J. Phys. Chem. B 2008, 112, 10915–10921
10915
Solvent-Induced Porosity in Ultrathin Amine Plasma Polymer Coatings Krasimir Vasilev,* Leanne Britcher, Ana Casanal, and Hans J. Griesser Ian Wark Research Institute, UniVersity of South Australia, Mawson Lakes, SA 5095, Adelaide, Australia ReceiVed: April 28, 2008; ReVised Manuscript ReceiVed: June 23, 2008
Plasma polymers deposited from n-heptylamine onto silicon wafers have been found to form a porous microstructure when immersed in water and other solvents, with pores of dimensions and densities that vary considerably between coatings deposited under different plasma conditions. This solvent-induced pore formation was found to correlate with the observed percentage of extractable material. With low radio frequency (rf) power inputs, the resultant softer coatings possess considerably more extractable material than coatings deposited at higher applied power levels. The porosity is thus proposed to result from the formation of voids created by the extraction of soluble low-molecular-weight polymeric material, which produces shrinkage stress that the coating, firmly attached to the substrate, cannot relieve by macroscopic contraction. The microscopic contraction of plasma polymer volume creates voids that appear to span the entire film thickness. The effect of aging plasma polymers in air was also investigated. For films deposited at low power it led to reduced extraction of soluble material and different pore morphology, whereas for films deposited at higher rf power levels, the extracted amounts and pore formation were the same for aged coatings. It was also found that the density of surface amine groups was lower for films deposited under the two lowest power settings, in contrast to the commonly held belief that the use of minimal applied rf power aids retention of functional groups. These porous plasma polymer coatings with surface groups suitable for further interfacial chemical immobilization reactions may be useful for various membrane and biotechnology applications. Introduction Plasma polymerization is a convenient and versatile technique for the fabrication of thin polymer films ranging in thickness from a few nanometers to a few micrometers. Plasma polymerization has many advantages for the fabrication of polymeric coatings: it is a solvent-free technology, numerous monomers can be polymerized by this technique, and it is generally fast and a one-step process. In addition, plasma polymer films can be deposited on a wide range of substrate materials and often coating conditions are readily transferable between substrates. The process is not limited to flat substrate surfaces but can also be used with three-dimensional shapes and porous materials such as membranes. All those advantages have led to numerous investigations and applications of coatings prepared by plasma polymerization, for example, as protective coatings1 and insulating layers,2 thin conducting films,3 sensors,4-6 surface modification of membranes7-10 for diverse applications, fuel cells,11,12 and biomaterial coatings.13-16 The fabrication of surfaces containing amine groups by plasma techniques has attracted considerable attention, particularly for biomaterial applications, due to the feasibility of subsequent covalent attachment of biological or biologically relevant molecules. Amine plasma polymers have been employed as a functional interlayer for the fabrication of nonfouling coatings,17-20 for the attachment of proteins,10,16,21 and as a support for lipid bilayers,10,22 lipid vesicles, cells,23-25 and DNA.26 In addition, in aqueous media, the positively charge amine groups allow electrostatic attraction of negatively charged species27 onto the coated surface. * Author to whom correspondence should be addressed. Present address: Mawson Institute, University of South Australia, Mawson Lakes, SA 5095, Adelaide, Australia. E-mail:
[email protected].
A number of monomers have been used to fabricate aminefunctional plasma polymers, including allylamine, propylamine, propargylamine, butylamine, and ethylenediamine, as well as other nitrogen-containing process gases. The fabrication, characterization, biomedical relevance, and aging of aminefunctional plasma polymers have recently been reviewed by Siow et al.28 One highly suitable monomer for the fabrication of aminefunctional plasma polymer films is n-heptylamine. While the density of amine functional groups on its surface29 is less than for other plasma polymers, such as from allylamine, it is worth noting that a high density of surface groups is not necessary to immobilize biological macromolecules such as proteins, because their dimensions still greatly exceed the average spacing between surface amine groups on such plasma polymer interlayers. An attractive feature of this monomer is that, with our system, the plasma polymerization of n-heptylamine provides good cohesive films over a wide range of plasma conditions; the process is much less sensitive to small parameter changes than other amine monomers. In our experience, the surface density of amine groups is rarely a limiting factor, and other issues such as the stability of the coating in aqueous media are more important. Compared with allylamine plasma polymer, n-heptylamine plasma polymer coatings manifest less swelling in water, less extraction of water-soluble material, and less oxidation.30 Moreover, heptylamine is not as volatile as and is less harmful31 than other amine monomers; thereby presenting reduced health and environmental issues. These features, plus the ease of deposition, have led us to utilize n-heptylamine plasma polymer coatings as interlayers for various interfacial immobilization reactions.32-34 While the aging behavior (postplasma oxidation) of n-heptylamine plasma polymers (HApp) has been well-documented,35,36 other chemical and physical aspects require further documenta-
10.1021/jp803678w CCC: $40.75 2008 American Chemical Society Published on Web 08/07/2008
10916 J. Phys. Chem. B, Vol. 112, No. 35, 2008
Vasilev et al.
Figure 1. Rate of deposition as a function of the applied power.
TABLE 1: XPS Elemental Composition of HApp Films Deposited at Different Radio Frequency Power Levels power (W)
%C
%N
%O
10 20 30 40 50
90.1 88.0 86.8 86.2 86.5
7.6 9.6 10.7 11.5 11.6
2.4 2.3 2.5 2.4 1.9
tion in order for these coatings to be used in applications with sufficient predictability and control. The extraction of watersoluble material (presumably lower molecular weight oligomers) from amine plasma polymers has recently been reported.30 We report here that the stability and structure of HApp coatings upon immersion in various solvents varies substantially with deposition conditions, probably via differing cross-link densities. Of particular interest, however, is that such solvent extraction produces topologically nonuniform changes in the films, in that HApp coatings can, in both aqueous and organic solvents, form porous structures. This novel effect may open up new possibilities for engineering functional porous plasma polymer films as ultrathin membranes. To understand the formation of pores, the influence of polymerization conditions and solvent exposure on HApp films 50-60 nm thick was examined by X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared spectroscopy (FTIR) as well as by ellipsometric thickness measurements, and a mechanism is proposed for pore formation and pore-size regulation in the HApp films. Experimental Section Materials. n-Heptylamine (98%, Fluka) and potassium bromide (IR grade, Aldrich) were used as supplied. Silicon wafers (MMRC Pty. Ltd.) were cut to size and then cleaned with piranha solution (Caution: this is strongly exothermic and appropriate care and procedures must be followed), followed by copious rinsing with Milli-Q water. For all dissolution experiments reported here, high-purity water was used. It was produced by the following sequential treatments: reverse osmosis, two stages of mixed-bed ion exchange, two stages of active carbon treatment, and a final filtering step through a 0.22 µm filter. The conductivity was less than 0.5 µS/cm, and the surface tension was 72.8 mN/m at 20 °C. Plasma Polymerization. Plasma polymerization was carried out in a custom-built reactor described elsewhere37 with a 13.56 MHz plasma generator (Advanced Energy). The deposition in all cases was carried out at a pressure of 0.2 Torr. Input power levels of 10, 20, 30, 40, and 50 W were used. The duration of deposition was 90 s in the cases of 10, 20, and 30 W and 40 s for 40 and 50 W. The reason for using different deposition times for the different power levels was to obtain films with thick-
nesses in the range of 40-60 nm, based on preliminary experiments that established deposition rates for each power setting. Dissolution. HA plasma polymer films deposited at different power levels were immersed in high-purity water [extraction of material also occurs in other solvents (unpublished data) but the present data with water are sufficient to illustrate the phenomenon]. Samples were either immersed immediately after plasma deposition or left to age in air for specified time intervals prior to immersion. After soaking in water for selected times, the samples were dried with a stream of clean nitrogen gas. No peeling off the silicon wafer occurred with any of the films; the interference coloring of these films at the thicknesses used for this study provides a ready assessment of any patchy or complete delamination. Characterization. Ellipsometry. An IA commercial imaging ellipsometer (Beaglehole Instruments) was used. The measurements were carried at a constant wavelength of 600 nm as the angles of incidence and reflected light detection were varied between 40° and 85°. A refractive index of 1.55 was used for all samples. This value is based on preliminary unpublished work and is consistent with the refractive indices of similar thin films, as well as providing the best fit of the experimental data. The natural oxide layer on the silicon wafers was measured independently to be 1.57 nm thick, and this was accounted for in the polymer film thickness evaluation. X-ray Photoelectron Spectroscopy. XPS spectra were recorded on a Kratos AXIS Ultra DLD spectrometer with a monochromated Al KR radiation source (hν ) 1486.7 eV) operating at 15 kV and 10 mA. The elements present were identified from a survey spectrum recorded over the energy range 0-1100 eV at a pass energy of 160 eV and a resolution of 1.0 eV. The areas under the photoelectron peaks in the spectrum were used to calculate the percentage atomic concentrations. The error associated with quantification is on the order of 5-10%.38 High-resolution (0.1 eV) spectra were then recorded for pertinent photoelectron peaks at a pass energy of 20 eV to identify the chemical states of each element. All the binding energies (BEs) were referenced to the C 1s neutral carbon peak at 285 eV, to compensate for the effect of surface charging. The analysis area was 700 × 300 µm. Processing and component fitting of the high-resolution spectra was performed with CasaXPS with peak assignments based on previously reported work.38,39 Atomic Force Microscopy. An Asylum MFP-3D atomic force microscope (Asylum Research) was used for inspecting the surface topography of the plasma polymer films. The imaging was done in tapping mode, by use of noncontact “golden” silicon cantilevers (NT-MDT) with a resonance frequency of 255 kHz and a spring constant of 11.6 N/m (both specified by the manufacturer). Infrared Spectroscopy. A Nicolet Magna-IR 750 spectrometer was used for FTIR analyses. HApp films deposited on silicon wafers were immediately placed in an aqueous solution overnight. The solution was mixed with KBr and the aqueous medium was evaporated at 40 °C. The measurements were carried out in reflection mode. Results and Discussion Initial observations of the solvent-induced formation of pores in HA plasma polymer films prompted a systematic study of the phenomenon and its cause; to this end, a series of HApp coatings with different properties, produced by varying plasma deposition conditions, was required. Accordingly, before we
Solvent-Induced Porosity in Plasma Polymers
J. Phys. Chem. B, Vol. 112, No. 35, 2008 10917
Figure 2. N/C atomic ratio as a function of power input for HA plasma polymers.
discuss the elution of low molecular weight polymeric material from HApp coatings and the resultant microstructural changes, we briefly report experiments ensuring the production and characterization of well-controlled, reproducible coatings. Figure 1 shows the rate of deposition of n-heptylamine plasma polymer film as a function of the applied radio frequency (rf) power. The deposition rate was obtained by measuring the film thickness by ellipsometry after deposition at a given power for a specified time. As a general trend, the deposition rate increased with increasing power. At 10 W the deposition rate was ∼0.3 nm/s, and the rate increased to ∼1.1 nm/s at 50 W. The fact that the deposition rate increased with the power input implies that between 10 and 50 W the plasma polymerization process is in the power-limited regime and constructive processes of film build-up dominate over ablative processes. The elemental composition of HApp films deposited at various rf input power settings, obtained by XPS, are summarized in Table 1. The element oxygen, due to the quenching of trapped carbon-centered radicals with atmospheric oxygen following exposure to air,35,36 was present only as a minor component, with the O/C atomic ratio remaining below 0.03 for all of the deposited films. In our laboratory, exposure to ambient air for ∼0.5 h is unavoidable in the transfer from the plasma reactor to the XPS instrument. However, given the very similar O contents of the HApp samples, the O incorporation can be neglected when samples produced under different plasma conditions are compared. Figure 2 shows that the N/C ratio increased with power and at the higher input power levels approached the N/C ratio of the monomer (0.14). This behavior is interesting, as it is normally assumed that there is higher retention of heteroatom functional groups at low plasma power settings; the absence of substantial cleavage and loss of amine groups even at 50 W provides a clue about the reactions and mechanism involved in HApp formation, which is the subject of ongoing study. This is in contrast to allylamine plasma polymerization30,40 and suggests that hydrocarbon fragmentation (homolytic C-H bond scissions) and combination reactions dominate the polymerization. The longer hydrocarbon chain appears to be beneficial for amine retention, by providing a suitable frame for hydrocarbon polymer formation to build the plasma polymer; the probability of a radical being produced by abstraction of the amine group is much reduced compared with lower alkylamines. Thus, n-heptylamine enables the generation of plasma polymers with a density of surface amine groups sufficient for many applications and with a deposition rate higher than for other amine monomers that require low power settings,30 and hence low deposition rates, for analogous retention of the amine functionalities.
Figure 3. C 1s spectra of HApp films deposited at different power settings.
Figure 4. N 1s spectra of HApp films deposited at different power settings.
The C 1s spectra also reflected the change in composition with power, as the component-fitted C 1s spectra (Figure 3) showed an increase in the C-O/C-N component up to 30 W and then changed very little. We did not fit for separate C-N and CdN components at 286 and 286.5 eV, respectively, as had been done by others,39,41 as such fitting is, with the present spectral resolution, prone to relatively large uncertainty and ambiguity. The N 1s peak maxima (Figure 4) increased from 399.0 eV, characteristic of amine groups, for the 10 W sample to 399.5 eV for all other HApp films, which is attributed to an increase in the number of C-N species generated by deposition at higher power levels.35,39,42 However, given the close proximity of the components and the experimental resolution, component fitting is subject to considerable uncertainty, and we prefer to use the spectra of Figure 4 as a qualitative indication of the shift in relative importance of amine and imine groups. These five different HApp coatings were investigated for their behavior when exposed to immersion in water. As it is known that HApp films undergo ambient postplasma oxidation upon storage in air,35,36 it was also tested whether the length of exposure to air would affect the response to immersion. Some films were immersed within 3 h after deposition, while others were immersed after aging in air for 3 days. The film thickness was measured by ellipsometry before and after immersion in water, and the resultant decrease in thickness was calculated. Figure 5 shows the reduction in film thickness as a function of immersion time for films prepared at different rf input power settings. These films were placed in water immediately after deposition. To facilitate comparison, the data were normalized
10918 J. Phys. Chem. B, Vol. 112, No. 35, 2008
Figure 5. Thickness decrease (%) as a function of time of immersion in Milli-Q water for samples prepared with different power settings and immersed in water within 3 h after deposition.
Figure 6. Thickness decrease upon immersion in water of HA polymer films prepared at different power settings. (2) Immersed in water within 3 h after deposition; (b) immersed in water 3 days after deposition.
by plotting the thickness decrease as a percentage rather than as absolute thickness values, since the initial thickness of the coatings was not identical, being in the range 40-60 nm. Samples prepared at low power (10 W) lost up to 55% of their thickness after 1000 min of immersion. Samples prepared at higher power levels were more stable; films fabricated at 20, 30, and 50 W decreased in thickness by 25%, 10%, and
