Experimental design and response surface methodology
The application of the experimental CCD design and the response surface methodology allowed the researchers to know the conditions that had a better inhibition of Staphylococcus aureus by using nanoparticles of TiO2 modified and deposited with Ag as an antibacterial agent. The zone of inhibition, the halos present in the S. aureus plate (mm), obtained by CCD design established are shown in Table 2. The sizes of the halos are in a range of 7.32 to 11.26 mm.
The application of the RSM generated the second-order regression polynomial (Eq.1), which represents an empirical relationship between the response (length of the inhibition halos with Staphylococcus aureus) and the evaluated variables. The statistical significance of each of the coefficients was evaluated by the Student's t-test and the p-value. Table 3 shows the variables and interactions that have the highest incidence on the deposition process in Staphylococcus aureus inhibition. In Table 3, a significant variable is found when the p-value is less than 0.05, which indicates that the effect is significant with 95% confidence. Therefore, the significant effect corresponds to AgNO3 (x2) concentration, while the TiO2 solids in solution and pH are not significant in the process, being pH the least influential. The model shows a good fit with the experimental data, having a coefficient of determination (R2) of 0.86, implying that 86.00% of the variables fit the model and that it only does not explain 14% of the total variation.
The analysis of the response surface, illustrated in Fig. 1, shows the behavior of the different combinations of effects on the size of S. aureus inhibition halos, in which the largest size occurs at intermediate levels, pH of 10 and 5 g of TiO2 solids in solution. In contrast, for AgNO3 concentration, high levels are required (10% w/w). The most suitable working ranges were determined from the contour graph, Fig. 2; the results obtained were pH 9 to 11, a TiO2 amount between 4 and 6 g, while the range of AgNO3 concentration covers a region above 11% w/w. From the second-order model (Eq. 1), the canonical model (Eq. 2) was obtained, and the characteristic values, with alternating signs (λ1= -0.2459, λ2= -0.1464, λ3= 0.0115), determine that the results show a saddle point.
Based on the results obtained, a second equi-radial experimental CCD design was established with two factors (Design 2), considering as independent variables: AgNO3 concentration concerning TiO2 (X1) (25, 35, and 45% w/w) and the amount of NPs of TiO2 in solution (X2) (3, 5 and 7 g/250mL), and now with constant pH in 10. The response variable remains the same, size of the halo or zone of inhibition of the disks (diameter of halos in mm). Table 4 shows the results obtained from Design 2 of the disk test. The ranges of the S. aureus halo of this second proposed design are between 13.16 and 22.09 mm, and compared to the first experimental design (7.32 to 11.26 mm), it presents an increase of 96.18%, which indicates that there is an enhancement of almost the double in halo size.
Statistical analysis of the results using the Student's t-test and calculation of the p-value provided the estimated coefficients for the quadratic model. Table 5 shows the variables and interactions present in the deposition process in Staphylococcus aureus inhibition. As a result of Design 2, the p-value shows that the TiO2 solids in solution remain as a non-significant variable, while the AgNO3 concentration is the significant variable in the deposition process due to the p-value was found below 0.05.
The response function (length of the inhibition halos with Staphylococcus aureus) was represented using an RSM through a second-degree polynomial equation (Eq. 3), where the two independent variables were considered. As a result, the quadratic model shows an acceptable fit of experimental data, with a coefficient of determination (R2) of 0.82, which implies that 82.00% of the experimental data fit the model; therefore, 18% of the total variation was unexplainable.
In the response surface graph, Fig. 3, it was observed that the maximization of S. aureus inhibition occurs at a low concentration of AgNO3 (20-25% w/w) with high amounts of TiO2 solids. (7-8 g), observing the highest levels of inhibition (26-28 mm) highlights the importance of the addition of TiO2 solids. In the statistical analysis, the X1X2 interaction was significant; this was observed in the curvature shown by the surface graph. The significance of this interaction indicates that within the experimental domain, synergistic interaction effects commonly attributed to the dispersion of TiO2 particles were observed when they are dispersed in an aqueous medium in the Ag deposition process.
The most suitable working ranges were determined through the contour graph, Fig. 4, having a value between 21 and 24% w/w for the AgNO3 concentration, with a TiO2 quantity ranging from 7.5 to 7.8 g. The results obtained from the second experimental design allowed determining that an AgNO3 concentration between 20-25% w/w with amounts of silanized TiO2 between 7-8 g and pH of 10 achieves an inhibition zone of 28 mm in Staphylococcus aureus. This value has improved 300% regarding the zone of inhibition, reported by [64], who analyzed the antibacterial behavior of an Ag/TiO2/PVC composite membrane tested in the same bacterial strain. Likewise, an increase of just over 200% was found in the inhibition of the same microorganism, given that they tested Ag and TiO2 modified polycaprolactone nanofibers (PCL/TiO2-Ag), showing an antibacterial impact of 9.2 mm in the diameter of the zone of inhibition [66]. In other research [36], a 40% increase in the zone of inhibition was obtained compared to unmodified Ag/TiO2 nanoparticles since they showed a maximum zone of inhibition of 20 mm when working with a similar strain of S. aureus (methicillin-resistant).
The most surprising results emerging from the data are that these Ag/TiO2 showed excellent inhibition against S. aureus since inhibition is better than previous reports. For example, Emam et al. [7] found no antibacterial activity against this gram-positive bacterium using AuNPs obtained sing starch polymer. While Ahmed et al. [12] obtained nanoparticles using hydroxyethyl cellulose as nanogenerator and surfactant, they reported an inhibition zone of only 16 mm using AgNPs obtained, 15 mm with AuNPs, and 18 mm with Ag-Au bimetallic nano-alloy.
Characterization of TiO2 nanoparticles
Surface-modified TiO2 nanoparticles were used in the Ag deposition process and subsequently used in the antibacterial test. The modification of the surface of TiO2 was carried out with APTES; these particles underwent a series of tests to confirm the presence of silica (present in APTES) on the oxide surface. One of the characterizations performed was the FTIR, which provides information about the functional groups present in the samples to be analyzed. Fig. 5 shows the results of the FTIR spectra for unmodified (black line) and modified TiO2 (red line). It is essential to mention that new bands appeared in the particles modified with APTES; symmetric and asymmetric stretching vibration of the C-H bond was observed in the methylene group at 2960, 2927, and 2858 cm-1 [29, 66-67].
The peak spanning between 1630 to 1640 cm-1 is attributed to the stretching vibration of absorbed water and the hydroxyl (OH-) groups present on the surface of the nanoparticles [29, 66, 52, 68]. The small peak at 1641 cm-1 corresponds to the stretching vibration of the (NH)C=O group [66]. Other evidence of the modification with APTES on the nanoparticle surface is the peak at 1560 cm-1 that corresponds to the flexion of the amino-functional group (-NH2) [66, 68]. The peak shown between 1420 and 1490 cm-1 corresponds to the elastic band of the C-H organic group bond [29, 32, 52]. The peak observed at 1378 cm-1 was assigned to the C-N aromatic amine group [65]. The stretch band of the C-O group bonds is also observed at 1259 cm-1 [62] and the carbonyl group at 1300 cm-1 [52]. Finally, the last peaks around 1120, 1130, and 1044 cm-1 are attributed to the Si-O-Si bond, indicating polymeric siloxane as a product of the organosilane precursor [29, 66, 52, 68]. Similarly, peaks at 1045, 1075, and 1130 cm-1 are attributed to Ti-O-C bonds [29, 66, 52]. The superficial modification of the TiO2 nanoparticles with the coupling agent (APTES) is confirmed with the above described.
On the other hand, to analyze size and morphology, transmission electron microscopy is a suitable test for this purpose. The images provided by TEM gave further evidence of the superficial modification of the TiO2 NPs with the coupling agent APTES. Fig. 6 (a) shows TiO2 without modification. These NPs do not have a uniform surface, and it is appreciated how they tend to agglomerate. This behavior is due to the surface/particle size ratio that causes strong agglomerates that limit its suspension stability. Besides, this is attributed to Van der Waals forces, and the attracting effect tends to decrease the dispersion of the nanoparticles. While in Fig. 6 (b), the micrograph is observed after performing the silanization process. It is possible to observe an organic coating due to the APTES bound to the TiO2 surface with an average of 2.82 nm thick. All this information confirms the presence of the coupling agent on the TiO2 surface.
However, it was necessary to verify that the superficial modification of the TiO2 led to an improvement in the dispersion and colloidal stability, which is necessary to facilitate the interaction in the aqueous medium of these nanoparticles in the deposition process with Ag. The profiles of Z potential as a function of the pH of pure TiO2 (unmodified) and silanized TiO2 (modified) are observed in Fig. 7. After surface modification of the NPs, the Zeta potential of the silanized TiO2 NPs increased considerably due to the amino groups (NH2) that appeared in the outer layer of the surface of the NPs.
The increase in the positive charge generated by the protonation of the amino groups caused an increase in the positive Zeta potential in the acid region [67], increasing the repulsion between the NPs. Likewise, this increase in the Z potential is favorable because it prevents instability (IEP) and agglomeration [41]. The IEP value for the unmodified oxide is approximately at pH 3.5, lower than for modified TiO2, at pH 5.6. This shift in IEP is attributed to the alkaline characteristics of NH2 groups [52] or the easy desorption of hydrogen (H+) protons found on the NPs surface of silanized TiO2. All the aforementioned indicates that the surface behavior of the TiO2 NPs changed very markedly and that at the same time, the dispersion of the nanoparticles favors the deposition process with Ag since there is a higher surface ratio of TiO2 nanoparticles dispersed that come into contact and interact with the metal favoring the process and the amount of silver deposited on the surface. Values of Z potential from -30 to 30 mV indicate instability in the system; values below -30 and above 30 mV indicate an increase in stability, increasing with higher absolute values of the Z potential [41].
Another essential characteristic sought when developing these types of materials is that they acquire the ability to withstand high temperatures and that there is no degradation thereof, to such a degree that the material is not lost: at the same time, it maintains its characteristics and properties. The TGA analysis corresponding to unmodified TiO2 is shown in Fig. 8 (black line), in which it is possible to observe that it remains practically stable from 25 °C to 800 °C, unlike the functionalized system present losses of weight. When analyzing the thermal degradation of the coupling agent APTES on the surface of the TiO2, Fig. 8 (blue line), it can be distinguished that the modified material has thermal stability at high temperatures, such as the temperatures used in industrial polymer processes, for example, since at approximately 98.5 °C only 0.15% by weight of the material is lost, at 563 ° C a total of 0.38% is lost. Finally, until 734 ° C, a total of 0.56% of material is lost in weight. This result is incredibly low since little more than 99.4% of the total is conserved; the loss in weight is related to the decomposition of the organic chains of the coating with APTES.
The characterization of the NMR spectrum for 13C, Fig. 9, shows the existence of 3 peaks in the spectrum that correspond to signals 12.40, 27.84, and 45.76 ppm, they are identified in Fig. 9 as C1, C2, and C3, these signals refer to the three carbons of the propyl group (a group that is present in the coupling agent APTES), these signals are approximately the same indicating that the carbons are in the same amount on the surface of the TiO2. As already mentioned, the signals belong to carbons. However, the difference between each one is related to the type of group to which it is bond, Carbon 1 is the one that is bond to Silicon, Carbon 2 is linked to two other carbons, and Carbon 3 would be the one that is linked to the amino group (NH2). On the other hand, in Fig. 10, the NMR spectrum for 29Si denotes signals at -67.43 and -59.45 ppm, identified as T3 and T2, respectively, attributed to the silica found on the surface of TiO2. T3 is assigned to the bond between silica in the APTES and three oxygens on the TiO2 surface. In comparison, T2 silicon refers to silica bonds with only two oxygens on the TiO2 surface and one remaining oxygen as hydroxyl (OH) group.
Characterization of TiO2 particles deposited with Ag considering the highest and the lowest result of the response variable of the two-factor experimental CCD design.
The results obtained from implementing the experimental design on the microorganism inhibition allowed the researchers to know the highest and lowest size of the inhibition zone. The results of Table 4 made it possible to determine that the experimental unit 6 presented the maximum size of inhibition (22.09 mm). While sample 1 determined that the halo size (13.16 mm) was not favored under these conditions. Nevertheless, the inhibition of the microorganism exists in all the experimental units of the CDC. Therefore, the characterization was carried out using TEM/EDS to explain the behavior of these two samples.
Fig. 11 illustrates the TEM micrograph of experimental unit 6 of Table 4. This sample was made up of 49.14% w/w AgNO3 concentration and 5g of TiO2, that is, the points (+1.4142, 0) of the CCD of two-factor, founding the maximum length of inhibition for S. aureus (22.09 mm). On the other hand, Fig. 12 shows the micrograph of experimental unit 1, which obtained the shortest length of the microorganism's inhibition halo (13.16 mm). This sample was prepared with 25% w/w AgNO3 concentration and 3g of TiO2, that is, the axial points (-1, -1) of the two-factor CCD. Morphologically in both experimental units, TiO2 nanoparticle with Ag dots on the surface can be seen; the difference is distinguished in the amount of Ag that is superficially distributed on the TiO2 nanoparticle. This difference is mainly attributed to the AgNO3 concentration that was used in each sample.
The chemical analysis provided by TEM-EDS of the Ag-deposited TiO2 is shown in Fig. 13 and Fig. 14 of the experimental units 6 and 1, respectively. From this analysis, the chemical composition of each sample was obtained. The results of this analysis confirmed the presence of silver (b), oxygen (c), and titanium (d), which interact during the deposition process with the metal. Furthermore, it is possible to observe that Ag is distributed on the surface of TiO2 (same that is confirmed in Fig. 11 and Fig. 12), this indicates a correct dispersion of AgNO3 during the deposition process with the oxide because an ideal chemical interaction was assumed by improving the dispersion of TiO2 after the silanization process with APTES. The different compositions of the samples are found in Table 6.
The results of the drop-test method carried out to the composites against S. aureus are shown in Fig. 15. The values indicate inhibition activity in the PLA composites prepared with 0.1% TiO2 nanoparticles deposited with Ag. As observed, better inhibition was obtained for the system with the highest result of the response variable of the two-factor experimental CCD design, M6, with an 18.75% of inhibition rate. Nevertheless, the lowest result, M1, also presented inhibition activity but with less efficiency, only 6.25%. Therefore, with this information, it is possible to report that it is possible to confer the antimicrobial property from the TiO2 deposited with Ag to the PLA composites [1]. Last, with this antibacterial test was possible to corroborate the applicability of the experimental CCD design since the best inhibition result for the composite (PLA-M6) aggress with the highest value obtained for the TiO2 deposited nanoparticles in the disc diffusion method.
Compared with the literature, although these results may look lower values than those reported previously, it is important to mention that the filler (Ag/TiO2) in the PLA matrix is exceptionally low. For example, Emam [11] reported a high reduction percentage in S. aureus (96-100); nevertheless, he stated that the antibacterial efficacy was attributed to the bi-action of two nanoparticles: silver and gold; in addition, they worked with a high amount of nanoparticles (0.5 mL/20 mL). At the same time, Ahmed et al. [12] reported great inhibition rates against S. aureus of approximately 60%, but with approximately 0.3% of Au content in silk fabrics. With all this information, it is possible to state that the composites prepared at 0.1% of Ag/TiO2 in PLA are consistent and even better than previous reports regarding the nanoparticle content.
XRD-patters were used to know the crystalline structure of the TiO2 particles deposited with Ag. Fig. 15A shows by X-ray diffraction (XRD) the results, and it can be observed the characteristic peaks for the structures of anatase and rutile in both samples; these results are due to both phases of titanium dioxide are mixed during their industrial origin [5]. Besides, it is possible to observe the characteristic peaks of the silver nanoparticles at 32.12° (101) and 38.01° (111) [5]. While Fig. 15B shows the structures of the PLA nanocomposite films were analyzed to study the effect of nanoparticle content on the crystallinity of the PLA matrix. Both XRD patterns showed an amorphous curve indicating a low degree of crystallinity of PLA and a maximum intensity at approximately 2θ = 17° attributed to the characteristic peaks of PLA [6][7]. On the other hand, the absence of TiO2 peaks or new diffraction peaks can be seen, which can be attributed to the few numbers of functionalized nanoparticles (only 0.1%) immersed in the PLA matrix. This information suggests that the structure of PLA was not changed by incorporating the TiO2 particles deposited with Ag [8].