Introduction

Among all synthetic polymers, polypropylene (PP) is one of the most popular thermoplastic polymers used in textiles, medical devices, food packaging and many other products [1]. Since PP is a low-cost polymer with excellent mechanical properties and high chemical resistance, it has become the best candidate for hygiene, healthcare and medical products, especially in the form of non-woven fabrics that are moisture-proof, breathable, flexible, lightweight and thin. Key products include liquid-proofing disposable surgical and isolation gowns, drapes, face masks, surgical caps, disposable bed sheets, pharmaceutical filter media, and diapers [2]. PP here takes advantage of its inertness, low toxicity and lack of interactions with biological media, but in specific biomedical applications such as regenerative medicine or tissue engineering, the interactions of PP-based substrate with cells and tissues can be desired. However, to retain the beneficial bulk properties of PP, it is necessary to use a simple, fast, effective and gentle surface modification technique to produce a highly reactive substrate from completely inert material.

Non-thermal plasma treatment is the most frequently used process today to modify or activate the surface of polymers [3]. Plasma treatment can be used to graft different kinds of chemical groups to the surfaces, which will improve adhesion and hydrophilicity, but this process of surface functionalization is usually incidental. However, for special applications such as medical engineering and biotechnology, well-defined surfaces are necessary [4]. In medical applications, the most required group is a carboxylic acid group (–COOH), since the high-density incorporation of COOH groups allows good adhesion and proliferation of fibroblast and other cells [5]. Acrylic acid (AAc) is often used as a precursor to produce polyacrylic acid (PAAc) films, coatings, or hydrogels with a high amount of COOH groups on the surface. Carboxylic-rich surfaces, having a good adhesion, have found applications in biomedicine as support for the immobilization of biomolecules, biosensors, biocompatible surfaces for implants as well as cell culture substrates.

There are several known methods for grafting the monomer using plasma. It is possible to graft from the liquid phase by dipping the activated polymer into the solution containing the reactive precursor or from the gas phase by exposing the activated substrate to vapour precursor. This method is so-called post-irradiation grafting. On the other hand, the monomer can also be added directly to the carrier gas, which comes into contact with the high-energy particles of plasma to form a plasma-polymerized coating on the surface. Such a plasma polymerization process involves monomer fragmentation, subsequent recombination, cross-linking and incorporation into the substrate in one step and depends on many plasma operation parameters (power modulation, power supply, carrier gas, monomer flow rate, temperature etc.). The deposition of the plasma polymerized AAc (ppAAc) layer on different polymer substrates was usually performed with plasma discharges operated at low pressure in old studies [6,7,8,9,10,11]. Argon or helium plasma is widely used for generating the radicals on the surface able to initiate the grafting of monomers similar to the work of Johnsen et al. [6] and Wang et al. [8]. They first activated the polymeric substrate with RF plasma, then exposed it to air to form peroxides on the surface that can initiate the graft polymerization of AAc. Another paper describes the activation of non-woven PP textiles by low-pressure argon plasma follow by immersing in AAc solution and additionally plasma curing to achieve polymerization. The obtained samples were used for the immobilization of biomolecules (gentamicin, heparin) to improve their anticoagulant and antibacterial properties [12]. To reduce the initial cost of the process, recent publications have used medium or atmospheric-pressure plasma sources. Recently, an atmospheric pressure AC excited dielectric barrier discharge plasma reactor was used for the deposition of AAc-rich films on plasma-pretreated substrates in different carrier gases (Ar, air, O2, Ar + O2). Depending on the gas used, the resulting ppAAc layers differed in their physicochemical properties and cell adhesion capacity. Layers prepared from a mixture of argon and oxygen significantly stimulated cell viability compared to other ppAAc coatings [13]. Cools et al. used a parallel-plate dielectric barrier discharge and attempted to optimize the plasma polymerization process by changing the discharge power and monomer flow rate to obtain the desired stable coatings suitable as support for cells adhesion and proliferation [14].

In our study, we used a new approach for AAc plasma polymerization and grafting of nonwovens. We used a pulsed underwater diaphragm electrical discharge generated directly in an aqueous AAc solution. This resulted in a one-step process of plasma surface activation and simultaneous plasma grafting of the PP nonwoven fabric. Moreover, the treatment of the nonwoven substrate could be performed in a continuous regime, which implies the possibility of a future up-scaling of this technique to meet industrial requirements. This type of discharge generated directly in distilled water has previously been attempted for the hydrophilic surface treatment of PP non-woven fabric [15] and generated in aqueous AAc solution for AAc coating but only to PP multifilament fibres [16], not to nonwoven fabrics. We have successfully customized and optimized the pulsed underwater diaphragm discharge to achieve the plasma-induced polymerization and grafting of AAc on the surface of PP nonwoven fabrics, representing a significantly wider material than fibres. The modified PP textile surface has been investigated by several diagnostics methods: the critical wetting surface tension method (CWST), the standard method of strike-through time (STT) measurement and the change in adhesion properties was studied by adhesion „tape-peel wheel“ test. The chemical composition of the grafted PP surface was investigated by X-ray Photoelectron Spectroscopy (XPS) and Fourier transform infrared spectroscopy in Attenuated total reflectance mode (ATR-FTIR), and changes in morphology were evaluated by Scanning Electron Microscopy (SEM). The plasma polymerized AAc layers were repeatedly verified after some weeks to monitor the surface stability. The present new method of AAc polymerization in underwater plasma discharge is suitable for continuous modification of polymer substrates to increase wettability permanently and incorporate -COOH groups for subsequent immobilization of biomolecules or utilization in regenerative medicine.

Experimental

Experimental Set-up

The plasma treatment was carried out using a diaphragm discharge reactor which is illustrated schematically in Fig. 1a. The main part of the reactor is a rectangular 500 ml plastic container, in which thin planar metal electrodes made from stainless non-magnetic steel (EN 1.4306) resistant to the corrosive solutions are attached on the opposite sides. One of the electrodes is grounded, the other is connected to a pulsed high-voltage source (HV). In the middle of the reactor, between the electrodes, there is a dielectric diaphragm made of 3 mm thick Plexiglas with a narrow horizontal slit of 29 mm. The reactor is filled with a conductive solution so that both electrodes and the slit are immersed (Fig. 1a). The reactor is equipped with a winding device with two rotating regular rollers whose diameter allows the speed of sample processing to be adjusted.

The diaphragm discharge was fed by a thyratron high-voltage pulse source (developed by the Institute of Plasma Physics, the Czech Academy of Sciences, Prague, Czech Republic). The electrical parameters of the discharge were measured by Pearson current probe Model 4100 and two Tektronix P6015A high voltage probes (1000:1), using the Tektronix TDS 2014B digitizing oscilloscope in aqueous solution of NaCl (1 g/l) with a specific conductivity of 1442 µS/cm and all studied concentrations of AAc aqueous solution. The frequency, peak voltage, rise time and half-width of the pulse were 100 Hz, 25 kV, 75 ns and 400 ns, respectively. A picture illustrating the visual appearance of the discharge in tap water, NaCl solution and AAc solution are shown in Fig. 1b, c and d, respectively.

Fig. 1
figure 1

a Schematic draw of experimental arrangement of the diaphragm discharge; b photo of the plasma generated in the tap water, c 1% NaCl, d 20% AAc solution

Full size image

Material

Acrylic acid polymerization initiated by diaphragm discharge plasma was studied on the samples of industrial spun-bonded polypropylene nonwoven textile (PPNT) 15 g/m2 from Pegas Nonwovens s.r.o. (Znojmo, Czech Republic). Since it is a nonwoven fabric, the individual fibres do not have any regular arrangement, so the PPNT surface is considerably porous and thus non-uniform. PPNT was cut into long narrow strips that corresponded to the size of the slit (diaphragm aperture) with a length of about 80 cm and a width of 29 mm. The used monomer was commercially available stabilized acrylic acid, supplied by Merck comp. s.r.o, CAS No. 79-10-7, EC No. 201-177-9 (Germany).

Plasma Treatment

The stripped PPNT samples were passed through the slit of the diaphragm discharge reactor by both guide rollers. Depending on the rotation speed of the winding device, the processing time was adjustable to 3.5 and 5 s, which corresponded to the movement speeds of the non-woven fabric ˜ 0.9 mm/s for 3.5 s and 0.6 mm/s for 5.0 s.

Four different concentrations of AAc solution in deionized water were investigated, namely 5%, 10%, 20% and 30%. The aim of the selected concentrations was to cover a wider conductivity range of the solutions and thus to examine the effect of the AAc concentration in the diaphragm discharge solution on the process of plasma-initiated polymerization on the PPNT surface. The electrical conductivity of the used AAc water solutions was measured with an Eutech CON 150 conductometer. The value of the specific electrical conductivity of deionized water was 1.19 µS/cm.

After the diaphragm plasma treatment, half of the treated samples were rinsed twice in distilled water (at room temperature) to remove unbound AAc as well as its homopolymer. The other half remained without rinsing. All samples were dried in an oven at 60 °C. The non-rinsed samples were dried for 20 min and rinsed samples up to 30 min. After drying, the samples were analyzed by the selected surface diagnostic methods to determine changes in the surface chemistry and morphology.

Surface Diagnostics

Wettability Measurements

The easiest way to verify the changes in hydrophilicity of the AAc grafted PPNT samples is to use strike-through time (STT) measurement. This measurement is based on ISO 9073-8 technical standard, which defines the experimental equipment.

The device consists of a 500 g flow plate with a cylindrical hole. Below the flow plate is the sample to be measured placed, which is pressed onto five filter papers (MH989 Ahlstrom Barcelona SAU) with dimensions of (100 × 100) mm. The cylindrical hole contains two electrodes connected to an electrical timer and a conductive solution ensures circuit switching. The conductive solution of NaCl salt in distilled water (9 g/l - physiological solution) with a volume of 5 ml is poured into the hole. Consequently, the time corresponding to the switch-on is equal to the time of strike-through required to flow the solution through the sample.

Modified PPNTs that have a strike-through time below 5 s are considered ideally hydrophilic.

Using the critical wetting surface tension (CWST) method, we can determine the surface energy of the investigated PPNT samples. The method requires several test measurements. 10 drops of a test liquid with a predetermined surface tension are dropped onto the sample, and after 10 min, it is evaluated whether the surface has been soaked by the liquid. This evaluation depends on the number and shape of the drops. If at least 9 of the 10 drops wet the surface, it means that their contact angle is less than 90°, and then the surface tension of the test liquid is considered to be the lower limit of the PPNT surface energy. Conversely, if at least 9 of the 10 drops do not wet the surface, their contact angle is greater than 90°, the surface tension value of the test liquid represents the upper limit of the surface energy, and thus the surface tension of the liquid exceeds the surface energy of the sample. The measurements are repeated with other test liquids, which differ from each other by 2–4 mN/m to achieve the narrowest difference between the upper and lower surface tension limits. The obtained cut-off values are then averaged to obtain a surface tension value of the liquid, which correlates with the surface energy of the PPNT.

Tape-peel Wheel Test

We investigated the adhesion between the plasma polymerized AAc (ppAAc) layer and the PPNT materials using the tape-peel wheel test with an Instron 4301 (UK) tensile tester. A 15 mm wide adhesive tape (Aerotape 24-1533) was applied to the control and grafted samples. The end of the adhesive tape was gripped into the jaws of the dynamometer and peeled with a pulling speed of 10 mm/min at a 90° angle. The output of the measurements is the dependence of the average value of the peel strength and the standard deviation obtained from the PPNT test strip surface treated under certain conditions.

Scanning Electron Microscopy (SEM)

The changes in the morphology of PPNT after grafting of ppAAc layer were studied by scanning electron microscopy (SEM), which was performed using Vega II SBH (Tescan, CZ). We investigated three concentrations of AAc solution (5%, 20% and 30%) with the same treatment time of 5 s. The samples were rinsed in deionized water in an ultrasonic bath for 5 min. To obtain SEM images of the cross-section of the fibres, treated PPNT samples were frozen with liquid nitrogen and then cut transversely. All samples were plated before the measurements.

X-ray Photoelectron Spectroscopy (XPS)

We used XPS surface analysis as a suitable method to determine the chemical composition of the surface as well as the chemical bonds and the electronic state of the individual elements on the surface. The XPS measurements were performed using the ESCALAB 250Xi device (Thermo Fisher Scientific, East Grinstead, United Kingdom at CEPLANT, the Faculty of Science, Masaryk University in Brno). An X-ray beam with a power of 200 W (650 microns spot size) was used. The survey spectra were acquired with a pass energy of 50 eV and energy step of 1 eV. High-resolution scans were acquired with a pass energy of 20 eV and an energy step of 0.1 eV. In order to compensate for charges on the surface, an electron flood gun was used. The spectra were referenced to the hydrocarbon component C 1s, which was set to a binding energy of 284.8 eV. Spectra calibration, processing and fitting routines were done using Avantage software.

Samples of ppAAc layers on PPNT prepared from all AAc solution concentrations in 5 s plasma process time were used for the XPS measurements. XPS was measured at two points of the sample surface, whose values of a given elemental representation were averaged. The obtained data were measured 24 h after processing.

Attenuated Total Reflectance-Fourier Transform Infrared (ATR-FTIR) Spectroscopy

The infrared spectra were measured on the surface of the samples with the Bruker Vertex 80 V spectrometer in ATR mode using a diamond crystal. Fine contact between the ATR crystal and the sample was achieved by using a pressure clamp with a well-defined pressure regulator. To suppress unwanted effects due to ambient humidity and CO2, all measurements were performed at a vacuum of 2.51 hPa. Each spectrum was scanned 50 times and repeated at three different spots. The spectra were evaluated with the Bruker OPUS software (version 6.5) using the rubber band baseline correction.

Results and Discussion

Plasma Characteristics

The pulsed underwater diaphragm electrical discharge was characterised by electrical measurements for investigated aqueous solutions of AAc with different electrical conductivities compared to 1% NaCl solution. This solution was determined as the reference one without AAc presence, in which the best hydrophilic treatment of PPNT using the diaphragm discharge was obtained within the preparatory experiments. Typical HV pulse and current waveforms displayed in Fig. 2 illustrate the behaviour of the diaphragm discharge ignited in the narrow-slit region. Figure 2a shows the time evolution of the voltage and current of the diaphragm discharge generated in the aqueous solution of NaCl (1%), revealing two almost equal current peaks. The first peak reached 8 A with a rising time of 57 ns, the second with 7.5 A occurred at 104 ns. The shape of the time evolution of the voltage and mainly the current of the diaphragm discharge generated in different concentrations of AAc water solutions (Fig. 2b, c, d and e) differed from that generated in the NaCl solution (Fig. 2a). The differences were mainly because the current waveforms did not show equivalent current peaks. At each concentration of the AAc solution, the value of the first maximum was almost half the value of the second maximum. At the same time, the magnitude of the second maximum was twice as high as the first maximum for the burst of discharge in the 5% and 30% AAc solution; it increased up to three times for the 10% and 20% solutions. The differences in the current waveforms related to the different conductivity of the studied AAc solutions and the 1% NaCl solution, which affects the overall burning regime of the diaphragm discharge in terms of the number of filaments, their length, as well as the widespread distribution in the diaphragm horizontal slit, as can be visually seen in Fig. 1b-d. It was confirmed that as the conductivity of the used solution increases, the current peaks also increase reaching the highest values for 10% and 20% AAc solutions, which correspond to the highest conductivity values of 2.39 mS/cm and 2.14 mS/cm compared to the conductivity values of 1.85 mS/cm for 5% AAc and 1.7 mS/cm for 30% AAc, respectively. Based on the measured current and voltage waveforms and the definition of the electric current power, it can be estimated that the discharge power value reached the level of several watts with a fluctuation depending on the used aqueous solution. The maximum estimated power values were 10–15 W for the 20% and 10% AAc solutions.

Fig. 2
figure 2

Voltage and current waveforms vs. time for the diaphragm discharge burning in a 1% NaCl solution, b 5% AAc, c 10% AAc, d 20% AAc and e 30% AAc water solution

Full size image

Wettability and Surface Free Energy (SFE) Measurements

Strike-through Time Measurements

The measured strike-through time (STT) values of the PPNT samples after plasma-initiated polymerization of AAc were evaluated as a function of the AAc concentration in the solution both for the exposure times of 3.5 s, 5 s, and for different time intervals after treatment (different ageing times). STT of the untreated PPNT sample (reference) has the value of 9.2 s ± 1.2 s which corresponds to an insufficiently hydrophilic surface. After the diaphragm discharge treatment in AAc aqueous solution, the STT decreased significantly for all measured samples. As Fig. 3 illustrates, the most significant decrease in STT was observed immediately after treatment for the non-rinsed sample treated in a 20% AAc solution. When treated for 3.5 and 5 s treatment, the STT was ˜ 2 s ± 0.2 s. Since the STT value of 5 s is considered as the cut-off value between the poor and highly hydrophilic samples, the STT values measured for the treated samples refer to the highly hydrophilic surface. Similar values in the range of 2-3.5 s were obtained at concentrations of 10, 20 and 30% AAc solution. Only at applied concentration of 5% at processing time 3.5 s, the resulting STT values are higher than hydrophilicity threshold of 5 s. For samples prepared using the low concentration of AAc (5%), we can predicate the insufficient coating of PPNT considering the large area of the substrate. Accordingly, inhomogeneous coating using the low AAc concentration led to big error bars compared to higher concentrations. Further, the results show that STT of PPNT samples rinsed 2 times in distilled water increased slightly as excess unbound acrylic acid was removed from the surface of the tested PPNT samples. The samples prepared in AAc solution with concentrations of 10–30% measured after 4 weeks retained their hydrophilicity with values of STT in the range of 1.9-3 s. In the case of plasma treated polymeric substrate, the induced hydrophilicity usually decreased during long storage due to the hydrophobic recovery phenomenon. The retention of the achieved STT values after 4 weeks in our case is an indirect proof of the successful coating of the PPNT substrate with plasma-polymerized AAc layer, which was stable during the storage. On the other hand, the samples produced from 5% AAc solution, which were not ideally hydrophilic immediately after plasma operation, increased in hydrophilicity after 4 weeks (Fig. 3b and d). Such fluctuation in wettability during storage under unspecified laboratory conditions can generally occur due to oxidation by air humidity or contamination of the samples with organic material. Since we don’t have the chemical characterization of prepared ppAAc layers after 4 weeks of storage, we are not able to exactly explain this phenomenon to avoid speculations.

Fig. 3
figure 3

The strike-through time of PPNT samples after plasma-initiated polymerization of AAc by diaphragm discharge as a function of the concentration of AAc in solution for the treatment time a 3.5 s immediately after plasma treatment, b after 4 weeks ageing, and for c 5 s immediately after plasma treatment, d after 4 weeks ageing

Full size image

CWST Method of Surface Free Energy Evaluation

The surface free energy (SFE) value for the PPNT reference sample was (36 ± 2.8) mJ/m2. After plasma-initiated grafting of AAc on PPNT using underwater diaphragm discharge, the SFE increased. As can be seen in Fig. 4a, the largest changes occurred in non-rinsed PPNT samples treated with 20% and 30% AAc, although it was not possible to determine the exact SFE value of these samples. The highest available value of the surface tension of the test liquid was 96.6 mN/m when the liquid immediately wetted the surface. This means that there was a significant increase in surface energy after treatment compared to the reference PPNT. However, after rinsing these samples, a significant decrease in SFE was observed to ˜ 50 mJ/m2, which was lower than the SFE values of the samples prepared with 5% and 10% AAc solution. The obtained values of surface energy correlate with the measured STT values. Therefore, for samples where the STT was decreased, the surface energy of the investigated surface increased.

Tape-peel Wheel Test of Adhesion Properties

Figure 4b shows the results of a peel strength measurement using a tape-peel wheel test. The PPNT reference sample had an average peel strength of the test tape of 3.8 N/m, which can be considered a relatively weakly adherent surface. Already after a simple treatment of PPNT by the plasma generated using underwater diaphragm electrical discharge in 1% NaCl solution, the peel strength increased twice compared to the reference. Additionally, after AAc polymerization and grafting onto the PPNT surface, a significantly higher average peel strength was observed for all treated samples. The most significant increase was measured for non-rinsed PPNT treated with 5% AAc solution plasma with an average peel strength of ˜ 8.9 N/m. This sample even showed detachment of individual fibres from the fabric when the tape was peeled (Fig. 4c). This is due to a significant increase in the surface adhesion of the PPNT fibres treated with diaphragm discharge ignited in AAc water solution, which adheres to the test tape with a significantly greater force than the strength of their bonding with other fibres. The measured peel strength of all rinsed PPNT samples was lower by an average of 1.6 N/m compared to non-rinsed samples. This difference could be due to the fact that some of the unbound AAc was removed from the surface of the PPNT.

Fig. 4
figure 4

a The surface free energy of PPNT immediately after plasma-initiated polymerization of AAc depending on AAc concentration using plasma processing time 5 s; b the average peel strength of the test strip of plasma polymerized PPNT surface in all AAc concentrations compared to peel strength achieved for PPNT treated with distilled water and 1% NaCl solution; c picture of pulling off fibres from PPNT surface after plasma-induced polymerization in 5% AAc solution (non-rinsed sample) demonstrating a significant increase in the surface adhesion of the PPNT fibres

Full size image

Scanning Electron Microscopy (SEM)

SEM measurements provided visual evidence of the successful formation of the ppAAc layer attached to the PPNT fibres (Fig. 5). The image of the reference sample showed separated PPNT fibres with a smooth surface. In contrast, after plasma polymerization using diaphragm discharge ignited in AAc solutions, a ppAAc layer formed on the individual PPNT fibres even at the low concentration (5%). After the plasma polymerization, the originally independent, separated fibres were interconnected through the ppAAc layer. Thus, from the highly porous nonuniform PPNT surface, we obtained a surface that had higher homogeneity after treatment. To confirm the stability of the prepared ppAAc layer, the samples were additionally rinsed with the ultrasonic cleaner. Figure 5c and f clearly show that the ppAAc layer has not been removed even by two rinsing cycles in the ultrasonic cleaner. Thus, the diaphragm discharge-initiated polymerization of acrylic acid can be considered stable. When PPNT was treated in 30% aqueous AAc, it was also possible to measure the thickness of the ppAAc layer formed, which was 248 nm after applying the rinsing cycles (Fig. 5d).

Fig. 5
figure 5

SEM images of a reference PPNT, b ppAAc layer from 5% solution (non-rinsed), c ppAAc layer from 5% solution (rinsed), d ppAAc layer from 30% solution after 2 rinsing cycles with marked thickness of the layer, e ppAAc layer from 30% solution (non-rinsed), f ppAAc layer from 30% solution (rinsed)

Full size image

Surface Chemical Composition

We performed ATR-FTIR and XPS measurements to monitor the process of AAc polymerization and grafting to PPNT fabric during diaphragm discharge operated in the water solution of different AAc concentrations (5–30%). Due to the high density of liquid medium and high electrical conductivity, but lower diffusiveness and mobility of ions and molecules on the other hand, the electrical discharges generated in a liquid medium present an appropriate environment in which the physical and chemical reactions can take place in several reaction channels. The generation of underwater plasma discharge induces the formation of active species as •H, •O and •OH radicals as well as molecular H2O2 [17]. These oxidizing agents react with each other or with surrounding molecules and offer a wide range of possible reactions responsible for functionalization, crosslinking, bond cleavages, etc. In our experiment, the aqueous solution consists of AAc molecules and PPNT substrate, so the chemical reactions during plasma ignition and the mechanism of AAc grafting to PPNT are likely to be very complex.

ATR-FTIR Spectroscopy

The FTIR spectra shown in Fig. 6a clearly indicate that AAc is incorporated into the structure of PPNT. As the concentration of the AAc solution used as the medium for the diaphragm discharge increases, the intensity of the characteristic peaks increases. In all spectra of the formed layer (5–30%), an intense band in the range of wavelengths 2800–3000 cm-1 is visible, which is assigned to the asymmetric and symmetric vibrations of the functional groups CH3 and CH2 of the PPNT substrate. The broad band in the range 2700–3300 cm-1 visible in the spectrum of AAc, which is assigned to the -OH group of carboxylic acid, is narrower and shifted (3000–3300 cm-1) in the spectra of grafted layers formed at different AAc concentrations. This band corresponds to the -OH group originating from the alcohols (H-bonded), and the new band at 3500–3800 cm-1 is assigned to the free -OH groups, albeit both with low intensity. The possible consumption of -OH groups from the carboxylic acid can be explained by esterification or other reactions via the -COOH group that may occur during plasma discharge in water.

Fig. 6
figure 6

a The whole range of FTIR spectra of grafted PPNT prepared with a diaphragm discharge in an aqueous solution with different AAc concentrations (5–30%) compared to PPNT and AAc references, b FTIR spectra in the range 1800 − 800 cm− 1 comparing the samples prepared in a 30% AAc solution at two different processing times (3.5 and 5 s) before and after washing

Full size image

The FTIR range of 1800 − 800 cm− 1 with the assigned peaks for layers grafted from 30% AAc solution before and after washing is depicted in Fig. 6b. As you can see, application of different treatment times (3.5 or 5 s) did not affect the composition of the ppAAc layer, and the situation is similar for all concentrations used (data not shown). All PPNT samples were rinsed in distilled water to remove the unbound material (AAc monomer and short ppAAc polymer chains) and to stabilise the layer on the PPNT fibres. After washing, only layers prepared from 30% AAc solutions show a detectable FTIR signal of sufficient intensity (Fig. 6b). Layers prepared from lower AAc concentrations are below the detection limit of the FTIR spectrometer (micrometre range) probably due to lower thickness (data not shown).

The spectral line of the grafted ppAAc before rinsing (30%_5s, 30%_3.5s) contains the C = O bond signal at 1719 cm− 1 shifted from the original signal of 1699 cm− 1 in initial AAc. Such a shift was also observed in the work of Villegas et al. after plasma polymerization of AAc [18]. Moreover, this peak is quite broad, indicating the overlap of different C = O bonds originating from the AAc monomer, plasma polymerized AAc and the possibly formed ester functional groups. The reactions during plasma operation take place not only at the vinyl group but possibly also at the carboxyl group of AAc. This is also evidenced by the presence of C = C bond signals at 1635 and 983 cm− 1 in the grafted layer spectra. The band at 1297 cm− 1 is called the elongation band CO and peak at 1243 cm− 1 may be assigned to the CO stretching vibrations. This band at 1297 cm− 1 is seen in the spectra of AAc, plasma polymerized AAc as well as after washing, albeit with less intensity. The band at 1243 cm− 1 is absent in the spectra of the grafted ppAAc layer, but another peak at 1185 cm− 1, which can be attributed to C-O-C bond, becomes broader in the grafted ppAAc layer compared to AAc and shifts to lower wavelengths. In the work of Sipaut et al. [19]. they assigned the peak at 1158 cm− 1 to C-O-O functional group of peroxyl group which along the broad band in the range 1250–1280 cm− 1 could indicate the presence of peroxyl groups. The intense band at 1435 cm− 1 corresponds to so-called in-plane OH bending vibration. In the AAc spectrum, this band overlaps with the shearing mode of CH2, but after plasma polymerization, only an intense peak at 1410 cm− 1 appeared in the spectra. In the spectra of washed samples appeared the broad signal for carboxylate ion at 1560 cm− 1 which usually occurred along with another peak with lower wavelength (around 1400 cm− 1), however here the second carboxylate peak can be overlapped with CH3 symmetrical bending peak from PPNT. The new broad band appearing in the range 1020–1100 cm− 1 can be attributed to vinyl ether or ester.

XPS Measurements

The chemical changes in the few nanometres thick layer of the grafted ppAAc coating were analysed by XPS measurements. The data are shown in Fig. 7 (Table 1). The elemental composition of the bare PPNT consists of 99% carbon and 1% oxygen originating from air contamination. All samples contain traces of Ca and Si ranging from 0.8 to 4.7%, but their contribution is neglected. After grafting ppAAc onto the PPNTs, the O/C ratio increased from 0.01 for the bare PPNTs to values in the range of 0.17–0.35 before rinsing, which indicates successful attaching of ppAAc layer. Compared to the hypothetical O/C ratio of ppAAc (0.67), our achieved values are only half as high; however, the similar O/C ratios were achieved in different previous studies. Ramkumar et al. [13] observed values in range 0.2–0.35 for acrylic acid polymerized films on LDPE substrate. A similar lower amount of oxygen in the ppAAc layer was observed in the study by Cools et al. [14], where different discharge powers of the medium-pressure DBD used resulted in different degrees of AAc fragmentation. A high level of fragmentation during plasma exposure results in fewer oxygen-based groups being incorporated into the layer, but at the same time, the cross-linking of the layer required for stability increases.

Fig. 7
figure 7

C1s bonds concentration (columns) and O/C ratio values (yellow circles) measured on the surface of all prepared rinsed and non-rinsed samples (5, 10, 20, 30% AAc) compared to bare PPNT as a reference sample (REF)

Full size image
Table 1 Atomic concentration and C1s deconvolution to appropriate bonds
Full size table

The deconvolution of the C1s peak (Fig. 8) to the corresponding bond signals revealed the presence of a small amount of C-O bonds on the bare PPNT surface accounted for 8%, while the bulk of 92% was accounted for by C-C/C-H bonds, which represent the main polymer chain of polypropylene. The successful grafting of ppAAc onto the PPNT surface is evidenced by appearing of new peaks correspond to carbonyl C = O and carboxylic COO groups, which are absent in the PPNT spectrum. The amount of COO groups is proportionally lower than expected after AAc polymerization (33%). The samples before washing contained 7–14% of COO groups, while after the washing their amount decreased to 2–5%. The lower amount of COO after plasma polymerization can also be explained by the considerable fragmentation of AAc and the subsequent crosslinking occurring during plasma polymerization. The incorporation of other bonds, such as C = O and C-O, which are not present in the precursor, supports the theory of strong fragmentation during the underwater plasma deposition of AAc, which converts a high amount of the carboxylic acid groups into a mixture of ketones, aldehydes, alcohols and ethers [20, 21].

Fig. 8
figure 8

The deconvolution of the C1s peak to the corresponding bond signals on the bare PPNT surface, and on the PPNT surface with the ppAAc layer

Full size image

The Suggested Mechanism of Plasma-induced AAc Polymerization and Grafting to PPNT

Most polymer chemists are wary of the term “plasma polymerization” because the chemical reactions in plasma are random, and the resulting “plasma polymers” are quite different from their standard counterparts. Polymer layers produced by plasma are highly cross-linked, functionalized, and entirely amorphous compared to defined polymer structures synthesized by a strictly controlled polymerization process. In our case, underwater plasma discharge represents an initiation tool that forms active sites not only on the vinyl bond of AAc as it is common during radical polymerization, but also on the chemical structure of PPNT. More than 50 chemical reactions have been proposed in the literature for underwater plasma discharges, which are responsible for the production of primary active species, intermediates, and stable by-products from their recombination. These reactions are nicely summarized in the article of Joshi [22]. The underwater plasma discharge could dissociate the water into •H and •OH radicals, while in the next step the hydroxyl radical •OH can dissociate further into an oxygen atom O and another •H radical. These active species can induce -H abstraction and form C radicals on the PPNT backbone. The weakest C-H bond in the PP structure is the tertiary bond, which is preferentially cleaved (Fig. 9), but the dissociation energies of primary, secondary and tertiary C-H are very close, so powerful plasma can disrupt them all. Subsequently, these active sites can be functionalized preferentially by hydroxyl groups due to presence of high concentration of •OH in the underwater plasma [23]. According to theoretical and experimental findings the •OH and H2O2 are the most important products/intermediates of underwater plasma discharges. The following selected reactions (1–6) represent the most likely pathways for hydrogen peroxide formation, together with the molecular oxygen that is also produced in this process, although in much less amount [24]:

$${H_2}O\,\buildrel {under\,{\rm{water}}\,{\rm{plasma}}} \over\longrightarrow \,H\, \bullet \, + \, \bullet \,OH$$
(1)
$$\text{OH}\, \bullet \, + \, \bullet \,\text{OH}\, \to \,{\text{H}_2}\,{\text{O}_2}$$
(2)
$${\text{H}_2}\,{\text{O}_2}\, + \, \bullet \,\text{OH}\, \to \, \bullet \,{\text{O}_2}\,\text{H}\, + \,{\text{H}_2}\,\text{O}$$
(3)
$$\bullet \,\text{OH}\, + \, \bullet \,{\text{O}_2}\,\text{H}\, \to \,{\text{H}_2}\,\text{O}\, + \,{\text{O}_2}$$
(4)
$$\bullet \,{\text{O}_2}\,\text{H}\, + \, \bullet \,{\text{O}_2}\,\text{H}\, \to \,{\text{H}_2}\,{\text{O}_2}\, + \,{\text{O}_2}$$
(5)
$$\bullet \,\text{H}\, + \, \bullet \,{\text{O}_2}\,\text{H}\, \to \,{\text{H}_2}\,{\text{O}_2}$$
(6)

In addition, this molecular oxygen dissolved in the water can trigger an auto-oxidation process that produces hydroperoxide, which can decompose into various oxidation products attached to the polypropylene backbone (see Fig. 9) [23, 25].

Fig. 9
figure 9

Possible mechanism of -H abstraction induced by active species generated in underwater plasma discharge and auto-oxidation caused by presence of traces of O2 molecules in water resulting in formation of various functional groups

Full size image

On the other hand, formed C radicals on the polypropylene backbone are capable of initiating AAc polymerization, which then proceeds by the unspecified mechanism to produce ppAAc grafted layers. Moreover, the hydroperoxyl radical formed on the PPNT surface can initiate AAc polymerization similarly to the radicals formed in solution at the water-plasma interface. We can conclude that the whole process of plasma-triggered polymerization of AAc on the PPNT by underwater plasma discharge is very complex. The proposed mechanism of PPNT activation and functionalization explains how the polymerization of AAc could be initiated. However, considering the fragmentation and rearrangements phenomenon that occurre during AAc plasma polymerization depending on the plasma operation parameters (discharge power, exposure time), it is very difficult to estimate the precise polymerization process [26].

Conclusion

The present study investigates the possibilities of using pulsed underwater diaphragm electrical discharge generated in aqueous solution for the polymerization and grafting of acrylic acid onto non-woven polypropylene fabric (PPNT). Under the selected plasma operating conditions, increased surface water uptake was achieved for all AAc concentrations used, which was evaluated by measuring the strike-through time. The strike-through time was measured in the range of ˜ 2–4 s, and no significant increase was observed even after 2 months of storage. The wettability of PPNT was tested using the critical wetting surface tension (CWST) method, in which the surface energy value of the non-rinsed PPNT samples treated in 20% and 30% AAc solution increased by three times the reference value. An increase in surface energy was also observed in the rinsed samples, indicating that the ppAAc layer was stably bound to the fibers. The presence of a ppAAc layer on the PPNT fibres was also confirmed by increased adhesion compared to the bare fibres as well as visual evidence of the ppAAc layer on the SEM images. Chemical analyses by FTIR and XPS showed the formation of new functional groups such as carbonyl, carboxyl or peroxyl on the PPNT surface as well as the consumption of double bonds in the AAc monomer, indicating successful polymerization. However, the lower amount of free carboxyl groups incorporated into the layer is due to strong fragmentation during plasma irradiation, while at the same time the cross-linking of the layer required for stability increases. Based on the results obtained, we can say that our new method of plasma-initiated polymerization and grafting of AAc to fibrous substrate is fast and efficient and it is possible to be performed in the continuous regime. Functionalization and hydrophilization of chemically inert PPNT surface is stable during few months and formatted functional groups open possibilities to used such modified material for biomedical applications.