Xe-VUV/O2 being the most effective (Figure 6). Discussion A range of energy transfer mechanisms are operational during low-pressure rf electrical discharge modification of solid surfaces; these include electron acceleration within the bulk of the plasma, electron deflection from sheath potentials, and ion

14064 J. Phys. Chem., Vol. 100, No. 33, 1996

Hopkins et al.

TABLE 2: Summary of Oxidation Treatments for Polystyrene % functionalities treatment

CxHy

-C-CO2

-C-O

〉CdO

OsCdO

-OsC(O)dO

π-π*

O2 plasma O2-VUV/O2 Ar-VUV/O2 Kr-VUV/O2 Xe-VUV/O2

58.8 ( 0.3 89.9 ( 0.5 67.7 ( 1.6 71.7 ( 3.1 63.4 ( 2.5

4.0 ( 0.7 0.0 ( 0.0 5.9 ( 0.3 1.9 ( 0.3 3.6 ( 0.7

11.9 ( 0.1 3.7 ( 0.1 9.1 ( 0.5 10.7 ( 0.2 10.5 ( 0.5

10.4 ( 0.7 0.0 ( 0.0 5.6 ( 0.1 5.5 ( 1.7 8.1 ( 0.3

4.0 ( 0.7 0.0 ( 0.0 2.5 ( 0.9 1.9 ( 0.3 3.6 ( 0.7

8.8 ( 0.1 0.0 ( 0.0 6.0 ( 0.2 5.7 ( 1.4 8.0 ( 0.2

2.6 ( 0.5 6.3 ( 0.4 3.1 ( 0.3 2.7 ( 0.8 2.7 ( 0.2

TABLE 3: Most Intense Vacuum UV Emission Lines for Inert Gas Plasmas8,31

Figure 5. Compilation of C(1s) XPS spectra for the various oxidation treatments of polystyrene: (a) clean; (b) O2 plasma treatment; (c) O2VUV/O2; (d) Ar-VUV/O2; (e) Kr-VUV/O2; and (f) Xe-VUV/O2.

Figure 6. Variation in O:C ratio with surface treatment of polystyrene.

and electron acceleration in the wall boundary sheaths.16 Electron collisional processes determine the density of ions, radicals, metastables, and photons present in the plasma. In terms of surface modification, the governing parameters are the nature, the arrival rates, and the angular and energy distributions of the glow discharge species impinging upon the surface.17 In the case of a pure O2 plasma, ions, atoms, ozone, and metastables of atomic and molecular oxygen, as well as electrons and a broad electromagnetic spectrum, can all participate in surface reactions18 leading to the observed incorporation of oxygenated functionalities accompanied by the continual evolution of oxidized volatile molecules (e.g., CO, CO2, H2O, etc.).19 Oxygen atoms20 along with VUV irradiation21 are widely considered to be the two key participants in such reactions, although it is not clear precisely what their relative roles are. Direct abstraction of hydrogen atoms from ground state aliphatic polymer chains by ground state molecular oxygen is an endothermic reaction,22 and therefore it is thermodynamically unlikely.23,24 However, hydrogen abstraction by atomic oxygen, and reaction of oxygen at VUV generated polymer radical sites via either direct attachment of atomic oxygen or addition of molecular oxygen followed by O-O bond rupture, are possibilities.25

noble gas (M)

M I emission lines/nm

M II emission lines/nm

Ar Kr Xe

104.8, 106.7 116.5, 123.6 131.2, 147.0

92.0, 93.2 91.7, 96.4 110, 124.5

The spectral characteristics of the photon flux emitted by lowpressure plasmas8 are mainly dominated by emission lines originating from the relaxation of excited neutral and singly ionized species back down to their ground states. VUV wavelengths correspond to the region of the electromagnetic spectrum where radiation is absorbed by air (below 180 nm), and hence experiments need to be carried out under vacuum.26 Such photons are sufficiently energetic as to be capable of initiating electronic transitions and photoionization within macromolecular systems.20 Saturated polymers require fairly high photon energies in order to generate active sites. The longest wavelength absorption band for polyethylene in the VUV region is associated with σ f σ* transitions (λ < 160 nm)27,28 which can lead to dissociative electronic excitation of C-C and C-H σ bonds, i.e., the formation of free radicals;24 a 100-fold increase in VUV photon absorption cross section is reported to occur below 120 nm which corresponds to the onset of photoionisation.28 Since the difference between the HOMO and LUMO states is smaller for unsaturated compared to saturated polymers,25 the former can absorb at longer wavelengths; quite often photochemistry can proceed at λ > 200 nm.24 The hydrocarbon polymer backbone contained in polystyrene exhibits similar photoabsorption characteristics to those described for polyethylene, while the pendant phenyl rings can undergo π f π* transitions below 280 nm leading to the formation of reactive free radical centers27,29 in conjunction with the desorption of molecular fragments.30 On comparing the relative distributions of oxidized carbon centers (Tables 1 and 2), it can be seen that -C-O functionalities are the most prominent species in all cases; and polyethylene has a stronger tendency to form ester (OsCdO) linkages, whereas polystyrene produces a higher proportion of carbonate (O-CO-O) centers. The important VUV emission lines for low-pressure noble gas plasmas are listed in Table 3. In general, the M I emission lines are the most intense (where M is the noble gas);8,31 these arise from transitions between the lowest lying electronically excited states and the ground state of the atom (e.g., for Ar 3s23p54s1 f 3s23p6).32 The excited ion M II resonance lines are usually the next strongest (e.g., for Ar+ 3s13p6 f 3s23p5). In both cases, spin-orbit coupling results in pairs of lines (Table 3). The background emission present in the UV/visible region consists of a radiation continuum produced by excited inert gas molecules M2*, but the intensity is at least 2 orders of magnitude lower than for the vacuum UV region.8 Photoexcitation of the oxygen molecules within the vicinity of the polymer substrate could potentially take place during plasma oxidation. The variation in molecular oxygen absorption behavior can be summarized in terms of the following threshold wavelengths:33

Synergistic Oxidation at the Plasma/Polymer Interface

O2(X3Σg-) f O2(A3Σu+)

250-300 nm

O2(X3Σg-) f O2(B3Σu-)

175-200 nm

O2(X3Σg-) f O(3P) + O(3P)