pubs.acs.org/Langmuir © 2010 American Chemical Society
Effects of Surface Properties of Different Substrates on Fine Structure of Plasma-Polymerized SiOCH Films Prepared from Hexamethyldisiloxane (HMDSO) Chia-Hao Lo,† Kuo-Sung Liao,† Manuel De Guzman,† Vincent Rouessac,‡ Ta-Chin Wei,† Kueir-Rarn Lee,*,† and Juin-Yih Lai† †
R&D Center for Membrane Technology, Department of Chemical Engineering, Chung Yuan University, een des Membranes- ENSCM/UM2/CNRS UMR5635, Chung-Li 32023, Taiwan, and ‡Institut Europ Universit e Montpellier 2-CC047, 2 Place Eug ene Bataillon, 34095 Montpellier cedex 5, France Received July 9, 2010. Revised Manuscript Received September 10, 2010
In this study, Doppler broadening energy spectroscopy (DBES) combined with slow positron beam was used to discuss the effect of substrate types on the fine structure of a plasma-polymerized SiOCH layer as a function of depth. From the SEM pictures, the SiOCH films formed on different substrates showed hemispherical macrostructures, and the deposition rate was dependent on the mean pore size. It appears that the morphology of the plasma-polymerized SiOCH films was associated with the porosity-related characteristics of the substrate such as the size/shape of pores. As deposited on the MCE-022 substrate (mixed cellulose esters membrane with a mean pore size of 0.22 μm) with a nodular structure, the SiOCH films had pillar-like structures and high gas permeabilities. DBES results showed that the SiOCH films deposited on different substrates were composed of three layers: the SiOCH bulk layer, the transition layer, and the substrate. It was observed that the microstructure of the SiOCH films was affected layer by layer; a higher surface pore size in the substrates induced thicker transition layers with higher microporosities and led to thinner bulk layers having higher S parameter values during the plasma polymerization. It was also observed that the change in O2/N2 selectivity was consistent with the DBES analysis results. The gas separation performance and DBES analysis results agreed with each other.
Introduction In gas separation, gas permeability generally decreases with an increase in the membrane thickness, whereas selectivity shows the reverse tendency. Fabrication of composite membranes, which consist of a permselective thin top layer with high cross-linking structures and a porous substrate, has been regarded as a means to improve gas permeability and simultaneously retain high selectivity.1,2 For the preparation of this thin layer, plasma polymerization has been considered to be an appropriate method because of the following specific properties of the resulting plasma-polymerized membranes: ultrathin (from nanometers to micrometers), highly cross-linked, pinhole-free, and strong adhesion to substrates.3,4 Moreover, the plasma polymerization technique usually operates at low temperatures (70-150 °C) on the substrate side, and as such, thermal damage on the polymeric substrate can be avoided.5 In plasma polymerization, various plasma parameters such as plasma power, monomer flow rate, and system temperature can affect the deposition mechanism and result in several types of plasma-polymerized film structures.6 During the plasma polymerization process, the surface of substrates is in direct contact with the plasma. The surface property of substrates, such as pore size, roughness, and surface free energy, can affect the deposition *To whom correspondence should be addressed. Tel: þ886 3 2654190. Fax: þ886 3 2654198. E-mail:
[email protected].
(1) Huang, S. H.; Li, C. L.; Hu, C. H.; Tsai, H. A.; Lee, K. R.; Lai, J. Y. Desalination 2006, 200, 387. (2) Ulbricht, M. Polymer 2006, 47, 2217. (3) Roualdes, S.; der Lee, A. V.; Berjoan, R.; Sanchez, J.; Durand, J. AIChE J. 1999, 45, 1566. (4) Bosc, F.; Sanchez, J.; Rouessac, V.; Durand, J. Sep. Purif. Technol. 2003, 32, 371. (5) Inagaki, N.; Tasaka, S.; Hiramatsu, H. J. Appl. Polym. Sci. 1999, 71, 2091. (6) Yasuda, H. Plasma Polymerization; Academic Press: New York, 1985; pp 244250. (7) Rochat, G.; Leterrier, Y; Plummer, C. J. G.; Ma˚nson, J.-A. E.; Szoszkiewicz, R.; Kulik, A. J.; Fayet, P. J. Appl. Phys. 2004, 95, 5429.
17470 DOI: 10.1021/la102759b
efficiency. Rochat’s study7 pointed out that because SiOCH films were deposited on a polyamide substrate with a spherulite structure, localized delamination was observed at the boundary between the SiOCH film and the spherulite zone with the use of local-acceleration microscopy (SLAM). This is because interspherulitic zones are softer than spherulitic zones, leading to strain concentrations at the spherulite boundaries. The study of Roualdes et al.2 found that SiOCH was deposited on different substrates at various rates; the more hydrophobic the substrate, the more difficult the growth rate. However, supported films for gas separation are generally thin (few micrometers or less), which makes it hard for their microstructure and especially the depth profile to be characterized by conventional techniques. To date, there is little research that discusses the effect of the depth profile in plasma-polymerized film structures for gas separation.8,9 In recent years, positron annihilation spectroscopy (PAS), including Doppler broadening energy spectroscopy (DBES) and positron annihilation lifetime spectroscopy (PALS), has been developed as a useful tool for probing the microscopic structure of polymeric materials.10-14 One of the great advantages of PAS characterization is the direct determination of the polymer free volume and (8) Jean, Y. C. Microchem. J. 1990, 42, 72. (9) Peng, Z. L.; Olson, B. G.; McGervey, J. D.; Jamieson, A. M. Polymer 1999, 40, 3033. (10) Jean, Y. C., Mallon, P. E., Schrader, D. M., Eds.; Principle and applicaion of posiron and posironium chemisry; World Scientific Publishing: Singapore, 2003. (11) Jean, Y. C.; Mallon, P. E.; Zhang, R.; Chen, H.; Li, Y.; Zhang, J.; Wu, Y. C.; Sandreczki, T. C.; Suzuki, R.; Ohdaira, T.; Gu, X.; Nguyen, T. Radiat. Phys. Chem. 2003, 68, 395. (12) Zhang, J.; Zhang, R.; Chen, H.; Li, Y.; Wu, Y. C.; Suzuki, R.; Sandreckski, T. C.; Ohdaira, T.; Jean, Y. C. Radiat. Phys. Chem. 2003, 68, 535. (13) Zhang, J.; Chen, H.; Li, Y.; Suzuki, R.; Ohdaira, T.; Jean, Y. C. Radiat. Phys. Chem. 2007, 76, 172. (14) Chen, H.; Hung, W. S.; Lo, C. H.; Huang, S. H.; Cheng, M. L.; Liu, G.; Lee, K. R.; Lai, J. Y.; Sun, Y. M.; Hu, C. C.; Suzuki, R.; Ohdaira, T.; Oshima, N.; Jean, Y. C. Macromolecules 2007, 40, 7542.
Published on Web 10/08/2010
Langmuir 2010, 26(22), 17470–17476
Lo et al.
Article Table 1. Polymeric Membranes Used As Substrates for Plasma Deposition
hole properties on an atomic scale. The variable energy positron beam with an adjustable energy and a narrow energy distribution allows depth-resolved measurements that are useful in plasmapolymerized film fine structure studies. Such method has the characteristic of accurately probing the free-volume variation from the topmost layer (∼0.1 nm) of the surface down to the bulk layers (several micrometers).11-14 In our previous study,15 the SiOCH films deposited on an MCE substrate consisted of two layers (bulk and interfacial layers), which were probed using the slow positron beam technique; the interfacial layer showed larger pore sizes than the bulk layer because of the etching and redeposition on the MCE substrate in the early deposition stage. In this study, the slow positron beam was used to discuss the effect of the substrate type on the fine structure of a plasmapolymerized SiOCH layer as a function of depth. Also, the relationship between the gas separation performance and the pore structure was discussed.
Experimental Methods Materials. Hexamethyldisiloxane (HMDSO), supplied by Aldrich, was used without further purification as the precursor. To discuss the effect of different substrates on the structure of plasma-polymerized SiOCH films, four substrates with different morphologies were chosen: mixed cellulose esters (MCE-0025, MCE-022), polyacrylnitrile (PAN), and polytetrafluoroethylene (PTFE-02). Their chemical structures and other details are given in Table 1. To facilitate the description of each type of substrate, a simple nameplate consisting of alphabets and numbers was used in this study. The alphabets represent the type of the polymer and the numbers refer to the pore size. For example, MCE-0025 denotes that the substrate is a mixed cellulose esters polymer with a mean pore size of 0.025 μm. PECVD Process. SiOCH films were polymerized on different polymeric substrates by plasma deposition at 13.56 MHz using a homemade capacitively coupled parallel-plate reactor, the description of which was given in our previous study.16 The experimental procedure for preparing the composite membranes was as follows: first, the reactor was evacuated to 0.023 Torr; second, the HMDSO vapor was fed into a reactor without any carrier gas through a vapor mass flow controller (MKS type 1150) at a constant flow rate of 20 sccm; and third, as the system pressure was adjusted to 0.2 Torr, the RF power was turned on and set at 200 W. In this process, the deposition time was the major operating parameter that caused an adjustment in the thickness of the SiOCH film. (15) Lo, C.-H.; Huang, J.-K.; Hung, W.-S.; Huang, S.-S.; Guzman De, M.; Rouessac, V.; Li, C.-L.; Hu, C.-C.; Lee, K.-R.; Lai, J.-Y. J. Membr. Sci. 2009, 337, 297. (16) Tu, C.-Y.; Wang, Y.-C.; Li, C.-L.; Lee, K.-R.; Huang, J.; Lai, J.-Y. Eur. Polym. J. 2005, 41, 2343.
Langmuir 2010, 26(22), 17470–17476
Doppler Broadening Energy Spectrum (DBES). A newly built slow positron beam with a variable mono energy at the R&D Center for Membrane Technology,17,18 Chung Yuan University, Taiwan was used for this study to determine the mean depth of the membrane between 0 and ∼10 μm. (The mean depth was calculated by using an established equation19 based on the positron incident energy from 0 to 30 keV.) This new radioisotope beam uses 50 mCi of 22Na as the positron source. The DBES spectra were measured using an HP Ge detector at a counting rate of ∼2000 cps. The total number of counts for each DBES spectrum was 2.0 million. The obtained DBES spectra were characterized in terms of the S, W, and R parameters. The S parameter, representing the low momentum part of the positron-electron annihilation radiation, is sensitive to the change of the positron and positronium (Ps) states due to microstructural changes.10,14 Because the positron and Ps are trapped in a hole with a finite size, the observed S parameter is a measure of the momentum broadening according to the uncertainty principle: a larger hole results in a larger S parameter value. Two other parameters from DBES were also reported here: (1) W parameter, which represents the high momentum part in the DBES spectra. It is an indication of the types of the chemical elements or defect types in the DBES data. (2) R parameter, which is defined as the 3γ to 2γ annihilation ratio, can provide information about the existence of large pores (nanometers to micrometers). It is from o-Ps undergoing 3γ annihilation, whereas S and W are from p-Ps and o-Ps undergoing 2γ radiation (pick-off annihilation) in free volumes (angstroms to nanometers). Therefore, the pore is larger, and the opportunity for the 3γ annihilation to take place increases, resulting in a higher R parameter. A more detailed description of these parameters can be found in the literature.9,14
Results and Discussion Figure 1 shows the SEM surface morphology and cross-section of the substrates used in this study. MCE-0025 and PAN substrates displayed surfaces with a lot of nanopores, whereas MCE022 and PTFE-02 showed surfaces with porous structures. The latter substrates had different surface pore structures: MCE-022 surface was of the nodular-type structure and PTFE-02 surface had many nodes connected to each other by fiber-like filaments. Although the manufacturer provided some information about the substrates, it only referred to bulk properties such as mean pore size. It was necessary, therefore, to analyze the surface properties (17) Hung, W.-S.; Lo, C.-H.; Cheng, M.-L.; Chen, H.; Liu, G.; Chakka, L.; Nanda, D.; Tung, K.-L.; Huang, S.-H.; Lee, K. R.; Lai, J.-Y.; Sun, Y.-M.; Yu, C.-C.; Zhang, R.; Jean, Y. C. Appl. Surf. Sci. 2008, 255, 201. (18) Huang, S.-H.; Hung, W.-S.; Liaw, D.-J.; Li, C.-L.; Kao, S.-T.; Wang, D. M.; DeGuzman, M.; Hu, C.-C.; Jean, Y. C.; Lee, K. R.; Lai, J. Y. Macromolecules 2008, 41, 6438. (19) Schultz, P. J.; Lynn, K. G. Rev. Mod. Phys. 1988, 60, 701.
DOI: 10.1021/la102759b
17471
Article
Lo et al.
Figure 1. SEM images of substrates used (a) MCE-0025, (b) MCE-022, (c) PAN, and (d) PTFE-02.
of the substrates in more detail. The image analysis program (SimplePCI, v5.3.1, HAMAMATSU Company) was used to determine the surface porosity and surface mean pore size of the substrates. The choice of pores in this program was dependent on the shading value of the SEM pictures of the substrates (Figure 1). The fitting results of the substrates, surface water contact angle, and deposition rate of SiOCH films are tabulated in Table 2. MCE-0025 and PAN had the lowest surface porosities of ∼14% but indicated different mean pore sizes. MCE-022 and PTFE-02 had the highest surface porosities of >40% but gave the lowest and highest water contact angles, respectively. In addition, MCE-022 and PTFE-02 showed different surface mean pore sizes: MCE-022 had a pore size or diameter of 0.128 μm, whereas PTFE-02 pore diameter was 0.179 μm The surface of the SiOCH films covering the MCE-0025 and PAN substrates gave an intrinsic water contact angle of 105°. Large hemisphere-like SiOCH films deposited on porous substrates (MCE-022 or PVDF-022) induced high roughness and improved surface hydrophobicity (126° for the SiOCH on MCE-022). Figures 2 shows the surface and cross-sectional images of the SiOCH films deposited on four different substrates at 200 W for 60 min. As seen in the SEM surface pictures, the SiOCH films formed can be described as dome-shaped.20 These SiOCH surfaces were composed of densely packed, spherical clusters with (20) Teshima, K.; Inoue, Y.; Sugimura, H.; Takai, O. Thin Solid Films 2002, 420-421, 324.
17472 DOI: 10.1021/la102759b
well-defined boundaries. For all SiOCH/substrate composite membranes, the common observation was that the SiOCH films on all substrate surfaces (i.e., regardless of the raw substrates) were descriptive of hemispherical structures. However, the morphology of the plasma-polymerized SiOCH films was strongly dependent on the substrate. For example, the SiOCH cluster formed on MCE-022 with a nodular structure was composed of many distinct hemispheres, whereas the SiOCH layer on the PTFE-02 substrate had both fiberlike and nodular structures, as can be seen in Figure 2b,d. It is interesting that the SiOCH deposited on the MCE-022 substrate with a nodular morphology had a pillar-like structure, whereas the SiOCH deposited on PTFE-02 showed a uniform and dense structure (SEM cross-sectional images, Figure 2d), although the PTFE-02 had a high surface porosity of 43% (Table 2). In addition, one can see that the SiOCH films on the MCE-0025 and PAN substrates with a lower surface porosity, in comparison with the films on the porous substrates, had more closed surfaces and clear interface between the SiOCH film and the substrate. These results on the deposition of the SiOCH films on the porous substrates could be understood and explained by the “pore-filling mechanism”. This means that the SiOCH films deposited preferentially on the solid part, then in the pore on the substrate surface. As the HMDSO fragments contacted the pores on the substrate surface, the fragments would penetrate into the pores and then result in a decrease in the SiOCH thickness, as observed in the SEM image. If the HMDSO Langmuir 2010, 26(22), 17470–17476
Lo et al.
Article Table 2. Surface Properties of Substrates and SiOCH Film substrate mean pore diameter (μm)
maximum diameter (μm)
surface porosity (%)
MCE-0025 0.043 ( 0.008 0.160 ( 0.011 13.4 ( 0.1 MCE-022 0.128 ( 0.013 4.428 ( 0.054 44.1 ( 0.6 PAN 0.016 ( 0.003 0.085 ( 0.016 15.2 ( 0.1 PTFE-02 0.179 ( 0.010 4.050 ( 0.043 43.3 ( 0.4 a SiOCH films were prepared at 200 W for 60 min. b C.A.: contact angle.
water C.A. (deg)b 78 ( 1 32 ( 1 57 ( 2 121 ( 1
SiOCH filma water deposition rate C.A. (deg)b (nm/min) 105 ( 1 126 ( 1 105 ( 1 108 ( 1
40.9 ( 1.8 39.6 ( 8.7 53.7 ( 3.7 29.6 ( 7.5
Figure 2. SEM images of SiOCH films deposited on different substrates at same deposition times of 60 min. (During the PECVD process, RF power and system pressure were fixed at 200 W and 0.2 Torr, respectively.)
fragments directly contacted the solid part of the substrate, the fragments would deposit and form SiOCH films. Therefore, the morphology of the plasma-polymerized SiOCH films could be seen to follow the substrate morphology (or surface porosityrelated characteristics). For example, as the SiOCH deposited on PTFE-02 with a fiber-like structure, it would preferentially deposit on the solid part such as nodes and fibers, and the resulting surface would then show a morphology similar to that of PTFE-02. The deposition rate of the SiOCH films decreased with an increase in the mean pore size in the substrate (Table 2). It Langmuir 2010, 26(22), 17470–17476
appears that the morphology of the plasma-polymerized SiOCH films is associated with the substrate surface porosity-related characteristics, such as the size/shape of pores. On the basis of the SiOCH structure difference, which is mainly due to the substrate morphology, we can expect that these composite membranes will have different gas-separation properties. Gas-Separation Performance. SiOCH films were deposited on all substrates under the same plasma conditions of 200 W for 60 min, and the resulting composite membrane was used for the measurement of O2 and N2 gas permeabilities. The data on the DOI: 10.1021/la102759b
17473
Article
Lo et al.
Table 3. Gas-Separation Performance of SiOCH Films on Different Substrates substrate
SiOCH thickness ( μm)
PO2 (barrer)
O2/N2
2.45 ( 0.11 2.36 ( 0.52 3.22 ( 0.22 1.78 ( 0.45
58.8 ( 4.5 -a 80.5 ( 18 4840 ( 1048
2.89 ( 0.02 1.12 ( 0.08 1.90 ( 0.06
MCE-0025 MCE-022 PAN PTFE-02 a -: overflow.
thickness and O2/N2 gas-separation performance of the SiOCH films are given in Table 3. It can be seen that the SiOCH films on different substrates had very different O2 permabilities and O2/N2 selectivities. For MCE-022, the SiOCH films did not completely cover the pores on the substrate surface because of the high surface porosity; therefore, overflow occurred during the gas permeability test. The pillar-like structure of the SiOCH films on MCE022 (Figure 2b) seems to have a high porosity and to display a high gas permeability. In addition, on the basis of the SEM image, the SiOCH films deposited on MCE-0025, PAN, and PTFE-02 were observed to show no characteristic porosity, but they displayed very different O2 permeabilities (PO2) and O2/N2 selectivities. SiOCH/MCE-0025 showed the lowest PO2 of ∼60 (barrer) and the highest O2/N2 selectivity of 2.89, whereas SiOCH/PAN indicated a low PO2 of 81 (barrer) and the lowest O2/N2 selectivity of 1.12. Compared with these composite membranes, SiOCH/ PTFE-02 showed an unusually high PO2 of 4840 (barrer) and an O2/N2 selectivity of 1.90. The reason for this may be that the fast deposition rate for SiOCH/PAN would result in a looser packing of HMDSO fragments, as compared with SiOCH/MCE-0025 and SiOCH/PTFE-02. These results appear to indicate that the microstructure of SiOCH films was strongly affected by the substrate surface. However, on the basis of the SEM images, it is difficult to explain the relationship between the changes in the fine structure of the plasma-polymerized SiOCH films depending on the substrate and the gas-separation performance of the three composite membranes. To be able to have an understanding of the microstructure based from the depth profile of plasmapolymerized layers, it is helpful to choose an appropriate substrate morphology where the fabricated plasma-polymerized layers have the optimum thickness for gas separation. Structure Characterization of SiOCH Films on Different Substrates. Because the results on the surface morphology of SiOCH films based on the SEM pictures are insufficient to explain the gas-separation behavior, the SiOCH film structure in regard to the variation in the microstructure of the SiOCH films was investigated using PAS coupled to a slow-positron beam. Doppler broadening of energy spectra (DBES), one of the positron annihilation techniques, was used in this study. This technique is a powerful method for determining the chemical composition and physical microstructural change in materials. Figure 3 shows the S parameter as a function of the positron incident energy (S-E plot) data for four SiOCH/substrate composite membranes. The S parameter value relates to the free volume or pore size; the larger the S value, the bigger the free-volume size. These composite membranes consisted of three layers, as labeled in Figure 3a. Three major regions for the SiOCH/PAN composite membrane can be identified: (1) near the surface and the bulk of the SiOCH layer (region I), (2) the transition layer (mixed layers of SiOCH and substrate, region II), and (3) the polymer substrate (region III). In region I, the S parameter near the surface indicated a typically low value due to the back diffusion of the positron. As the positron penetrated deeper in the plasma-polymerized SiOCH layer, a large fraction of the positronium (Ps) could be trapped in the holes. This positronium was then annihilated, resulting in a sharp 17474 DOI: 10.1021/la102759b
Figure 3. S-E plots for plasma-polymerized SiOCH/substrate composite membranes with different substrates. (a) S-E plot for raw SiOCH/PAN composite membrane; (b) S-E plots for SiOCH films deposited on four different substrates (one of which is the enlarged portion of S-E plot from part a).
increase in the S parameter with the mean depth. When the positron penetrated further into the bulk of the plasma-polymerized SiOCH, the S parameter displayed an approximately constant value. These values are similar to those published in other literature.15,21 It is seen that the SiOCH films deposited on different substrates showed different S values (Figure 3b), which means that their microstructures are different. As the positron kept penetrating into the transition layer, which corresponds to the mixed layers of the SiOCH film and the substrate (region II), the values of the S parameter showed a marked difference from those in the bulk layer. When the positron energy was >17 keV, the positron reached the polymer substrate, and the value of the S parameter displayed a decreasing trend. This is because the S parameter value of the polymer is usually lower than that in SiO2like materials. (The S parameter for each polymeric substrate is discussed in the Supporting Information.) To understand clearly the variation in the S parameter in each layer, a computer program VEPFIT22 was used to deconvolute (21) Yu, R. S.; Ito, K.; Hirata, K.; Sato, K.; Zheng, W.; Kobayashi, Y. Chem. Phys. Lett. 2003, 379, 359. (22) VanVeen, A.; Schut, H. H.; de Vries, J.; Hakvoort, H. A.; Ijpma, M. R. AIP Conf. Proc. 1990, 218, 171.
Langmuir 2010, 26(22), 17470–17476
Lo et al.
Article Table 4. Multilayer Analysis of S Parameter for SiOCH/Substrates Composite Membranes SiOCH/substrate composite membrane
boundary length (nm)
substrate
S1
S2
S3
L1
L2
MCE-0025 MCE-022 PAN PTFE-02
0.5444 ( 0.0003 0.5495 ( 0.0003 0.5498 ( 0.0003 0.5477 ( 0.0003
0.5531 ( 0.0007 0.5670 ( 0.0002 0.5560 ( 0.0008 0.5541 ( 0.0016
0.4359 ( 0.0010 0.4386 ( 0.0011 0.4862 ( 0.0012 0.4463 ( 0.0010
800 ( 150 613 ( 135 607 ( 167 501 ( 173
2448 ( 625 2385 ( 256 3182 ( 105 2361 ( 533
Figure 4. Schematic diagram of three-layer structure obtained from VEPFIT fitting results. Top layer is bulk SiOCH layer, second layer is transition layer mixing of SiOCH and substrate, and bottom layer is polymeric substrate (MCE-0025, MCE-022, PAN, and PTFE-02).
the S parameter in each SiOCH/substrate composite membrane layer. The best fitted results as judged from good χ2 values are tabulated in Table 4 and plotted in Figure 4. On the basis of Figure 4, the SiOCH deposited on MCE-0025 with a surface porosity of 13.4% had the lowest S values among the SiOCH films, indicating that such SiOCH had the smallest microporosity or pore size. Therefore, this composite membrane showed a lower O2 permeability and the highest O2/N2 selectivity, consistent with the observation of the S values for the SiOCH films. Compared with the SiOCH films deposited on MCE-0025, those on porous substrates (MCE-022 and PTFE-02) had thicker transition layer and thinner bulk layer at a higher S parameter, which means that the microstructure of the SiOCH films was affected layer by layer. A higher surface porosity of the substrates induced thicker transition layer with a higher microporosity, resulting in a thinner bulk layer with a higher microporosity in the process of plasma polymerization. The SiOCH films deposited on MCE-022 with a nodular surface showed the highest S value in the transition layer, indicating that these SiOCH films had the highest microporosity. On the basis of Figure 2b, it appears that the SiOCH films with a pillar structure (Figure 2) had too high of a porosity or pore size to improve the O2/N2 selectivity. Compared with SiOCH/MCE-022, the SiOCH films deposited on the PTFE-02 substrate with a fibrous surface showed lower S values in the transition layer and the SiOCH bulk layer. We infer that a slow deposition rate induced a high packing density and that a fiber-like morphology suppressed the formation of a pillar structure, and these results were reflected on a lower S parameter in the transition layer. Therefore, the SiOCH/PTFE-02 composite membrane showed a high PO2 and an O2/N2 selectivity of 1.9. For the SiOCH/PAN membrane, a fast SiOCH deposition rate of 53.7 (nm/min) led to a loose packing and a SiOCH film having a thicker transition layer Langmuir 2010, 26(22), 17470–17476
Figure 5. R-E plots for plasma-polymerized SiOCH/substrate composite membranes with different substrates.
with a higher S value in comparison with the SiOCH on PTFE-02, even when PAN had a lower surface porosity of 15%. Thicker SiOCH films with a thickness of 3.22 μm caused a lower PO2, whereas those with higher microporosity (higher S parameter) reduced the O2/N2 selectivity. From a comparison of Tables 3 and 4, it can be deduced that the O2/N2 selectivity was inversely proportional to the S parameter for the SiOCH film bulk layer. Figure 5 shows plots of the R parameter as a function of the positron incident energy for four SiOCH/substrates composite membranes. A high R value can be observed at the surface because of back diffusion because R represents the relative amount of 3γ annihilation from large pores (>1 nm) or vacuum.14,23 In (23) Mariazzi, S.; Toniutti, L.; Patel, N.; Brusa, R. S. Appl. Surf. Sci. 2008, 255, 191.
DOI: 10.1021/la102759b
17475
Article
Lo et al.
specific S and W values indicate a significant interface between the plasma-polymerized SiOCH layer and the substrate; this can help us obtain good VEPFIT fitting results easily. On the basis of the marks indicated in Figure 6, we can clearly observe that the characteristics of each layer in the plasma-polymerized SiOCH/ substrate composite membranes are composed of the SiOCH bulk layer, the transition layer, and the substrate.
Conclusions
Figure 6. S-W plots for plasma-polymerized SiOCH/substrate composite membranes with different substrates.
Figure 5, one interesting phenomenon that can be observed is that as the positron penetrated further into the porous MCE and PTFE substrates, R did not increase. This is because a polymer substrate with highly electronegative atoms such as oxygen and fluorine would suppress the formation of positronium; this would reduce the relative amount of 3γ annihilation. However, in the SiOCH bulk layer, the variation in the R parameter can be clearly observed. The SiOCH films deposited on MCE-0025 and PTFE02 gave a lower R value, indicating fewer big pores in these SiOCH films in comparison with the SiOCH films on MCE-022 and PAN. The results are in agreement with the S parameter analysis. Other than the S parameter, we also have the W parameter, which is sensitive to the chemical elements surrounding the annihilation site in materials.9 Therefore, the S-W plot can exhibit the variation in the positron annihilation characteristics in the plasma-polymerized SiOCH/substrate composite membranes. In Figure 6, the S-W plots for two composite membranes show three straight lines with different slopes, indicating three distinct characteristics of positron annihilation in the SiOCH/substrate composite membranes. The plots display two different slopes for the MCE-0025 and PAN polymeric substrates; MCE-0025 had a lower S parameter. Normally, most polymeric materials display similar slopes in S-W plots. But in this case, the MCE-0025 substrate contains positron acceptors (carbonyl groups), which can trap positrons and inhibit the formation of positronium, resulting in a decrease in the S parameter9,24 and different positron annihilation characteristics compared with the PAN substrate. For the SiOCH films, the S-W plots display similar slopes despite the fact that the SiOCH films were deposited on different substrates (MCE-0025 and PAN), which means that both plasmapolymerized SiOCH films had similar chemical structures. Two lines plotted for the SiOCH films and substrates intersecting at (24) Shantarovich, V. P.; Suzuki, T.; He, C.; Gustov, V. W. Radiat. Phys. Chem. 2003, 67, 15.
17476 DOI: 10.1021/la102759b
In this study, the slow positron beam was used to discuss the effect of substrate types on the fine structure of a plasmapolymerized SiOCH layer as a function of depth. As seen in the SEM pictures, the growth of the SiOCH films on different substrates led to hemispherical macrostructures, and the deposition rate was dependent on the surface porosity-related characteristics. The bigger the surface mean pore size in the substrate (MCE-022 and PTFE-02), the smaller the deposition rate of the SiOCH films. It is interesting that the SiOCH deposited on the MCE-022 substrate with a nodular surface had a pillar-like structure. The results of the DBES analysis indicated three positron annihilation characteristics of the plasma-polymerized SiOCH/ substrate composite membranes, which consisted of SiOCH films, a typical polymer PAN, and substrates with positron acceptors such as MCE and PTFE. All SiOCH/substrates composite membranes showed three layers, as labeled in the S-E plot: the SiOCH layer, the transition layer, and the polymer substrate. It was found that a higher surface porosity of substrates induced a thicker transition layer with higher S values (larger microporosity) and resulted in a higher O2 permeability. For the SiOCH films packed on a hydrophobic PTFE-02 with a fiber-like surface, as compared with MCE-022, a slower deposition rate could produce SiOCH films with a higher packing density and suppress the formation of a pillar structure, leading to SiOCH/PTFE-02 membranes with a high PO2 and an O2/N2 selectivity of 1.9. For the SiOCH/PAN membrane, a fast SiOCH deposition rate of 53.7 (nm/min) produced a loose packing, and the resulting SiOCH films had a thick transition layer with a higher S value in comparison with SiOCH on PTFE-02, although PAN had a low surface porosity of 15%. Thicker SiOCH films with a thickness of 3.22 μm caused a lower PO2, whereas those with a higher microporosity (higher S parameter) reduced the O2/N2 selectivity. Therefore, the microstructure of the SiOCH films was affected layer by layer; a higher substrates surface porosity induced a thicker transition layer with a higher microporosity and led to a thinner bulk layer with a higher microporosity during the plasma polymerization. It was observed that the O2/N2 selectivity was inversely proportional to the S parameter. Supporting Information Available: FTIR spectra for SiOCH films deposited on different substrates under the same plasma conditions and S-E plots for four different substrates. This material is available free of charge via the Internet at http://pubs.acs.org.
Langmuir 2010, 26(22), 17470–17476
