The Effect of DLC Surface Coatings on Microabrasive Wear of Ti-22Nb-6Zr Obtained by Powder Metallurgy

Abstract

Titanium alloys have a high cost of production and exhibit low resistance to abrasive wear. The objective of this work was to carry out diamond-like carbon (DLC) coating, with dissimilar thicknesses, on Ti-22Nb-6Zr titanium alloys produced by powder metallurgy, and to evaluate its microabrasive wear resistance. The samples were compacted, cold pressed, and sintered, producing substrates for coating. The DLC coatings were carried out by PECVD (plasma-enhanced chemical vapor deposition). Free sphere microabrasive wear tests were performed using alumina (Al₂O₃) abrasive suspension. The DLC-coated samples were characterized by scanning electron microscopy (SEM), Vickers microhardness, coatings adhesion tests, confocal laser microscopy, atomic force microscopy (AFM), and Raman spectroscopy. The coatings did not show peeling-off or delamination in adhesion tests. The PECVD deposition was effective, producing sp2 and sp3 mixed carbon compounds characteristic of diamond-like carbon. The coatings provided good structural quality, homogeneity in surface roughness, excellent coating-to-substrate adhesion, and good tribological performance in microabrasive wear tests. The low wear coefficients obtained in this work demonstrate the excellent potential of DLC coatings to improve the tribological behavior of biocompatible titanium alloy parts (Ti-22Nb-6Zr) produced with a low modulus of elasticity (closer to the bone) and with near net shape, given by powder metallurgy processing.

1. Introduction

Over the past few years, materials science has investigated different types of biomaterials and their applications to replace or restore the function of compromised or degenerated tissues or organs [1]. Every year, over 13 million prostheses/medical devices are implanted in the US alone [1,2]. Thus, biomaterial helps to improve the quality of life and longevity of human beings. The field of biomaterials has shown rapid growth to keep up with the demands of an aging population [3]. An acceptable reason for the increase in revision surgeries is the longer life expectancy [3,4]. Implants are expected to work much longer or until the end of life without failure or revision surgery [3]. Thus, developing appropriate materials with a high longevity and excellent biocompatibility becomes essential. Titanium and its alloys have been widely used in numerous biomedical applications due to a unique combination of desirable properties [5,6]. Such mechanical properties include excellent corrosion resistance, low density, high toughness, and excellent biocompatibility [5,6,7]. Titanium alloys have proven to be superior in terms of biocompatibility when compared to stainless steel and cobalt alloys [8,9], thus being the most promising biomaterials for implants [3,10]. The titanium alloy Ti-6Al-4V is the most commonly used for application as an implant material [11]. This alloy was originally developed for other applications, such as the aerospace industry [12]. However, in biomedical applications, both Al and V released into the bloodstream are related to long-term health problems [9,12]. It has been reported that Al is an element involved in serious diseases such as Alzheimer’s disease and bone metabolism (osteomalacia) [4,9]. Some V-free Ti alloys for biomedical applications, such as Ti-6Al-7Nb and Ti-5Al-2.5Fe, have been developed [9,13]. The modulus of elasticity of Ti-6Al-4V alloy (~110 GPa) is much lower than that of stainless steel and Co-based alloys (~180 and 210 GPa, respectively) [13]. However, its modulus of elasticity is significantly greater than that of bone tissue (10–40 GPa), causing the formation of tension shielding that can potentially cause bone resorption and eventual implant failure [12]. Thus, the development of low modulus Ti alloys for biomedical applications has evolved in recent years [13,14]. New alloys with high biocompatibility, low elastic modulus, and tensile strength superior to pure titanium are promising candidates for implant application [15,16,17,18]. Titanium has high reactivity, especially with oxygen [19]. The cost of machining these alloys is relatively high. The powder metallurgy process can produce parts with the final shape very close to the desired one, known as “near-net-shape”, effectively reducing the cost of producing titanium alloys [20,21]. The powder metallurgy technique also enables the production of parts with controlled porosity and modulus of elasticity closer to the bone, and, consequently, reduces the tension between the implant and the bone [14,22,23]. Despite the good mechanical properties, Ti and its alloys exhibit low tribological performance, such as a high and unstable wear coefficient, severe adhesive wear, and low abrasive wear resistance, limiting their application [7,24]. Proper surface treatment expands the use of titanium and its alloys, being one of the most effective methods for improving wear resistance [25,26]. DLC (diamond-like carbon) coatings show good behavior, with excellent mechanical, tribological, and biocompatible properties [27]. They are excellent candidates for use as anti-wear coatings due to their extreme mechanical strength, low friction coefficient, high stability, and excellent biocompatibility [28,29]. They have, therefore, a potential medical application to suppress the generation of particles in implants arising from the implant/bone movement [29]. The accumulation of wear residues at the implant/bone interface can produce an adverse cellular response leading to inflammation, the release of harmful enzymes, osteolysis, infection, implant loosening, and pain [29]. High friction and consequent wear of artificial hip implants after 10–15 years of implantation are the major issues leading to revision surgery [30,31]. Therefore, surface modification techniques can be developed to improve the implant quality considering the lifespan of younger patients. DLC coating has an amorphous and chemically inert structure composed of two types of carbon hybridization (sp2 and sp3) that provide high hardness, low friction coefficient, biocompatibility, and, in addition, it is a solid lubricant [32,33,34]. As a solid lubricant, the so-called transfer layer of the graphitic fraction is deposited on the counterpart, preventing wear and providing negligible wear rates for the DLC coating under tribological conditions [33,34,35]. The DLC coating is bio-inert and has good cell adhesion, unlike most biomaterials [36,37,38,39]. In this work, diamond-like carbon (DLC) coating, with dissimilar thickness, was carried out on Ti-22Nb-6Zr titanium alloys produced by powder metallurgy. Niobium (Nb), titanium (Ti), and zirconium (Zr) have attracted much attention as implant materials due to their excellent mechanical properties and biocompatibility. [40]. Although Zr is considered a neutral element in relation to the alpha and beta phases, some studies demonstrate that Zr is a stabilizer of the beta phase in the Ti-Nb-Zr ternary system [41]. In the Yudin et al. study, [42] a powder of Ti-18Zr-15Nb biomedical alloy with spongy morphology and with more than 95% vol. of β-Ti was obtained by reducing the constituent oxides with calcium hydride. The influence of the synthesis temperature, the exposure time, and the density of the charge (TiO₂ + ZrO₂ + Nb₂O₅ + CaH₂) on the mechanism and kinetics of the calcium hydride synthesis of the Ti-18Zr-15Nb β-alloy was studied.

2. Materials and Methods

2.1. Production of the Sample of Ti-22Nb-6Zr by Powder Metallurgy

The powders of titanium and zirconium elements were produced from the hydrogenation process. Sponge fines previously washed with organic solvent (acetone) and air-dried were used. The hydrogenation step for all metals was carried out at 500 °C, in a Thermal Technology Astro Series 1000 high vacuum oven, Thermal Technology, 2221 Meridian Blvd Minden, NV 89423, USA, with a maximum temperature of 2500 °C, for approximately 3 h, with a pressure of 100 kPa. After cooling, the friable material was ground under a 133 × 10⁻³ Pa mechanical vacuum at room temperature. A stainless steel mill was used, coated with titanium plates and containing titanium balls, aiming to avoid contamination. All powders were used in their hydrogenated state, aiming to achieve greater activation of the sintering process through the atomic movement of hydrogen during the process and reduce costs since the dehydrogenation step is expensive and time-consuming. The average particle size for the powders was 20 μm. In preparing the substrate of the Ti-22Nb-6Zr alloy (SBTNZ), the specimens of the alloy under analysis were obtained by the blended-elemental (BE) technique from the mixture of hydrogenated elemental powders, followed by a sequence of uniaxial and cold isostatic pressings and vacuum sintering. These methodologies aimed to achieve the maximum possible densification and optimize the process parameters. A Mettler Toledo analytical balance model PB3002l, Mettler-Toledo Indústria e Comércio Ltda, Barueri-SP, Brazil, with a precision of 0.01 g was used to weigh the powders in the stoichiometry of the alloy. Then, grinding and mixing were carried out for 15 min in a mechanical shaker. Cylindrical samples measuring 8 mm in diameter by 4 mm in height were prepared for the Ti-22Nb-6Zr alloy using steel dies with a floating jacket. Compaction was performed using a uniaxial hydraulic press (Marconi, model 0981, Marconi- Equipamentos para Laboratórios LTDA, Piracicaba-SP, Brazil). The uniaxially cold compacted specimens were encapsulated under vacuum in flexible latex molds and introduced into the cylindrical pressure vessel of a cold isostatic press. A pressure of 450 MPa for 30 s was applied. A Paul Weber KIP 100 E isostatic press, Paul-Otto Weber GmbH, Fuhrbachstraße 4-6/73630 Remshalden, Germany, was used, with a capacity of 100 t, equipped with a cylindrical chamber with a diameter of 50 mm, a useful height of 160 mm, and a maximum pressure of 500 MPa. A Thermal Technology Inc vacuum furnace, model 1000-3060-FP 20, Thermal Technology, 2221 Meridian Blvd Minden, NV 89423, USA, with a graphite resistive element and maximum temperature of 2500 °C was used for sintering. The samples were sintered at 1400 °C under a 10⁻⁷ Torr vacuum with a 20 °C/min heating rate. Upon reaching the specified temperature, the samples remained at this level for two hours. The samples, after sintering, were progressively ground in 240, 400, 600, 800, 1200, and 2400 sandpaper. Polishing was performed in alumina solution, with a final granulometry of 0.05 µm.

2.2. Deposition of the DLC Coating

The polished substrates were first cleaned by ultrasonic cleaning in distilled water for 30 min and then in acetone for 20 min before being placed in the deposition chamber. Next, the DLC coatings were deposited on the titanium alloy substrates using the PECVD technique—plasma-enhanced chemical vapor deposition. The vacuum chamber for coating deposition was assembled in Brazil, with an internal volume of 130 L, with a pumping system composed of a mechanical pump of 90 m³/h and a 2000 L/s diffuser pump. First, the substrates were sputter cleaned using argon plasma for 30 min, a flow of 10 sccm, a pressure of 0.15 Pa, and a self-polarization of −0.6 kV. This treatment with argon plasma allowed for the elimination of the oxide layer on the metallic surfaces. Then, the DLC coatings were deposited using acetylene (C₂H₂) as a precursor gas, with a gas flow of 7.5 sccm, using an applied constant voltage of −0.75 kV and pressure of about 6.6 Pa. For greater understanding concerning carrying out the tests and analyzing the results found, the alloy samples under analysis were identified according to the treatment condition of the DLC coating, as shown in

Table 1. Identification of the sample nomenclature and its description.

Sample Identification	Description
SBTNZ	Sample of uncoated Ti-22Nb-6Zr alloy (substrate).
TNZ1	Sample of Ti-22Nb-6Zr alloy with DLC coating thickness of 0.487 ± 0.06 µm.
TNZ4	Sample of Ti-22Nb-6Zr alloy with DLC coating thickness of 4.23 ± 0.08 µm.

.

2.3. X-ray Diffractometry

In this work, X-ray diffractometry was carried out to identify the crystalline phases present in the SBTNZ substrate. A Shimadzu diffractometer, model XRD-6000, Shimadzu corporation, Nishinokyo Kuwabara-cho, Nakagyo-ku, Kyoto, 604-8511, Japan, was used, with a wavelength of 1.54 Å, generated by a Cu-Kα tube with a 2θ interval between 30° and 80°, with a step of 0.01° and a counting time of 1.5 s per step.

2.4. Vickers Microhardness

The samples of Ti-22Nb-6Zr alloys with and without DLC coating were submitted to the Vickers microhardness test to evaluate the top microhardness (perpendicular to the coating deposition surface). For this purpose, Microhardness Emco Test DuraScan equipment, EMCO-TEST Prüfmaschinen GmbH, Kuchl, Germany, was used.

2.5. Scanning Electronic Microscopy

One of the primary functions of scanning electron microscopy (SEM) was the measurement of the thickness of the DLC coating, consisting of a nanometer-scale for one deposition condition (TNZ1) and a micrometer-scale for another condition (TNZ4), which generated two coatings with different thicknesses. In addition, SEM was also used in the following analyses:

-

Microstructural evaluation of the samples, including energy-dispersive mapping;

-

Evaluation of the wear mode observed inside the wear caps, considering the micro-scale analysis.

A JEOL JSM 7100 FA Field Emission Scanning Electron Microscope, Tokyo, Japan, was used.

2.6. Raman Spectroscopy

Raman spectroscopy was used to characterize the DLC coatings, identifying the types of bonds. We sought to demonstrate the effective formation of the diamond-like carbon coating. The equipment used to perform the Raman spectroscopy measurements was a Jobin-Yvon triple spectrometer, model T64000, HORIBA Advanced Techno, Co., Ltd, Tokyo, Japan, in the subtractive configuration using an optical microscope (50× objective—spot around 5 μm). The Raman signal was detected by a CCD (charged couple device) cooled by liquid nitrogen. The sample was excited using a coherent CW (continuous wave) argon-ion laser tuned to the 532 nm line with a power of 10 mW on the sample. All spectra were obtained at room temperature.

2.7. Microabrasive Wear

The microabrasive wear tests of samples with and without deposition of DLC coatings were carried out with the CSM free sphere equipment, Anton Paar GmbH, Graz, Austria, using the Calowear model. Its configuration is a free sphere made of 100 Cr6 steel, 20 mm in diameter, which rotates continuously on the sample’s surface at a constant speed. The abrasive medium was a suspension composed of alumina (Al₂O₃) particles in distilled water at a concentration of 0.40 g of abrasive per cm³ of water. We sought to use abrasive particles with a hardness higher than the calcium particles present in the debris from bone degeneration in orthopedic implants. The objective was to test the wear resistance of the samples under test conditions theoretically more rigorous than those existing in prostheses implanted in the human body. The mean particle size was 1 µ. The abrasive suspension was continuously stirred throughout the test by a magnetic stirrer coupled to the microabrasion device to prevent the decanting of abrasive particles. The mixture was pumped to the sphere–sample interface using a peristaltic pump connected to the equipment. The abrasive flow rate was set at approximately one drop every 8 s. The rotation of the drive shaft was maintained at 280 rpm, generating a velocity between the surface of the sphere and the sample of approximately 0.195 m s⁻¹. Wear tests were performed, adjusting to the sliding distance traveled by the ball. The initial test times for the substrate samples (SBTNZ) and the samples (TNZ1) were set at 5 min intervals, with the first time being 15 min and the last time being 40 min. Thus, for the times used, the first sliding distance was 175.66 m and the last 468.44 m. Therefore, with increments of 58.55 m between intervals.

Table 2. Sliding distance and respective times of each test for samples SBTNZ and TNZ1.

Test Time (min)	Sliding Distance (m)
15	175.66
20	234.22
25	292.77
30	351.33
35	409.88
40	468.44

presents the sliding distance and the respective times for each test.

For samples with the thicker coating (TNZ4), test times were set at 15 min intervals, with the first time being 95 min and the last time being 170 min. The increments were 175.66 m between intervals.

Table 3. Sliding distance and the respective times of each test for the TNZ4 samples.

Test Time (min)	Sliding Distance (m)
95	1112.54
110	1288.20
125	1463.86
140	1639.53
155	1815.19
170	1990.86

presents the sliding distance and the respective times of each test.

2.8. Wear Volume

The wear volume after each interval of the ball sliding over the sample’s surface was determined using expression (1):

$$V\cong {\displaystyle \frac{\pi ·b^4}{32·\mathsf{\varphi }}}b\ll \mathsf{\varphi }$$

(1)

where:

-

b is the diameter of the wear crater;

-

ϕ is the diameter of the test sphere.

This equation is used to calculate the spherical crater since its volume is minimal concerning the volume of the sphere.

2.9. Wear Coefficient (K)

The Archard equation was used to calculate the wear coefficient K of the samples in the uncoated condition (SBTNZ) and for the coating + substrate systems (TNZ1 and TNZ4) considering the set (coating + substrate). Archard’s equation for calculating the sample wear coefficient is shown in expression (2).

$$K={\displaystyle \frac{\pi ·b^4}{32·\mathsf{\varphi }·\mathrm{S}·\mathrm{N}}}$$

(2)

where:

-

b is the diameter of the wear crater;

-

S is the slip distance;

-

ϕ is the diameter of the test sphere;

-

N is the normal force to the sample.

2.10. Confocal Laser Microscopy

The reconstruction of the crater generated in the microabrasive tests was carried out using a confocal laser microscope Olympus model LEXT OLS 4100, Olympus, Tokio, Japan.

2.11. VDI Indentation Test

The study of peeling-off regions (loss of adhesion) of the coatings was carried out through a load applied with a Rockwell Durometer. In the Rockwell indentation images, the damage caused to the coatings by the indentation was compared to reference standards described in the VDI 3198 standard, which has a scale from HF1 to HF6. According to that standard, an acceptable coating deposition is considered the qualitative verification of classes HF1 to HF4. However, above class HF4, the coating does not have ideal adhesion, and consequently, there are peeling-off regions in the deposited layer. Therefore, using the standard, it is possible to classify the adhesion of the coatings qualitatively.

3. Results and Discussions

3.1. X-ray Diffractometry Results

Figure 1 illustrates the X-ray diffractometry analysis of the ternary alloy (Ti-22Nb-6Zr) before coating (SBTNZ substrate), showing the α-Ti phase, the β-Ti phase, and niobium.

No intermetallics from the TiNbZr system were detected at the resolution level of the diffractometer used.

Table 4. Lattice parameters and phase percentages using Rietveld for the Ti22Nb6Zr substrate.

Phases	% wt Calculated	Lattice Parameters (Å)
		a	B	c
Tiβ	56	3.3099	3.3099	3.3099
Tiα	43	2.9789	2.9789	4.7662
Nb	1	3.3369	3.3369	3.3369

presents the results of the quantitative evaluation of the Tiα, Tiβ, and Ni phases present and the lattice parameters of the alloy developed in this work, using the Rietveld method. The substrate Ti22Nb6Zr presented 56 wt% Tiβ, 43 wt% Tiα, and 1% Nb.

3.2. Microstructural Analysis

Figure 2 shows the SEM micrograph of the Ti-22Nb-6Zr alloy produced by powder metallurgy, still without coating. The grain boundaries are well defined, and the matrix is composed of β-phase and α-phase in the form of lamellae, with low porosity levels and high densification.

Figure 3a–c shows the mapping of elements present in the Ti22Nb6Zr alloy obtained by EDS, where it is possible to observe the lower affinity of the α phase for the Nb and Zr elements compared to the β phase.

3.3. DLC Coatings

The DLC coating procedure was carried out under the different processing conditions defined in the methodology, giving rise to DLC coatings with different thicknesses. Figure 4 presents the image for the coating produced on the Ti-22Nb-6Zr alloy in a treatment condition that generated a thinner coating (sample TNZ1).

Figure 5 depicts the cross-section of a sample of Ti-22Nb-6Zr alloy coated with DLC, showing the substrate and the coating layer. The lighter area at the top of the image shows the thicker coating surface (TNZ4 sample) seen in perspective. The wavy surface area under the coat represents the substrate base material.

The detailed image of the TNZ4 sample shows the difference in texture between the substrate and the coating with greater magnification (Figure 6). Being a more ductile material, the substrate has a rougher aspect, while the coating section has a thinner and more regular appearance.

The thickness measurements of the coatings were carried out in the SEM software and Image J (version 1.5h).

3.4. Raman Spectroscopy

Analyses by Raman spectroscopy were carried out on specimens coated with DLC coatings with a laser wavelength of 532 nm. The D and G bands are obtained by Gaussian deconvolution, which makes it possible to determine the location of the bands, the full width of the G band at half maximum (FWHM (G)), and the determination of the ID and IG intensities. The spectrum could be deconvoluted using two Gaussian components for the present work. Raman spectra were determined by Gaussian fitting in the 850–1800 cm⁻¹ and are shown in Figure 7 and Figure 8.

The Raman spectrum showed two characteristic bands (D and G) indicative of the DLC phase formation [43]. In the DLC Raman spectra, the D band is attributed to the deformation of disordered aromatic rings in the graphitic phase. Its appearance indicates that it is an amorphous coating [44]. The G band is associated with C = C bonds (crystalline graphite) located at sp2-hybridized carbon sites. It is common to define D and G bands in the Raman spectrum of carbon-based materials [44]. The position values of the D and G bands, the full width at half maximum (FWHM), and the ratio of ID/IG integral intensities are shown in

Table 5. Position of D and G bands and band intensity ratio.

Sample	D Band Position [cm−1]	G Band Position [cm−1]	FWHM (G) [cm−1]	Ratio ID/IG
TNZ1	1371.10	1544.24	160.51	1.01
TNZ4	1383.59	1550.35	149.46	1.37

.

As described in Table 5 and through the deconvolution of the Raman spectrum, it is possible to identify the presence of the D band centered at 1371 cm⁻¹ for the TNZ1 condition and 1380 cm⁻¹ for the TNZ4. They are characteristics of graphite-like materials due to the symmetrical E_2g vibrational mode in graphite materials [6,7]. Thus, the analysis of the Raman spectrum classifies the layers obtained in this work as diamond-like carbon (DLC) by both bands typical for DLC-type films. Therefore, the results suggest that the structure produced was composed of mixed carbon sp2 and sp3, characteristic of DLC materials.

3.5. Vickers Microhardness

The microhardness values for the substrates and the DLC coatings are shown in

Table 6. Vickers microhardness of the substrate (SBTNZ) and the TNZ1 and TNZ4 coatings.

Vickers Microhardness (HV) (50 gF)
SBTNZ	TNZ1	TNZ4
324 ± 0.22	1170 ± 0.35	1750 ± 0.27

. The Ti-22Nb-6Zr alloy substrate showed a microhardness of approximately 324 HV. As expected, it can be seen that the uncoated alloy (SBTNZ) has microhardness values below all coated samples analyzed, regardless of the thickness of the coating used.

3.6. Microabrasive Wear of the Substrate (SBTNZ) and Coated Samples (TNZ1 and TNZ4)

Two-body abrasive wear is caused by rubbing of a softer surface by a hard rough surface, while three-body abrasive wear is caused by hard particles entrapped between two sliding surfaces [45]. Both for the analysis of the microabrasive wear of the Ti-22Nb-6Zr alloy substrate and the samples coated with DLC, the measurements of the wear craters were carried out using confocal laser microscopy and scanning electron microscopy.

3.6.1. Diameter of the Craters

For each wear crater produced, the determination of its external diameter was carried out through the average of five diameters. Four diameters were measured in different directions (Figure 9a). The fifth diameter was obtained by inserting a circle by the confocal microscope analysis program, as exemplified in Figure 9b.

3.6.2. Abrasive Wear Mechanisms

Detailed images of the wear craters were obtained by scanning electron microscopy to identify the predominant abrasive wear mechanisms in each sample. Micro-rolling abrasive wear was identified between the grooving abrasive wear risks in the SBTNZ samples and for the thinnest coating (TNZ1). As for the thicker coating (TNZ4), the predominant wear mechanism was rolling, as shown in Figure 10, Figure 11 and Figure 12.

For grooving abrasive wear, the particles must be embedded in the specimen or in the sphere (body). In the samples that showed grooving abrasive wear and micro-rolling between the grooving abrasive wear risks (SBTNZ and TNZ1), the characteristic grooves of micro-grooving prevailed, where it is possible to identify the displacement of material adhered to the edges of the groves. However, for samples with a thicker film, TNZ4, the mode of wear of the substrate material under the test conditions was by rolling.

3.6.3. Volumes and Wear Coefficients

The wear volume as a function of the sliding distance for the Ti-22Nb-6Zr alloy substrate sample without any coating (SBTNZ), with the thinner coating (TNZ1), and with increased DLC film thickness (TNZ4) are presented in Figure 13.

Through the analysis of Figure 13, it was possible to notice a successive increase in the amount of material worn for greater sliding distances. When compared, the substrate samples show a more significant volume loss at all sliding distance measurement points, i.e., the worn volume of the substrate is always greater than that of coated samples (film + substrate system), regardless of the coating thickness, indicating the effectiveness of the protective treatment. One way to verify if the steady-state of wear has been reached is through the analysis of the graphs of the wear volume (V) as a function of the sliding distance (S) (V = f(S)). If the wear volume shows a linear variation with the sliding distance, it is considered that the steady-state wear regime has been reached. A constant (permanent) wear regime is obtained for the condition with constant normal force when the wear volume is linearly dependent on the sliding distance. In all graphs, the linearity of the V x S relationship is observed for the results of the data obtained experimentally, indicating the condition of the steady-state wear regime for the substrate with and without coatings. Based on Figure 13, which characterizes trends close to a straight line and the consideration of obtaining the steady-state wear regime, the wear rate was determined, as shown in

Table 7. Wear rate for the Ti-22Nb-6Zr alloy substrate and for substrate and systems with DLC coatings (TNZ1 and TNZ4).

Sample SBTNZ
Equation V = f(S) − V(m3) and S(m)	Wear Rate $\left(\mathit{Q}\right)-$ Converted to [mm3/m]
$V=3.61528\times {10}^{-14}S$	3.62 × 10−5 or 0.0000362
Sample TNZ1
$V=6.89591\times {10}^{-15}S$	6.90 × 10−6 or 0.0000069
Sample TNZ4
$V=1.06325\times {10}^{-15}S$	1.06 × 10−6 or 0.00000106

. As previously described, the wear rate of wear was obtained by deriving the equation of the curve referring to each of these figures $\left(\mathrm{Q}={\displaystyle \frac{\mathrm{d}\mathrm{V}}{\mathrm{d}\mathrm{s}}}\right)$.

The wear rate obtained (Q) is the volume of material removed per unit of sliding distance between the ball and the specimen. Considering the wear rate, the coated samples showed better wear resistance. The most significant reduction in the wear rate relative to the substrate was obtained with the highest coating thickness (TNZ4), as shown in

Table 8. Reduction in wear rate (%) in relation to the substrate.

Sample	Reduction in Wear Rate (%)—in Relation to the Substrate
TNZ1	80.93
TNZ4	97.06

.

From the graphs constructed to verify that the analyses were carried out under steady-state wear, it was possible to calculate the substrate’s wear coefficient and substrate + DLC coating system through Archard’s equation (K = πb⁴/(32·Φ·S·N)).

Table 9. Values of the wear coefficient of the substrate (SBTNZ) and samples with DLC coating of different thicknesses (TNZ1, TNZ4).

Sample	Wear Coefficient $\left(\mathbf{K}\right)-$ [m3/N.m]	Standard Deviation
SBTNZ	7.20 × 10−13	1.08 × 10−13
TNZ1	1.30 × 10−13	1.48 × 10−14
TNZ4	2.08 × 10−14	5.54 × 10−15

presents the results of the wear coefficients of the samples coated with DLC and without coating for the test conditions used, while Figure 14 shows the graphic representation of these values.

It was observed that the DLC-coated samples showed superior performance when compared to the uncoated samples. As shown in Table 9 and Figure 14, the sample with the thicker coating (TNZ4) showed better wear resistance when compared to the respective substrate, with a wear coefficient 34.68 times lower. This progress was also observed in the TNZ1 sample, with a wear coefficient (k) 5.53 times lower than its substrate.

3.6.4. Adhesion of Coatings

In their different coating conditions, the samples were subjected to evaluation of adhesion of DLC films through Rockwell C indentation (HRC). Figure 15a shows a typical indentation generated on the surface of the TNZ1 sample. Very thin radial cracks can be detected. Small chips can be observed only at specific points on the edge of the indentation for the coating on the titanium alloy, as shown in the detailed image in Figure 15b. No circumferential cracks or significant delaminations around the edge were observed. Interfacial bonds are strong. Even in the region where the substrate accumulates due to deformation caused by indentation, minimal chipping is observed in the coating. Under these treatment conditions (TN1), the DLC film showed satisfactory adhesion, with an adhesion quality index of the coating–substrate pair related to the HF2 condition provided for in the VDI 3198 standard.

Typical indentations of Rockwell C hardness from the adhesion test of the DLC coating on the TNZ4 sample are shown in Figure 16. Very thin radial cracks can also be observed on the edge of these indentations. For both the thinner (TNZ1) and thicker (TNZ4) coatings, these cracks were favored because titanium alloy substrates are soft and produce great deformation under load. However, the resistance to DLC peeling-off proved to be quite efficient. Also, for TNZ4, the interfacial bonds are strong, with minimal DLC coating peeling off even in the area where the substrate accumulates (pile-up), i.e., where there is an amount of material displaced to the edge of the indentation due to surface plastic deformation during an adhesion test. Circumferential cracks were not observed around the edge. The thicker coating showed the adhesion quality index of the coating–substrate pair, which can be associated with the HF2 pattern, representing a good coating adhesion to the substrate.

4. Conclusions

• X-ray diffraction examinations of the substrate (SBNTZ) showed α-Ti and β-Ti phases. No intermetallics from the TiNbZr system were detected at the resolution level of the diffractometer used.
• Scanning electron microscopy analyses of these substrates showed homogeneous microstructures with low porosity and high densification. In both cases, lamellae and well-defined grain boundaries of alpha phase and matrix composed of β phase prevailed. Spectral mappings via EDS showed the concentration of the beta-stabilizing element (Nb) in the beta matrix
• Through the deconvolution of the Raman spectrum, it was possible to identify the presence of the D band centered at 1371 cm⁻¹ for the TNZ1 condition and 1380 cm⁻¹ for the TNZ4. They are characteristics of graphite-like materials due to the symmetrical E_2g vibrational mode in graphite materials. Thus, the analysis of the Raman spectrum classifies the layers obtained in this work as diamond-like carbon (DLC) by both bands typical for DLC-type films. Therefore, the results suggest that the structure produced was composed of mixed carbon sp2 and sp3, characteristic of DLC materials.
• Both the thinnest (TNZ1) and the thickest (TNZ4) coatings did not present with DLC delamination or peeling-off in the Rockwell C indentation adhesion tests provided for in the VDI 3198 standard. The PECVD process used in the deposition of the DLC films produced strong coating–substrate interfacial bonds. The quality indices of the film-substrate pairs reached the HF-2 condition of the standard, as mentioned earlier, indicative of excellent adherence.
• The PECVD-deposited DLC coatings using acetylene (C₂H₂) as a precursor gas, with a gas flow of 7.5 sccm and constant voltage of −0.75 kV, applied on Ti-22Nb-6Zr titanium alloy produced by powder metallurgy, showed excellent tribological performance in microabrasive wear tests compared to the results obtained for the wear resistance of uncoated substrates.
• The three-body wear prevailed for the thicker coating (TNZ4), giving rise to rolling wear with abrasive particles rolling during the test, which is less aggressive. Two bodies wear was predominant for the thinner coatings (TNZ1).
• According to the results obtained in this work, the DLC coatings in both thicknesses provided good structural quality, homogeneity, adhesion, high hardness, and resistance to microabrasive wear. Therefore, the deposited DLC coatings are very promising to improve the tribological behavior of Ti-22Nb-6Zr titanium biomedical alloys produced by powder metallurgy
• Future research aims to evaluate the influence of the microabrasive medium on wear resistance.