In Fig. 2, FT-IR spectra of the original LDHs, LDHs@PPy and LDHs@PPy-Ag show almost identical characteristics of typical LDHs (Liu et al., 2014) Compared with original LDHs, LDHs@PPy-Ag and LDHs@PPy show additional peaks at 1464 cm–1and 1128 cm–1, which attributed to the conjugated C—N stretching vibration in py ring (Li et al., 2013). And the peaks at 1284 cm–1 and 1071 cm–1 are related to the stretching and deformation vibration of C—H (da Silva et al., 2017). In addition, in the FT-IR spectrum of original LDHs, the bands recorded below 700 cm–1 are attributed to the vibration of metal-oxygen bonds (Liu et al., 2014). It is preliminary evidence that PPy is synthesized on the LDHs surface.
The successful synthesis of LDHs@PPy-Ag is further confirmed by XRD and UV-vis as shown in Fig. 3. As shown in Fig. 3a, the diffraction reflections of original LDHs are sharp and symmetrical, and the baseline is stable, indicating well-formed crystal layered structures. And the characteristic diffraction peaks at (003), (006) and (110) of typical LDHs (JCPDS file No. 38-0487) are easily identified in all the patterns (Liu et al., 2014). In addition to characteristic diffraction peaks of typical LDHs, (111) and (200) diffraction peaks of Ag are observed in the XRD pattern of LDHs@PPy-Ag (Wang et al., 2013). In Fig. 3b, original LDHs and LDHs@PPy both displays no visible absorption peak between 300 nm and 800 nm. After loading AgNPs onto the surface of LDHs@PPy-Ag, there is a clear peak at 430 nm (Xu et al., 2011). This is due to characteristic of AgNPs induced by surface plasmon absorption (Zhang et al., 2013), which further indicates the successful synthesis of LDHs@PPy-Ag.
Micromorphologies of LDHs@PPy are investigated by TEM under different magnifications, as shown in Fig. 4. Obviously, a soft coating (~15 nm) is observed on the layered particles. Furthermore, the average diameter of LDHs@PPy is ~1500 nm evidenced from TEM image. In order to obtain direct evidence of PPy coatings on the LDHs, chemical composition was also investigated by EDS (Fig. 4c). The large peaks for carbon, oxygen, magnesium, and aluminum in the EDS spectrum indicate the basic components of the LDHs. It is worth noting that the small peaks for nitrogen and iron indicate further the existence of PPy (Kim et al., 2015b).
Figure 4. The TEM images of LDHs@PPy at low magnification (a) and high magnification (b), EDX scans of LDHs@PPy (c)
The TEM and SEM are applied to further observe the micromorphology of LDHs@PPy-Ag directly, as shown in Fig. 5 and Fig. 6, respectively. After the reduction process during the chemical oxidative polymerization of Py, AgNPs are uniformly adsorbed on the LDHs@PPy-Ag, as shown in Fig. 5a and Fig. 5b. The thickness of the PPy coatings is ~6 nm. And the average diameters of AgNPs on the LDHs@PPy-Ag can also be estimated at ~20 nm. In Fig. 5c, LDHs@PPy-Ag possesses Ag (111) crystal planes with a lattice spacing of 0.24 nm (Zhao et al., 2015).
Figure 6. The TEM images of LDHs@PPy-Ag at low magnification (a) and high magnification (b), EDS results of LDHs@PPy-Ag (c), the elemental distribution mapping of LDHs@PPy-Ag (d)
In Fig. 6a and Fig. 6b, upon PPy-Ag coatings onto LDHs, AgNPs are uniformly distributed on the LDHs@PPy-Ag. Meanwhile, LDHs@PPy-Ag is evenly dispersed in the field of observation. In Fig. 6c and Fig. 6d, the uniform distribution of elements reflects the uniform distribution of Ag and PPy. It is revealed that original LDHs is well coated by PPy-Ag.
In order to quantitatively evaluate the antibacterial properties of original LDHs, LDHs@PPy and LDHs@PPy-Ag, bacterial colony counter was used to count the number of colonies in the petri dishes with different nanoparticles. And the antibacterial activity was also evaluated by the antibacterial rate. In Fig. 7a, original LDHs still shows a certain degree of antibacterial activity, and the antibacterial rate reaches 35.19%. Upon coating of PPy onto LDHs, the antibacterial activity of LDHs@PPy is significantly improved, and the antibacterial rate is up to 99.97%. Due to the dual antibacterial effects of PPy and AgNPs, the antibacterial rate of LDHs@PPy-Ag reaches about 100%. It is generally believed that Ag ions released from unstable AgNPs are responsible for its excellent antibacterial activity. A large surface area of AgNPs contributes to the release of more Ag ions, as a result of more exposure to bacteria. Therefore, LDHs@PPy-Ag exhibites the best antibacterial properties. In general, the antibacterial properties of nanoparticles are closely related to that of the nanocomposites prepared by them. In Fig. 7b, not surprisingly, the presence of original LDHs does not improve the antibacterial properties of PCL matrix. And the antibacterial rate of LDHs@PPy/PCL and LDHs@PPy-Ag/PCL nanocomposites both exceeds 99.99%. Especially for LDHs@PPy-Ag/PCL nanocomposites, its concentration of Escherichia coli is only 800 CFU/cm2. Therefore, LDHs@PPy-Ag/PCL nanocomposites show the best antibacterial properties.
Figure 8 shows DSC cooling and second heating curves of LDHs@PPy/PCL and LDHs@PPy-Ag/PCL nanocomposites. And the related results of thermal analysis are shown in Table 1. As shown in Fig. 8a and Table 1, there is a crystallization exothermic peak (Tc) at 29.9℃ in the DSC cooling curve of PCL, which is attributed to the α-polymorph (Pucciariello et al., 2007). With the addition of original LDHs, Tc and χ of LDHs/PCL nanocomposites occur at higher value. An increase in Tc and χ is due to the fact that original LDHs can act as the heterogeneous nucleating agent to promote the crystallization process of PCL (Peng et al., 2010). Compared with original LDHs, LDHs@PPy and LDHs@PPy-Ag show more significant heterogeneous nucleation because of higher Tc and χ. In Fig. 8b, the melting temperature (Tm) of pure PCL is 56.7℃ while Tm of these nanocomposites shows no obvious change (56.2℃–56.7℃). It shows that original LDHs, LDHs@PPy and LDHs@PPy-Ag do not affect Tm of PCL matrix obviously.
Figure 8. The DSC (a) cooling and (b) second heating curves of LDHs@PPy/PCL and LDHs@PPy-Ag/PCL nanocomposites
Sample ΔHm (J/g) Tm (℃) Tc (℃) χ(%) PCL 44.75 56.7 29.9 33.2 LDHs/PCL 49.70 56.7 31.6 36.9 LDHs@PPy/PCL 52.01 56.7 32.5 38.6 LDHs@PPy-Ag/PCL 51.85 56.2 32.4 38.5
Table 1. The related results of LDHs@PPy/PCL and LDHs@PPy-Ag/PCL nanocomposites according to thermal analysis
The XRD analysis of LDHs@PPy/PCL and LDHs@PPy-Ag/PCL nanocomposites further verified the results of DSC analysis, as shown in Fig. 9a. For all the nanocomposites, XRD patterns exhibit reflections at 2θ=21.16°, 21.79° and 23.41°, corresponding to (110), (111) and (200) planes of the orthorhombic crystal form for PCL, respectively. Compared with pure PCL, all the nanocomposites show higher diffraction peak intensity of PCL chains. It reveals that original LDHs, LDHs@PPy and LDHs@PPy-Ag all facilitate the crystallization of PCL chains as a result of an increase in crystallinity. This is consistent with the results of DSC analysis. In addition, in all the nanocomposites, (003) crystal plane of typical LDHs shows similar diffraction intensity and diffraction angle (~11.74°). This is consistent with the XRD analysis of LDHs@PPy and LDHs@PPy-Ag.
Figure 9. The XRD patterns (a) and TGA curves (b) of LDHs@PPy/PCL and LDHs@PPy-Ag/PCL nanocomposites
The TGA curves of LDHs@PPy/PCL and LDHs@PPy- Ag/PCL nanocomposites are shown in Fig. 9b. According to previous studies (Pucciariello et al., 2007; Costantino et al., 2009), the thermal stability of LDHs/polymer nanocomposites decreases gradually with the increase of LDHs. Sure enough, the thermal stability of all the nanocomposites is not as good as that of pure PCL. However, compared with LDHs/PCL nanocomposites, LDHs@PPy/PCL and LDHs@PPy-Ag/PCL nanocomposites exhibit better thermal stability. Owing to the high thermal stability of PPy relatively, core-shell structured LDHs@PPy and LDHs@PPy-Ag contribute to delay of thermal degradation (Mao et al., 2017b). Therefore, LDHs@PPy and LDHs@PPy-Ag improve the thermal stability of their nanocomposites in comparison to LDHs/PCL nanocomposites.
The mechanical parameters derived from the stress-strain curves are shown in Table 2. Compared with pure PCL, deformation at the yield point, stress at breaking, and deformation at breaking of all the nanocomposites are improved to some extent. Particularly, compared with pure PCL, LDHs@PPy/PCL nanocomposites show a 32% increase in tensile strength (41.3 MPa) and a 31% increase in elongation at break (830%). And compared with LDHs@PPy/PCL nanocomposites, the mechanical properties of LDHs@PPy-Ag/PCL nanocomposites are slightly reduced, but still significantly improved. At this low filler content, the improvement of mechanical properties is a typical feature of nanocomposites (Pucciariello et al., 2012).
Sample σy (MPa) ɛy (%) σb (MPa) ɛb (%) PCL 23.9±0.9 19.1±1.2 31.3±1.4 635±51 LDHs/PCL 24.7±1.1 19.4±1.8 32.4±1.6 714±36 LDHs@PPy/PCL 21.5±0.8 20.6±2.1 41.3±1.3 830±42 LDHs@PPy-Ag/PCL 22.8±0.7 23.9±1.9 36.5±1.1 822±53 Notes: σy, stress at the yield point; ɛy, deformation at the yield point; σb, stress at breaking; ɛb, deformation at breaking.
Table 2. Mechanical parameters of LDHs@PPy/PCL and LDHs@PPy-Ag/PCL nanocomposites according to stress-strain curves
To further understand the relationship between material structure and mechanical properties, SEM images of fracture surfaces for all the nanocomposites are shown in Fig. 10. According to our previous research (Mao et al., 2017b), overall fracture surface of pure PCL is rough. And a large number of bulges and voids can be observed in the region of fracture, which indicates plastic deformation. Similarly, results of Fig. 10a–10c also show rough and well stretched surface (numerous curled broken fibrils) after fracture, indicating the existence of original LDHs, LDHs@PPy and LDHs@PPy-Ag enhances the flexibility of PCL matrix (Du et al., 2010; Liau et al., 2014; Bai et al., 2016). Apparently, there are some interface defects between original LDHs and PCL matrix (Fig. 10d), and this indicates that the interfacial compatibility is relatively poor. However, as shown in Fig. 10e and Fig. 10f, LDHs@PPy and LDHs@PPy-Ag are both tightly combined with PCL matrix without visible interface gap. This indicates that the interfacial compatibility has been improved markedly, which can conclude that LDHs@PPy/ PCL and LDHs@PPy-Ag/PCL nanocomposites have better mechanical properties than LDHs/PCL nanocomposites (Mao et al., 2017b; Mao et al., 2018). In addition, the better interfacial bonding, the fewer the interfacial defects, which is also closely related to the gas barrier properties of materials.
Gas barrier properties of LDHs@PPy/PCL and LDHs@PPy-Ag/PCL nanocomposites are shown in Fig. 11. When the original LDHs content reaches 1 wt%, Rp (relative permeability=P/P0) decreases obviously to 67%. Upon PPy coatings onto LDHs, Rp of LDHs@PPy/PCL nanocomposites further decreases to 44% with the same addition. The results show that LDHs@PPy can significantly improve the oxygen barrier properties of PCL matrix. Only when clay is better dispersed in the PCL matrix and the interface compatibility between modified clay and PCL matrix is better, can it help to reduce interface defects and prolong the tortuous path of gas diffusion (Huang et al., 2014). The presence of AgNPs may reduce the interfacial compatibility between LDHs@PPy-Ag and PCL matrix, resulting in the decrease of the barrier properties for the final nanocomposites (Mao et al., 2019). However, the oxygen barrier properties of LDHs@PPy-Ag/PCL nanocomposites (Rp=52%) are still better than those of LDHs/PCL nanocomposites. Therefore, PPy and PPy-Ag coatings play important roles in reducing gas permeability in LDHs@PPy/PCL and LDHs@PPy-Ag/PCL nanocomposites.
According to considerable studies in references (Bharadwaj, 2001; Xu et al., 2006; Choudalakis et al., 2009; Zare-Shahabadi et al., 2011; Avérous et al., 2012; Wu et al., 2013; Genovese et al., 2014; Layek et al., 2015), the barrier mechanism of clay in the nanocomposites is mainly attributed to the bulk effect and barrier effect. As a rule, clay/polymer nanocomposites consist of an impermeable phase (layered clay) and a permeable phase (polymer matrix) that dispersed in polymer matrix. Three main factors of layered clay can affect the gas barrier properties of clay/polymer nanocomposites: volume fraction, micromorphology and orientation (Bugatti et al., 2010). Based on Nielsen's detour theory, classical relative permeability equation can be expressed as follow,
where ϕs represents the volume fraction of clay. Assuming clay is considered to be a disk with diameter (L) and thickness (W). The aspect ratio (L/W) of original LDHs, LDHs@PPy and LDHs@PPy-Ag is ~37.5, ~21.4 and ~28.8, respectively. For example, the average diameter of LDHs@PPy is ~1500 nm evidenced from TEM images. Estimated from our previous research, the lateral thickness of original LDHs and PPy coatings are ~40 nm and ~15 nm, respectively. Therefore, the lateral thickness of LDHs@PPy is ~70 nm. The aspect ratio of LDHs@PPy is thus ~21.4. The aspect ratio of LDHs@PPy-Ag is also estimated in this way.
According to the above equation, the predicted and experimental values of relative permeability are shown in Fig. 12. The results show that Rp of the predicted values is significantly higher than that of the experimental values. And the deviation (23.9%–49.7%) between predicted and experimental values is large in the whole region. The true cause lays in the fact that the above equation is based on a complete geometric analytical procedure. Two important factors should be considered: a final change of polymer matrix behavior due to the incorporation of clays and the interaction interface (Gain et al., 2005). It should be pointed out that the better interface compatibility between clay and PCL matrix, the greater deviation of Rp. Especially in our clay/polymer nanocomposites, surface modification of LDHs (LDHs@PPy and LDHs@PPy-Ag) can lead to very different surface-to-bulk ratios eventually. Meanwhile, it can also cause interactions of different types toward the polymer chains, resulting in more strong interface (Avérous et al., 2012; Wu et al., 2013; Genovese et al., 2014).