Zn:DLC films via PECVD-HIPIMS: Evaluation of antimicrobial activity and cytotoxicity to mammalian cells

The application of antimicrobial products on textile materials can contribute to health and environmental benefits. Its action can prevent or minimize microbial colonization and proliferation that generate unpleasant odors to the fabric or microbial transfer to other surfaces, contributing to the hygiene of individuals and environments. In addition, antimicrobial textile treatments can increase the durability of fabrics and reduce the frequency of fabric washing, contributing environmentally to water and energy savings.¹ Besides, recently with the coronavirus pandemic, the interest in medical textiles has further increased.² Antimicrobial textiles and “smart” protective materials can benefit not only healthcare professionals but also individuals, providing effective and sustainable control against bacteria, viruses, and fungal spores.^3,4 In view of this, the textile industry seems to be expanding the use of antimicrobial textile materials (woven and nonwoven) in novel applications such as water and air purification systems, domestic textiles (bedding and clothing fabrics), technical textiles (sports, multidisciplinary professionals), automotive textiles, food industries, and hygiene products (facial masks, gowns caps, cloths).^3–5 

There are several approaches to modifying the properties of textiles and producing high antimicrobial surfaces.^2,6 In recent years, the deposition of antimicrobial coatings on textile surfaces has been highlighted to interrupt the transmission route of microorganisms and ensure specific properties to the material.² Several types of thin films have been studied for application as antimicrobial coatings.^7–13 Diamond-like carbon (DLC) films have stood out for presenting attractive mechanical and tribological properties, biocompatibility, and chemical inertness. Despite the wide application of DLC films in the industry, only recently these coatings have received much attention for application in multifunctional textiles,¹⁴ expanding the applicability of these films in several areas such as self-cleaning textiles,¹⁵ water-oil separation systems,¹⁶ and as a flame retardant material.¹⁷ In particular, in the biomedical field, research has shown that the deposition of DLC films on textiles increases the material's antimicrobial protection, expanding its biomedical applications as coatings of commercial items.^18,19

One of the focuses of studies in recent decades is the incorporation of other elements in DLC films to improve their antimicrobial properties.^20,21 The incorporation of metals to modify other properties of the DLC (Me-DLC) has been carried out since 1987 when Dimigen et al. evaluated these films with different types of materials for the first time.²² Besides, the incorporation of a metallic and nonmetallic material in the DLC films can ensure other benefits such as lower intrinsic compressive stress, and better mechanical and tribological properties, among others.²³ Different elements have already been incorporated into the DLC structure, and their properties have been studied and expanded for different application fields^24–27 Also, several studies were carried out as a function of different deposition techniques.^9,21,28 The literature presents the results of films grown with metallic and nonmetallic materials via a plasma-enhanced chemical vapor deposition (PECVD), physical vapor deposition (PVD), or a combination of both techniques, resulting in coatings with different compositions, properties, and structural qualities.^25,29,30

Metallic and nonmetallic materials have been previously incorporated into the film structure to combine the excellent tribological and mechanical properties of DLC coatings with the antibacterial action of different nanoparticles.^19,31,32 The film's antimicrobial efficacy depends on the type and concentration of the material incorporated into its structure.^31,33 In addition, another parameter that influences the antimicrobial properties of the film is the deposition technique. The literature reports the study of the antimicrobial activity of DLC films deposited via different techniques including radio frequency magnetron sputtering (RFMS),^25,32 PECVD,^20,21,31 and thermionic vacuum arc (TVA).³⁴ Besides, the antibacterial activity of metal DLC films deposited via high-power impulse magnetron sputtering (HIPIMS) was also studied.⁹ 

HIPIMS presents several advantages concerning conventional magnetron sputtering techniques, such as higher plasma density and ionization level, and the deposition of coatings in complex geometry. These characteristics of the HIPIMS process improve the film's properties such as density, roughness, and electrical and optical properties.³⁵ In addition, the antibacterial activity of films grown via HIPIMS and DC magnetron sputtering (DCMS) has already been studied and their efficacy has been compared. Some studies performed with silver and copper films grown on textiles via both techniques showed that depositions via HIPIMS had higher antibacterial efficacy than those grown via DCMS.^2,8,10

In particular, zinc (Zn) is well known for its antimicrobial properties.^36,37 In this work, DLC films with zinc (Zn:DLC) were grown on textiles via a combined PECVD and HIPIMS technique. In this study, the antimicrobial property of the coatings against a Gram-negative bacterium (E. coli), a Gram-positive bacterium (S. aureus), and a fungus (C. albicans) was studied as a function of the peak current, incubation time, and the percentage of metal in the film matrix. Additionally, the cell viability of keratinocytes was evaluated after indirect exposure to these textiles.

Zn-DLC films were grown via a PECVD-HIPIMS technique with a BAS 450 machine. For the film deposition, a 2000/1000 BP MAGPULS generator charged by an Advanced Energy Pinnacle^™ was used. The substrate support was in the rotation mode, and the film growth was performed without applying a bias voltage. Initially, an Ar (99.999%) plasma was used to sputter-clean the samples and the target for 30 min. Thus, an Ar flow of 100 SCCM was set with a working pressure of 2.5 Pa and a voltage of −400 V. Afterward, a 200 nm thick chromium (Cr) interlayer was deposited via a direct current magnetron sputtering (DCMS) process. A chromium target with dimensions of 254 × 127 × 12 mm³ and a purity level of 99.95% were used in an atmosphere of Ar with a flow of 50 SCCM. The applied power and working pressure were 2000 W and 0.5 Pa, respectively. Finally, approximately 1 μm thick Zn-DLC film was grown, for that the deposition time was adjusted for each experiment.

As a source of metal, a Zn target with a purity level of 99.99% was used, with a dimension of 254 × 127 × 12 mm³, and located 16 cm from the substrate holder. Films were grown on silicon, glass, steel 100Cr6, and textiles samples in an atmosphere of Ar and C₂H₂ (99.6%). The working pressure was 0.5 Pa (controlled with a butterfly valve), starting from a base pressure of 10⁻³ Pa. The pumping system used was a mechanical and turbomolecular pump from Pfeiffer with a pumping speed of 1450 l/s. The average power was 500 W, the on-time was 100 μs, and the peak current varied from 50 to 175 A. In all processes, the Ar flow was set at 50 SCCM. The metal concentration in the film matrix was adjusted by the variation in the acetylene gas flow, which varied from 0 to 60 SCCM. Afterward, the film's properties were evaluated, and their antimicrobial activity and cytotoxicity were studied as a function of the concentration of metal in the DLC matrix and the peak current.

Wear analyses were performed with a kaloMAX NTII equipment using a 30 mm diameter steel ball with a speed of 5.6 m/min and a force of 0.5 N. For all tests, an abrasive of aluminum oxide in glycerin was used. Besides, the wear was measured in a controlled environment with 21 °C and relative humidity of 45%. Afterward, to calculate the worn volume, the depth of the worn region was measured using a DektakXT Bruker profilometer. Finally, the wear rate was calculated using the equation: W_v = V_W/Fs, where W_v is the wear rate, V_W is the worn volume, F is the applied load, and s is the sliding distance. Then, to measure the coefficient of friction (COF), a pin-on-disk tribometer was used with a 5 mm diameter 100Cr6 steel ball. The applied force was 3 N with a speed of 76 mm/s (23 °C, 43% rh). The tribological properties of the film were tested for depositions on substrate samples of steel 100Cr6.

To determine the process growth rate, the film thickness was also measured with a DektakXT Bruker stylus profilometer. The morphology of the films was evaluated using a TESCAN MIRA3 scanning electron microscopy with field emission gun (SEM–FEG). In addition, the topography of the coatings was studied with an atomic force microscope (AFM) with a Shimadzu SPM 9500 and the roughness of the film was measured in areas of 5 × 5 μm². Raman analysis was performed to evaluate the film's microstructural quality with a Horiba Scientific LabRAM HR Evolution equipment with a laser of 514 nm wavelength, previously calibrated with a diamond peak at 1332 cm⁻¹, and the curves' deconvolution was determined using software FityK. The mechanical properties were analyzed with a microhardness test instrument from Fischerscope with a maximum load of 1 mN and a depth of 10% of the total film thickness. Also, the film's phase was analyzed with x-ray diffraction (XRD), using PANalytical Empyrean equipment. Finally, to determine the chemical composition of the coatings, electron probe microanalysis (EPMA) was used with Cameca SX 100.

The evaluation of the antimicrobial activity of the films was carried out using two different standards. Initially, a screening test was performed with the ISO 22196 standard with modification. Afterward, the analysis of the antimicrobial activity on textiles was carried out following the AATCC TM100:2019 standard.

At first, a screening test was performed with the ISO 22196 standard with modification. These initial tests allowed us to perform a screening of the material to continue the experiments on other surfaces and spanning other microorganisms. Thus, the antibacterial activity of the Zn:DLC samples was evaluated as a function of the film composition and the peak current.

The Gram-negative bacterium Escherichia coli (Lederberg W1485) was seeded on NZ amine, casamino acids, and yeast extract medium (NZCYM) agar and incubated at 37 °C for 24 h, under aerobiosis. The culture plate was prepared with 21.98 g/l of NZCYM and 1.5% agar with a pH of 7.5. Before the test, all the samples were sterilized with ethanol and dried at room temperature. For the procedure, the bacterium suspension with a concentration of 2.5–10 × 10⁵ CFU/ml was prepared with an NZCYM culture medium with an optical density (OD) of 0.5. For the measurement, a BIOCHROM Ultrospec 10 photometer (λ_600nm) was used. Afterward, an aliquot of 50 μl of the bacteria suspension was dropped on the glass with and without the coating and covered with another glass with an area of 1200 mm². Then, different incubation times were applied: 1 min, 1 h, and 2 h, at 37 °C. The incubation time of 1 min was applied to validate the test, following the standard. Thereafter, the samples were washed with 10 ml of soybean casein digest lecithin polysorbate (SCDLP) solution. Then, an aliquot of 200 μl of the bacterium suspension was seeded on the culture plate of agar NZCYM and incubated at 37 °C for 24 h; finally, the counting of units of colony (CFU) was performed. The assay was executed in duplicate at different times.

The previous analysis allowed us to have the first impressions of the antimicrobial activity of the material and to select some parameters to continue the studies. Thus, after the screening test, the antimicrobial activity of the coatings was assessed on textiles (PET fabric). Therefore, the antimicrobial properties for both treated and untreated textile materials were evaluated against Gram-positive bacterial species Staphylococcus aureus (ATCC 6538), Gram-negative Escherichia coli (ATCC 10799), and fungal Candida albicans (ATCC 18804). The quantitative antimicrobial assessment was performed according to the methodology described by the AATCC TM100:2019 standard. At first, the microorganisms were seeded on agar brain heart infusion (BHI) for bacteria and agar Sabouraud dextrose (SD) for C. albicans and incubated at 37 °C for 24 h. Afterward, the microorganisms were cultivated in tryptic soy broth (TSB) at 37 °C for 18 h in a shaker at 150 rpm. Afterward, textile materials in standardized sections (3.8 × 3.8 ± 0.1 cm and mass equal to 1 ± 0.1 g) were disinfected in alcohol 70% for 10 min, followed by drying and sterilized by exposition to UV light for 15 min on each side. Then, the overlaid swatches were stored in a closed sterile container. An aliquot of 1 ml of the standardized suspension in TSB with Triton X-100 at a concentration between 1 and 3 × 10⁵ CFU/ml was added separately on the textile material and incubated at 37 °C for 24 h. Afterward, the swatches were neutralized with 100 ml of neutralizing broth, followed by vortexing for 1 min. Tenfold serial dilution was performed and seeded on nutrient agar, followed by incubation at 37 °C for 24 h. Finally, the number of colony formation unit per milliliter (CFU/ml) and the percentage reduction rate (R%) were calculated. The assay was performed in duplicate at different times.

The morphology of the textiles with the microorganisms was analyzed by a scanning electron microscope. At first, the microorganisms were fixed for 24 h with 2.5% of glutaraldehyde (Sigma) prepared with 0.1 M phosphate buffered saline (PBS, pH 7.4). Afterward, the swatches were washed four times in PBS and followed by an additional wash for 15 min. Then, a dehydration process was carried out with 50%, 80%, 90%, and 100% of ethylic alcohol for 15 min each and dried. Finally, a Denton Vacuum Desk II metallizer was used to deposit a thin layer of gold on the samples, the time was set to 60 s, and the current was set to 40 mA. Then, a TESCAN MIRA3 SEM–FEG was used to analyze the morphology of S. aureus, E. coli, and C. albicans on treated (samples with Zn:DLC) and untreated (negative control) textile materials after 24 h contact.

For this test, normal oral keratinocytes (NOK) were maintained at 37 °C and 5% CO₂ in Dulbecco's modified Eagle medium (DMEM), supplemented with 10% of fetal bovine serum and 1% of penicillin (100 U/ml)/streptomycin (100 mg/ml). The cells were seeded at a density of 8 × 10³ cells per well in 96-well plates and incubated for 24 h to allow the cell adhesion. Afterward, indirect contact of the cells with the extracts conditioned by the textiles was performed according to ISO 10993-5. For this, the textiles were soaked in DMEM for 1 and 7 days. After, the cells were exposed to the extracts for 24 h when the cell viability was analyzed. For this, 100 μl of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) was added to the wells and the plates were incubated for 1 h. Then, formazan crystals were dissolved with dimethyl sulfoxide (DMSO), and the resulting optical density of the solution was measured in a spectrophotometer at 570 nm. Absorbance data were normalized to the untreated control group (=100%). Two independent experiments were carried out using five replicates (n = 10). Cell viability below 70% was considered cytotoxic (ISO 10993-5). Additionally, the data were statistically analyzed by the Kruskal–Wallis test (p < 0.05).

At first, the antibacterial efficacy of the Me-DLC films was evaluated following the ISO 22196 standard. Thus, the antimicrobial activity of the Zn:DLC films was tested as a function of the peak current and the metal content. The maximum peak current was defined with preliminary tests, by observing the stability of the system. Therefore, the peak current varied from 50 A to a maximum value of 175 A. The metal content in the DLC matrix was varied by controlling the flow of acetylene gas, and to determine the flow values used, the hysteresis curve of the process was evaluated. In reactive HIPIMS, the reactive gas also interacts with the target, resulting in the formation of a compound material. This results in the modification of the sputtering yield and the yield of secondary electrons.³⁸ Therefore, by adding acetylene gas to the deposition chamber, the target mode is changed and can be evaluated by the hysteresis curve. Thus, for the experiments, the acetylene gas flow was defined up to the middle of the hysteresis curve from 0 to 60 SCCM.

Figures 1(a) and 1(b) show the film's growth rate and composition as a function of the peak current. The coatings were deposited with a fixed acetylene gas flow of 40 SCCM. The film's growth rate varies with the deposition parameters. Even for fixed target power and pressure, there is a modification of the metal sputtering rate depending on the HIPIMS peak current and the acetylene flow. Figure 1(a) illustrates a small increase in the growth rate with an increase in the peak current, varying from 27.8 to 34.6 nm/min. In addition, Fig. 1(b) shows that for an increase in the peak current, there was only a slight decrease in the Zn content in the films, varying from 55.5 to 51.1 at. %.

The peak current was adjusted by controlling the voltage that varies from 750 to 832 V. Setting a higher voltage leads to a higher electron density in the plasma and increases the peak current. This results in a higher ionization rate and increases the probability of the ions to return to the target, resulting in a lower deposition rate.³⁹ However, it is visible a different trend in Fig. 1(a). For reactive HIPIMS, there are competing effects. Thus, with increasing ionization, the target surface stays more metallic, i.e., less compound formation with a lower sputter yield on the target surface. But with the increasing peak current, the deposition rate will drop due to the resputtering of the target ions toward the target. As a third effect, less reactive gas is required to realize a similar working point.^40,41 Therefore, the combination of these effects seems to result only in a slight reduction in the Zn content and a small increase in the deposition rate with increasing peak current for fixed gas flow.

Afterward, the morphology and topography of the films were evaluated as a function of the peak current (Fig. 2). It can be observed in Fig. 2(a) that the films present a flat surface, and there is an indication of a smaller grain size for a higher peak current. Higher energy associated with the higher peak current could result in smaller cluster sizes.⁴² Besides, Fig. 2(b) shows the film topography. Note that the roughness of the film reduces with increasing peak current from 14.78 to 5.48 nm. The influence of the peak current of the HIPIMS process on the film's topography has already been demonstrated.⁴³ Alami et al. studied the effect of the peak current on the morphology of CrN films deposited via HIPIMS and without applying a substrate bias. The authors showed a modification in the film's structure and morphology and indicated shrinking grains size by increasing the peak current. In addition, the increase in the peak current also results in a smoother surface.^35,43

After analyzing some properties of the film, the antibacterial activity of the coating was evaluated as a function of the peak current. Thus, Table I shows the antibacterial activity of the Zn:DLC samples as a function of the peak current using the ISO 22196 method. By varying the peak current for an incubation time of 2 h, no variation in the antibacterial activity was observed and all the samples showed 100% of efficacy against E. coli; i.e., no CFU was counted. However, by reducing the incubation time to 1 h, an increase in the antibacterial activity is noted by increasing the peak current, and the antibacterial activity was 2.0 ± 0.7 and 4.0 ± 0.2 for 50 and 175 A, respectively. The statistical analysis of the coating deposited with 50 A showed that this sample was not significant, indicating that there is no statistical difference in relation to the control group. On the other hand, by increasing the peak current to 175 A, the statistical difference in relation to the control was significant (p ≤ 0.001), showing higher efficacy against E. coli. For 1 h of incubation time, the reference sample (negative control) had approximately 6.5 × 10⁵ CFU/ml, and for the Zn-DLC samples grown with 50 A, the number of CFUs was reduced to 1.3 × 10⁴ CFU/ml. Besides, by increasing the peak current to 100 A, this value reduces to 2.5 × 10³ CFU/ml and reaches 0 CFU/ml for 175 A. These results indicate an increase in the antibacterial activity of the coating by increasing the peak current.

Note in Figs. 1 and 2 that the variation in the peak current influences the film's roughness and composition. Jeyachandran et al. assumed that there are two main parameters related to the bacterium interaction with the material: surface roughness and chemical composition.⁴⁴ Regarding the composition, the literature shows that films with a higher percentage of metal have greater antibacterial activity.³⁶ However, note in Fig. 1 that the coatings grown with 50 and 175 A have a percentage of zinc of 55 and 51 at. %, respectively. Thus, the coating deposited with a 50 A of peak current shows slightly higher percentage of zinc. Therefore, despite presenting a higher percentage of zinc, the film grown with 50 A is less effective against E. coli. Furthermore, the analysis of the film's topography (Fig. 2) is an important parameter to evaluate the antibacterial effectiveness of the coatings. Several studies show how surface roughness can influence the biofilm formation since this parameter is related to the mechanical retention of the bacteria.^44–46 As shown in the AFM images, by increasing the peak current, the film roughness decreases. Thus, the higher antibacterial activity of Zn:DLC for the higher peak current could be associated with the smoother surface, films with higher roughness result in higher possibility of bacterial adhesion and lower efficacy against the microorganism.

However, the correlation between roughness and antimicrobial activity is more complex than that, and some studies indicate a roughness threshold (Ra: 200 nm) that could influence the microbial retention, and below this value, a reduction in the bacterial accumulation could not be expected.⁴⁷ On the other hand, some authors stated that the microbial retention is only influenced by the surface roughness when the roughness dimensions are similar to the microorganism.⁴⁸  E. coli has dimensions of approximately 1 μm, and the surface roughness is ranging from 5.48 to 14.78 nm. Thus, the reduction in antibacterial activity shown in Table I could not be related to the surface roughness.

Besides, increasing the peak current increases ionization, which affects some other coating properties, such as the grain size and consequently its surface area, as shown in the SEM/FEG images of the film's surface, where there is an indication of smaller grain size for higher peak current. This parameter also influences the antibacterial activity; i.e., a higher surface area results in higher antibacterial activity.^49,50 Therefore, it was observed that with increasing ionization in the process, at constant acetylene flow and constant average power and pressure, the antibacterial activity is increased.

By analyzing the results of Sec. III A 1, it was observed that films grown with higher peak current resulted in less rough coatings and better antimicrobial activity. Therefore, in this section, deposited Zn:DLC films using a peak current of 175 A were studied, as these films showed better results against E. coli. The literature shows that the antimicrobial activity of metal DLC films is strongly dependent on the concentrations of other elements in the film matrix.⁹ Therefore, in this stage of the tests, the antibacterial activity of the Zn:DLC films was evaluated as a function of the Zn content. Figure 3 illustrates the film's growth rate and composition as a function of the acetylene gas flow.

Note in Fig. 3(a) that adding acetylene gas to the process reduces the growth rate from 49 to 19.13 nm/min. Besides, the growth rate ranged from 19.13 to 23.68 nm/min for 60 and 20 SCCM, respectively. In general, increasing the acetylene gas flow is expected to have a higher deposition rate, since there are more hydrocarbon molecules in the reactor atmosphere. However, in reactive sputtering, the growth rate could vary with the process due to the target poisoning. The formation of a compound material on the target surface leads to a lower sputtering rate and hysteresis effect.⁵¹ This could explain why increasing the acetylene gas for the Zn-DLC decreases the growth rate. Figure 3(b) shows the metal content in the Zn-DLC film as a function of acetylene gas flow for a fixed peak current of 175 A. It is visible that with increasing acetylene gas flow, i.e., more hydrocarbon molecules available, the percentage of metal in the DLC matrix decreases, as expected. The minimum value of this investigation was reached with 43.4 at. % of zinc for 60 SCCM of acetylene flow.

The XRD analyses of the films with different metal content are shown in Fig. 4, and the measurements were performed from 10° to 60° (2θ angle). Note that the pure Zn coatings show a hexagonal structure (ICSD 03-065-3358) with the characteristic peaks at the angles: 36.29°, 38.99°, 43.22°, and 54.32° for the diffraction planes (002), (100), (101), and (102), respectively. Other studies also show similar results for the deposition of pure Zn via HIPIMS on glass substrates⁵² and via DC magnetron sputtering.⁵³ However, by adding acetylene gas to the process, the film presents an amorphous structure, and even with the lowest gas flow tested, it is no longer possible to visualize the characteristic peaks of zinc. Furthermore, it is possible to obtain an approximate value of the material grain size using the Scherrer equation, which relates the full width at half maximum (FWHM) of the XRD analysis with the dimension of the particle.⁵⁴ Thus, using the FWHM obtained from the (101) plane, the diameter of the particles of the Zn samples is approximately 36.4 nm. Nonetheless, as the Zn:DLC samples show an amorphous structure, it was not possible to calculate the grain size for these materials.

The SEM/FEG and AFM images of the films as a function of the acetylene flow are shown in Fig. 5. Note that the surface morphology of the pure Zn films is completely different than the Zn:DLC samples. In the SEM/FEG image, it was possible to identify that the pure Zn sample presents a morphology composed of clusters of Zn particles. Using ImageJ software, it was calculated the diameter of this particulate. Therefore, the particles shown in Fig. 5(a) have a diameter of approximately 71.3 ± 13.4 nm. Note that the grain size measured by XRD is of the same magnitude as the micrograph analysis. However, this result is just an estimation since it is assumed a spherical geometry for the particulate.

Furthermore, note in the SEM/FEG images that by reducing the Zn concentration to 62.6 at. % (20 SCCM), the surface morphology changes considerably, and the Zn clusters are no longer visible, showing a homogeneous surface. In contrast, previous studies related a different morphology for Zn:DLC films, showing that films with the Zn content between 6 and 62.5 at. % did not present a homogeneous surface but a morphology containing Zn-rich areas separated by carbon-rich regions. In addition, these particulates became more frequent with the increase in the percentage of metal. However, the films were deposited via a combination of magnetron sputtering of a zinc target and plasma source ion implantation, and here we use HIPIMS/PECVD^55–57 

It is visible in Fig. 5 that the coatings deposited with 20–60 SCCM presented a relatively similar surface morphology and topography. Also, Fig. 2(f) shows the AFM image of the film deposited with 40 SCCM. The film's roughness is 4.82, 5.48, and 5.89 nm for 20, 40, and 60 SCCM, respectively. Thus, there is a variation in the roughness of the films by varying the metal content. In addition, the pure Zn coatings show a higher roughness of 133 nm, which can be explained by the completely different morphology of this film. Furthermore, the literature shows that the film roughness increases by increasing the Zn content in the Zn:DLC samples. However, in this case, the Zn:DLC films show a different morphology with agglomerates of Zn on the DLC surface.⁵⁵ 

For these initial analyses of the films, the antibacterial activity was evaluated for the coatings of Zn and Zn:DLC deposited with 40 and 60 SCCM. Table II shows the antibacterial activity of the samples against E. coli for 1 and 2 h of incubation time as a function of the acetylene gas flow. Note that by varying the acetylene gas flow, the films showed high antibacterial activity (R > 3) for both incubation times tested. For 1 and 2 h, all films showed 100% of antibacterial efficacy. Also, it is visible in Table II that by varying the acetylene gas flow to the maximum value tested of 60 SCCM (43.4 at. % of Zn, lowest metal content), the antibacterial efficacy remained high. Thus, there are no significant differences in the antibacterial activity of the films against E. coli by varying the metal content.

Furthermore, note in Table II that the antibacterial activity was reduced with the decrease in the incubation time. The calculation of the antibacterial effectiveness was carried out taking as a reference the uncoated glass sample. Therefore, although for 1 and 2 h, the antibacterial activity was 100%, the uncoated sample had a higher E. coli concentration for a longer incubation time. Thus, as the calculation of the R-value in Table II is based on the uncoated glass, the 2 h sample showed higher effectiveness. The antibacterial activity of the Zn-DLC coating can be explained by the interaction of the ions with the bacterial cell wall and its rupture. Amna et al. studied the antibacterial activity of Zn-doped titania nanofibers against E. coli and S. aureus. This study showed that the antibacterial activity of the material is a result of the rupture of the cell membrane and consequently cell lysis. This occurs due to the interaction of the Zn²⁺ with the bacterium cell walls.⁵⁸ 

These initial evaluations of the Zn-DLC films showed high antibacterial efficacy by varying the composition of the films. Additionally, the study also showed an increase in the antibacterial activity with the increase in the peak current, indicating an effect of ionization on the film's antibacterial properties. These tests made it possible to have an overview of the film's antibacterial actions for the study of the antimicrobial efficacy of the coatings on textiles.

After performing the screening tests with the ISO 22196 standard, the films were deposited on textiles to evaluate their antimicrobial activity against a Gram-negative bacterium (E. coli), Gram-positive bacterium (S. aureus), and a fungus (C. albicans). From the previous results, it was observed that a higher peak current resulted in greater antibacterial efficacy against E. coli. Therefore, for this test, the antimicrobial properties were evaluated as a function of the metal content for fixed peak current. Thus, in this section, the peak current was set at 175 A, and the metal content in the DLC matrix was varied by controlling the acetylene gas flow. In the preliminary tests, pure Zn and Zn:DLC deposited with 40 and 60 SCCM of acetylene were evaluated. Note that in Table II, there was no difference in the antibacterial efficacy between the films for the different fluxes. Thus, the pure Zn sample was not tested at this stage and the gas flow varied from 20 to 60 SCCM.

This section is divided into three parts. First, the film's mechanical and tribological properties are analyzed, and the material's structural quality is studied as a function of the metal content in the DLC matrix. Afterward, the antimicrobial activity of the coatings is evaluated following the ATCC TM100:2019 standard; subsequently, the material's morphology with microorganisms is shown. Finally, the Zn:DLC's cytotoxicity is studied as a function of the metal content.

Figure 6 illustrated the textile morphology for the uncoated and coated samples. The difference between the surface of the textile with and without the coating is visible. Thus, the micrography confirms the formation of the films on the textile's surface. Figures 2 and 5 show that the morphology of the films did not show significant differences by varying the Zn concentration in the DLC matrix. A similar effect can be observed in the textile's surfaces by varying the metal concentration from 43.4 to 62.6 at. %, where the morphology of the coatings is similar for all fluxes tested and with a homogeneous surface.

The Raman spectra of the films deposited with different gas flows are presented in Fig. 7. In addition, Table III shows the intensity ratio of the G and D bands (I_D/I_G), the FWHM of the G band (FWHM_G), and the G and D band's positions, and these parameters are related to density, size, and structure of the sp² clusters.⁵⁹ This information was obtained by removing the background and fitting the curve using the Gaussian function. Figure 7 illustrates the characteristic spectrum of DLC films showing photoluminescence background, with the D band around 1350 cm⁻¹, which is related to the breathing mode of the aromatic rings, and the G band around 1580 cm⁻¹, which correspond to the stretching mode of the sp² sites pairs in aromatic rings and olefinic chains.^59,60

As seen in Table III, the I_D/I_G ratio of the Zn:DLC films increases by decreasing the flow of acetylene gas, i.e., higher metal content. The Zn:DLC coatings with a flow varying from 60 to 20 SCCM showed an increase in the I_D/I_G ratio from 0.8 to approximately 1. This result agrees with previous studies in the literature, where films with higher Zn content show an increase in the I_D/I_G ratio.⁶¹ Nevertheless, note that there are no significant changes in the FWHM_G and the G-band positions by varying the metal content. The literature commonly related the I_D/I_G ratio, the FWHM_G, and the Raman shift of the G band of DLC films with the ratio of sp² and sp³ hybridization. Several studies have shown a qualitative estimation of the sp³ content by analyzing the data of the Raman spectrum.^59,62 In particular, studies done with Zn-DLC indicated a decrease in the sp³ fraction associated with an increase in the I_D/I_G ratio, a blue shift of the G band, and a reduction in FWHM_G.⁶¹ In this case, to calculate the sp³ content, a quadratic model for H-free DLC films was used, which relates the sp³ content with the FWHM_G.^59,61 However, the relationship of the Raman parameters with the sp³ content is nonlinear and varies with the hydrogen content, i.e., the deposition conditions. Furthermore, these parameters only show meaningful measurements for specific sp³ contents.^59,63,64 Nevertheless, some authors have shown that the Raman spectroscopy technique performed at single wavelength causes a strong resonance with the vibration modes of sp² hybridization; consequently, the diamond-like character of DLC films is not able to be directly deducted from the spectrum.^65,66 In DLC films, higher I_D/I_G ratio (usually related to sp² aromatic rings size), dislocation of G band to higher values, and reduction in the FWHM_G (structural bond distortions) are commonly associated with an evolution of the graphite amorphization trajectory toward higher sp² (graphite-like) dominion.^67,68

In addition, the hydrogen content in the DLC films can be estimated by the slope of the photoluminescence (PL) background and its relationship to the G-band intensity, and the PL background increases with the hydrogen percentage.^62,64 Besides, the increase in the PL background could also be related to the incorporation of metal into the film matrix. Studies performed with Zn:DLC showed an increase in the PL background with increasing Zn content.⁵⁵ However, note in Fig. 7 that there are no significant differences in the PL background by varying the Zn content from 43.4 to 62.6 at. %. Perhaps, this relationship is more visible for lower concentrations of metal since the literature described this effect for films with the metal content from 2.5 to 62.5 at. %.⁵⁵ 

Table IV shows the film's mechanical and tribological properties such as COF, wear rate, hardness (H), elastic modulus (E), and the plastic index parameter (H/E) as a function of the acetylene gas flow and the metal content. The Zn:DLC films had a hardness between 2.3 and 3.3 GPa for a metal content from 43.4 to 62.6 at. %. Note in Table IV a small increase in the hardness by reducing the Zn content. The literature shows that ZnO:DLC films presented a hardness lower than 5 GPa for a coating deposition using a Zn 10 mol. %/C target.⁶⁹ In this study, we obtained hardness below 5 GPa for the metal content higher than 43.4 at. %, and this was the lowest concentration of Zn tested. Depending on the type of metal incorporated in the DLC matrix, the film hardness may increase or decrease. For example, by adding materials such as Mo and Ti, the film hardness can be increased up to 31.8 GPa.⁷⁰ However, previous studies show that the addition of soft materials such as Zn, even at low concentrations, results in films with lower hardness. Hatada et al. studied the properties of Zn-DLC films and found a reduction in the film hardness with increasing Zn content. For Zn:DLC films with 2.5 and 3.8 at. % of Zn, the hardness reduced from 9.5 to 8.8 GPa, respectively.⁵⁵ Thus, the results in Table IV agree with the literature. All the coatings showed low hardness; however, since a high Zn content was incorporated in the DLC matrix (higher than 43 at. %), this result was expected.

Furthermore, the hardness of DLC films is influenced by their structural quality. Films with a higher percentage of sp³ hybridization have relatively good mechanical properties.⁷¹ Thus, the increase in the film hardness with decreasing metal content is also a result of the change in the structural quality of the film.⁵⁵ Wong et al. varied the percentage of Zn on the DLC matrix and showed the influence of the Zn content on the sp³ fraction. Films with lower Zn concentrations have better structural quality with a higher percentage of sp³ hybridization.⁶¹ The Raman analysis shows a decrease in the I_D/I_G ratio with a decrease in the Zn content. Therefore, Fig. 7 and Table III suggest an increase in the sp² hybridization content with the Zn percentage increase, which results in coatings with slightly poorer mechanical properties, as shown in Table IV.

Also, Table IV presents the film's tribological properties as a function of the metal content. The incorporation of metals in the DLC matrix could modify its tribological properties. The COF of the film could vary depending on the type of material and its concentration in the coating. Note that the COF does not show significant changes with the variation in the Zn content. The literature shows that the increase of the Zn percentage on the DLC structure could increase the coefficient of friction of the coating; Zn:DLC with 25 at. % had a COF of 0.6, and for 2 at. %, this value reduces to 0.3.⁵⁵ However, although some studies show that a higher Zn content modifies the COF of the films, others demonstrate that the coefficient of friction remains constant.⁶¹ However, here, it is visible that varying the Zn percentage in the DLC matrix from 42 to 63 at. % does not significantly affect the COF that presents a COF between 0.3 and 0.4. Note that this value is higher than what is expected for pure DLC coatings, where conventional DLC coatings could present a COF lower than 0.1.⁷² 

In addition, note in Table IV, a small increase in the plastic index parameter of the material for lower metal contents. For 62.6 at. % of Zn, the H/E is 0.071, and by decreasing the metal content to 43.4 at. %, it increases to 0.087. Through the plastic index parameter, it is possible to have an indication of the wear resistance of the material, and higher values of H/E suggest a coating with better wear resistance.⁷³ Previous investigations with Zn:DLC films have shown that the increase in the Zn content in the DLC matrix results in lower wear resistance.⁵⁵ However, note that the wear decrease with the increase in the Zn content, varying from 39.1 to 27.6 × 10⁻⁶ mm³/Nm, by decreasing the flow from 60 to 20 SCCM, respectively. This result can be explained with a modified H/f.E criterion, considering f the coefficient of friction.^67,74 To better understand the wear behavior, it is important to consider the surface microroughness and the chemical composition, which are influencing the nature of contact and friction between surfaces. Moreover, the incidence of film transformations according to the amorphization trajectory, the effect of the surface oxidation and passivation, and the total compressive stress of the film influence the wear behavior of this type of DLC coating.⁷⁵ In general, it is possible to improve surface microroughness by reducing the deposition temperature and controlling the recrystallization effects during the nucleation and growth of DLC films.⁶⁷ As seen in Fig. 5, a higher film roughness was achieved for the film deposited at 60 SCCM. As a result, it is suggested that slightly lower wear rates were achieved for Zn:DLC containing a higher metal content due to the important role of surface roughness during the wear mechanisms.

Figure 8 highlights the antimicrobial activity of the material as a function of the amount of CFU/ml. The Zn:DLC films on textiles show an excellent antimicrobial activity with a reduction percentage of approximately 100%. Significative reduction was observed in S. aureus (p < 0.0001) and E. coli (p < 0.0001) for all samples tested when compared to the control group. In addition, the Zn:DLC film deposited with 40 SCCM of acetylene flow promoted a 99.6% of significative reduction (p ≤ 0.001) against C. albicans.

As stated in Sec. III A 1, the film's roughness influences the possibility of microbial adherence and colonization. Note in Figs. 2 and 5 that there is a variation in the roughness of the films by increasing the flow of acetylene gas, ranging from 4.82 to 5.48 nm for 20 to 60 SCCM, respectively. As already discussed, previous research stated that the roughness is a significant variable in the antimicrobial activity when the surface roughness dimension is similar to that of the microorganism.^46,48 As the microorganisms tested are greater than 500 nm and the surface roughness of the films is approximately 100 times lower, the roughness does not have a decisive effect on the antimicrobial effectiveness of the coatings. In addition, several other coating parameters can influence the antimicrobial activity of the film; e.g., the literature also reports an influence of the film structural quality on bacterial adhesion, indicating a discrete reduction in the antibacterial activity with the reduction in the I_D/I_G ratio.^2,46,76 Table III shows a small reduction in the I_D/I_G ration with increasing acetylene gas flow. However, no significant differences were observed in the effectiveness of the films by varying the gas flow, with all coatings showing an excellent antimicrobial activity. In addition, by varying the Zn composition in the film matrix to a minimum value of 43.4 at. %, no significant differences were noted against the microorganisms tested.

Bacteriostatic effect of Zn/ZnO particles on the glass surface against E. coli, S. aureus, and C. albicans has already been shown.⁷⁷ Besides, the antibacterial and antifungal activity of zinc complexes has also been demonstrated, including the effect on E. coli and S. aureus.⁷⁸ On the other hand, Cazalini et al. tested Zn:DLC films on polypropylene fibers prepared by magnetron sputtering technique. The antimicrobial action of the material evaluated by halo-diffusion showed an intermediate effect for C. albicans and no action against E. coli and S. aureus.⁷⁹ Antimicrobial action of zinc ions and zinc oxide (ZnO) has been explored. Zinc ions can inhibit different physiological pathways, and ZnO presents cellular damage through a photocatalytic process or by the action of reactive oxygen species (ROS) caused by nanoparticles in the cytoplasm or cell membrane.^80,81

Afterward, the surface of the textiles with microorganisms was analyzed. Figures 9 and 10 show the SEM/FEG images of C. albicans, E. coli, and S. aureus after 24 h of contact with the textiles. In a general trend, microorganisms showed high growth in the untreated textile material and growth in the interstices of fibers. Growth of C. albicans on the uncoated material [Figs. 9(a)–9(c)] showed yeast and filamentous forms. Fungal images on Zn:DLC films produced in different gas flows showed only the yeast morphology with cell surface alterations [Figs. 9(d)–9(f)]. The images in Figs. 10(a)–10(c) demonstrate the remarkable bacterial growth of the E. coli biofilm on the untreated tissue fibers, while S. aureus showed sparse growth between the fibers. No bacteria were found on the surfaces of Zn:DLC textiles, corroborating the counts of viable cells, as seen in Fig. 8.

Figure 11 shows the analysis of the cell viability after 24 h of cell exposition to the extracts, which were kept in contact with the textiles for 1 and 7 days. A pure form (100%) and a diluted form of each extract (10%) were analyzed. The red dashed line indicates the normative limit of cell viability (70%). All textiles with Zn-DLC were highly and similarly cytotoxic when using 100% of the material extract. However, when using 10% of the extract, we could better visualize the differences between the groups. Note that the reduction in the Zn percentage, i.e., increase the flow of acetylene gas, led to an increase in the cell viability with no statistical difference of these groups for the uncoated textile (UNC). It was previously demonstrated that the deposition conditions of Zn:DLC films affect the Zn content released in a medium. Zn:DLC films deposited with a duty cycle of 100% releases 15 times more Zn than with 40%.⁸² Moreover, Zn may be toxic to osteoblasts depending on the concentration,⁸² which is in agreement with our findings. However, small concentrations, in addition to nontoxic, may favor some important biological properties such as osteogenesis with possible application in bone biology.^82,83

Additionally, a slight reduction in the cell viability was observed when the cells were put in contact with extracts exposed to the textiles for 7 days. Probably, higher amount of metal-based material was released into the medium over time, which could explain this result. It was previously demonstrated that the amount of Zn in the culture medium really increases as the time of exposure to the Zn:DLC becomes higher.⁸³ In this way, a control of Zn release should be performed in a range that is not toxic to mammalian cells.^82,83

Finally, Fig. 12 summarizes the results of this investigation. Note the correlation between the mechanical, chemical, structural, and antimicrobial properties and cell viability of the coatings as a function of the acetylene gas flow.

Recent epidemic events reinforced the need for the development of high-touch surfaces designed to control the transmission of viruses, bacteria, and fungus. In this work, Zn:DLC films were grown via HIPIMS/PECVD and their antimicrobial activity was evaluated as a function of peak current and composition. Initially, a screening test was performed to assess the antibacterial efficacy of the films against E. coli. The films showed high antibacterial efficacy by varying composition and incubation time. Furthermore, the antibacterial activity increased as the peak current increased, indicating an effect of ionization on the film's effectiveness against E. coli. Afterward, the antimicrobial activity of the coatings on textiles was evaluated as a function of the metal composition in the film matrix. The antimicrobial activity was evaluated against E. coli, S. aureus, and C. albicans. All parameters tested showed high antimicrobial activity with almost 100% of efficacy. Subsequently, the analysis of the cytotoxicity of the material showed that the films were toxic when testing 100% of the extract. However, by reducing to 10% of the extract, an increase in the cell viability is observed for a lower metal content. This indicates an influence of the composition of the coatings on their cytotoxicity. Besides, the Zn:DLC samples with a metal content higher than 43.4 at. % did not show significantly difference regarding the cell viability to the uncoated group. In addition, other properties of the films were investigated such as microstructural quality, tribological and mechanical properties, morphology, and composition. The analysis of the microstructural quality of the films suggested an increase in the size and number of the sp² dominion by increasing the Zn content, and consequently, the film hardness reduced. Also, it was observed that the variation in the Zn content did not change the friction coefficient of the coatings. However, a modified plastic index suggested that films with a lower metal content had poorer wear behavior due to the surface roughness and chemical composition of the surface.

This research was funded by FAPESP (Grant Nos. 2021/00046-7, 2020/12450-4, 2019/25652-7, 2017/08899-3, and 2012/15857-1), CNPq (Grant Nos. 132884/2020-8 and 309762/2021-9), Capes, the Fraunhofer Internal Programs under Grant No. Anti-Corona 046-600051, and the German Federal Ministry of Education and Research. The content of this paper was presented at the International Conference on Fundamentals and Applications of HIPIMS in virtual format, June 2021, and in Sheffield, UK, June 2022. In addition, the authors want to acknowledge Dr. Walter Toshi for helping with the AFM images.

The authors have no conflicts to disclose.

Rebeca F. B. de O. Correia: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Writing – original draft (equal); Writing – review & editing (equal). Aline G. Sampaio: Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Writing – original draft (equal); Writing – review & editing (equal). Noala M. Milhan: Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Validation (equal); Writing – original draft (equal). Ariel Capote: Investigation (equal); Methodology (equal); Writing – original draft (equal). Holger Gerdes: Conceptualization (equal); Investigation (equal); Methodology (equal). Kristina Lachmann: Conceptualization (equal); Investigation (equal); Methodology (equal); Supervision (equal). Vladimir J. Trava-Airoldi: Conceptualization (equal); Data curation (supporting); Formal analysis (supporting); Funding acquisition (lead); Investigation (equal); Methodology (equal); Project administration (equal); Resources (lead); Supervision (lead); Validation (lead); Writing – review & editing (equal). Cristiane Yumi Koga-Ito: Conceptualization (lead); Funding acquisition (equal); Investigation (lead); Methodology (lead); Resources (equal); Supervision (lead); Validation (lead); Writing – review & editing (equal). Ralf Bandorf: Conceptualization (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.