Wetting Characteristics of Plasma-Modified Porous Polyethylene

Georgia Institute of Technology. , ‡. Porex Corporation. , Corresponding author. E-mail: rina.tannenbaum@ mse.gatech.edu. Cite this:Langmuir 2003, 1...
2 downloads 0 Views 135KB Size
Langmuir 2003, 19, 5869-5874

5869

Wetting Characteristics of Plasma-Modified Porous Polyethylene George Greene,†,‡ George Yao,‡ and Rina Tannenbaum*,† School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, and Porex Corporation, 500 Bohannon Road, Fairburn, Georgia 30213 Received December 2, 2002. In Final Form: April 16, 2003 This paper examines the wetting characteristics of porous polyethylene surfaces modified by exposure to reactive oxygen glow discharge gas plasma, through the direct measurement of the wicking properties of the modified material. It is well-known that oxygen plasma can be used to chemically alter the surface of polyethylene to enhance wetting properties. Chemical and physical modification of a polymer’s surface is the consequence of reactions initiated by the collision of high-energy species in the plasma with the polymer surface. The conditions of the plasma treatment, such as electric field strength (power), exposure time, and chamber pressure, govern the frequency and energy of collisions and, thus, determine the nature and degree of the chemical modification of the polyethylene surface. Comparisons of the chemical modification of sintered porous polyethylene surfaces achieved through treatments with reactive oxygen glow discharge gas plasmas generated at various powers, chamber pressures, and times of exposure were made by measuring the wicking rate of distilled and deionized water in the modified materials. A strong correlation was observed between the electric field power used to generate the plasma and the degree of chemical modification of the polyethylene surfaces. In addition, the rate of chemical modification was also found to be a function of the electric field power.

1. Introduction The use of low-pressure, reactive gas plasmas to enhance the wetting properties and reactivity of polymers by altering the chemical composition of the surface is a wellestablished technology used extensively throughout the fields of chemistry and polymer science.1 Plasma modification has been used successfully in improving the adhesive properties, printability, and permeability of polymer surfaces.2-5 In addition, plasma can be employed to alter the molecular weight, melting temperature, and solubility of polymer surfaces while, at the same time, leaving the properties of the bulk polymer unchanged.1,6 Though the majority of polymer applications utilizing plasma modification involve solid substrates such as thin film coatings and webbing, this technology can also be applied to modify the surface properties of porous polymeric materials. Improved wetting properties can be achieved through the plasma modification of porous polymer substrates.6,7 Plasma modification can be applied to sintered porous polyethylene (UHMWPE) membranes and three-dimensional substrates to successfully trans* Corresponding author. E-mail: mse.gatech.edu. † Georgia Institute of Technology. ‡ Porex Corporation.

rina.tannenbaum@

(1) Boenig, H. V. Applications of low-temperature plasma technology: Advances in Low-Temperature Plasma Chemistry, Technology, Applications; Technomic Publishing Company, Inc.: Lancaster, PA, 1984; Vol. 1, pp 1-10. (2) Hargreaves, M. S.; Hussey, D. S.; Leuchtner, R. E. Mater. Res. Soc. Symp. Proc. 1999, 544, 291-296. (3) Choi, D. M.; Park, C. K.; Cho, K.; Park, C. E. Polymer 1997, 38 (25), 6243-6249. (4) Plawky, U.; Londschien, M.; Michaeli, W. Acta Polym. 1996, 47 (2-3), 112-118. (5) Wu, D. Y.; Gutowski, W. S. Mater. Sci. Forum 1995, 189-190, 221-228. (6) Inagaki, N. Plasma surface modification and plasma polymerization; Technomic Publishing Company, Inc.: Lancaster, PA, 1996; pp ix-xi, 2, 9-10, 22-33, 70-74, 88-98. (7) Steen, M. L.; Jordan, A. C.; Fisher, E. R. J. Membr. Sci. 2002, 204, 341-357.

form a material initially impervious to water into an effective wicking medium. The chemical and physical modification of a polymer’s surface is the consequence of reactions initiated by the collision of high-energy species in the plasma with the polymer surface. Because the chemical modification of a polymer is the consequence of reactions initiated by the collision of species in the plasma with the polymer surface, the extent of chemical modification is a function of the frequency and energy of the impinging species. For a flat, solid surface, the average kinetic energy of the species colliding with the surface is equivalent to the average kinetic energy of the species in the plasma. When a microporous structure is considered, the situation becomes more complicated. Internal surfaces in such a structure cannot be reached via a direct path by a particle initially outside the structure. Inside the porous matrix, the distances between the internal surfaces are much smaller than the mean free path of the species in the plasma (∼400 µm for air at 120 mTorr pressure and 27 °C compared to a 10-µm average pore diameter for sintered porous polyethylene),8 thus preventing the ionized species already within the structure from acquiring sufficient kinetic energy to initiate a reaction before encountering a surface. Chemical modification of the internal surfaces, therefore, is dependent on the diffusion of the external plasma species into the porous matrix. The inelastic nature of collisions between plasma particles and polymer surfaces depletes the kinetic energy of the particle as it progresses through the structure. As a result, the extent of the chemical modification of porous polyethylene surfaces varies with the depth of penetration.7 In this work, we use the migration rate of deionized water through the porous matrix to measure the interfacial energy of a plasma-modified polyethylene surface. Traditional methods of comparing the surface energies through direct contact-angle measurements cannot be (8) Ohring, M. The Material Science of Thin Films; Academic Press: New York, 1992; pp 51-53.

10.1021/la026940+ CCC: $25.00 © 2003 American Chemical Society Published on Web 05/31/2003

5870

Langmuir, Vol. 19, No. 14, 2003

applied to porous materials because the surface roughness hinders the free dispersion of the testing fluid. Moreover, liquid penetration into the porous structure further distorts the contact-angle measurement. As a result, the contact-angle measurements of porous surfaces are unreliable and cannot be used to accurately gauge the interfacial energy.9 The porous structure of sintered porous polyethylene can be envisioned as an idealized array of capillaries with radii equivalent to the average pore radius. Under this assumption, the volumetric flow rate of a fluid through the porous media is equivalent to that observed in an idealized capillary with a circular cross section. Because the volumetric flow rate is directly proportional to the capillary pressure, which in turn is directly proportional to the contact angle, the migration rate of a fluid as a means to measure the contact angle of the surface in a porous material is, thus, justified. The relationship of the surface contact angle to the liquid volumetric flow rate in an idealized capillary is given by the Lucas-Washburn equation.10 4

cos θ ∝

[(V/t)(8η/πr ) + Fg]rL w 2γ(lv) [(V/t)(8η/πr4) + Fg]rL (1) cos θ ) K 2γ(lv)

where V/t ) volumetric flow rate; r ) capillary radius; F ) fluid density; η ) fluid viscosity; g ) gravitation constant; L ) penetration distance (capillary length); γ(lv) ) liquid vapor interfacial energy; and cos θ ) contact angle. A derivation of the Lucas-Washburn equation is presented by Zhmud, Tiberg, and Hallstensson who discuss, in further detail, the implications of surface interfacial energy to the migration of fluids within a porous matrix.10 The previously mentioned relationship between the contact angle and the volumetric flow rate makes possible the correlation of the migration rate with the surface energy. Note that the proportionality sign has replaced the equality sign as a result of the recognition that the surface does not necessarily behave as an ideal capillary, and, hence, K represents the correction factor to adjust for this deviation from ideality (K < 1). If the pore size and pore size distribution of the porous media being tested and the wicking fluid being used are kept constant, liquid-migration-rate measurements can be used to gauge the surface interfacial energy of the porous media. However, as was denoted by Chilbowsky and PereaCarpio, the geometric uncertainty arising from anisotropy in the pore matrix of the membrane represents a significant source of error that casts doubt upon the accuracy of the contact angles calculated by this method in nonideal systems.11 Because of this uncertainty, this study uses migration-rate measurements as a means of making qualitative comparisons of the surface interfacial energies between membranes instead of as a way of quantifying the surface contact angle. Modification of a polyethylene surface through exposure to reactive oxygen plasma increases the solid-liquid interfacial energy relative to the solid-vapor interfacial energy through the implantation of oxygen containing (9) Chan, C. M. Polymer Surface Modification and Characterization; Hanser/Gardner Publications: Cincinnati, OH, 1994; pp 45-49. (10) Zhmud, B. V.; Tiberg, F.; Hallstensson, K. J. Colloid Interface Sci. 2000, 228, 263. (11) Chilbowsky, E.; Perea-Carpio, R. Adv. Colloid Surf. Sci. 2002, 98, 245.

Greene et al.

Figure 1. SEM image of porous polyethylene accompanied by the pore size profile, as was obtained from mercury porosimetry measurements.

functional groups at the polymer’s surface. The difference between the solid-vapor and solid-liquid interfacial energies constitutes the work of dispersion, which governs the dispersion of a liquid on a surface, which in turn governs the pressure of a liquid in a capillary and, thus, the wicking of a liquid in a porous matrix.

W(d) ) γ(sv) - γ(sl)

(2)

where W(d) ) work of dispersion; γ(sv) ) solid-vapor interfacial energy; and γ(sl) ) solid -liquid interfacial energy. For the process of wicking to occur, the work of dispersion for a surface must be negative. Therefore, as the degree in which the surface of polyethylene is chemically altered, the difference between the solid-vapor and solid liquid interfacial energies becomes more negative. As a result, the migration rate, or volumetric flow rate of fluid through the porous matrix, will increase as the extent of chemical modification increases. 2. Experimental Section 2.1. Materials. The porous polyethylene sheet material (UHMWPE, 400-µm thickness) used in this experiment was obtained from the Porex Corporation (Fairburn, GA). The porous polyethylene sheets used all possessed a 40-45% pore volume, a 10-µm mean pore size, and a similar distribution of pore sizes between 1-14 µm, as was measured by mercury porsimitry and shown in Figure 1. 2.2. Surface Activation of Porous Polyethylene via RF Glow Discharge Reactive Gas Plasma. Before exposure to reactive gas plasma, the porous polyethylene sheets were washed with excess absolute ethanol to remove low-molecular-weight impurities, residual mold release agents, and other contamination such as dirt or oil. The sheets were then dried in a vacuum oven at 60 °C and 25-mmHg vacuum for 1-2 h until dry. All the plasma treatments were carried out in a Europlasma CD600M/PC plasma chamber. The washed porous polyethylene sheets were placed in the plasma chamber, which was then pumped down to a baseline pressure of 0.1 mTorr. The sheets were held at the baseline pressure for 30 min to allow the trapped gases to evacuate the porous structure. The plasma treatment used for this experiment consisted of two steps. First, the porous polyethylene sheets were exposed to inert argon plasma to prepare the surface by burning off any residual contamination not removed by the alcohol wash. This step was kept constant for all the plasma treatments used in this experiment and was conducted at a chamber pressure of 120 mTorr of argon, a power of 200 W RF (radio frequency), and an exposure time of 1 min. After the inert argon plasma exposure, the chamber was then pumped back down to a baseline pressure of 0.1 mTorr. The second step of the plasma treatment was initiated once the baseline pressure was reached. In this step of the treatment process, the porous polyethylene sheets were exposed to a reactive oxygen plasma to chemically modify the properties of the surface. For this experiment, the porous polyethylene sheets were exposed to various oxygen plasmas generated at pressures ranging from 60 to 120 mTorr, at electric field powers ranging from 63 to 500 W RF, and for time periods between 1 and 14 min.

Plasma-Modified Porous Polyethylene X-ray photoelectron spectroscopy (XPS) measurements of the plasma-treated polyethylene films were performed on a Surface Science Model SSX-100 with small spot ESCA spectrometer, at a takeoff angle of 45°. 2.3. Measurement of the Migration Rate in a Chemically Modified Porous Polyethylene Membrane. The migration rate is used as a basis for comparing the extent of chemical modification in porous polyethylene sheets exposed to reactive oxygen plasmas. All the migration-rate measurements of porous polyethylene media were conducted immediately after removal from the plasma chamber. The migration-rate measurements were conducted on 1 by 5 cm samples cut from the plasma-treated sheets. All the migration-rate measurements were made using room-temperature (27 °C) deionized water. Deionized water was chosen for the wicking fluid because its inability to wet unmodified porous polyethylene allows for the assumptions that the observed wetting properties are the sole product of chemical modification and that variations in the wetting properties are a reflection of the extent of chemical modification. Testing of the migration rate was carried out on porous polyethylene samples mounted vertically to a stationary bar held above a shallow dish of deionized water. With the use of an adjustable platform, the dish of deionized water was raised to a height where the surface of the water makes initial contact with the end of the suspended sample. A stopwatch was used to measure the time required for the water to wick through the porous sample to a height of 4 cm after the initial contact between the sheet and the water. The reported migration rate represents the average rate over the 4-cm distance and not the actual rate of liquid migration as it exists at the 4-cm position. In addition, each reported average migration rate was determined by averaging the migration rates of six individual samples of membrane. The inverse dependence of the migration rate on the height to which the fluid has traveled in the membrane results in the rapid decline of the migration rate over the 4 cm. Geometric variations in the membrane structure on a localized level make single-point rate measurements unreliable given the importance of the height and pore radius to the migration of water through the membrane. However, over a sizable area, geometric variations both within and between membranes become insignificant. Averaging the migration rate over a 4-cm distance reflects the migration rates at all points of the membrane, thus negating the influence of localized geometric variations, and, likewise, enables the wetting properties of the membrane surface to be reliably assessed.

3. Results and Discussion The goal of this work was twofold: (a) to characterize the extent and type of polymer functionalization as a result of the plasma treatment and (b) to determine the relationship between the parameters of oxygen glow discharge gas plasmas (i.e., electric field strength, chamber pressure, and exposure time) and the surface wetting and wicking properties of 10-µm porous polyethylene material. With the understanding of how each specific variable of the oxygen plasma treatment influences the surface properties, this research attempts to establish a method to chemically modify the surface of sintered porous polyethylene with controllable wicking properties and chemical reactivity. To study the effect of oxygen plasma on the chemical composition of polyethylene surfaces, XPS analysis was conducted on a series of polyethylene films treated for 0, 1, 2.5, and 5 min with glow discharge oxygen plasmas generated at a fixed chamber pressure (120 mTorr) and power density (200 W/m2). Polyethylene films were used for this XPS study in lieu of sintered polyethylene membranes because of complications arising from the porous structure and irregular surface topography of the sintered porous polyethylene membrane. Before the oxygen plasma treatment, the surfaces of the films were first washed with alcohol and then exposed to an inert argon plasma for 1 min generated at a 120-mTorr chamber pressure and 200 W RF power to remove any surface

Langmuir, Vol. 19, No. 14, 2003 5871

Figure 2. Low-resolution XPS survey spectra of polyethylene films exposed to plasma generated at a 200 W/m2 power density and a constant chamber pressure of 120 mTorr: (a) untreated polyethylene film and (b) after exposure to plasma treatment for 5 min. Please note that the small peak visible at 700 eV, corresponding to fluorine atoms, is not a product of a reaction with the plasma but rather a residue of the mold release agent and, hence, should be disregarded.

contamination. It should be noted that the use of an inert argon plasma in the preparation of the surface is not without consequences. In addition to the removal of dirt, oil, low-molecular-weight polyethylene waxes, and other contamination from the film surface, chain-scission reactions initiated by the argon plasma alters the makeup of the polyethylene surface; however, the modification effects primarily the physical properties of the surface (morphology, cross-link density) with only a minor impact to the chemical composition (i.e., the formation of carbon radicals). The XPS spectra of the polyethylene sheets before and after 5 min of exposure to oxygen plasma show the appearance of various oxygen-containing species on the surface. The low-resolution survey XPS spectra at a 45° takeoff angle before plasma treatment (Figure 2a) show only a small presence of oxygen (∼2%) on the surface, most likely due to low-molecular-weight organic molecules in the ambient environment and polymer chain oxidation produced during film processing. After the exposure to plasma (Figure 2b), the concentration of the oxygencontaining species on the surface increased dramatically (∼18%). It is interesting to note that oxygen plasma exposure effectively increases the oxygen percentage of the surface from an initial value of ∼2% to a value of ∼18% after just 1 min of exposure time, and prolonged exposure does not result in additional formation of oxygenconatining species on the surface, as is shown in Figure 3a. Moreover, the distribution of the various species on the polymer surface, after plasma exposure exceeding 1 min, is also stable, as is shown in Figure 3b. The high-resolution XPS spectra of both C(1s) and O(1s)core level electrons are shown in Figure 4. Before plasma treatment, the carbon peak at 281.28 eV may be decon-

5872

Langmuir, Vol. 19, No. 14, 2003

Greene et al.

Figure 3. Relative concentration of oxygen and oxygencontaining species as a function of exposure time to plasma generated at a 200 W/m2 power density and a constant chamber pressure of 120 mTorr: (a) percent change in the relative concentration of oxygen and (b) percent change in the relative composition of the various chemical species on the surface of the film.

voluted into only one main peak corresponding to the aliphatic carbon of the polyethylene chain (Figure 4a). The percent Gaussian and the full width at half the peak’s maximum (fwhm) were varied until the best fit was obtained. These parameters were then fixed and used exactly as such in all the subsequent spectral deconvolutions. After the exposure to plasma treatment, the C(1s) core level photoemission spectrum may be fitted with a minimum number of four peaks (Figure 4b): the original aliphatic carbon peak at 281.22 eV and three additional peaks at 281.88, 283.38, and 285.13 eV, corresponding to a CsO single bond, CdO double bond, and COOH group, respectively.6,12,13 On the basis of the number of oxygencontaining species identified via analysis of the C(1s) peaks, the O(1s) core level photoemission spectrum should be fitted with a minimum number of three peaks (Figure 4c) at 528.15, 529.32, and 530.45 eV, corresponding to a CdO double bond, COOH group, and CsO single bond, respectively. The carbon-oxygen bonds in the COOH group are resonative, and, hence, they fluctuate between first and second order. A summary of all the peaks and their properties is shown in Table 1. It is interesting to note that, from a similar analysis of high-resolution C(1s) and O(1s) spectra obtained for different times of exposure to plasma, a similar composition of the various species is observed, as is shown in Figure 3b. The kinetic energy of the plasma species colliding with the surface impacts the type and frequency of reactions initiated at the polymer surface. Theoretically, the average (12) Kostandinidis, F.; Thakkar, B.; Chakraborty, A. K.; Potts, L.; Tannenbaum, R.; Tirrell, M.; Evans, J. Langmuir 1992, 8, 1307-1317. (13) Tannenbaum, R.; Hakanson, C.; Zeno, A. D.; Tirrell, M. Langmuir 2002, 18 (14), 5592-5599.

Figure 4. High-resolution X-ray photoemission spectra of C(1s) and O(1s) core electrons of polyethylene films before and after exposure to plasma generated at a 200 W/m2 power density and a constant chamber pressure of 120 mTorr: (a) photoemission spectrum of the C(1s) core electrons of the unmodified polyethylene film; (b) photoemission spectrum of the C(1s) core electrons of the polyethylene film exposed to plasma treatment for 5 min; and (c) photoemission spectrum of the O(1s) core electrons of the polyethylene film exposed to plasma treatment for 5 min.

kinetic energy of the ionized species in the plasma is independent of the chamber pressure and only governed by the electric field strength used to generate the plasma, which in turn is determined by the power density supplied to the plasma generator electrodes. Figure 5 shows the wicking rates of porous polyethylene sheets (400-µm thick, 10-µm average pore size) measured for each chamber pressure of 60, 120, and 240 mTorr and at various power densities of 63, 100, 200, 300, 400, and 500 W/m2. The exposure time for each treatment was kept constant at 14 min. Little change in the wicking properties is observed as a result of varying the chamber pressure, confirming the independence of the kinetic energy of the plasma from the pressure. On the other hand, the wicking rate increases rapidly as the power density is increased from 63 W/m2 (the minimum power density necessary to spark the

Plasma-Modified Porous Polyethylene

Langmuir, Vol. 19, No. 14, 2003 5873

Table 1. Summary of the Peak Characteristics Observed in XPS Analysis of the Plasma-Treated Polyethylene Films speciesa (band no.)

peak position

δ (eV)

fwhm (eV)

% Gaussian

assignment of atoms

2 4 3

O(1s) (After Plasma Treatment) 530.45 2.30 1.73 84 529.32 1.17 1.79 87 528.15 0.00 1.69 91

CsOH HOsCdO CdO

4 3 2 1

C(1s) (After Plasma Treatment) 285.13 3.81 1.51 98 283.38 2.16 1.59 100 281.88 0.66 1.69 100 281.22 0.00 1.69 94

HOsCdO CdO CsOH aliphatic

1

C(1s) (Before Plasma Treatment) 281.28 0.00 1.69 94

aliphatic

a

The band numbers correspond to the specific peaks identified in Figure 3.

Figure 6. Schematic representation of the outcome of the collisions of the plasma molecules with the polyethylene surface. Such collisions can result either in an ion-implantation process or in surface etching, depending on the kinetic energy of the plasma molecules and the frequency of collisions.

Figure 5. Average wicking rates of 10-µm-pore-size porous polyethylene sheets modified at a fixed exposure time with oxygen plasma generated at various power densities and three different chamber pressures.

plasma) to 200 W/m2. Above 200 W/m2, a gradual decrease in the wicking rate is observed. A faster wicking rate is associated with an increase in the surface energy because of the increased chemical modification of the surface. The fastest wicking rate, and, thus, greatest chemical modification of the surface, was observed at a 200 W/m2 power density. Collisions of the plasma species with the polyethylene surface can result in one of three outcomes, depending upon the kinetic energy of the plasma species. If the kinetic energy of the charged plasma species is too low, a collision between the plasma species and the surface will fail to initiate a reaction. If the kinetic energy is too high, the collisions between the plasma species and the polymer surface will result in the scission of the polymer chain and the formation of radicals at the polymer-chain ends. Radicals formed through chain scission are capable of undergoing cross-linking reactions. However, if the process of chain scission occurs very rapidly and the rate of chainterminating radical formation is faster than that of crosslinking reactions, etching of the polymer surface will be observed. Only collisions of plasma species with polyethylene surfaces having the right amount of kinetic energy are capable of initiating ion-implantation reactions responsible for the chemical modification of polymer surfaces. The general schematics of this process are shown in Figure 6. Collisions of this type result in the formation of radicals without chain scission through the displace-

ment of hydrogen atoms directly from the polymer chain. Radicals of this type, formed within the polymer chain itself, can either react with nearby chains to form crosslinks, undergo a rearrangement that results in chain scission, or recombine with reactive species in the plasma, such as ions and free radicals, to generate chemical functionalities.14-20 The increase in the wicking rate of porous polyethylene modified with plasmas generated using power densities below 200 W/m2 is the result of the decrease in the number of species in the plasma colliding with the surface with the kinetic energy necessary to initiate an ion-implantation reaction. The observed decrease in the wicking rate as the power density is increased above 200 W/m2 is due to the increased frequency of the chain-scission reactions and surface etching. Though increasing the power density above 200 W/m2 also increases the number of species colliding with the surface with enough kinetic energy to cause ion-implantation reactions, the number of collisions resulting in chain scission also increases, thus shifting the equilibrium between surface creation and surface destruction.16,18,19 Because maximum chemical modification is determined by the equilibrium between ion-implantation and surfaceetching reactions,19 changes to the relative reaction rates of these two processes will shift the equilibrium and, thus, the extent of chemical modification of the surface. Figure 7 shows how the average wicking rate for 10-µm-poresize porous polyethylene modified with oxygen plasma at 120 mTorr pressure varies as a function of the exposure time and power density used to generate the plasma. Clearly, for each power density used, the wicking rate reaches a maximum value, at a time beyond which further chemical modification of the surface is not possible through continued exposure to the plasma. This is supported by (14) Murata, Y.; Aradachi, T. J. Electrost. 2001, 51-52, 97-104. (15) Wheale, S. H.; Barker, C. P.; Badyal, J. P. S. Langmuir 1998, 14 (7), 6699-6704. (16) Coates, D. M.; Kaplan, S. L. Plasma Processing of Advanced Materials. In MRS Bulletin; Collins, G. A., Rej, D. J., Eds.; Materials Research Society: Warrendale, PA, 1996; Vol. 21, Chapter IV. (17) Takeda, H.; Murata, Y. Jpn. J. Appl. Phys., Part 1 1996, 35 (9a), 4791-4792. (18) Meichsner, J.; Nitschke, M.; Rochotzki, R.; Zeuner, M. Surf. Coat. Technol. 1995, 74-75, 227-231. (19) Normand, F.; Granier, A.; Leprince, P.; Marec, J.; Shi, M. K.; Clouet, F. Plasma Chem. Plasma Process. 1995, 15 (2), 173-198. (20) Inagaki, N.; Tasaka, S.; Hibi, K. Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 1990, 31 (2), 380-381.

5874

Langmuir, Vol. 19, No. 14, 2003

Figure 7. Variation of the average wicking rates for 10-µmpore-size porous polyethylene modified with oxygen plasma at 120 mTorr pressure as a function of the plasma exposure time for several power densities used to generate the plasma.

the XPS experiments of the solid polyethylene films, in which the constant concentration of the plasma-generated oxygen-containing species that was observed after ∼2 min of plasma exposure (for the experiment at 120 mTorr and 200 W/m2 shown in Figure 3). A relationship is also observed between the electric field power used to generate the plasma and the time required for the system to reach equilibrium. At a power density of 63 W/m2, the rate of chemical modification is very slow, and only after an exposure time of 14 min does the wicking rate appear to reach a maximum value, signaling that equilibrium has been established. In contrast, at a power density of 500 W/m2, the highest power density tested, chemical modification occurs much more rapidly, reaching an equilibrium state after only ∼2 min of exposure to the plasma. Between these two extremes, the exposure time necessary to reach an equilibrium state decreases steadily as the power density is increased from 63 to 500 W/m2, but the changes between 200 and 500 W/m2 are essentially negligible. 4. Conclusions The experimental evidence suggests a strong correlation between the extent that a porous polyethylene surface can be chemically modified by a reactive oxygen plasma and the electric field strength (power) used to generate

Greene et al.

the plasma as well as the exposure time of the surface to the plasma. In addition, the extent of chemical modification was shown to be insensitive to the chamber pressure used to generate the plasma. The surface modification of porous polyethylene by oxygen plasma is a product of the two competing processes of ion-implantation and surface-etching reactions. A limit to the degree of chemical modification is established when these two competing processes come into equilibrium. The power density of the electric field used to generate the plasma strongly influences this equilibrium. The average kinetic energy of the plasma, a direct function the electric field power density, determines the frequency and energy of the collisions and, thus, governs the relative kinetics of these two processes. Altering the power density will shift the average kinetic energy and, consequently, will result in a shift in the equilibrium. To realize the maximum modification of the surface properties in a porous polyethylene substrate, it is necessary to maximize the rate of ion-implantation reactions relative to that of surface-etching reactions. In this experiment, the wicking properties and, thus, chemical modification of porous polyethylene surfaces were maximized under plasma generated at 120-mTorr pressure and a 200 W/m2 power density. Because the wicking properties decayed rapidly below 200 W/m2 and gradually above 200 W/m2, the energy of the collisions that produce ion-implantation reactions and that of the collisions resulting in chain scission and surface etching must be quite similar. The fact that a limit in which a surface can be chemically modified is observed over the entire range of the electric field indicates that the process of ion implantation cannot be isolated from that of surface etching. Acknowledgment. This research was supported by and conducted with the cooperation of the Porex Corporation in Fairburn, Georgia, and the NSF, through the Packaging Research Center at Georgia Tech. We would like to give special thanks to the Porex Corporation for supplying the sintered porous polyethylene materials for this experiment, to Dr. George Yao and Dr. Gary Mao for their collaboration in this effort, and to Dr. Brent Carter from the School of Materials Science and Engineering at Georgia Tech for his valuable comments and advice regarding the surface analysis of polyethylene. LA026940+