1. Introduction
Biomaterials are widely used in medical procedures for the replacement of tissues, organs, or bodily functions. Depending on the application, these materials differ in terms of structure, chemical composition, and processing. Under certain conditions, they may present limitations because they can suffer deformation, corrosion, and wear, increasing the possibility of failure of the implanted material [
1].
There are important concerns regarding the effects of metallosis caused by the infiltration of debris generated by the wear of metallic implants. Metallosis can occur in soft tissues and bones, with effects including infection and tissue necrosis, due to material corrosion that is usually associated with the improper use of metal in permanent implants in contact with body fluids. The problem of metallosis can necessitate implant replacement after periods shorter than ten years [
2].
Among the various types of metals used in implants, the present work investigates 316L stainless steel and Ti6Al4V alloy, which are both used in the femoral arches of a total hip arthroplasty. These metals are currently used because their mechanical properties are favorable for the machining of implants due to their low deformability, and they have moderate resistance to corrosion [
1,
2].
Table 1 shows the chemical compositions of the 316L steel and the titanium alloy, including metallic, nonmetallic, and semi-metallic (such as silicon) elements. It can be seen that SS316L contains 22% chromium, which, during the deterioration of implants, may be delivered to the body in the form of ions including Cr6+, which can cause the development of cancer [
3]. Other chemical elements present in lower concentrations include nickel, carbon, silicon, manganese, cobalt, and molybdenum. It has been suggested that the aluminum and vanadium found in titanium alloy may be responsible for Alzheimer’s disease [
4]. Nitrogen, carbon, hydrogen, iron, and oxygen are also present as contaminants in Ti6Al4V.
SS316L presents a high rate of biocorrosion when in contact with body fluids, requiring the use of surface treatments. In particular, diamond-like carbon (DLC) has attracted attention as a coating due to its low friction coefficient and good biocompatibility [
5].
Considering that SS316L and Ti6Al4V are still present in most of the implants used in the Brazilian Unified Health System (SUS), the aim of the present work was to perform a comparative study to evaluate the improvement of these metallic materials using DLC coatings to protect against tribocorrosion and to reduce cell damage. Some alternatives for metal protection have been reported and include using ceramic materials; for example, hydroxyapatite grown on a metal surface can act as a protective coating, as was explained very well by Nahum et al. [
1] and Hwartz et al. [
2]. The DLC advantage over other coatings is related to the low friction coefficient of DLC coatings and the corrosion protection observed in previous works [
3,
4,
5].
Diamond-like carbon [
6] is a metastable form of amorphous hydrogenated carbon (a-C: H) [
7] with sp
3 bonds, sp
2 carbon atoms, and hybridized hydrogen, as can be seen in the ternary diagram available elsewhere [
8]. DLC can be deposited using various processes with different gaseous or liquid precursors, such as acetylene, benzene, ethylene, methane, or hexane. Briefly, deposition consists of three steps: plasma reactions (dissociation or ionization), interaction of the plasma with the surface, and reactions of the substrate with the coating [
9]. The quality of DLC coatings is directly related to the deposition parameters, the type of source used, and the energy of the ions present in the chemical reactions that determine the percentages of hydrogen and carbon bound in chains and in graphite rings [
10].
The reactor configuration for the deposition process and parameters is set up such that the voltage, current, power supply, working pressure, and precursor type should provide a plasma environment with high ion density and a pulsed power supply to favor a low, free path between the ions and the substrate surface, avoiding intense collisions [
11]. The aim is to achieve an optimal collection of charged particles with ideal electrical conductivity, which collectively responds to the electromagnetic forces and chemical reactions during deposition. Inside the plasma reactor, the gas precursors undergoing ionization may take the form of bundles of neutral ions, while also acting as highly charged gas clouds that may contain particulate material in the form of dust. During the deposition process, these particles can be deposited onto the surface of the coating, causing the formation of porous dimples or pinholes [
11,
12] that act as sites for the permeation of corrosive liquids or vapors through the layers onto the substrate. The pores generated by dust from the deposition process are starting points for corrosion that reduce the lifetime of the coating. The use of DLC coatings with high hardness, high resistivity to etching, high electrical resistivity [
13,
14], a low friction coefficient [
15], and low pore content can be one way to protect the metal substrate and avoid corrosion.
3. Results and Discussion
3.1. DLC Chemical Structure: Raman and XPS Analyses
Figure 3 shows the Raman spectra of the DLC coatings deposited on the SS316L and Ti6Al4V substrates. The original D and G bands were fitted by the red and green lines, respectively, with the 1350 and 1580 cm
−1 positions corresponding to the original D and G band centers. The spectra for the DLC coatings were obtained with laser excitation at 514.5 nm. The spectra indicated the predominance of the G band, shown by the peak centered at 1580 cm
−1, together with a small shoulder centered at 1350 cm
−1, related to the D band. The green and red lines in the spectrum are the deconvolutions of the bands used to determine the I
D/I
G ratio.
The Raman analysis of the DLC coating microstructures indicates that the coatings were formed by micro-graphitic rings and open chains, as shown by the I
D/I
G ratios, calculated as described previously by Robertson [
25]. The I
D/I
G ratios were 0.62 and 0.58 for the coatings deposited on SS316L and Ti6Al4V, respectively, using the same deposition parameters for the two substrates. The physical properties of the DLC coatings are directly related to the I
D/I
G ratio. The D (disorder) band corresponds to the vibrational modes associated with sp, sp
2, and sp
3 hybridizations. A right shift of the D band is associated with the presence of clusters with aromatic carbon rings and carbon bonds in open chains in the coatings. The G band is due to a tangential elongation of the sp
2 hybridization mode [
26]. A right shift of the G band corresponds to the existence of sp
2 bonds of a mixture of carbon chains with aromatic rings. The intensity of the G band is directly proportional to the content of the sp
2 bonds [
27].
Figure 4 shows the XPS spectra for the DLC coatings. The spectra were dominated by the C1s bands, providing information on the chemical bonding environment and the components of the DLC. The SS316L spectrum presented three main bands, corresponding to C–C/C–H at 284.8 eV, C=O=C at 285.8 eV, and C–O at 287.4 eV [
28]. The Ti6Al4V spectrum showed main bands corresponding to C–C/C–H bonds at 284.6 eV, C–O–C at 285.9 eV, and O=C=O at 287.4 eV [
28,
29]. The binding energy values found for the sp
2 and sp
3 components of the C1s spectra were consistent with the binding energies reported in the literature for the peaks corresponding to graphite and diamond [
30]. The presence of oxygen in the chemical bonds found in the analysis was due to residual air in the vacuum process applied before deposition of the DLC coatings. The presence of oxygen in DLC films could lead to bonds resulting from chemical corrosion during tribocorrosion tests.
3.2. DLC Film Thickness and Roughness Measurements
Figure 5 shows the results of the optical profilometry measurements of the thicknesses of the DLC coatings on the SS316L and Ti6Al4V substrates, determined from the step heights in the line profiles shown in
Figure 5. The thickness values were similar, since the deposition parameters employed were the same, with deposition during 120 min at Vb of −700 V. The thicknesses of the DLC coatings deposited on SS316L and Ti6Al4V were 1.59 and 1.33 μm, respectively. The deposition rates were approximately two-fold higher than those obtained using DLC in a previous study by the same research group [
11]. The growth rates observed here were 0.013 and 0.011 μm/min for the SS316L and Ti6Al4V substrates, respectively. Bradley and co-workers reported the relation between the deposition and the sputtering rate due to ion beam bombardment during the deposition process, which is also common in PVD processes [
31].
The images in
Figure 6a show the Ti6Al4V+DLC profile. The colors indicate the peaks and valleys on the surface (the dimensions are shown in the color bar on the right side of the images).
Figure 6b shows the surface morphologies of the SS316L and Ti6Al4V substrates with and without DLC coatings. Roughness analyses were performed using areas measuring 301 × 229 μm. For SS316L, the roughness values were 52.49 nm with the DLC coating and 52.20 nm in the absence of the coating. For Ti6Al4V, the corresponding values were 21.4 nm and 12.76 nm, respectively; hence, despite the nanometric scale, the surface roughness was increased by the presence of the DLC coating in both cases.
3.3. Adherence and Hardness of the DLC Coatings
Figure 7 shows the results for adherence, which was evaluated using the scratch test. The test was performed with a diamond tip, in accordance with ASTM C1624-05 [
32]. An optical microscopy image of the scratch track on the SS316L substrate coated with DLC is shown in
Figure 7a. The first crack occurred after 8 s, at a critical load (Lc1) of −3 N, indicated by the red line, with the image showing progressive failures in the form of waves in that region. The first critical load (Lc1), which is indicated by the first sign of damage or cracks in the coating, without substrate exposure, was evidenced by the increase in the acoustic emission signal (green line), as well as by failures shown in the optical microscopy image. The second critical load (Lc2) is related to the complete failure of the coating [
7,
8]. The DLC-coated SS316L sample showed no substrate exposure (Lc2) up to a force of 20 N. In the case of the DLC-coated Ti6Al4V sample, there was also no substrate exposure (Lc2) up to 20 N, although the Lc1 occurred at −2 N.
Hardness testing of the DLC coatings was performed using eight indentations to obtain an average. The diamond tip penetrated to a depth of only 10% of the thickness of the DLC coatings in order to avoid any influence of the hardness of the substrate. The literature reports DLC hardness values in a range from 10 to 80 GPa [
25].
Table 4 shows the hardness values obtained from an average of eight measurements on the surfaces of the SS316L and Ti6Al4V samples with and without DLC coatings. It can be seen that the hardness values were considerably higher for the coatings, compared to the substrates. These hardness values might suggest that the DLC film could have scratched the counter body; however, this was unlikely, considering the low friction coefficients obtained for the DLC coatings (
Figure 8).
3.4. Friction Tests
Figure 8 shows the friction coefficient results obtained in an air environment, using the tribological pairs consisting of a sphere and the substrate plates with and without DLC coatings. The friction coefficient measured for SS316L without the DLC coating was 0.50 ± 0.16, while the value for the corresponding coating was 0.10 ± 0.01, representing a reduction of 81%. The friction coefficient obtained for the Ti6Al4V substrate was 0.42 ± 0.07, while a value of 0.08 ± 0.01 was obtained for the corresponding DLC coating, representing a reduction of 80%. The lower friction coefficients measured for the DLC coatings demonstrated the excellent lubrication properties of DLC, making this material highly suitable for reducing friction [
33]. Furthermore, in addition to conferring a low friction coefficient, the use of DLC coatings can increase the shelf life of metallic implants.
3.5. Tribocorrosion
Figure 9a–d shows the friction coefficient and open circuit potential (OCP) curves obtained for the tribocorrosion tests using Ringer’s lactate solution. The OCP was measured at a steady state under static conditions in order to obtain the equilibrium potential. Fluctuation of the OCP was then measured during the sliding tests. The average friction coefficient is shown by the red line, with the values being used for comparison of the friction in an air environment and in Ringer’s lactate solution.
Figure 9a shows the friction coefficients for the tribological pairs consisting of the SS316L sphere and the SS316L plates with and without DLC, immersed in Ringer’s lactate solution. The friction coefficient for the SS316L in the solution was 0.34 ± 0.11, while the value decreased to 0.04 ± 0.02 for the pair composed of the SS316 sphere and the SS316L+ DLC plate, representing a reduction of 88%.
Figure 9b shows that during the tribocorrosion in reciprocating mode, the bare SS316L sample (blue line) presented an abrupt decrease of the OCP from 0 to −0.35 V, indicating the removal of passive protection due to wear damage, with an increased tendency for corrosion on the wear track. This effect was also observed for the SS316L+DLC sample, although the OCP decrease was only −0.1 V, demonstrating that the DLC coating provided increased resistance to corrosion.
Figure 9c shows the friction coefficient plots for the pairs consisting of the Ti6Al4V sphere and the Ti6Al4V plates with and without DLC, in Ringer’s lactate solution. The mean friction coefficient obtained using the plate without DLC was 0.35 ± 0.05, while a value of 0.05 ± 0.02 was obtained using the DLC-coated Ti6Al4V plate, representing a reduction of 85%.
The friction test results for the Ti6Al4V and SS316L substrates with and without DLC coatings show that the protective DLC coating reduced the friction in both cases, with similar values obtained, considering the error bars.
For the DLC-coated Ti6Al4V sample, the initial OCP was around +0.4 V, with a slight change to around +0.3 V during the tribocorrosion test (
Figure 9d, black line). The positive potential value showed that corrosion protection was maintained after the sliding test. This indicates that the Ti6Al4V sample had better corrosion protection than the SS316L substrate, since the latter presented a rapid decrease in OCP from 0 to −0.35 V, indicating the removal of passive protection and susceptibility to corrosion.
In terms of tribocorrosion susceptibility, the results (
Figure 9b,d) show that the Ti6Al4V was two times more susceptible to corrosion than SS316L, while the presence of the DLC coating provided greater protection to Ti6Al4V, compared to SS316L, since the potential showed a more significant decrease for SS316L compared to Ti6Al4V. The different behaviors of the DLC films as protective coatings could be related to an effect of the surface roughness, which differed for the two substrates (
Figure 6).
Titanium alloy (Ti6Al4V) is more resistant to corrosion in Ringer’s lactate solution, compared to stainless steel (SS316L). This is due to the formation of a passive layer on the surface of the titanium alloy, which prevents further corrosion. On the other hand, SS316L can corrode in Ringer’s lactate solution because chloride ions react with the iron in stainless steel, leading to pitting corrosion. As shown in
Table 1, SS316L is composed of 59.46% iron, indicating a probable interaction with chloride ions during the reciprocating mode tribocorrosion tests.
In previous work with titanium nitride thin films grown on titanium during different deposition times, it was observed that an increased thickness provided better corrosion protection and wear resistance [
34]. Corrosion protection is related to an absence of pores in the coating. In thin films, discontinuities and the presence of high pore concentrations or dimples provide corrosion pathways that lead to the failure of the coating.
3.6. Crystal Violet Assays
Figure 10 shows the results of the crystal violet assays of cell viability after 24 and 48 h for the control, SS316L, SS316L+DLC, Ti6Al4V, and Ti6Al4V+DLC groups. At 24 h, the Ti6Al4V group showed a statistically significant difference from the control group (
p < 0.0001); however, there was no significant difference between the Ti6Al4V+DLC and control groups, indicating similar cellular growth. For the 48 h exposure, the Ti6Al4V group presented a statistically significant difference to the control group (
p < 0.0001), with decreased cell viability. The higher cell viability value at 48 h reflected cell division that occurred after the 24 h period. The Ti6Al4V+DLC group was not statistically different from the control, indicating that these groups remained constant regarding cell proliferation.
For the 24 h period, the cell viability of the SS316L group was significantly different in comparison to the control group (p < 0.0001); however, the 316L+DLC group showed no statistical difference from the control group, indicating that cell proliferation remained unchanged.
For the 48 h period, the cell viability of the SS316L group was significantly lower compared to the control group (p < 0.0001), with the lower value at 48 h indicating a toxic environment that led to cell death. The SS316L+DLC group showed no statistically significant difference from the control, consequently remaining the same regarding cell proliferation. For both metal substrates, the presence of the DLC film increased the cell viability, resulting in values equal to the control. The significantly lower (p < 0.0001) cell viability values for the uncoated SS316L and Ti6Al4V indicate that these materials were unsuitable for cell growth.
3.7. Mitochondrial Activity
Figure 11 presents a plot illustrating Mitochondrial Activity at 24 and 48 h, evaluated using the MTT assay for both Ti6Al4V and Ti6Al4V+DLC samples. The results demonstrate that there were no statistically significant differences in cell death compared to the control group during the 24 h evaluation period. Similar outcomes were observed during the 48 h period, suggesting that cell metabolism, which involves chemical reactions and energy production for cell function, was comparable across these groups.
On another hand, the mitochondrial activity data highlight that the SS316L and SS316L+DLC groups exhibited statistically significant differences from the control group (p < 0.0001) for both the 24 and 48 h periods. The results for the SS316L group were consistent with the crystal violet assay, indicating an increased level of cell death. Conversely, the results for the SS316L+DLC group indicated that, although the cells showed behavior similar to the control group in the crystal violet assay, with no notable difference in cell death, mitochondrial activity occurred at a lower level, implying cellular adaptation to the environment.
3.8. Micronucleus Assays
Many materials can cause damage to the complex structure of DNA, resulting in the formation of micronuclei, which are small chromosomal fragments dispersed outside the nucleus. Ethyl methanesulfonate (EMS) is a mutagenic substance that can cause the formation of micronuclei after contact for extended periods, enabling its use as a positive control [
35].
The results of the micronucleus assays are shown in
Figure 12. For the 24 h period, in comparison to the control group, the Ti6Al4V+DLC, SS316L, and EMS groups showed statistical differences (
p = 0.0190,
p < 0.0001, and
p < 0.0001, respectively), while no statistical differences were observed for the Ti6Al4V and SS316L+DLC groups. This heterogeneity of the data was due to a period of adaptation of the cells to the exposure environment. The EMS group (control for micronuclei formation) presented statistical differences when compared to the control, SS316L, and SS316L+DLC groups (
p < 0.0001,
p < 0.0001, and
p < 0.0001, respectively); however, a possible explanation for this observation is the absence of micronucleus formation during the 24 h exposure period, since the action is closely linked to the cycle time of the animal cell, which is typically around 24 h.
For the 48 h period, a statistically significant difference was observed between the Ti6Al4V group and the control group (p < 0.0001); however, the Ti6Al4V+DLC, SS316L, and SS316L+DLC groups showed no significant difference when compared to the control group. In comparison to the EMS group, all the other groups were significantly different, evidencing probable non-genotoxicity, considering the substantial micronucleus formation in the EMS group, while much lower micronucleus formation occurred in the other sample groups.
Figure 13 shows SEM images, at 3000× magnification, of the L929 cells on the Ti6Al4V, Ti6Al4V+DLC, SS316L, and SS316L+DLC substrates. The yellow arrows indicate filopodia, while the red arrowheads indicate lamellipodia. The lamellipodia and filopodia are protuberant extensions originating from the membrane of a cell. Lamellipodia are membrane extensions with several filopodia joined by actin. These structures are formed by the action of a protein that forms microfilaments, with the presence of actin leading to the filopodia becoming sealed in a sheet-like network structure. According to Mattila [
36], the difference between lamellipodia and filopodia is in the size of the extension. The filopodia are primary finger-like structures that extend up to 0.3 μm and have the function of cell mobility and spreading. The lamellipodia are structures containing sealed filopodia in the presence of actin. It can be seen from
Figure 13 that the L929 cells were more consolidated in the Ti6Al4V samples (with and without DLC). In the case of the bare SS316L sample, the cells presented low spreading and appeared to grow in a vertical direction. A comparison of the images for the DLC-coated substrates indicated that there was cell preference for the Ti6Al4V+DLC coating.
4. Conclusions
The deposition of DLC coatings on 316L steel and Ti6Al4V alloy, using the same parameters, and with acetylene gas as a carbon precursor, produced coatings composed of hydrogenated amorphous carbon containing 80% graphite rings and open chains in its structure.
The carbon bonds presented sp
2 hybridization, with graphite-type characteristics, while sp
3 hybridization was obtained from C–C/C–H and C–O–C bonds. Both sp
2 and sp
3 were detected by XPS in the regions of 284 and 285 eV [
28,
29,
37,
38]. The thickness of the DLC coating on SS316L was 1.59 μm, with a growth rate of 0.013 μm/min. For the Ti6Al4V substrate, the DLC coating thickness was 1.33 μm and the growth rate was 0.011 μm/min. Both growth rates were as expected for DLC films produced by PECVD processes.
The hardness of the DLC coatings was in the range of 23–26 GPa, which provided mechanical resistance up to −20 N in scratch tests, without substrate exposure.
Friction tests showed that the DLC with 80% of the carbon in its structure distributed between carbon rings and open chains increased the mechanical strength and provided lubrication. The friction coefficient was reduced by an order of magnitude, compared to the samples without coatings.
Tribocorrosion analyses performed using the samples with and without DLC coatings and immersed in Ringer’s solution demonstrated that the coatings provided corrosion protection in both static and dynamic modes, reducing the corrosion potential and the friction coefficient.
The use of in vitro biological analyses showed non-cytotoxicity of the DLC-coated samples during periods of 24 and 48 h; however, exposure to the uncoated Ti6Al4V or SS316L led to cell death. In addition, a micronucleus analysis showed minimal genotoxic potential, especially for the DLC coatings. It could be concluded that the DLC coatings presented suitable properties of interaction and biocompatibility when exposed to the cellular environment, together with low friction and high hardness, making them suitable for use on fixed implants. It is necessary to control the thickness and roughness of the coating in order to avoid pinholes and increase the corrosion protection of implants. The use of these DLC coatings with low friction coefficients could facilitate the fixation of implantable pins and screws, including Kirschner wires.
Finally, is important to note that this paper shows a novel approach because the functioning of joints is very complex; there is a lot to study to improve the joint range movement that decreases with increasing age. Flexibility needs to be maintained with collagen production, as it keeps the skeletal system flexible. Many aged people do not replace hormones and their reduction contributes to collagen reduction, as well as limited range of movement. In addition, physical exercise is mandatory for people of all ages to maintain their range movement and help with natural hormone repositioning.