3.1 X-Ray diffraction of clay nanofillers
The XRD pattern of initial VER (Fig. 1) shows VER basal reflections observed at 2θ 6.18° and 7.02° (correspond to d-values 1.428 nm and 1.258 nm). Admixture phase of tremolite (ICSD PDF card no. 00-013-0437) is evident for all samples about 2θ 10.6° (d = 0.836 nm). Description of VER is given in [37]. Intercalation of CH into the VER interlayer gallery is confirmed by shift of VER basal reflections to 2θ 2.50°, 3.94° and 7.91°. This means expansion of VER interlayer space to d = 3.157 nm, 2.239 nm and 1.117 nm, respectively, as a result of the intercalation of CH molecules into the VER interlayer. The XRD pattern of VER_OCT shows a basal reflection at 2θ 2.41° (d = 3.660 nm) and subsequent reflections at 4.62°, 4.05° and 7.94° (d-values 1.911 nm, 1.448 nm, 1.253 nm and 1.112 nm). These indicate disordered VER structure with intercalated OCT molecules in the interlayer.
Moreover, neither CH nor OCT reflections weren’t confirmed on the VER surface based on the XRD patterns, so both CH and OCT are completely intercalated into the interlayer.
The XRD pattern of ZnOVER (Fig. 2) shows basal reflections at 2θ 7.22° and 7.78° (d = 1.224 nm and 1.136 nm) as a result of ZnO formed in the VER interlayer space [40]. Moreover, reflections at 2θ 8.83° and 9.77° (d = 1.001 nm and 0.905 nm) confirm partially dehydrated VER structure after calcination at 350°C. Expansion of the interlayer space in the ZnOVER_CH sample is evident based on the basal reflections at 2θ 2.82°, 3.83° and 7.85° correspond to d = 3.130 nm, 2.303 nm and 1.125 nm, respectively, due to the intercalation of CH molecules into the interlayer. Intercalation of OCT into the ZnOVER interlayer confirm, although with some differences in intensity distribution, a series of new reflections at 2θ 2.85°, 3.88°, 4.96°, 7.67° and 8.02° (d = 3.10 nm, 2.275 nm, 1.782 nm, 1.152 nm and 1.101 nm).
The differences in the expansion of the VER interlayer space after CH and OCT intercalation are caused by the different structural arrangement of molecules and thus also their different arrangement in the VER interlayer.
The XRD pattern of ZnOVER and organically modified ZnOVER (Fig. 2) show reflections about 2θ 31.87°, 34.35°, and 36.34° (d = 0.281 nm, 0.261 nm, and 0.247 nm) which correspond to ZnO nanoparticles formed and anchored onto the VER particle’s surface (ICSD PDF card no. 01-078-2585).
The XRD patterns of PCL composite films are shown in Fig. 3. Pure PCL shows characteristic PCL reflections of the semicrystalline orthorhombic structure [37] at 2θ 15.70°, 21.34°, 21.95° and 23.64° (d = 0.563 nm, 0.416 nm, 0.404 nm and 0.376 nm).
The relative intensity of VER reflections for all PCL samples is very low because only 3 wt.% of nanofiller is used. The XRD pattern of PCL/VER display reflections at 2θ 6.08°and 7.05° (d = 1.452 nm and 1.252 nm) and PCL/Ver_CH at 2θ 2.80°, 3.85° and 7.99° (d = 3.153 nm, 2.300 nm and 1.106 nm). In case of PCL/ZnOVER, this one shows reflections at 2θ 5.12°, 7.21°, and 7.26° (d = 1.492nm, 1.226 nm and 1.139 nm) and PCL/ZnOVER_CH at 2.74°, 3.86°, and 7.74° (d = 3.219 nm, 2.290 nm and 1.141 nm). Reflections of these PCL composites are close to the original nanofiller’s reflections; therefore, no significant change has occurred compared with the original nanofillers. However, PCL/VER_OCT shows reflection at 2θ 6.10° (d = 1.449 nm) and reflection at 2θ 2.41° (d = 3.660 nm) disappeared. PCL/ZnOVER_OCT shows shift to 2θ = 2.31° (d = 3.826 nm). This can mean that PCL chains are intercalated into the VER interlayer and nanocomposite structure is formed.
3.2 FTIR spectroscopy of clay nanofillers
The FTIR spectrum of the initial VER (Fig. 4), which was used as one of the nanofillers for thin PCL nanocomposite films, shows a band at 3671 cm− 1 in O-H stretching region attributed to structural OH groups, the band at 3416 cm− 1 corresponding to O-H stretching vibration of adsorbed water thereto O-H bending vibration at 1637 cm− 1 characterized adsorbed water. Furthermore, the intensive band at 999 cm− 1 is assigned to Si-O stretching vibrations together with Si-O bending vibration at 459 cm− 1 [41]. The FTIR spectrum of antimicrobial organic component CH (Figs. 4, 5) shows characteristics bands at 3338 and 3172 cm− 1 corresponded to the asymmetric and symmetric N-H stretching vibrations, two bands at 2934 and 2860 cm− 1 are assigned to asymmetric and symmetric C-H stretching bands of CH [42, 43]. The C-N stretching vibration of imine group appears at 1643 cm− 1 and it is followed by the bands in the 1582–1490 cm− 1 interval which belong to a N-H bending vibration of secondary amine and imine groups [42, 43]. Absorption at 1415 cm− 1 belongs to the C-C stretching vibrations of an aromatic ring, and at 822 cm− 1 to the C-H out-of-plane deformation vibrations of a 1,4-disubstituted aromatic ring.
In the FTIR spectrum of VER_CH nanofiller (Fig. 4), we can observe a shift in the region of characteristic valence vibrations of the OH groups (3349 cm− 1) from the initial VER and an enlargement in this region. These changes should confirm the binding of CH, probably by hydrogen bonding between the structural OH groups of VER and NH groups of CH.
The FTIR spectrum of second antimicrobial organic compound OCT (Figs. 4, 5) shows characteristics bands at region 3472 − 3034 cm− 1 belong to N-H stretching vibrations, further at 2928 and 2858 cm− 1 asymmetric and symmetric C-H stretching vibrations, at 1655 cm− 1 stretching vibrations of C = N group. Absorption at 1222 cm− 1 belongs to C-N (vibration of aromatic ring group) and at 724 cm− 1 to C-Cl stretching vibration.
In the FTIR spectrum of VER_OCT nanofiller (Fig. 4) we can observe a shift in the region of characteristic valence vibrations of the OH groups (3404 cm− 1) from the initial VER and an enlargement in this region. As well as in the case of VER_CH we can assume the binding of OCT through hydrogen bonding between the structural OH groups of VER and NH groups of OCT.
The FTIR spectrum of ZnOVER nanofiller (Fig. 5) shows, besides characteristic bands originating from VER, bands at 1500 and 1377 cm− 1 corresponding to C = O vibrations belonging to traces of ZnCO3, which is an intermediate in the ZnO synthesis reaction. At the same time, when comparing the spectra of VER and ZnOVER, a decrease in the intensity of characteristic stretching and bending vibrations of OH groups corresponding to adsorbed water can be seen, which is due to the reduced water content in ZnOVER sample caused by calcination at 350°C.
Similar to the previous VER nanofillers, on the basis of shifts of characteristics vibrations of OH bonds of VER and NH bonds of CH and OCT in FTIR spectra of ZnOVER_CH and ZnOVER_OCT nanofillers (Fig. 5), the binding of organics by means of hydrogen bonds is evident.
3.3. Morphology characterisation and particle size distribution of clay nanofillers
Figures 6 and 7 show SEM images of the clay nanofillers when each sample is displayed at a magnification of 2 500x and 8 000x. Particle size distribution (PSD) curves are also shown for each clay nanofiller.
The initial vermiculite (VER) with bimodal particle size distribution (dm1 = 0.131 µm and dm2 = 3.420 µm) and ζ-potential value − 46.0 mV was used as the input material for the preparation of clay nanofiller samples (Fig. 6). The VER particles had a predominantly oval shape with a smooth surface.
The intercalation of the VER particles by antimicrobial organic compounds (CH and OCT) caused an increase in the size of the main fraction (dm2) to 5.8 µm in both cases and the formation of a new fraction (dm3) also with identical values 174.6 µm. For the VER_CH sample, there was a slight increase in the size of the finest fraction (dm1) from 0.131 µm to 0.296 µm. The intercalation of organic components leads to a positive ζ-potential value, when for the VER_CH sample ζ = +30.7 mV and for the VER_OCT sample ζ = +21.9 mV. These values correspond to both the degree of intercalation of the organic component and the mutual interaction of the individual size fractions. The VER_CH nanofiller showed a non-uniform shape with curved edges. The surface of the particles showed a broken layered structure with exposed individual layers of the original particles. At the same time, smaller particles of uniform size agglomerated on the surface and edge of larger particles. In contrast, the particles of the VER_OCT sample showed a smooth surface with sharp edges. Fragments of individual layers of particles were located on the longest sides of the particles in the perpendicular direction. The smallest fractions were located separately in the sample volume.
Figure 7 show SEM images and PSD curves of the ZnOVER type of clay nanofillers. The sonochemical preparation of ZnO nanoparticles on the original VER particles did not cause significant changes in the particle size of the ZnOVER sample. The original bimodal character of the distribution curves (PSD), which confirms the presence of two size fractions with values dm1 = 0.226 µm and dm2 = 3.905 µm, was maintained. The bigger fractions of ZnOVER particles had a non-uniform shape with sharp edges. The smaller fractions had a rounded shape and were equally dispersed in the sample volume. The ZnOVER particles did not agglomerate, which is also confirmed by the ζ-potential value − 34.9 mV.
The intercalation of the ZnOVER particles with the CH compound had no effect on changes in the particle size distribution compared to ZnOVER_CH sample. The particles conserved their original smooth surface, their edges were rounded, and the layered structure was more visible. The positive value of ζ-potential + 24.6 mV corresponds to the presence of an organic component and the tendency of the smaller fraction (dm1 = 0.197 µm) to interact with the edges of the larger fraction (dm2 = 3.905 µm). On the contrary, the intercalation of the OCT component contributed to a significant decrease in the value of ζ-potential (+ 8.6 mV for the ZnOVER_OCT sample), which was showed by a strong agglomeration of particles. This resulted in the formation of a new unstable fraction of dm3 = 101.46 µm. The original fine fractions (dm1 = 0.259 µm) agglomerated with each other and simultaneously interacted with the surfaces of the original fraction at the edges, leading to an increase in their size to dm2 = 5.122 µm.
3.4. Surface morphology of thin PCL/clay nanocomposite films
The Fig. 8 shows SEM images of thin PCL film and thin PCL/clay nanocomposite films based on VER type of nanofillers, when individual images are shown at magnification of 500x and 1 000x. Further, the surface wettability of these samples was evaluated based on water contact angles (WCA). The light microscopy (LM) images of water drops are shown in Fig. 8 as well together with evaluated WCA.
In case of the control PCL film we observe a rather smooth structure with only a minimum of cracks, formed by more or less regular spherulite hexagons, which are characteristic for some types of polymers and are formed during their crystallization [44, 45]. The thin PCL film has a hydrophilic surface with WCA 83.3°. After the addition of VER nanofiller to PCL matrix, the reduction in the size of individual spherulitic grains is evident. These are rather irregular than in pure PCL with larger cracks between the individual spherulite grains. There is also reduction in WCA to 67.3°, which is the lowest value from all measured samples. Presence of organic phases CH or OCT smoothed out the films surface structure and we can observe slightly higher WCA values; for PCL/VER_CH 71.9°and for PCL/VER_OCT 70.9°. More detailed view (magnification 1 000x) shows the presence of the clay nanofiller inside the polymer matrix.
In the case of ZnOVER as nanofiller for PCL (Fig. 9) smaller irregular spherulitic grains with a greater number of larger cracks throughout the sample compared to the pure thin PCL film (Fig. 8) could be observed. There is a reduction in WCA value even to 68.9°. For the thin PCL/ZnOVER_CH nanocomposite film we can see from SEM images that presence of organic phase (CH) caused the smoothing of the film surface and reduction in the number of cracks between spherulitic grains. In the case of PCL/ZnOVER_OCT, there is also very smooth surface, moreover, we can observe the disappearance of cracks between the spherulitic grains. OCT in structure as well as in previous samples caused slightly higher WCA (69.9°, PCL/ZnOVER_OCT) than CH (68.7°, PCL/ZnOVER_CH).
3.5. Thermogravimetry and mechanical properties of thin PCL/clay nanocomposite films
The Fig. 10 shows TG and DTG curves of thin PCL film and thin PCL/clay nanocomposite films. The mass loss is determined in the temperature interval 210–520°C (Fig. 10a). The highest mass lost was observed for sample PCL/ZnOVER_CH (97.64%), the lowest for sample PCL/ZnOVER_OCT (94.37%) and values for another samples ranged around 95% (Table 1). As it is seen from the DTG curves (Fig. 10b), the mass loss in the above-mentioned temperature interval is realized at one step for all samples. After the introduction of ZnO type nanofillers into the PCL, we can observe decreasing of the thermal stability. The most significant decrease, up to 46°C, is apparent for sample PCL/ZnOVER compared to PCL (Table 1). For PCL/ZnOVER_CH (394.7°C) and PCL/ZnOVER_OCT (397.4°C) are values slightly higher due to presence of organic compounds CH and OCT in nanofillers structure. This phenomenon was described by Malakpour et. al. [46], who claimed that ZnO can catalyse the decomposition of surrounding carbons from PCL. On the other hand, presence of OCT in the structure slightly increases thermal stability of thin PCL/VER_OCT film compared to PCL.
Table 1
Thermal data (∆ m, Td and Tmax) and tensile test results (E, Rm and Smax) of thin PCL film and thin PCL/clay nanocomposite films.
Sample | ∆ m (%) | Td (°C) | Tmax (°C) | | E (MPa) | Rm (MPa) | Smax (mm/mm) |
PCL | 96.02 | 380.3 | 412.9 | | 177.5 | 19.8 | 6.42 |
PCL/VER | 95.71 | 377.4 | 410.1 | | 187.3 | 14.3 | 4.16 |
PCL/ZnOVER | 95.88 | 317.7 | 367.3 | | 165.8 | 12.5 | 0.74 |
PCL/VER_CH | 95.09 | 374.5 | 411.3 | | 230.7 | 12.4 | 0.62 |
PCL/ZnOVER_CH | 97.64 | 356.0 | 394.7 | | 219.0 | 12.4 | 0.72 |
PCL/VER_OCT | 94.78 | 378.4 | 413.6 | | 213.2 | 14.1 | 0.94 |
PCL/ZnOVER_OCT | 94.37 | 367.5 | 397.4 | | 207.1 | 14.3 | 0.88 |
The mechanical properties of the thin PCL film and thin PCL/clay nanocomposite films were evaluated on the basis of the tensile tests. Table 1 summarised the measured and evaluated parameters: Young’s modulus (E), tensile strength (Rm) and maximum strain (Smax).
The results point to the fact that VER particles in PCL matrix slightly increase the Young's modulus (E) from the original 177.5 MPa (PCL sample) to 187.3 MPa (PCL/VER sample), while ZnOVER nanofillers cause a decrease to 165.8 MPa (PCL/ZnOVER sample). The intercalation of organic compounds CH and OCT had a significant effect on increasing the E values. The highest value was measured for the PCL/VER_CH sample (230.7 MPa), which showed the finest spherulitic structure (Fig. 7) and whose VER_CH nanofiller showed the highest ζ-potential value (+ 30.7 mV). Conversely, the lowest E values were measured for the PCL/ZnOVER_OCT sample (207.1 MPa). This value corresponds to the fact that this nanocomposite film was formed by a spherulitic structure with non-uniform grain size and shape (Fig. 8), resulting from the lowest stability of ZnOVER_OCT nanofiller (ζ-potential = + 8.6 mV) and non-uniform particle size (PSD). These findings are also consistent with the results of the decrease in thermal stability after the addition of ZnOVER-type nanofillers. The strength limit values also confirm the fact that nanofillers reduce the values of mechanical parameters. PCL/VER nanocomposite film with ultimate strength of 14.3 MPa (Rm) and PCL/VER_OCT (14.1 MPa) and PCL/ZnOVER_OCT (14.3 MPa) nanocomposite films with the presence of OCT in the nanofiller structure were evaluated as the most resistant to tensile load (strongest compared to the original PCL film, Rm = 19.8 MPa). In the case of maximum strain (Smax), PCL/VER (Smax = 4.16 mm/mm) and PCL/VER_OCT (Smax = 0.94 mm/mm) were the most mechanically resistant compared to the original PCL film (Smax = 6.42 mm /mm).
Measurements of mechanical parameters correspond to the results of Lepittevin et al. [47] and confirm that the mechanical properties of PCL/clay nanocomposites are related to the structure of the PCL matrix. Depending on the PCL/clay preparation method, the composites remain rather ductile and the tensile stress decreases when the clay content is lower than 5 wt. %. Overall, it can be assumed that organically intercalated clays have higher stiffness and exhibit ultimate strength properties.
3.6. Antimicrobial tests of thin PCL/clay nanocomposite films
Antimicrobial activity of selected thin PCL/clay nanocomposite films with antimicrobial clay nanofillers was studied by a certified method ČSN ISO 22196 (Measurement of Antibacterial Activity on Plastics and other Non-porous Surfaces). The results shown in Table 2 are average of two determinations of each sample and are expressed as the mean of calculated logarithms of bacteria. The value of antibacterial efficiency (A) is given in logarithms (as a difference of lg of the number of bacteria in the test sample and lg of the number of bacteria in the control sample). If the value of (A) is in the range 2 ≤ A ≤ 3, the effectiveness is evaluated as significant. If the values of A ≥ 3, the effectiveness is rated as strong.
Table 2
Antimicrobial efficiency of selected thin PCL/clay nanocomposite films.
Sample | S. aureus | E. coli |
A [lg] | time [h] | effectiveness | A [lg] | time [h] | effectiveness |
PCL/ZnOVER | 2.43 | 48 | significant | 2.21 | 48 | significant |
PCL/VER_CH | 2.43 | 24 | 2.04 | 24 |
PCL/VER_OCT | 2.03 | 48 | 2.41 | 24 |
PCL/ZnOVER_CH | 2.43 | 24 | 2.41 | 24 |
PCL/ZnOVER_OCT | 2.01 | 24 | 2.41 | 24 |
Table 2 shows that all tested samples are significantly effective against the tested bacteria strains. In the case of S. aureus, the highest antibacterial effect A = 2.43 in the fastest time of 24 h was achieved by samples PCL/VER_CH and PCL/ZnOVER_CH. Compared to the other samples, the thin PCL/VER_OCT nanocomposite film showed the lowest activity with value of A = 2.03 after 48 hours, but in general, this result is considered very favourable and does not lose its significance.
Almost all samples are very effective against E. coli in shorter time interval of 24h with value of antibacterial efficiency A = 2.41, 2.04. Otherwise sample PCL/ZnOVER showed good activity for both bacteria strains, but in longer time period.
3.7. Degradation assays: morphology analysis and weight loss
In this chapter, the morphology of the films after 45 days of degradation in PBS buffer is described. Films before degradation are shown in Figs. 11 and 12 only for visual comparison with degraded samples. Their morphology has already been described in Chap. 3.4.
From the SEM images (Fig. 11) for the thin PCL film after 45 days of degradation we can observe that the individual original spherulitic grains have been kept, but there are significant cracks emerging from their tops. In the case of the thin PCL/VER nanocomposite film, the structure of the individual spherulitic grains has disappeared, as well as the cracks. On the other hand, protruding particles of nanofiller from the polymer matrix and also visible particles of recrystallized polymer are visible. In thin PCL/VER_CH and PCL/VER_OCT nanocomposite films, the original structure was kept, only in the case of the sample with CH small cracks appear from the centre of the spherulitic grains, and in both samples nanofiller particles of recrystallized polymer can be observed on the film surface.
From the SEM images (Fig. 12) of the thin PCL/ZnOVER nanocomposite film after 45 days of degradation it can be seen that the structure of the individual spherulitic grains remains relatively unchanged, only the lines between the grains are less sharp and also there are less amount of cracks. In the case of PCL/ZnOVER_CH film there are only a few small cracks on the surface as well as visible particles of recrystallized polymer, otherwise the film remains in a proper state. In contrast, the sample PCL/ZnOVER_OCT after degradation shows visible grooves in the film structure, apparently with nanofiller particles protruding from the polymer matrix.
Figure 13 shows the weight loss of the thin PCL film and thin PCL/clay nanocomposite films against incubation time in PBS solution (pH = 7.4) at 37°C. Generally, degradation rates of the polymer were affected by their structure, molecular weight, and other structural characteristics. The degradation of semi-crystalline polymer PCL in aqueous media has been reported to occur in to steps [48], when in the first step water starts diffuse into amorphous, less organized regions, which are easier allowed for water penetration to polymer matrix. The second step starts after the degradation of amorphous regions, then a hydrolytic degradation process follows from the edge toward the centre of the polymer crystalline domains. From that reason, in this case weight loss was small in range between 0.35–2.32%, and it depends on a degradation time and used nanofiller.
At early degradation state (first 5 days) the degradation process was faster, especially for samples with inorganic nanofillers only - PCL/VER and PCL/ZnOVER. The sample PCL/VER_CH showed the highest weight loss (2.32%) after 30 days of degradation. Other samples showed comparable or lower values compared to control thin PCL film. Only PCL/VER nanocomposite film showed increasing tendency much better than pure thin PCL film. This behavior can be attributed to the high relative hydrophilicity of the clay, allowing an easier permeability of water into the material thus accelerating the hydrolytic degradation process. These results are in good agreement with SEM analysis and WCA measurements. In other words, the nanofillers with the highest hydrophilic character are responsible for the fastest degradation of the polymer. Possibly, for longer degradation time periods, swelling of the crystalline phase would take place and higher values of weight loss would be obtained.
3.8. Degradation assays: differential thermal analysis
The Fig. 14 shows comparison of crystallinity of thin PCL film and thin PCL/clay nanocomposite films before and after 5, 15, 30 and 45 days of degradation in PBS solution based on differential thermal analysis (DSC). Before degradation test, adding clay nanofillers into PCL matrix has some nucleating effect causing crystallization rate slightly higher than of pure thin PCL film [49], except sample PCL/VER_CH with the lowest crystallinity (44.3%). Likewise, this sample possesses the greatest increase in crystallinity (18%) after 45 days of degradation to value of 62.3%. In the case of thin PCL film, the crystallinity increases almost uniformly during whole degradation time. Except PCL/VER_OCT, the largest increase in crystallinity occurs in the first 5 days of degradation (especially for samples PCL/ZnOVER, PCL/VER_CH and PCL/ZnOVER_OCT this is evident), after which the increases are gradual. Since the amorphous part of thin PCL films is degraded faster than the crystalline part, the crystalline phase predominates after degradation. This suggests that the degradation process starts on the amorphous or less ordered areas in the initial stages of degradation, as the solution ions are able to migrate more easily to the less ordered regions.