Influence of Si Addition on the Chemical and Tribological Performance of TiAlCrN Coating Deposited by Co-Sputtering

Abstract

In this work, nanostructured TiAlCrN coatings were deposited on a WC-Co substrate using a co-sputtering process varying the silicon composition on the coatings. The influence of silicon content on the mechanical, chemical, and tribological performance of the coatings was studied. The hardness increases from 11 to 16 GPa with the Si content; also, Young’s modulus increases from 260 to 295 GPa. The H/E ratio, which is a measure of materials’ ability to take the strain before deformation, is also increased with the increase in Si content, suggesting increased toughness. XPS analysis reveals that the coatings present titanium, aluminum, chromium, and silicon nitrides. The tribological behavior of the coatings was conducted through ball-on-disc tests, in which the results show that the coefficients of friction range from 0.15 to 0.55, with the lowest for the samples with the highest Si content. This behavior is benefited by the formation of oxynitride species, identified by XPS, which acts as lubricating layers and diffusion barriers. TiAlCrSixN coating presents a potential application for severe wear owing to its tribological performance.

1. Introduction

The development of multifunctional coatings on substrates by physical vapor deposition (PVD) processes is increasingly recognized as an important component in modifying the critical properties of materials. The aspect of these films is to control the physical and chemical properties of the couple coating/substrate and, subsequently, apply the new material to the industry in specific applications [1,2]. The PVD co-sputtering technique has achieved important improvements in carbide and nitride nanocomposite coatings [2,3]. In industrial and scientific fields, there is a particular interest in the increase in the coating’s performance associated with enhanced cutting parameters, leading to higher cutting speed and cutting volume that contributes significantly to an increase in productivity [4,5].

Recently, there has been increasing interest in designing machining tools for hard-to-cut materials commonly used in aeronautical, automotive, or biomedical applications, where materials detach due to wear with a cutting tool [6,7]. Those machining processes constitute approximately 15% of the economic cost of the product in industrialized countries [8]. Nanostructured coatings are a major area of interest within the field of coatings, extending the lifetime of classical cutting tools, which can be used to machine those materials. However, the rapid changes in developing new materials are affecting innovation in machining operations where high loads are present [9].

Metal nitride coatings have been applied to machining operations due to their enhanced mechanical, thermal, chemical, oxidation resistance, and tribological properties. Thin films, such as TiN, CrN, and TiAlN, have been used as conventional nanocomposite coatings [10,11]. Moreover, the AlCrN has been considered as a substitute for (Ti, Al)N, which has been used in recent years as a standard coating for machining. Thus, some weaknesses of TiAlN coating can be solved with the ternary AlCrN system because it shows lower friction coefficients and wear rates [4].

In addition, quaternary systems, including the incorporation of several metals such as Cr, Y, Si, etc., have improved cutting performance, mostly due to hardening, grain refinement, and formation of stable oxides [12,13]. Additionally, the AlCrN presented better characteristics of oxidation resistance thanks to the formation of complex Al-Cr oxides in the tool surface during the tribochemical reactions [14]. The effect of silicon on the contents is up to 7% [15,16] and is known to increase the hardness as a result of the microstructural refinement produced by the blocking of the sliding bands by the grain boundaries [17,18].

TiAlCrN-based coatings with silicon addition have also been investigated [19]. These coatings have been developed for dry high-speed machining of hard-to-cut materials, such as hardened tool steels, nickel-based superalloys, titanium alloys, etc., where high mechanical loads are present; therefore, the tool surface can be coated to improve wear and fatigue resistance [20,21]. Such coatings exhibit high oxidation stability at elevated temperatures, high resistance against abrasive and adhesive wear, and increased crack performance [20]. The most used PVD process to deposit these coatings is the cathodic arc technique; however, it does not offer the versatility to deposit materials such as ceramics or aluminum alloys. Moreover, the sputtering technique provides better control of coating composition.

This work considers the development of TiAlCrNSi-based coatings by the sputtering process deposited on the cemented carbide K20 substrate of tungsten carbide, which are widely used as inserts in processes of machining hard materials. The inclusion of silicon in the range from 0.61 to 1.59 at% was investigated on the mechanical, chemical composition, and tribological performance.

2. Materials and Methods

For this study, K20 hard metal pieces of 16 × 16 mm and 3 mm height were used to explore the influence of the application of TiAlCrN-based coatings on its surface in terms of the mechanical and tribological characteristics. The substrate surface was previously prepared until it reached a mirror-like aspect and then cleaned with isopropanol and acetone to remove contaminants on the surface.

The synthesis of the coatings was carried out using a co-sputtering system equipped with RF and DC power supplies. A 99.999% pure TiAl (TiAl = 36:64 wt.%) and Cr targets with 4 inches of diameter were used. The deposition parameters are depicted in

Table 1. Deposition parameters of TiAlCrN-based coatings with DC-RF co-sputtering system.

Parameter	Value(s)
Power applied to Cr target (W)	110
Power applied to TiAl target (W)	270
Deposit time (min)	70
Vacuum pressure (mbar)	5.7 × 10−6
Work pressure (mbar)	4.0 × 10−3
N2 flux (sccm)	3.8 sccm
Ar flux (sccm)	14 sccm
Substrate temperature (°C)	300 ± 5
Substrate rotation speed (rpm)	10

.

To evaluate the influence of the silicon inserts on the coating’s performance, small pieces of pure silicon were located on the Cr target in the four configurations of zero, two, four, and six pieces, as can be seen in Figure 1. This arrangement sets a co-sputtering system in which the targets, the Ti, Al, Si, and Cr atoms, are supplied, and from the gas mixture the N atoms are expected to react and form compounds.

To determine whether the surface morphology changed with the silicon content, the samples were analyzed through a scanning electron microscope equipped with an energy-dispersive X-ray spectroscopy module for elemental analysis of the sample. Quantification of surface elements and chemical species was carried out by XPS. Analyses were performed on the XPS/ISS/UPS-A.Centeno surface characterization platform (SPECS).

The measurements were carried out with a pressure of around 1 × 10⁻⁹ mbar. For each sample, the spectra were recorded using monochromatic Al Kα radiation (hv = 1486.6 eV) operated with 100 W and 12 kV. The following sequence of spectra was recorded: survey spectra, C 1s, O 1s, N1s, Si 2p, Cr2p, Ti2p, and C 1s again to verify the stability of the load compensation in the function of time. Survey spectra were recorded at a pass energy of 100 eV, while the high-resolution spectra were recorded at a pass energy of 20 eV.

The acquired spectra were analyzed with the CasaXPS program (Casa Software Ltd., Devon, UK, Version 2.3.25) using the SPECS Prodigy-ACenteno library provided with R.S.F. values (Response Sensitivity Factor) determined by the manufacturer. A Shirley baseline was used. The binding energy (BE) scale of the spectra was corrected by reference to the C-(C, H) component of the C 1s peak at 284.8 eV.

The ball-on-disc method was carried out to evaluate the tribological response of the coatings. A Universal tribometer Bruker nanosurface was employed with a stationary ball in the air under ambient laboratory conditions test using 200 g of load with a rotation speed of 100 mm/s during 70 s with a 6 mm in diameter ball of Al₂O₃.

The ball-on-disc tests were performed at 78% relative humidity and a temperature of 20 °C. The mass loss was calculated according to the Equation (1).

$${\mathrm{d}\mathrm{i}\mathrm{s}\mathrm{k} \mathrm{v}\mathrm{o}\mathrm{l}\mathrm{u}\mathrm{m}\mathrm{e} \mathrm{l}\mathrm{o}\mathrm{s}\mathrm{s},\mathrm{m}\mathrm{m}}^3=\frac{\mathsf{\pi }\ast \left(\mathrm{w}\mathrm{e}\mathrm{a}\mathrm{r} \mathrm{t}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{k} \mathrm{r}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{u}\mathrm{s},\mathrm{m}\mathrm{m}\right)\ast {\left(\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{k} \mathrm{w}\mathrm{i}\mathrm{d}\mathrm{t}\mathrm{h},\mathrm{m}\mathrm{m}\right)}^3}{6\ast \left(\mathrm{s}\mathrm{p}\mathrm{h}\mathrm{e}\mathrm{r}\mathrm{e} \mathrm{r}\mathrm{a}\mathrm{d}\mathrm{i}\mathrm{u}\mathrm{s},\mathrm{m}\mathrm{m}\right)}$$

(1)

This is an approximate geometric relation that is correct at 1% for (wear track width/sphere radius) < 0.3 and is correct at 5% for (wear track width/sphere radius) < 0.8.

Data of nanoindentation, which describe the nanohardness and elastic modulus of the coatings, were collected using a Hysitron (TI 750) equipped with a Berkovich-type diamond tip using an indentation force of 2.0 μN. The penetration depth of indentation was controlled within 10% of the coating thickness to avoid the substrate effect with a holding time of 10 s. For each sample (i.e., each combination of coating and substrate), a set of 10 tests was carried out with a minimum distance of 30 μm between indentations. Those tests that show anomalies in the load vs. displacement curves were nonvalid data.

3. Results

3.1. Structure and Morphology

According to a previous paper [5,23], the X-ray diffraction patterns (XRD) of coatings deposited with different dopant concentrations have a mixed orientation with planes belonging to the FCC-NaCl structure located at diffraction peaks observed at 2θ values of 37.5°, 43.9°, and 78.04° correspond to the (111), (200), and (222) planes of TiAlCrN with different silicon content, respectively. Some researchers [24,25] have shown that the addition of Si in the nitride structure can be given by (i) the substitute Si atoms in the crystal structure, and (ii) the formation of silicon nitride when there is an excess of silicon, which is an amorphous structure.

Figure 2 shows the micrograph of the morphology of the TiAlCrN with four pieces of silicon as an example of the surface morphology obtained in all the coatings. These micrographs reveal that the morphologies are homogenous and compact, showing small pores or droplets (circled in red). These processes have been attributed to the growth mechanism of the coatings, which has been widely discussed in the literature [26,27,28,29]. It has been found that silicon interrupts columnar growth when it segregates into the grain and column boundaries. Large agglomerates can then be formed, thereby obstructing the continuity of the column and promoting the re-nucleation of the nitrides formed. Furthermore, in various studies, silicon has been added to increase the density of the coating so that the microstructure can change from Zone 1 to Zone T or Zone 2 [30,31]; however, transmission electron microscopy measurements or cross-section SEM micrographs are necessary to verify this structural change.

3.2. Chemical Analysis

Table 2. Elemental chemical compositions for the coating specimens.

Sample	%Atomic by XPS
	C	O	N	Ti	Al	Cr	Si
TiCrAlSiN-0P	31.38	16.31	18.93	4.10	25.37	3.91	ND+
TiCrAlSiN-2P	26.92	13.17	20.58	5.16	28.53	5.04	0.60
TiCrAlSiN-4P	36.50	14.57	18.22	3.85	22.92	3.11	0.83
TiCrAlSiN-6P	19.42	15.36	24.19	5.03	29.38	5.0	1.59

ND⁺: Element not detected.

shows the elemental chemical composition of the coatings obtained on the surface of the coatings, around 10 nm of penetration.

In these XPS analyses, the main elements of the coatings, along with carbon and oxygen were studied. The variation of different pieces of silicon was evident in the change in silicon content as can be shown in Table 2. In the configuration with two silicon pieces, the at% was 0.60, for four pieces it was 0.83; finally 1.59 was obtained for six pieces.

The Si2p XPS spectra are depicted in Figure 3, in which one peak at the binding energy in the range of 101.66–102.06 eV was identified for the coatings produced with 2, four, and six silicon pieces. This result suggests the existence of Si₃N₄ [32]; however, a new not well-known species was identified at similar energies. Hence, this peak fits with Si-N and Si-N-O species, similar, to that found in [33,34]. That species plays an important role in the wear behavior, mainly because the Si inclusion can be helpful in the stabilization of TiN at high temperatures [32].

It is worth noting that the definition of the peak increases with silicon content increment, as well as the full width at half maximum (FWHM) gets narrower being the most intense and defined peak, the one at 1.6 at% (102 eV). The increase in silicon intensity is directly linked to the number of atoms incorporated into the films with the respective oxidation state. It is directly proportional to atoms in a particular chemical state.

The XPS high-resolution spectra of Ti, Al, and Cr 2p are depicted in Figure 4.

The high-resolution XPS spectrum shows the presence of nitride-based phases for titanium, aluminum, and chromium. In the case of titanium, two asymmetric peaks belonging to TiN were identified in the range between 455 and 462 eV of binding energy and correspond to Ti 2p_3/2 state [26]. In this range of energies, the low energy peak at ∼456.5 eV can be associated with Ti-N bonds and the shoulder peak at ∼458 eV can be related to the Ti-N-O bonds. In the same way and at higher binding energies the Ti2p1/2 peaks can be seen at ∼462 eV for the Ti-N bonds and at ∼463.5 eV for the Ti-N-O bonds.

Regarding the Al signals, a single peak at ∼73.8 eV is assigned to the Al 2p spectrum, and it represents the formation of Al-N bonds [35,36,37,38] compounds. The shoulder peak at a binding energy of ∼74.7 eV for the sample without silicon addition corresponds to the formation of Al-O bonds in the γ-Al₂O₃ compound due to the ionic characteristic bonds, and ∼75 eV for the sample with six pieces of silicon, suggesting the presence of the Al-O or Al-N-O bonds and, subsequently, the small shift of this peak could be attributed to the formation of nonstoichiometric AlOx [26,39].

Finally, the nitride of chromium presents two peaks, one of them approximately at 575 eV and the other at 585 eV. In the literature, there is much discussion about these peaks and Milosev [40,41] establishes that the binding energy of a completely clean surface of CrN should be ∼574.5 eV; however, Zhou et al. depict that this signal should be centered at 575.8 eV [28,42] now that it can attribute the presence of our peak to the Cr-N formation with a slightly metallic Cr characteristic, indicating that the CrN external layer has low thickness. The second peak found in the deconvolution process and centered at ∼577 eV corresponds to the presence of oxy-nitrides of chromium (Cr-N-O) formed by the substitution of oxygen atoms in nitrogen atoms sites [40]. The presence of those metal oxy-nitrides for the four specimens is shown in Figure 5.

From these data, it can be seen that oxynitrides were identified in all the samples with different atomic percentages, probably due to variable residual oxygen in the chamber during the four-deposition process. Hence, Al-N-O species keep constant for all four coatings; titanium increases in all coatings except in the coating with four pieces of silicon; similar behavior is observed with chromium. In the case of Si, the increment of Si pieces located on the target surface during the sputtering process leads to the formation of Si-N and Si-N-O species, as already discussed in XPS analysis, which acts as solid lubricants potentiated [28]. This lubrication performance is also enhanced by the presence of Ti, Al, and Cr oxynitrides.

3.3. Mechanical Properties

The mechanical properties of the coatings were evaluated by measuring their hardness and the elastic modulus (Figure 6a), as well as the (H/E) and the (H³/E²), which were calculated based on the hardness and elastic modulus values. There is an increase in hardness and Young’s Modulus as the silicon content rises. Furthermore, the ratios of H/E and H³/E² reflect the elastic strain to failure and plastic deformation resistance of a material, respectively, Figure 6b.

The hardness increases by about 29% when two pieces of silicon are added to the TiAlCrN; moreover, Young’s modulus increases from 260 GPa to 280 GPa. Despite the hardness with four pieces of silicon increasing to 14.1 GPa, the young modulus decreases slightly to 275 GPa; however, the margin of error for this measurement falls within the values for two and six pieces of silicon. This behavior can be related to defects generated in the deposition process because of the high energy of the atoms impinging on the substrate surface. These impacts produce dislocations and vacancies, which result in changes in both hardness and Young’s modulus, deteriorating the crystalline integrity weakening the atomic bonding and thus decreasing Young’s modulus [43]. The best results of mechanical properties were obtained with six pieces of silicon; that is, 1.59 at%. of silicon. Hardness increased up to 15.5 GPA and Young’s Modulus reached 295 GPa.

Both H/E and H³/E² ratios increased with the addition of silicon in the films deposited. The H/E and H³/E² ratios show a relevant effect on the wear behavior of the coatings; thus, an increase in these values is correlated with superior wear resistance, associated with higher plastic deformation resistance and fracture toughness [44]. The coatings deposited with silicon pieces from zero to six exhibit Si at% lower than 3%, presenting the Hall–Petch effect of grain fine hardening and, which is a consequence of grain boundaries acting as barriers to dislocation motion [35].

Consequently, toughness is related to the energy required to create a crack and the energy spent on resisting the crack propagation until fracture [37]. Hence, coatings with superior toughness exhibit high resistance to the formation of cracks typical in applications under extreme mechanical loads, leading to severe wear. This high capacity of energy absorption prevents crack propagation, making these coatings a good alternative for high-wear applications.

3.4. Wear Behavior

Figure 7 shows the relationship between wear rate and coefficient of friction for the TiAlCrSiN-based coatings.

In the initial stage of the tests, the COF increases to 0.32 for all the coatings and drops until approximately 0.13, which corresponds to the running-in period, then slowly increases until it reaches the steady-state up to the end of the test. This running-in behavior is attributed to the interaction of the initial surface roughness, which produces a momentary rise in friction and lasts until the surface area gets bigger and more stable. The average friction coefficient of all coatings was calculated between 20 and 30 s due to the stability of the films and avoiding the substrate-coating effect, which could affect the real values associated exclusively with the coatings.

The best performance was 0.18 for TiAlCrN with four pieces of silicon and very close to it and almost without variation is located the coating TiAlCrSiN with six pieces of silicon with a COF of 0.20, which shows relatively smooth and stable curves, in this sense, the silicon presence tends to reduce COF of the coatings. The worst behavior was obtained in the coatings without and with two pieces of silicon with friction coefficients of 0.52 and 0.30, respectively; showing that silicon addition reduces the friction coefficient of TiAlCrN coatings [12,45,46]. Generally speaking, the constant increase in COF values in all the samples at the steady-state regime could be attributed to the delamination of the coatings in some parts, which promotes the contact with substrate areas and debris thus becoming a more irregular surface of contact, which increases COF values.

The unstable behavior of the COF belonging to the coating with two and without pieces of silicon has been reported and is related to large deformation and wear phenomena causing the delamination of the coating during the friction tests; thus, the fluctuation of the COF corresponds to the substrate and remains debris of the coatings [47,48]. Additionally, low hardness in these samples can also promote low adhesion and a higher delamination process, which increases the COF values. However, it is well known that the friction force of the coatings can be correlated with the morphology, hardness, and chemical composition. In general, hard coatings can have excellent tribological properties, due to the high hardness and lubricating oxides formed even at room temperature during the wear test. In this case, it shows that during the sliding test, the strong interfacial conditions could have generated local flash temperatures of the magnitude necessary for the formation of oxides. The coatings’ worn surfaces can produce a tribo-layer with the presence of Cr₂O₃, Al₂O₃, and TiO₂. In this case, the growth of the TiO₂ was expected to be faster than that of the Cr₂O₃ and Al₂O₃. The addition of Si can improve the tribological properties. For example, Si can promote notable amounts of silicon oxides, which are excellent compounds for significantly reducing the coefficient of friction. In the present investigation, the lowest COF was obtained in the TiAlCrSiN coatings, which have four and six pieces of Si, which is attributed to their high hardness, H/E, and H³/E². Another possible factor that could have controlled the COF was the nature of the tribo-layer formed at the interface. The Si content in the coating and the sliding temperature produced greater amounts of SiO₂ and Si(OH)₂, which were more lubricous than the other oxides and provided better wear resistance. Moreover, the low shear strength of TiO₂ has also been proposed as being responsible for the reduction in the COF and the wear rate.

Additionally, the wear rate versus the silicon pieces shows a decrease with the increase in silicon content. With two pieces of silicon, which corresponds to 0.60 at% of silicon, the wear rate decreases around 60% compared with the reference coating, that is, with the effect of silicon. In the case of coatings with four and six pieces of silicon belonging to 0.83 at% and 1.59 at% of silicon, respectively, the drop is almost 100% in both cases. This behavior is associated with the solid self-lubricating capabilities favored for oxide formation and mechanical properties and low coefficient of friction.

In consequence, the wear performance of the coatings is related to the synergic effect of mechanical properties and oxide formation. Thus, TiSiN coatings have been extensively studied [49]. Jiang et al. [32] found a hardness of 33–35 GPa at a 9% Si concentration; meanwhile, Mei et al. [50] found a maximum hardness value of 34 GPa at a 7% Si concentration. Diserens et al. [49], like the previous authors, evaluated the hardness, finding a maximum value at a Si concentration of 5%. These values vary with a small silicon modification, which shows the remarkable effect of silicon addition on mechanical properties, as is evident in the present investigation.

Hsu and collaborators found a growth in the wear resistance as a function of hardness for tool steel substrates using CrAlSiN coatings [51], evidencing that silicon positively affects the anti-wear performance of the coating due to oxides formation in the wear track [52,53]. Furthermore, the addition of silicon to TiAlN has been shown to improve hardness and oxidation resistance, which provide excellent wear behavior. Therefore, TiAlSiN is frequently employed as a coating for mainly dry machining applications [12,45,54]. Consequently, the wear resistance of the coatings under analysis is influenced by the hardness and the ratios of H/E and H³/E²; in this sense, the decrease in the wear rate is mainly owing to the increases in the above-mentioned mechanical properties shown in Figure 6.

Figure 8a,b are representative behavior of the tribo wear track following the ball-on-disc test for TiAlCrSiN with two pieces of silicon and six pieces of silicon, respectively.

In Figure 8a the wear track of TiAlCrSiN with less amount of silicon is shown. A detail of that zone is depicted in Figure 8c. In this case, the delamination is the main wear mechanism. It is typical for this type of mechanism to have transverse micro-cracks, partial film delamination, and substrate material signals over the tracks. Based on the ball-on-disc test, it is expected that the counterpart slides on the coating surface, and a tribo-oxide layer is formed between the ball and the surface. The tribo-oxide layer results from the reaction of environmental oxygen with the elements that have the coating; nevertheless, the continuous counterpart’s sliding on the film’s surface causes compressive stress in front of the ball and stress behind it, which favors a mechanism of contact fatigue. According to the literature [55], fracture toughness is the stress that a material can support without undergoing failure. Therefore, the high silicon films had the highest H/E values, which surely improved their fracture toughness to absorb the cyclic stress during counter-sliding. Additionally, the cyclic movement promoted the propagation of cracks and film delamination as a consequence of an increase in wear debris during the sliding test. The wear debris increased the actual contact area between the ball and the disc, resulting in severe third-body abrasive wear of films deposited with low silicon content [56].

On the other hand, Figure 8b shows the worn width track for the coating with the highest content of silicon; the worn track shows a significant reduction pattern, also a slight formation of abrasion grooves without coating delamination is observed on the surface. Figure 8d shows the concentration behavior of elements through the wear track where the at% of aluminum, titanium, and chromium remain stable from EDS points 1 to four. Nevertheless, the silicon concentration decreases; in point 1, in the as-deposited coating, silicon concentration is 2.26 at%. In point two, the outer zone of the wear track, the silicon concentration drops to 1.77 at%; in point three, just in the worn zone of the wear track, the silicon does not change. In the inner zone of the wear track, silicon concentration was reduced to 1.55 at%. This behavior could be associated with the silicon oxynitride formation leading to COF and wear rate reduction, which, as has already been mentioned, is due to its self-lubricant properties associated with oxides and oxynitride formation.

It is relevant to point out that EDS analysis penetration on the coating surface is higher than XPS; in fact, XPS is a surface-dedicated technique that analyzes the core level to obtain information on the chemical and electronic state of a coating’s elemental components to a penetration depth of 10 nm. This is the reason why values change when comparing both techniques.

4. Conclusions

TiAlCrN coatings were deposited using a reactive co-sputtering process. The Si concentrations were controlled by adding pure silicon pieces on the TiAl target high-erosion race track.

According to XPS analysis, the presence of silicon in TiAlCrN coatings ranges from 0.60 to 1.59 at%. The high-resolution XPS spectrum shows the presence of nitride-based phases for titanium, aluminum, and chromium. Also, oxides and oxynitrides from the elements that compose the coatings were detected.

The inclusion of silicon produces an increase in hardness and Young’s modulus as a result of the microstructural refinement produced by the blocking of the sliding bands by the grain boundaries; also, the values of H/E and H³/E² exhibited the same variation as H, which results in higher plastic deformation resistance and fracture toughness benefiting the tribological response, that is, the wear rate and coefficient of friction.

The wear rate decreased from 1.31 × 10⁻⁶ without silicon pieces to 1.28 × 10⁻⁹ for six pieces, showing a positive promising effect of silicon addition on those properties; similar results were found in the coefficient of friction evaluation, where values of 0.18 and 0.20 were obtained for four and six pieces of silicon, respectively.

This behavior is attributed to the formation of different oxides and oxynitrides on the coating surface; according to XPS, Cr, Ti, and Si-oxynitrides were identified on the surface, as well as aluminum oxide. These oxides act as lubricating layers to prevent high-temperature degradation of the coatings and act as a diffusion barrier that reduces the chip, making it an interesting option of coatings for machining abrasive materials.