SEARCH

SEARCH BY CITATION

Keywords:

  • biomaterials;
  • ESCA/XPS;
  • FT-IR;
  • plasma polymerization;
  • pulsed discharge

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Part
  5. Results and Discussion
  6. Conclusion
  7. Acknowledgements

Low-pressure downstream microwave plasma was used to deposit thin water resistant PPAAm and PPAAc films on titanium substrates. Film stability against dissolution was tested under sonication in ultrapure water. Variation of duty cycle on mean power was applied to evaluate the sensitivity of film properties on plasma conditions. Strong cross-linked and therewith water resistant PPAAm and PPAAc films correlate with a relatively low but for biomedical applications sufficient density of functional groups. Furthermore, some similarity exists between process parameter and dissolution stability of deposited plasma polymer films prepared by microwave and reported for radio frequency plasmas.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Part
  5. Results and Discussion
  6. Conclusion
  7. Acknowledgements

Plasma polymer deposition is a promising technique for the preparation of interface layers for the purposes of permanent immobilization of biologically active molecules in different biomedical applications1–3 and of improvement in the performance of living cells at different substrates, e.g., on tissue culture dishes or metal implants.4–6 This is due to the specific surface chemistry of these layers, the ability to deposit very thin pinhole free films on complex surfaces, and the good adherence on substrates with different surface chemistries. Besides these advantages, it is still a challenging task to obtain films with desired chemical surface properties that are stable enough in these applications. While chemical aging effects can be tolerated or simply minimized by pre-aging techniques, insolubility of coatings in aqueous biological solutions is indispensable. Detachment of films by dissolution or leaching of unbound materials has to be avoided.7, 8

Two classes of plasma polymer films are in the focus of biointerface-related research activities: amino group containing and carboxylic group containing films. In both cases, certain precursors receive special attention. Several research groups8–10 have reported on plasma polymerized allylamine by pulsed low-pressure RF plasmas. They have obtained high densities of amino groups along with good structure retention of the monomer due to less fragmentation under very mild plasma conditions. Probably, this effect is due to the double bonds in allylamine, which encourages deposition by a combination of plasma polymerization and conventional free radical polymerization.2

Carboxylated surfaces are a convenient platform for the immobilization of bioactive amino group containing molecules (proteins) via carbodiimide coupling chemistry. A useful plasma-assisted approach to carboxylation is the plasma polymerization of acrylic acid in RF plasmas.11–13 This precursor is reported to result in plasma polymers with more linear chain structures and, therefore, higher carboxylic group densities than other plasma polymers. The reason is a noticeable tendency to thermally induced polymerization.2, 11, 14 High fragmentation at high discharge power of RF plasmas results in greater cross-linking, which is reported to be the basic prerequisite for stable films.15–17 Indeed, at the costs of COOH group retention, insoluble films could be obtained under such conditions. The O[BOND]C[DOUBLE BOND]O group of the C1s XPS peak is between 4% at high discharge power (20 W) and 22% at low power (2 W).18

While the influence of RF plasma conditions on plasma polymer dissolution in water is rather well investigated, little is known about similar investigations for plasma polymers deposited with the help of low-pressure microwave plasmas. Compared to RF plasmas, microwave plasmas typically exhibit higher electron densities, lower mean electron energies, and higher fluxes of ions to surfaces but lower energies of these ions.19 Therefore, effects of neutral radicals can be expected to be more dominant during microwave plasma-assisted deposition. On the contrary, energetic ions could have a strong influence on RF plasma polymer deposition. The latter was demonstrated for both acrylic acid20 and allylamine plasma polymer depositions.21 These are the reasons why these different plasma excitation methods may lead to different film properties. Furthermore, differences in technical processing performance arise that can be useful in either instance.

In preceding investigations, we could demonstrate the usefulness of microwave plasma deposited allylamine polymers in cell culture applications.22–24 Similar results were reported by others.25, 26 Investigations of microwave plasma deposited acrylic acid polymers are underway. Here, we summarize our experiences on dissolution stability for both types of films, which is the prerequisite for all further investigations.

Experimental Part

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Part
  5. Results and Discussion
  6. Conclusion
  7. Acknowledgements

Materials and Methods

Substrates

Polished discs of chemically pure titanium (TiP, 1–3 cm diameter) were used as substrates. While a criss-cross pattern of larger scratches with somewhat less than 1 µm width was still present on the surfaces, the mean roughness of the plain surface areas was Ra = 0.019 µm.

Deposition Procedure

For deposition, a commercial microwave plasma reactor (2.45 GHz) was used that generated disc-like planar plasmas (Plasma Processor V55G, PlasmaFinish, Germany). Substrates were located in a downstream position, related to the plasma (9 cm distance to the microwave coupling window). TiP was decontaminated and activated by a continuous microwave (cw) oxygen plasma (500 W, 50 Pa, 100 sccm O2/25 sccm Ar, 60 s). Immediately thereafter, without breaking the vacuum, the discs were coated with about 20–150 nm thin layer of plasma polymerized allylamine (PPAAm) or acrylic acid (PPAAc). For deposition, cw and pulsed discharge regimes with different plasma-on times and duty cycles (DCs) were used. The DC is the ratio of plasma ton divided by the overall pulse duration ton + toff. Pulse durations and DCs were selected that in principle facilitated good stability of coatings. Typically, argon was used as a carrier gas. Carrier gas flow was controlled by mass flow controllers. The monomer vapor flow was controlled by a calibrated needle valve. Generally, deposition durations were chosen long enough to ensure sufficient layer thickness for XPS analysis (≫10 nm). Homogeneous closed large-area coatings (about 300 cm2) could be obtained for allylamine at a minimum microwave power density of 1 W · cm−2 and for acrylic acid at about 2 W · cm−2.

XPS

The elemental chemical surface composition and chemical binding properties of the surfaces were determined by XPS (AXIS ULTRA spectrometer, Kratos, UK) using the monochromatic AlKα line at 1 486 eV (150 W), implemented charge neutralization, and a pass energy of 80 eV for estimating the chemical elemental composition or of 10 eV for highly resolved C1s and N1s peaks.24 In accordance with the literature,27 for saturated hydrocarbons, such as PPAAm and PPAAc films, the C[BOND]C/C[BOND]H component of the C1s peak was adjusted to 285 eV. The peak positions27 were assigned as follows: C[BOND]NH at 285.7 ± 0.1 eV, C[BOND]OH, C[BOND]O[BOND]C, and C[DOUBLE BOND]N at 286.6 ± 0.2 eV, C[DOUBLE BOND]O at 287.9 ± 0.3 eV, N[BOND]C[DOUBLE BOND]O at 288.0 ± 0.3 eV, and only for PPAAc O[BOND]C[DOUBLE BOND]O at 289.2 ± 0.2 eV. Amino groups were labeled by reaction with 4-trifluoromethyl-benzaldehyde (TFBA) at 40 °C for 2 h in a saturated gas phase. Carboxyl groups were determined by O[BOND]C[DOUBLE BOND]O C1s peak fitting.

FT-IR

The films were investigated by the diamond ATR unit of an FT-IR spectrometer (Spectrum One, Perkin-Elmer, Germany). In this case, Au films, sputtered on small polystyrene, polyetheretherketone, or silicon wafers were used as a basis for PPAAm and PPAAc films to improve sensitivity.

Surface Profilometry

The thickness of the deposited plasma polymer film was determined by two independent methods in parallel. The mass difference due to coating was compared with the height of a step that was prepared on a silicon wafer and measured with a surface profiler Dektak3ST′ (Veeco, USA). These two methods showed consistent results (about 20% deviation).

Film Stability

To explore the film stability, samples were sonicated in an ultrasonic bath (Elma, Germany) in 50 mL ultrapure water for 10 min. In addition, some samples were stored for several hours in deionized water, which was agitated with the help of a wave platform shaker (Heidolph, Germany).

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Part
  5. Results and Discussion
  6. Conclusion
  7. Acknowledgements

The main objective of these investigations was to obtain films that are reliably stable in aqueous environment, while exhibiting acceptable chemical composition. Thus, the tests of film stability in water served as the guiding line for experimentation.

In Figure 1, first afferent results that were obtained by rinsing of PPAAm films with the help of the wave platform shaker are given. Only in this case ammonia was used as carrier gas (10 sccm AAm, 50 sccm NH3), and plasma with 0.1 s plasma-on time and 1.6 s pulse duration (1 200 W) was applied for 20 s total plasma-on time (teff). Apparently, a rapid decrease of primary amino groups occurs from 2.5% initially to about 0.5% within a short period of rinsing. In parallel, a similar loss of oxygen of about 4% was observed. We assume that these effects may be due to the removal of loosely bound oligomeric polymer fragments that were formed by a conventional radical-based chain polymerization with growth termination, e.g., by oxygen attachment. Note that these amino group densities are low compared to 18%, a maximum value reported for a special RF plasma.8 Within the limits of experimental uncertainty, the amino group content does not change anymore after this initial decrease. In contrast, the oxygen content steadily increases. This behavior suggests that the remaining NH2-groups are stably embedded in alkoxy and peroxy radicals containing hydrocarbon matrix, which can further oxidize. Primary amino groups are reported to hardly oxidize to nitrogen and oxygen containing functional groups. While oxidation of neighboring CH2-groups by peroxo biradicals seems to be the main pathway,2 the oxygen uptake is not yet completed after 24 h obviously.28 More than 100 h are reported as typical time spans for radical-related oxygen uptake of pure hydrocarbon plasma polymers until saturation. Saturation values between 5% O/C and 30% O/C are reported to depend on the used monomers.28–30 Hence, the oxygen uptake observed here is very strong. We assume that this effect results from additional hydrocarbon radical integration due to the presence of the additional atomic hydrogen from ammonia dissociation. Less oxygen uptake (from 5 to 15%) was observed in the case of argon carrier gas.

thumbnail image

Figure 1. Surface composition of plasma polymerized allylamine films (PPAAm) in dependence on storage in agitated water, determined by XPS.

Download figure to PowerPoint

While this rinsing procedure lasted for 24 h, similar results (about 1% NH2/C and <20% O/C) could be obtained very quickly within few 10 min in the ultrasonic bath. Certainly, gentle long-time rinsing preferentially puts chemical stress on films. In contrast, sonication adds mechanical stress. Therefore, sonication is a more meaningful test and was used throughout for more detailed investigations.

With respect to the rapid initial decrease of NH2-groups and oxygen, it seems to be of interest that the layer thickness of PPAAm varied with the plasma-on time at teff. For teff = 144 s (at 500 W, 50 Pa; 50 sccm allylamine, 50 sccm argon carrier gas), the smallest layer thickness was obtained in the cw mode (∼20 nm), followed by 100 ms plasma-on mode (∼10–60 nm). The thickest layer with ∼150 nm was observed for the short plasma-on time of 5 ms. This observation suggests the assumption that conventional radical chain polymerization and true plasma polymerization differently contribute to film growth; i.e., short pulse mode would foster the former and cw mode the latter. However, this seems not to be the case. A detailed study23 of DC dependence of chemical composition and stability for different plasma-on times (5, 10, 100, and 300 ms) showed that the loss of nitrogen due to 10 min sonication in ultrapure water is about the same for almost all conditions (∼5% N/C). Only conditions of low DC and short plasma-on time exhibited a slight tendency to instability. As for the results of Figure 1, a final increase of oxygen could be observed (∼5% O/C). However, it also showed no significant difference for the different conditions. Furthermore, only for one single condition, an indication could be detected for film ablation. Only the amino group retention of the as-deposited films and the overall nitrogen content showed certain dependence on deposition conditions. Generally, amino group retention was low. However, a clear influence of both pulse length and DC was observed that did not follow the rules observed for, e.g., RF plasmas.15, 31 These rules predict good retention at short pulse, low DC conditions. In contrast, very low retention (about 0.5% NH2/C) was observed here for both cw and such conditions. Surprisingly, there was a distinct maximum of retention (2.4% NH2/C) at intermediate plasma-on time and DC (100 ms, 0.125). The reason for this behavior is unclear. Anyway, it is a hint at more complex competing deposition mechanisms. Interestingly, complex mechanisms were also suspected for PPAAm deposition from RF plasmas.21

FT-IR investigations showed that as-deposited PPAAm films mainly consist of hydrocarbon, amino, and some other nitrogen-based building units as long as low DCs are used.23 This composition remains almost unchanged if the samples are sonicated and stored in air for a longer time (see Figure 2). Dominant features of the FT-IR spectrum are the stretching vibrations of the aliphatic C-H groups, ν-CH2,3 at 2 980–2 880 cm−1, and deformation vibrations of amines, δ-NH at 1 650–1 510 cm−1. Also, ν-NH stretching vibrations between 3 380 and 3 200 cm−1 are clearly visible. Amino groups were partially transformed into amide, imine, or nitrile functional groups by the plasma process. An indication for this is the new band between 2 300 and 2 200 cm−1 associated with the stretching vibrations of nitrile groups, ν-CN. During the sonication, an incorporation of water occurs and overlapped ν-NH and ν-OH stretching vibrations can be seen in the spectrum. This effect is reversible. After following 2–7 days storage on air, the initial FT-IR curve is restored demonstrating the outstanding film stability.

thumbnail image

Figure 2. FT-IR spectra of PPAAm films as deposited and after 10′ sonication [ultrasonic water bath (USB)] in ultrapure water and subsequent storage on air for 7 days.

Download figure to PowerPoint

A similar, very high stability against water was observed for microwave plasma deposited PPAAc films. Depositions with overall pulse duration of 100 ms, DC variations between 0.1 and 1.0, and teff of 60 s were investigated in more detail (700 W, 20 Pa, about 100 sccm acrylic acid, 20 sccm argon carrier gas). Under these conditions, the film thickness varied between 40 nm (DC = 0.1) and about 10 nm (DC = 1.0). Analysis of highly resolved C1s peaks showed a decrease of peaks related to C[BOND]OH, C[DOUBLE BOND]O, and COO with increasing DC. The carboxyl density was 6.4% COO/C at DC = 0.1 and 3.9% COO/C at DC = 1.0 (compare Figure 3). There were only very small differences between C1s peaks after preparation and after 10 min sonication in ultrapure water, if any. Generally, changes in carboxyl groups were low. A small loss of carboxyl groups at DC = 0.1 (from 6.4 to 5.6% COO/C) may be an indication of some solubility. The slight increase at DC 0.5 (from 3.6 to 4.0% COO/C) and 1.0 (from 3.9 to 4.3% COO/C) may be explained by an oxidation of the surface during sonication.

thumbnail image

Figure 3. Composition of high-resolved XPS C1s peak of plasma polymerized acrylic acid films (PPAAc) in dependence on the DC and the treatment – as deposited and after 10′ in an USB.

Download figure to PowerPoint

No delamination of PPAAc films was observed in all cases. In contrast to PPAAm, the FT-IR spectra showed major differences between acrylic acid monomer H2C[DOUBLE BOND]CH[BOND]COOH and PPAAc, which can be ascribed to strong cross-linking (not shown here).

The FT-IR spectra of PPAAc with DC 0.5 and 1.0 are very similar and confirm the XPS results (see Figure 4). Strong ν-O[BOND]H stretching vibrations at 3 650–3 200 cm−1 verify associated carboxylic acids. This band interferes with ν-CH2,3 stretching vibrations at 2 980–2 880 cm−1. The little so-called combination band at 2 650–2 630 cm−1 is also a characteristic for associated carboxylic acids. Furthermore, the ν-C[DOUBLE BOND]O stretching vibration is to be seen quite clearly at 1 740–1 650 cm−1. δ-CH2,3 and ν-CO and δ-C[BOND]OH vibrations are superimposed in the fingerprint region. Again, certain indications of slight DC dependence can be seen (e.g., a small shift of the ν-OH band). The position of different FT-IR bands such as ν-OH (3 650–3 200 cm−1), combination band (2 650–2 630 cm−1), and ν-C[DOUBLE BOND]O (1 740–1 650 cm−1) and the DC correlate with each other. The FT-IR spectrum of DC 0.1 differs only marginally in the fingerprint region due to a higher layer thickness.

thumbnail image

Figure 4. FT-IR spectra of PPAAc in dependence on the DC.

Download figure to PowerPoint

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Part
  5. Results and Discussion
  6. Conclusion
  7. Acknowledgements

Microwave plasma-assisted deposition was used to deposit thin water resistant PPAAm and PPAAc films on titanium substrates. The stability of these films against the dissolution was tested under the harsh conditions of 10 min sonication in ultrapure water. XPS and FT-IR absorption were used to characterize film properties. Pulsed plasmas with different DCs were applied to evaluate sensitivity of film properties on plasma conditions, especially on mean applied power, which is known to be a means for controlling the retention of chemical structure of the monomer. Under most deposition conditions, films were resistant to delamination in the ultrasonic bath. Likewise, only low retention of the characteristic functional group of the monomer could be observed for most conditions. For PPAAm films, a distinct maximum of NH2-groups at intermediate plasma-on time and DC could be observed. This may be a hint at more complex deposition mechanisms. In dissolution tests, PPAAm films showed a certain loss of nitrogen and an increase of oxygen, which was largely independent of deposition conditions. PPAAc films were rather stable for all conditions.

Altogether, the results show that there is some similarity between the processing parameter dependence of dissolution stability of microwave plasma deposited plasma polymer films and the respective dependences reported for radio frequency plasma deposited films. Also, the retention of characteristic groups might be similar for stable films.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Experimental Part
  5. Results and Discussion
  6. Conclusion
  7. Acknowledgements

The authors thank their technicians Urte Kellner, Uwe Lindemann, and Gerd Friedrichs for excellent assistance. Financial support by the program TEAM of Mecklenburg–Vorpommern in Germany (UR 0402210) is gratefully acknowledged.