Biopolymer Composites with Ti/Au Nanostructures and Their Antibacterial Properties

In this study, we have aimed at the preparation and characterization of poly-l-lactic acid (PLLA) composites with antibacterial properties. Thin bilayers of titanium and gold of various thickness ratios were deposited on PLLA by a cathode sputtering method; selected samples were subsequently thermally treated. The surface morphology of the prepared composites was studied by atomic force, scanning electron, and laser confocal microscopy. The chemical properties of the composites were determined by X-ray photoelectron and energy-dispersive X-ray spectroscopy in combination with contact angle and zeta potential analyses. The antibacterial properties of selected samples were examined against a Gram-negative bacterial strain of E. coli. We have found that a certain combination of Au and Ti nanolayers in combination with heat treatment leads to the formation of a unique wrinkled pattern. Moreover, we have developed a simple technique by which a large-scale sample modification can be easily produced. The dimensions of wrinkles can be tailored by the sequence and thickness of the deposited metals. A selected combination of gold, titanium, and heat treatment led to the formation of a nanowrinkled pattern with excellent antibacterial properties.


Introduction
Polymer metallization may lead to an internal tension, which is created between a film and a substrate [1]. Additional stress may be caused by external conditions, such as temperature changes, ambient humidity, external loads, or oxidation. Exceeding the critical tension can then lead to surface deformation and wrinkle formation [2]. The value of critical stress depends on the ratio of Young's modulus of a film and substrate. This ratio is very low for a polymer-metal system. The critical stress is too low and, therefore, it is very easy to produce a wrinkle-patterned surface structure. [3]. Since tension in a homogeneous film is uniform, the resulting wrinkles are randomly oriented. Only at the edges of the substrate, there is anisotropic stress, when the parallel component is significantly larger than the perpendicular one, and thus, the resulting wrinkles are directed perpendicularly to the edges of the film. With increasing distance from the edges, the anisotropy disappears [2]. A similar anisotropy of stress occurs in bends that may arise in the initial phase of sputtering when metal atoms directly bombard the substrate surface. This results in high compression stresses resulting in cross-linked bends, for example, for polydimethylsiloxane (PDMS). With the development of wrinkles, the bends gradually disintegrate and eventually completely disappear [4]. Sputtered thin films on created metallic dielectrics consisting of two metallic layers (Cr, Al) separated by a dielectric (SiO 2 ) layer able to change color were reported to rely on the viewing angle [36]. In recent years, there has been intensive research in the nanomedicine field. Nanocomposite thin films with high biocompatibility and optimal mechanical properties, which may be coated with established implants, are being investigated [37][38][39]. Increased attention has been drawn to amorphous thin-film alloys, which can serve as biocompatible and antimicrobial coatings of stainless steel articles, or improve their mechanical properties [40,41]. In addition to metal-based nanocomposite materials, thin layers of biopolymers are also used in the medical industry. The thin film of the polymer matrix is suitable for targeted drug delivery, which can greatly increase the efficacy of the medicinal product over application in another dosage form. Thus, cellulose, poly(vinyl alcohol), or poly(ethylene oxide) polymers are commonly used [42].
This work focuses on the preparation of thin metal bilayers on a polymer and their analysis and assessment of antibacterial properties. PLLA/Ti/Au and PLLA/Au/Ti systems with different ratios of thicknesses of individual metallic layers were created by a sputtering method on a PLLA substrate, at which the total thickness of metals was 50 nm and 30 nm, respectively. Furthermore, the effect of thermal stress on the surface morphology and properties of the prepared samples was investigated. Heating the PLLA with a thin metallic layer to the glass transition temperature of the polymer led to the formation of a wrinkled structure. To the best of our knowledge, a selective combination of titanium and noble metal nanolayers, which would lead to the formation of a wrinkle-like pattern with outstanding antibacterial properties, has never been studied so far, with titanium dioxide playing an important role in antibacterial nanolayer formation.

Materials and Modification
Poly-(L)-lactic acid (PLLA) in the form of a polymer film was used as a substrate, supplied by Goodfellow (film thickness 50 µm, density 1.25 g·cm −3 , glass transition temperature (Tg) = 60 • C, crystallinity between 60 and 70%). To prepare the gold and titanium layers, 99.999% pure metal targets were used.
The metals were applied to the polymer by sputtering by Quorum Q300T ES. Sputtering was performed in an Ar atmosphere (purity of 99.997%). The metal deposition was conducted at a pressure of 0.1 Pa. A current of 20 mA and a sputtering time of 20 to 700 s were set for the gold deposition. A current of 150 mA was set for titanium deposition; the target sample distance was 20 cm with no rotation. Its thickness was monitored using a crystal gauge. Before sputtering, the surface oxide layer was removed from the titanium by "blind" sputtering. Two series of samples were prepared: one with titanium as the primary metal onto which the gold was deposited so that the total thickness was 50 nm; the second series consists of samples with a gold base on which titanium was deposited so that the total thickness was 30 nm. All detailed Ti and Au layer thicknesses are given in the Table 1. Thicknesses were verified by a combination of a scratch test and atomic force microscopy. Selected samples were thermally stressed in a Binder Owen thermostat oven at 60 • C correspondings to the PLLA glass transition temperature for one hour and then cooled in the air to room temperature.

Analytical Methods
The surface structure was monitored using an Olympus Lext confocal microscope. The surface of heated and unheated samples was monitored using different objectives. The images were processed by OLS software by a PC microscope.
Sample surface morphology was examined by atomic force microscope (AFM) Dimension ICON (Bruker Corp., Billerica, MA, USA); ScanAsyst mode in the air was used for determination. Silicon Tip on Nitride Lever SCANASYST-AIR with a spring constant of 0.4 N·m −1 was used. NanoScope Analysis software was applied for data processing. The mean roughness values (Ra) represent the average of the deviations from the center plane of the sample. SEM scans were acquired by a scanning electron microscope (FIB-SEM, LYRA3 GMU, Tescan, Brno, Czech Republic). The applied acceleration voltage was 10 kV. The examined samples were covered with a Pt conductive layer of 20 nm thickness by the deposition from the Pt target (purity of 99.9995%, SAFINA, Vestec, Czech Republic). The deposition was performed using the diode sputtering technique (Quorum Q300T equipment, Lewes Road Laughton, UK). Focused ion beam (FIB) cuts were made by a Ga ion beam integrated into the adapted scanning electron microscope. The FIB-SEM images were taken at an angle of 54.8 • . The elemental composition was determined by energy-dispersive X-ray spectroscopy (EDS, analyzer X-MaxN, 20 mm 2 SDD detector, Oxford Instruments, Abingdon, UK). The accelerating voltage for SEM-EDS measurement was 10 kV.
The surface concentration of the oxygen and carbon in the modified substrates was examined by X-ray photoelectron spectroscopy (XPS). The spectra were measured using experience with polymer analysis by XPS, it means using relatively low power of X-ray source (75 W), using monochromatic X-ray radiation (1486.7 eV), and using very low energy of charge compensating electrons (typically 2 eV), all of these to protect the measured polymer surface against changes caused by radiation. The data were processed by the Casa XPS program and the measured spectra were compared with other our results and reference analysis. Peak fitting was based on possibilities of the Casa XPS program; background removal was applied by Shirley curve, which allowed us to respect the asymmetry of the analyzed peaks.
Sample wettability was characterized by goniometry using a SEE System 2020 (v. 2.1. SEE System, Advex Instruments, Brno, The Czech Republic). Drops of water and glycerol with a volume of 8 µL were applied to the samples using an automatic micropipette. Photographs of the drops were taken with the See software; the contact angle was determined by a three-point method. The surface energy of all samples was determined by the Owens-Wendt method. Furthermore, an electrokinetic analysis (zeta potential determination) was performed on a SurPASS Instrument (Anton Paar GmbH, Gratz, Austria) by two methods (streaming current and streaming potential) using two equations (Helmholtz-Smoluchowski, HS, and Fairbrother-Mastins, FM). Samples of 2 × 1 cm 2 were prepared for the measurement. The samples were studied inside an adjustable gap with an electrolyte of 0.001 mol·dm −3 KCl at a constant pH value of 7.0 and room temperature. Two samples of each type were measured four times with a relative error of up to 5%.

Antibacterial Activity
The antimicrobial properties of the selected samples were determined. Measurements were performed using an environmental Gram-negative bacterial strain of E. coli (DBM 3138). The evaluated samples were placed in Falcon tubes with bacterial suspension diluted with physiological solution (0.9% saline) and gently mixed (Shaker Fisher Vortex-Genie 2, Scientific Industries, Bohemia, NY, USA). The samples were left in contact with the bacterial suspension for 4 and 24 h. After the incubation period, the samples were gently mixed again and 25 µL drops were taken from each tube five times (per one plate) and plated on Petri dishes (in triplicates) containing Luria-Bertani (LB) agar. Then, the agar plates with the bacteria were cultivated for 24 h at room temperature, after which photographs of the plates were taken and the number of colony-forming units (CFU) was calculated using ImageJ software (v. 2.1., LOCI, University of Wisconsin, Madison, WI, USA).

Sample Surface Morphology by LCM Analysis
The main idea of this study was a development of a very simple but unique technique of how to prepare a wrinkle-like patterned PLLA substrate with gold and titanium nanolayers enhancing the antibacterial properties of the biopolymer surface. As part of the study of thermally stressed biopolymer composites, we firstly focused on the evaluation of the material surface morphology and changes in biopolymers caused by thermal stress. For this purpose, laser confocal microscopy (LCM) was chosen as the first preliminary method to study changes in the material surface morphology. Figure 1 shows images of the surfaces of unheated samples, in which a primary Au layer was applied to the samples, on which a titanium layer was deposited so that the total metal thickness was set to 30 nm. The images show the formation of a typical structure for titanium nanolayers, i.e., the formation of certain "cracks" in the layer, which occurred during sputtering. The major effect of the titanium layer, which has a thickness of 28 nm in Figure 1a and which gradually fades from 25 and 20 to 5 nm (Figure 1b-d), is evident. It should also be mentioned that the titanium particles have high energy due to the deposition parameters when they hit the substrate, which can also lead to a local increase in the temperature of the sputtering head. Figure 1d shows the most pronounced porosity of the PLLA surface after the application of the metallic layers. We have also studied the surface of unheated titanium-based samples on which gold was deposited to the total thickness of 50 nm (i.e., the metals were deposited in reverse order). The influence of the titanium layer on the material surface morphology was determined to be similar as in Figure 1, with titanium as the bottom layer and gold at the top layer, which led to the formation of a heterogeneous surface structure with microcracks, which are closely related to the thickness of the Au layer, respectively to the different mechanical properties of gold and titanium layer. At higher values of Au thickness, it achieves more homogeneous properties in terms of surface morphology compared to the Ti layer and the surface, therefore, has fewer surface irregularities. Figure 2 shows the surfaces of the samples with the same combinations of Au and Ti thicknesses as in the previous part, but these samples were subjected to thermal stress at 60 • C for 1 h. The images show a formation of a wrinkled structure, similar to the ones in a previous study by Jurik et al. [7,8]. At the same time, the microcrack-like structure was partially preserved, which, however, can also take on the character of cross-linked bends (see Figure 3), which are visible on images from SEM-samples as bright raised lines. For this reason, in the following text, these structures will be discussed as "bends," although in some images they may appear as simple "cracks" in the structure. The influence of these structures on the directionality of the waves is visible. In Figure 2a, in which this network is the densest, the ripples that formed were more disordered (Figure 2d). A similar ordered orientation of corrugations was visible even with certain combinations of metal thicknesses on samples with a titanium base layer. The stress-induced sputtering with high energy atoms in combination with slightly elevated temperature during the metallization process led to the formation of a primary randomly oriented wrinkled structure over a longer period of metal deposition. This structure is shown in Figure 3 (SEM analysis), which shows the surface of a sample with a 10 and 20 nm layer of gold and titanium, respectively. over a longer period of metal deposition. This structure is shown in Figure 3 (SEM analysis), which shows the surface of a sample with a 10 and 20 nm layer of gold and titanium, respectively.   Figure 3B shows the same sample after heating. From this image, it is apparent that the primary wrinkles remained even after the sample heating when a secondary wrinkled structure was formed (images from laser confocal microscopy). The PLLA film was probably heated due to the energy of the incident atoms, especially in the case of Ti. This caused local surface heating, where a process very similar to thermal stress took place, although only in a thin surface layer and with a slight increase in temperature towards

Sample Surface Morphology by AFM Analysis
Figures 4 and 5 show AFM scans of selected unheated and heated PLLA samples with Ti and Au nanolayers together with a 2, 5, 10, and 25 nm thick titanium bottom layer. The PLLA surface with the lowest Ti thicknesses was evenly covered with a uniform metallic cluster array. The image with a 10-nm gold layer shows a disordered primary wrinkle structure accompanied by a slight increase in surface roughness. This slight increase indicates only a partial transformation in the process of wrinkle structures, the height of these waves being several tens of nanometers.
The samples with Au as the primary layer after heating have shown an intense increase in surface roughness with the formation of wavy structures caused by thermal stress. A significant effect of titanium on the surface roughness is evident. The PLLA sample with the highest thickness of a titanium layer (28 nm) also exhibited the largest surface roughness. As the proportion of the titanium layer decreased and the proportion of gold increased, so did the surface roughness. The sample with the biggest thickness of the gold layer (25 nm) and 5 nm layer of titanium had almost 50% of the roughness of the roughest sample. Sample a, with a 10-nm gold layer, retained the primary wrinkled structure formed by the deposition even after heating, which also increased the roughness of this structure. Figure 4 shows the surface of unheated samples with a titanium bottom layer and a gold top layer. The formed primary wavy structure is apparent in the images with 5 and 10 nm titanium layers. The corrugated surface with a titanium layer of 10-nm thick-  Figure 3B shows the same sample after heating. From this image, it is apparent that the primary wrinkles remained even after the sample heating when a secondary wrinkled structure was formed (images from laser confocal microscopy). The PLLA film was probably heated due to the energy of the incident atoms, especially in the case of Ti. This caused local surface heating, where a process very similar to thermal stress took place, although only in a thin surface layer and with a slight increase in temperature towards Tg. The directionality of the stress in the polymer film structure, which leads to the formation of oriented wrinkle-like patterns [7,8], was not significantly disturbed, as the subsequent thermal stress led to further significant changes in surface morphology, as shown below. The presence of the primary ripples formed directly during deposition does not have a significant effect on the formation of secondary ripples during the subsequent heating process. However, the direction of the wave pattern created by the subsequent thermal stress was influenced by the presence of bends on the sample surface, in which the wrinkling stress was anisotropic. Therefore, in the area around the bend, the ripples were oriented perpendicularly to its edges, while outside this area, they were oriented randomly, see Figure 3C. cluster array. The image with a 10-nm gold layer shows a disordered primary wrinkle structure accompanied by a slight increase in surface roughness. This slight increase indicates only a partial transformation in the process of wrinkle structures, the height of these waves being several tens of nanometers.

Sample Surface Morphology by AFM Analysis
Pharmaceutics 2021, 13, x FOR PEER REVIEW 9 of 20 ness exhibited the largest roughness of the whole sample series. Samples with a base titanium layer after heat treatment are shown in Figure 5. A wrinkled structure was formed by thermal stress induced by a temperature close to the Tg of the polymer. The sample roughness was also affected by the ratio of metals, increasing with a growing thickness of the titanium layer. On the sample with only 2 nm of titanium, the wrinkled structure exhibited a regular arrangement. Samples with a 5-and 10-nm titanium layer retained the primary ripples even after heating. The sample with a 25-nm titanium layer had the highest surface roughness compared to the others, but the ripples were not regular.   Figure 6A,B represent the surface roughness values determined on the samples with different combinations of gold and titanium layer, while the sum of the deposited thickness remained constant and one of the deposited metals was changed. There is a signifi-  The samples with Au as the primary layer after heating have shown an intense increase in surface roughness with the formation of wavy structures caused by thermal stress. A significant effect of titanium on the surface roughness is evident. The PLLA sample with the highest thickness of a titanium layer (28 nm) also exhibited the largest surface roughness. As the proportion of the titanium layer decreased and the proportion of gold increased, so did the surface roughness. The sample with the biggest thickness of the gold layer (25 nm) and 5 nm layer of titanium had almost 50% of the roughness of the roughest sample. Sample a, with a 10-nm gold layer, retained the primary wrinkled structure formed by the deposition even after heating, which also increased the roughness of this structure. Figure 4 shows the surface of unheated samples with a titanium bottom layer and a gold top layer. The formed primary wavy structure is apparent in the images with 5 and 10 nm titanium layers. The corrugated surface with a titanium layer of 10-nm thickness exhibited the largest roughness of the whole sample series. Samples with a base titanium layer after heat treatment are shown in Figure 5. A wrinkled structure was formed by thermal stress induced by a temperature close to the Tg of the polymer. The sample roughness was also affected by the ratio of metals, increasing with a growing thickness of the titanium layer. On the sample with only 2 nm of titanium, the wrinkled structure exhibited a regular arrangement. Samples with a 5-and 10-nm titanium layer retained the primary ripples even after heating. The sample with a 25-nm titanium layer had the highest surface roughness compared to the others, but the ripples were not regular. Figure 6A,B represent the surface roughness values determined on the samples with different combinations of gold and titanium layer, while the sum of the deposited thickness remained constant and one of the deposited metals was changed. There is a significant roughness increase for heated samples due to the formation of the wrinkled pattern. For samples with a base gold layer, the surface roughness decreased with reducing the thickness of the upper titanium layer, which was especially evident for heated samples. The trend was the same for samples with a base layer of titanium. With its increasing thickness, there was a significant increase in surface roughness.  Figure 6A,B represent the surface roughness values determined on the samples with different combinations of gold and titanium layer, while the sum of the deposited thickness remained constant and one of the deposited metals was changed. There is a significant roughness increase for heated samples due to the formation of the wrinkled pattern. For samples with a base gold layer, the surface roughness decreased with reducing the thickness of the upper titanium layer, which was especially evident for heated samples. The trend was the same for samples with a base layer of titanium. With its increasing thickness, there was a significant increase in surface roughness.   The surface morphology was different for each sample. While the sample with the bottom layer of gold of 10-nm thickness formed a wrinkled structure over the entire surface, the sample with the bottom layer of titanium of the same thickness created irregular wrinkling and also individual metal clusters were formed over the surface layer. The highest roughness values were obtained for a sample with a 10-nm thick gold bottom layer and a top layer of titanium of 20-nm thickness with subsequent heating. A comparison of the determined dimensions of the corrugations and the thicknesses of the deposited layers shows that the dimensions of the primary corrugated structure depend more on the type of a deposited metal than on the total applied thickness. As titanium thickness increased, the dimensions of the waves expanded as well. Figure 7 shows the morphology of the selected heated samples with different combinations and sequences of the metals. The morphology of the primary ripples induced by the sputtering process in combination with the secondary wrinkling structure formed after the heating process is also apparent. As also introduced in Figure 6, the surface roughness of the heated samples increases with the growing amount of titanium. All width and amplitude valus of the wrinkle are included in Table 2.
of a deposited metal than on the total applied thickness. As titanium thickness increased the dimensions of the waves expanded as well. Figure 7 shows the morphology of th selected heated samples with different combinations and sequences of the metals. Th morphology of the primary ripples induced by the sputtering process in combination wit the secondary wrinkling structure formed after the heating process is also apparent. A also introduced in Figure 6, the surface roughness of the heated samples increases with the growing amount of titanium. All width and amplitude valus of the wrinkle are in cluded in Table 2.

Surface Chemistry
The chemical composition of the surface layer was determined by the XPS method Figure 8A shows the results for samples with titanium bottom layer and gold upper laye before and after the heating process. Due to the relatively high thickness of the gold (25 40, 45, and 48 nm), the presence of titanium appeared only on the samples with a Ti thick ness of 25 nm, and the concentration of gold on the surface was about 55%. After heating

Surface Chemistry
The chemical composition of the surface layer was determined by the XPS method. Figure 8A shows the results for samples with titanium bottom layer and gold upper layer before and after the heating process. Due to the relatively high thickness of the gold (25,40,45, and 48 nm), the presence of titanium appeared only on the samples with a Ti thickness of 25 nm, and the concentration of gold on the surface was about 55%. After heating, only a slight increase was detected in the concentration of carbon and oxygen from the polymer, probably due to morphological changes. The results of the samples with the gold bottom layer and the titanium upper layer are shown in Figure 8B. The thickness of the upper titanium layer was 5, 20, 25, and 28 nm. In the case of unheated samples, the presence of gold appeared only for the samples with the greatest Au thickness of 25 nm. The concentration of titanium was about onethird compared to samples with a gold surface. At the same time, there was a much higher proportion of oxygen, which indicates the oxide layer formed in the Ti nanolayer. A slight decrease in titanium concentration was observed on heated samples. The presence of pure metal or titanium dioxide was examined in more detail. Figure 9 shows measurements at two angles (9° and 90°) for an unheated sample with a sputtered 2-nm gold layer and 28 nm of titanium. It is apparent that when measuring at a sharp angle (9°, red curve), only titanium dioxide with a binding energy of about 459 eV was detected, while at a perpendicular angle measurement, a metal with a binding energy of about 454 eV also appeared in the analyzed layer. The measurement, thus, shows that titanium metal can also be found at a very small depth below the surface layer of the oxide. In a similar comparison for the other samples with the top layer of titanium, the presence of metal in the surface layer was detected on both unheated and heated samples, except for the combination of 25 nm of gold and 5 nm of titanium, where only the titanium oxide layer was detected. The results of the samples with the gold bottom layer and the titanium upper layer are shown in Figure 8B. The thickness of the upper titanium layer was 5, 20, 25, and 28 nm. In the case of unheated samples, the presence of gold appeared only for the samples with the greatest Au thickness of 25 nm. The concentration of titanium was about one-third compared to samples with a gold surface. At the same time, there was a much higher proportion of oxygen, which indicates the oxide layer formed in the Ti nanolayer. A slight decrease in titanium concentration was observed on heated samples. The presence of pure metal or titanium dioxide was examined in more detail. Figure 9 shows measurements at two angles (9 • and 90 • ) for an unheated sample with a sputtered 2-nm gold layer and 28 nm of titanium. It is apparent that when measuring at a sharp angle (9 • , red curve), only titanium dioxide with a binding energy of about 459 eV was detected, while at a perpendicular angle measurement, a metal with a binding energy of about 454 eV also appeared in the analyzed layer. The measurement, thus, shows that titanium metal can also be found at a very small depth below the surface layer of the oxide. In a similar comparison for the other samples with the top layer of titanium, the presence of metal in the surface layer was detected on both unheated and heated samples, except for the combination of 25 nm of gold and 5 nm of titanium, where only the titanium oxide layer was detected.
Pharmaceutics 2021, 13, x FOR PEER REVIEW 13 of Figure 9. Representation of titanium oxide in the surface layer determined by XPS; two different angles, 9°, and 90°, were used for an unheated sample with sputtered 2 and 28 nm layers of gold and titanium, respectively. The red and green curves show measurements at the angles of 9° and 90°, respectively.
The composition of the sample in weight and atomic percentages was also monitore by EDS. Samples with a bottom layer of gold or titanium with a thickness of 5, 10, and 2 nm were measured, either as-sputtered or after the heating process. The results of the ED analysis are shown in Tables 3 and 4. For both series of samples, the increasing thicknes of the lower metallic layer and the changing ratio of the deposited films led to a corr sponding change in the percentage of gold and titanium. The heat treatment of sample had no significant effect on the atomic concentration of the metals, even though the su face morphology was changed drastically and the glass transition temperature of the po ymer was reached. This phenomenon was probably due to the depth of reach of this an lytical method, which extended up to 50 nm, in contrast to the XPS method, where it wa about 1 nm and, therefore, even changes on the very surface were determined by XPS; n such dramatic changes were determined by EDS.  The composition of the sample in weight and atomic percentages was also monitored by EDS. Samples with a bottom layer of gold or titanium with a thickness of 5, 10, and 25 nm were measured, either as-sputtered or after the heating process. The results of the EDS analysis are shown in Tables 3 and 4. For both series of samples, the increasing thickness of the lower metallic layer and the changing ratio of the deposited films led to a corresponding change in the percentage of gold and titanium. The heat treatment of samples had no significant effect on the atomic concentration of the metals, even though the surface morphology was changed drastically and the glass transition temperature of the polymer was reached. This phenomenon was probably due to the depth of reach of this analytical method, which extended up to 50 nm, in contrast to the XPS method, where it was about 1 nm and, therefore, even changes on the very surface were determined by XPS; no such dramatic changes were determined by EDS.   Figure 10A shows the change in wettability for unheated and heated samples with a titanium bottom layer and a gold upper layer with a total thickness of 50 nm. For a sample, at which 50 nm of gold was deposited without a titanium bottom layer, the measured water contact angle was 63.8 • ; for glycerol it was 61.8 • . The influence of the lower layer of titanium was already very significant when its thickness reached 2 nm. For samples with 2 nm of titanium as the upper layer and unheated samples, the water contact angle increased by 25%; a less significant increase was observed for glycerol. The increasing thickness of the titanium layer led to an increase in the contact angle for both liquids. The presence of a titanium bottom layer in the samples induced a decrease in the water contact angle for all studied samples. The opposite effect was observed for samples with a gold bottom layer ( Figure 10B). For subsequently deposited Ti, the water contact angle decreased for a lower Ti layer, followed by contact angle restoration at values still lower than for deposited samples without Ti. The subsequent heating of PLLA with an Au bottom layer stabilized the water contact angle at a narrow interval of 86-90 • .  The results of the zeta potential measurements are shown in Figure 11. The values were obtained for gold and titanium bottom layer thicknesses of 5 and 25 nm, respectively, both for unheated and heated samples. The unheated samples with 5-nm gold and 25-nm titanium layers and with 5-nm titanium and 45-nm gold layers had a lower zeta potential compared to unmodified PLLA. In both cases, an increase in the thickness of the metallic bottom layers led to an increase in the zeta potential. The heated samples with a 5-nm metallic bottom layer had a higher zeta potential compared to unheated samples; the increase in thickness of the bottom layer led to a decrease in the zeta potential. For unmodified PLLA, the values did not vary significantly. If we compare the determined zeta potential and contact angle, it can be concluded that with an increasing contact angle, the zeta potential increased for unheated samples with a gold bottom layer as well as with samples with a titanium bottom layer.

Antibacterial Activity of the Samples
Antibacterial properties were monitored on unheated and heated samples with a top Ti layer with thicknesses of 5, 20, and 25 nm. For comparison, the CFU of E. coli was determined for unmodified PLLA (designated pristine) and in the control group. A decrease in the CFU of E. coli was evaluated after 2 and 24 h of incubation with the samples. These two incubation periods were selected to determine the immediate and long-term antibacterial activity. The results are shown in Figure 12A,B. After a 2-h incubation of the samples with E. coli, it was apparent that there was no immediate antibacterial effect. However, the E. coli colonies grown on agar plates after incubation with samples of 20-and 25-nm thick titanium layers in the surface were about half the size of the untreated control (E. coli in physiological solution for 2 h), even though the number of CFU was comparable for the samples as well as the controls. The smaller colonies suggest that there was a bacteriostatic effect. The heating of the samples did not affect the antibacterial activity in any manner. thick titanium layers in the surface were about half the size of the untrea coli in physiological solution for 2 h), even though the number of CFU w for the samples as well as the controls. The smaller colonies suggest that th teriostatic effect. The heating of the samples did not affect the antibacterial manner. However, the samples showed a significant antibacterial effect after 2 (in contact) with the E. coli bacterial suspension. The results of the mea shown in Figure 12B. The partial antibacterial activity was also manifested PLLA. However, the effect was much more pronounced on modified pol With the increasing thickness of the upper titanium layer, a pronounced a fect was detected. Additionally, it can be concluded that for samples wi layers, there was no noticeable effect of heating. The observed decrease in CFU of E. coli during 24-h exposure of the samples compared to the contro marized in Table 5. The antibacterial effect is supported by the presence of However, the samples showed a significant antibacterial effect after 24-h incubation (in contact) with the E. coli bacterial suspension. The results of the measurements are shown in Figure 12B. The partial antibacterial activity was also manifested on unmodified PLLA. However, the effect was much more pronounced on modified polymer samples. With the increasing thickness of the upper titanium layer, a pronounced antibacterial effect was detected. Additionally, it can be concluded that for samples with "thicker" Ti layers, there was no noticeable effect of heating. The observed decrease in the number of CFU of E. coli during 24-h exposure of the samples compared to the control group is summarized in Table 5. The antibacterial effect is supported by the presence of TiO 2 (Figure 9), where despite that, no further excimer irradiation (ultraviolet wavelength) was performed; the ultraviolet part of the ambient irradiation likely supported the results.  Figure 13 represents images of E. coli CFU on LB agar plates inoculated after incubation with selected samples. Figure 13 shows a reduction in bacterial growth after 24 h of contact with PLLA-modified samples. The sample, which was thermally stressed, exhibited an effect on the shape and size of the bacterial colonies. Compared to other images, the boundaries of the colonies are blurred. The same difference in shape for the bacteria that were in contact with the samples was observed for all combinations of metal thicknesses used before and after heating. The photocatalytically active layer of titanium dioxide formed on their surface after sputtering due to contact with air, which enhanced the antibacterial properties of the samples. Because daylight with no further irradiation supplies the samples with relatively low energy and TiO 2 is present in the sample in small amounts, the photocatalytic activity was not very high. After 2 h of contact, no bactericidal effects were observed and, therefore, a longer, 24-h treatment was required. The results, thus, agree with the data found in the literature. where despite that, no further excimer irradiation (ultraviolet wavelength) was performed; the ultraviolet part of the ambient irradiation likely supported the results.  Figure 13 represents images of E. coli CFU on LB agar plates inoculated after incubation with selected samples. Figure 13 shows a reduction in bacterial growth after 24 h of contact with PLLA-modified samples. The sample, which was thermally stressed, exhibited an effect on the shape and size of the bacterial colonies. Compared to other images, the boundaries of the colonies are blurred. The same difference in shape for the bacteria that were in contact with the samples was observed for all combinations of metal thicknesses used before and after heating. The photocatalytically active layer of titanium dioxide formed on their surface after sputtering due to contact with air, which enhanced the antibacterial properties of the samples. Because daylight with no further irradiation supplies the samples with relatively low energy and TiO2 is present in the sample in small amounts, the photocatalytic activity was not very high. After 2 h of contact, no bactericidal effects were observed and, therefore, a longer, 24-h treatment was required. The results, thus, agree with the data found in the literature.

Conclusions
Heat treatment of the PLLA with thin metallic layers (Au/Ti, Ti/Au) to the polymer Tg led to the formation of a wrinkled pattern, which was caused by relaxation of the stress present inside the polymer film during heating. The structure showed an ordered direction over larger areas. The stress release in the polymer occurred already during sputtering, probably due to the high energy of titanium. This relaxation was manifested by various mechanisms, namely the formation of secondary structures, for example, crosslinked bends, structures with higher porosity compared to other samples, or randomly oriented primary wavy structures with low surface roughness. The formation of a wrinkled structure after heat treatment of the polymer led to an intense increase in the surface roughness of all samples. The roughness was affected by the type of deposited metal, where titanium leads to significantly higher values of surface roughness than the gold of the same thickness. XPS analysis of the surface layer composition showed the formation of a titanium dioxide layer on samples with a deposited gold bottom layer and titanium upper layer. The titanium layer deposited below the gold layer led to a decrease in wettability, while the contact angle for water and glycerol increased with the growing thickness of a titanium layer. After longer action of 24-h incubation with the samples, the decrease in CFU of E. coli was up to 85% (compared to control) for PLLA with a titanium layer of the biggest thickness. Overall, the antibacterial effect significantly increased with the growing amount of Ti on the surface. The bactericidal properties were induced by the formation of titanium dioxide on the surface of the modified PLLA, which was confirmed by XPS analysis. As a future outlook, we plan to combine titanium with other noble metals, such as silver and palladium, or titanium and carbon nanolayers. Moreover, we plan to study the wrinkling phenomenon for these combinations as well as the antibacterial properties of activated surfaces. Additionally, we plan to study the influence of a high-energy excimer laser on the prepared pattern.