HUNGARIAN JOURNAL OF INDUSTRIAL CHEMISTRY VESZPREM Vol. 30. pp. 275-280 (2002) SURFACE PROPERTIES AND TISSUE COMPATIBILITY OF NH3-PLASMA MODIFIED POL YETHYLENTEREPHTALAT :MEMBRANES ernstitute of Chemistry, GKSS Research Centre, Kantstrasse 55, D-145613 Teltow, GERMANY 2Institute of Chemical Engineering, University ofVeszprem, Egyetem u. 10, H-8200 Veszprem, HUNGARY) Received: October 14, 2002 Microwave surface plasma treatment was used to improve the tissue compatibility of.polymeric me~branes. Chang~s of surface properties, adhesion and spreading of human skin fibroblasts were studied on ammoma-plasr_na modtfied polyethylentetephtalat (PET) membranes. The variation of the microwave (MW)-plasr_na par~eters -_duratiOn, f~ed-gas composition (Ar/NH3) and microwave power- caused a significant change of t~e physicocheffilcal surface properties .. An enhanced cell-surface interaction was observed based on the number of adhenng fibroblasts, but also on morphological criteria including overall cell morphology. It was found that fibroblasts adsorbed better on plasma modified hydrophilic surfaces with a relatively high amine content than on the original membrane. Keywords: NH3-plasma, surface functionalisation, wettability, tissue compatibility, cell morphology Introduction Polymeric membranes play a crucial role for the development of biohybrid organs in tissue engineering and also in biotechnological.applications. In addition to specific bulk characteristics like permeability and selectivity they must possess an excellent biocompatibility. Substitutes for cell cultures have to support cell attachment and proliferation as well as to assure the delivery of nutrients and oxygen and the removal of metabolites [1, 2, 3]. There are several polymeric materials available for the production of synthetic membranes with diverse mechanical properties and permeation characteristic according to special needs, but most of them possess poor characteristics regarding to cellular interaction and function f4]. Attachment, spreading, and proliferation of anchorage dependent cells are highly dependent on the physicochemical properties of the biomaterial surface [5]. Several authors have already reported on enhanced cell adhesion on hydrophilic surfaces [1, 6, 7]. The presence of certain functional groups (eg. amines) [8, 9], or immobilised adhesive proteins [10], surface charge [11], and morphology [12] were also found to regulate cell adhesion. A number of physical and chemical techniques have already been applied for the modification of the topmost layer of polymeric materials ranging from methods of Contact information: E-mail: Thomas.Weigel@gkss.de conventional wet chemistry to novel methods of plasma chemistry [13]. In the last decade techniques using electromagnetic irradiation to induce chemical reactions are getting even more frequent. Such methods like plasma treatment, plasma polymerisation, and ion irradiation, are very attractive, and favourable in modifying samples with a chemically resistive surface or complex shape [6, 14, 15]. If a surface is exposed to a non-polymer forming plasma reactive plasma-species interact with the polymer surfaces and new functional groups are formed. As a result, such impl.antation reactions lead to remarkable changes of physical and chemical surface properties, and are frequently used to improve adhesion properties and wettability. Ammonia plasma treatment of polymeric. s.urfaces leads to . the incorporation of nitrogen contatmng groups - amme, amide, imin etc. - resulting a hydrophilic surface and supports cell adhesion and growth [9, 16, 17]. In frames of our present work polyethylene- terephtalate (PET) track-etched membrane~ were modified by low-temperature NH3-plasma to tmprove the ceU adhesive properties. Polyesters are widely used in the biomedical praxis for catheters) vascular grafts {15] or for joint protheses {18]. Because of the relatively low production costs, mechanical properties. uniform pore size distribution track--etch~ PET -me~bra~s could be very attractive for tissue engmeermg applications such as the production of biohybrid organs. We investigated the influence of different process 276 parameters on the amine concentration of polyester surfaces. Samples having the highest amine concentration were further characterised by means of X- ray photoelectron spectroscopy (XPS), and fluorescence labelling. The wettability of samples chosen for biocompatibility tests was also investigated by water-air contact angle (captive bubble) measurements. Biocompatibility was evaluated by studying adhesion and proliferation of human skin fibroblasts on modified PET surfaces. Cell adhesion was further characterised studying the overall cell morphology of adhering cells. Methods and materials PET-membrane PET track-etched microfiltration membranes with an effective thickness of 20 f.!m and nominal pore diameter of 1 f.!m were purchased from Oxyphem (GroJ3erkmannsdorf, Germany). All membranes were cleaned ultrasonically for 5 min in pure ethanol bath and dried in an exsiccator for 15 min before every treatment. MW-plasma treatment The plasma treatment was performed in a flow type cylindrical MW -plasma reactor operating at 2.45 GHz (modified Plaslan 500, JE PlasmaConsult, Wuppertal, Germany) [19]. Samples were placed in the post discharge region, 6 em downstream from the plasma chamber. The reactor was evacuated down to the base pressure of 1 o-3 bar for 10 minutes, followed by a purge with Ar for 10 other minutes. The flow rate of the AriNHrmixture was then adjusted and the plasma ignited. Surface chemical characterisation The chemical composition of the sample surfaces was determined by X-ray photoelectron spectroscopy (XPS) [20]. The total amine (primary, secondary. and ternary) concentration of surfaces was determined by colorimetric staining with Acid Orange II (AO) [21] and primary amines were labelled with Fluram® to obtain relative concentrations [22]. Surface physical characterisation Wettability was determined by water air contact angle measurements using the captive bubble methAr= 250 seem respectively. Deviation within 10% composition on the amine concentration of the polymer surface was studied. It was observed that at high MW- power (1.2 kW) the amine concentration increased proportionally to the NH3 content in the process gas. The highest concentration was reached using a pure NHrplasma. In this case the coupled energy was high enough to assure an appropriate ion density and energy of the reactive gas ammonia, in the plasma. On the contrary at low (0.4 kW) energies when plasma was operated with pure ammonia the coupled energy might not have been enough for an adequate ionisation of the reactive gas, indicated by the relatively low amine concentrations obtained. The mixing of 160 °0 -0 140 ·;; & 120 .s = .s ~ -~ 80 Q.l ~ 60 t,, ~ 40 ~ = ~ 20 0 -'----o-T·- . 0 r- -- Oo .. ,..."~--.. -- 277 420 440 460 480 500 520 540 560 580 600 Emission wavelength (nm) Fig.3 Fluorescence emission spectra of plasma modified PET membranes with Fluram® labelling of primary amines (excitation at 335 nm, emission at 467 nm) argon to the process gas up to 80 % promoted the implantation of amine functionalities possibly resulted by a higher ion density in the plasma (Fig. I). Treatment time dependency of amine concentration is depicted in Fig.2. Already after a short treatment amine groups were incorporated in the samples, and parallel with a longer treatment time an increase of the amine concentration was observed. It is clearly shown that iiTespective of the plasma composition (MW- power, and process gas composition) a maximum (15 ± 0.8 nm/cm2) concentration was reached. ThiSc occurred however earlier when samples were treated with high energetic plasma (1.2 kW) - containing more high energetic reactive species - than modified al "milder" conditions (0.4 kW). Nevertheless a further increase of the treatment time led to a loss of amine functionalities. Surface modification by low-pressure plasma can be described as a dynamic equilibrium 1.)f competing functionalisation and degradation processc:~ [14]. Our results indicate that after a longer treatmem surface degradation, while in the initial phase implantation was predominant. Samples (PETl, PET2 and PET3) - treated under different plasma conditions - having a relatively high amine concentration were further characterised regarding their primary amine content (Fig.3) and elemental composition. The highest p-amine concentration was found on sample PET l, modified at 0.4 kW whereas treatments at higher energies led to a considerably lower p-amine content (Table 1). Results of XPS-measurements are shown in Table 1 ° After exposure to high energetic plasma a relativel~ high amount of nitrogen was found on PET3 (3.6 at%) and only 1.4 at% on sample PETl (Table/). Oxygen abstraction was also observed. the total oxygen content decreased drastically in all cases. From 28 at% of the m1treated sample it was reduced to I 1.4 at% on PET3 and to 13.6 at% on PETl. It is well known when ammonia is exposed to a plasma. fragmentation, ionisation, excitation and radical 278 70 60 ·~50 Q; ~ ~ 40 t:>J) =