Role of Nitrogen and Yttrium Contents in Manufacturing (Cr, Y)Nx Film Nanostructures

The high-power impulse magnetron sputtering (HiPIMS) technique was applied to deposit multilayer-like (Cr, Y)Nx coatings on AISI 304L stainless steel, using pendular substrate oscillation and a Cr-Y target and varying the nitrogen flow rate from 10 to 50 sccm. The microstructure, mechanical and tribological properties were investigated by scanning and transmission electron microscopy, X-ray photoelectron spectroscopy, X-ray diffraction, instrumented nano-hardness, and wear tests. The columnar grain structure became highly segmented and nanosized due to pendular substrate oscillation and the addition of yttrium. The deposition rate increased continuously with the growing nitrogen flow rate. The increase in nitrogen flow from 10 to 50 sccm increased the hardness of the coatings (Cr, Y)Nx, with a maximum hardness value of 32.7 GPa for the coating (Cr, Y)Nx with a nitrogen flow of 50 sccm, which greatly surpasses the hardness of CrN films with multilayer-like (Cr, Y)Nx coatings architecture. The best mechanical and tribological performance was achieved for a nitrogen flow rate of 50 sccm. This was enabled by more elevated compressive stresses and impact energies of the impinging ions during film growth, owing to an increase of HiPIMS peak voltage with a rising N2/Ar ratio.


Introduction
Nowadays, advances in sustainable energy and technologies are critically dependent on designing materials with improved properties. In another way, thin-film development is essential for discovering and improving new technologies [1]. Studies on the growth mechanisms of thin films deposited using sputtering techniques have led to innovation and a better comprehension of the fundamental aspects and the physicochemical and technological properties of coatings, in addition to improvements in the process technologies [1][2][3].
The physical vapor deposition technique is widely explored in various industrial areas and used in several coating depositions. Magnetron sputtering is a plasma-based coating technology in which inert gas atoms (usually Ar) are ionized and accelerated due to the potential difference between the negatively polarized target (cathode) and the anode.

Coating Deposition Process
The (Cr, Y)N x coatings manufactured in this work were deposited on AISI 304L stainless-steel substrate disks (30 mm diameter × 10 mm thick). Previously, the 304L stainless-steel substrates were sanded with SiC sandpaper; polished with diamond suspensions and colloidal silica; cleaned in ultrasonic baths, with acetone, before the deposition; and blow-dried.
A CrY alloy target (98-2 at.%, 220 mm × 110 mm × 12 mm) was used for film deposition with a distance between the substrate and target of 80 mm in the 0 • position. When oscillating to 15 • , the distance between target and substrate is 119 mm, as shown in Figure 1, and connected to the power supply of HIPIMS. The deposition experiments were performed using a sputtering spray system consisting of a HiPIMS TruPlasma 4004 (TRUMPF Hüttinger, Ditzingen, Germany) feed source installed in a reaction chamber-HiPIMS-250 (Plasma LIITS, Sao Paolo, Brazil) of octagonal shape with inner chamber dimensions of 600 mm base diameter and 660 mm height. A Pinnacle MDX power supply (Advanced Energy, Fort Collins, CO, USA) was added to supply −180 V substrate bias voltage. The sample holder performs oscillatory movements in front of the target, with a time interval of 30 s and an angular period of ±15 • . The substrates were allowed to oscillate with an amplitude of ±15 • , with position 0 • corresponding to the sample surface parallel to the target surface. The entire cycle from 0 to −15 • to +15 • and back to 0 • took 120 s, with a nominal angular speed of approximately 0.6 rpm. The (Cr, Y)N x coatings depositions were carried out in an Ar atmosphere, at a constant flow rate of 50 sccm and varying nitrogen flow (10,20,30,40, and 50 sccm), except for ion etching and base layer, where it was under an atmosphere of Ar only with 50 sccm. The working pressure was set to 0.266 Pa at 400 • C, and a substrate bias of −180 V was applied. The HiPIMS pulse frequency was 500 Hz, and the ton was 200 µs. Initially, an ion etching with Cr + ions was carried out for 1 h, using HiPIMS with an average peak power density of 3.0 W/cm 2 , 100 Hz, and 50 µs of ton and substrate polarization of −800 V, thus allowing for cleaning and superficial ionic implementation. Subsequently, the deposition of the (Cr, Y) base layer was carried out for twelve minutes, using the same substrate bias of −180 V applied later during the deposition of the (Cr, Y)N x films. The parameters for deposition of (Cr, Y)N x were kept constant: the peak power HiPIMS density was 90 W/cm 2 , the peak current density over the target area was 203 mA/cm 2 , and deposition was 2 h. Such experimental conditions were employed in agreement with our previous work [20].
A CrY alloy target (98-2 at.%, 220 mm × 110 mm × 12 mm) was used for film deposition with a distance between the substrate and target of 80 mm in the 0° position. When oscillating to 15°, the distance between target and substrate is 119 mm, as shown in Figure 1, and connected to the power supply of HIPIMS. The deposition experiments were performed using a sputtering spray system consisting of a HiPIMS TruPlasma 4004 (TRUMPF Hüttinger, Ditzingen, Germany) feed source installed in a reaction chamber-HiPIMS-250 (Plasma LIITS, Sao Paolo, Brazil) of octagonal shape with inner chamber dimensions of 600 mm base diameter and 660 mm height. A Pinnacle MDX power supply (Advanced Energy, Fort Collins, CO, USA) was added to supply −180 V substrate bias voltage. The sample holder performs oscillatory movements in front of the target, with a time interval of 30 s and an angular period of ±15°. The substrates were allowed to oscillate with an amplitude of ±15°, with position 0° corresponding to the sample surface parallel to the target surface. The entire cycle from 0 to −15° to +15° and back to 0° took 120 s, with a nominal angular speed of approximately 0.6 rpm. The (Cr, Y)Nx coatings depositions were carried out in an Ar atmosphere, at a constant flow rate of 50 sccm and varying nitrogen flow (10,20,30,40, and 50 sccm), except for ion etching and base layer, where it was under an atmosphere of Ar only with 50 sccm. The working pressure was set to 0.266 Pa at 400 °C, and a substrate bias of −180 V was applied. The HiPIMS pulse frequency was 500 Hz, and the ton was 200 µs. Initially, an ion etching with Cr + ions was carried out for 1 h, using HiPIMS with an average peak power density of 3.0 W/cm 2 , 100 Hz, and 50 µs of ton and substrate polarization of −800 V, thus allowing for cleaning and superficial ionic implementation. Subsequently, the deposition of the (Cr, Y) base layer was carried out for twelve minutes, using the same substrate bias of −180 V applied later during the deposition of the (Cr, Y)Nx films. The parameters for deposition of (Cr, Y)Nx were kept constant: the peak power HiPIMS density was 90 W/cm 2 , the peak current density over the target area was 203 mA/cm 2 , and deposition was 2 h. Such experimental conditions were employed in agreement with our previous work [20]. Schematic representation of the sputtering system using CrY target (98-2 at.%) and 304L substrates for deposition of (Cr, Y)Nx coatings varying the nitrogen flow. The minimum distance (dismin) between target and substrate surface is 80 mm, carousel height (hc) is 400 mm, and carousel diameter (dc) is 300 mm. Figure 1. Schematic representation of the sputtering system using CrY target (98-2 at.%) and 304L substrates for deposition of (Cr, Y)N x coatings varying the nitrogen flow. The minimum distance (dis min ) between target and substrate surface is 80 mm, carousel height (h c ) is 400 mm, and carousel diameter (d c ) is 300 mm.

Coating Characterization
The morphology of the multilayer-like (Cr, Y)N x coatings was investigated using a Field Emission Scanning Electron Microscope FEG-Inspect, F-50 (FEI, Eindhoven, The Netherlands). Scanning Ion Microscopy (SIM) was also performed at 30 kV and 18 pA (Thermo Fisher Scientific, Waltham, MA, USA), using an FEI-Helios 650 dual-beam equipment. Electron microscopy characterization techniques (Transmission Electron Microscopy-TEM, Scanning Transmission Electron Microscopy-STEM, Selected Area Electron Diffraction-SAED, Energy-Dispersive X-ray Spectroscopy-EDS) were carried out on a JEM-2100F (JEOL Ltd., Tokyo, Japan) and a Double Spherical Aberration Corrected Titan Cubed Themis (Thermo Fisher Scientific) microscope, at 200 and 300 kV, respectively, at the LNNano. Cross-section lamellae were prepared by Focused Ion Beam (FIB) on a Dual Beam Scios 2 (Thermo Fisher Scientifics) microscope at the LNNano. Roughness measurements and surface mapping were performed in tapping mode using NanosurfFlex Atomic Force Microscope (AFM) (Nanosurf, Liestal, Switzerland).
X-ray photoelectron spectroscopy (XPS) was performed to determine chemical depth profiles using a Versaprobe PHI 5000 (Chanhassen, MN, USA), using 0.2 eV energy resolution and 15 eV Al kα radiation and an area with a diameter of approximately 200 µm.
X-ray diffraction (XRD) was applied to identify phases and residual stress in (Cr, Y)N x films. All XRD patterns were conducted on a Panalytical MRD-XL X-ray diffractometer (Almelo, The Netherlands), using Mo Kα radiation (0.7093 Å).
Nanohardness measurements were performed with a PB1000 mechanical tester (Nanovea, Irvine, CA, USA). A load of 50 mN was used. The nanohardness measurements were performed with a Berkovich dyad tip for each (Cr, Y)N x coating. Using the Oliver-Pharr [27] relationship, the average hardness and elastic modulus, along with standard deviations, were calculated.
Wear tests were performed by linear reciprocal sliding in accordance with ASTM G113, using the PB1000 mechanical tester. For the wear test, Al 2 O 3 spheres were used as a counter body; the tests were performed without lubrication and at room temperature. The normal load was 5 N, and the sliding speed was 60 mm/s along a 1 mm-long track. Five hundred cycles were performed, and, at the end of the measurement, the slip distance was 1000 mm. Subsequently, the wear measurements in the (Cr, Y)N x films were measured by using a noncontact optical profilometer, in which six cross-section profiles were obtained according to ASTM G133-05, and we obtained the average area of material loss from the area of the calculated cross-section, which, when multiplying by the sliding distance, provided the wear volume; the surfaces of the Al 2 O 3 counter body were also measured, and for the analysis, the theoretical sphere was subtracted to observe the material adhered to the sphere's surface.

Results and Discussion
FEG-SEM micrographs were obtained of the top surface and cross-section of the (Cr, Y)N x coatings fabricated with different N 2 /Ar flow rates. From the top surface micrographs ( Figure 2), one notices that, by increasing the nitrogen flow rate, a slight reduction in the size of the columnar structure is observed. The (Cr, Y)N x films in the fracture crosssection showed a dense columnar microstructure (Figure 2), which becomes segmented due to the oscillatory motion of substrates. The grains with the columnar structure are 100 to 500 nm for a nitrogen flow rate of 50 sccm. As the nitrogen flux increased, an increase in thickness was observed, having the m thickness at 50 sccm, as demonstrated by scanning ion microscopy (SIM) (Figure 3). Nitrogen flow was responsible for grain growth; as the nitrogen flow rate increased, the thickness increased and reached its maximum value with a flow of 50 sccm (see Figures 2 and 4). Such behavior is because the negative HiPIMS peak voltage increases (from −718 to −757 V) with the N 2 flow rate, and, consequently, the ion bombardment is enhanced. The deposition rate and stress increase show that the coating deposition (Cr, Y)Nx is in metallic mode [28]. The base layer was the average oscillation of the nanostructure, which was around 230 nm, which is expected when using ±15 • and an oscillation period of 120 s. deposition rate and stress increase show that the coating deposition (Cr, Y)Nx is in metallic mode [28]. The base layer was the average oscillation of the nanostructure, which was around 230 nm, which is expected when using ±15° and an oscillation period of 120 s.   deposition rate and stress increase show that the coating deposition (Cr, Y)Nx is metallic mode [28]. The base layer was the average oscillation of the nanostructure, wh was around 230 nm, which is expected when using ±15° and an oscillation period of s.    The XPS technique was used to study the chemical composition of the multilayer-like (Cr, Y)Nx films. The XPS spectra of Cr2p and N1s binding energies are displayed in Figure  5a-c for the (Cr, Y)Nx films fabricated under different N2 flow rates (10, 30, and 50 sccm). The Cr2p spectra contain two peaks, 2p1/2 and 2p3/2 ( Figure 5a). There was no significant difference in the binding energy of the 2p1/2 peak (theoretical peak at 584 eV), verified at 583.9 eV for the N2 flow rates of 10 and 50 sccm and 584.2 eV for 30 sccm. Similarly, the 2p3/2 peak (theoretical peak at 575 eV) was observed at approximately 574.6 eV for all the nitrogen flow rates studied [6,29,30]. In contrast, the Cr2p peak intensities, which correspond to the Cr content in the films, are at the maximum for 10 sccm, decrease to a minimum at 30 sccm, and increase again for 50 sccm. At an N2 flow rate of 10 sccm, the Cr2p binding energies are higher due to the formation of both ( Figure 5a) CrN (theoretical peaks at 575 eV) and Cr2N (theoretical peaks at 576.4 eV). These higher values are due to the greater distance between the Cr, Y, and N atoms, decreasing the bond strength between the elements. The nitrogen atoms can be placed in positions previously occupied by the Cr atoms. Figure 5b shows the spectra of N1s, and there is a single broad peak with significant variations in peak intensity. The N2 flow rate of 30 sccm leads to maximum nitrogen content, while 10 sccm promotes a minimum and 50 sccm, an intermediate nitrogen concentration. The N1s peaks lie at about 397.4 eV for all nitrogen flow rates, i.e., 10, 30, and 50 sccm. This binding energy is associated with the transition metal nitride CrN, exhibiting the theoretical peak at 396.7 eV. The lower binding energies for 10 and 50 sccm can be attributed to the increased distance between the Cr, Y, and N atoms and the decreasing atomic bonding. The N atoms can be placed in positions previously occupied by the Cr atoms [6,29]. For the Y3d spectra (Figure 5c), two peaks, 3d3/2 and 3d5/2, slightly shifted. The 3d3/2 and 3d5/2 binding energies lie at approximately 158.8 and 156.6 eV, respectively, compared to the theoretical values of 160 and 158 eV. This small energy shift is due to the (Cr, Y)Nx coatings bond [30,31].
From the spectra of atomic concentrations of Cr, Y, and N as a function of nitrogen flux (see Figure 5d), we analyzed that nitrogen concentrations increase in (Cr, Y)Nx films from 10 to 50 sccm. However, the nitrogen flow rate of 30 sccm leads to a maximum atomic concentration of nitrogen. This can be attributed to increasing the nitrogen flow. For the (Cr, Y)Nx coatings deposited by HiPIMS, the flux of the ionized species increases as the negative peak voltage of the HiPIMS increases (from −718 to −757 V) [30]. The XPS technique was used to study the chemical composition of the multilayerlike (Cr, Y)N x films. The XPS spectra of Cr2p and N1s binding energies are displayed in Figure 5a-c for the (Cr, Y)N x films fabricated under different N 2 flow rates (10, 30, and 50 sccm). The Cr2p spectra contain two peaks, 2p 1/2 and 2p 3/2 (Figure 5a). There was no significant difference in the binding energy of the 2p 1/2 peak (theoretical peak at 584 eV), verified at 583.9 eV for the N 2 flow rates of 10 and 50 sccm and 584.2 eV for 30 sccm. Similarly, the 2p 3/2 peak (theoretical peak at 575 eV) was observed at approximately 574.6 eV for all the nitrogen flow rates studied [6,29,30]. In contrast, the Cr2p peak intensities, which correspond to the Cr content in the films, are at the maximum for 10 sccm, decrease to a minimum at 30 sccm, and increase again for 50 sccm. At an N 2 flow rate of 10 sccm, the Cr2p binding energies are higher due to the formation of both (Figure 5a) CrN (theoretical peaks at 575 eV) and Cr 2 N (theoretical peaks at 576.4 eV). These higher values are due to the greater distance between the Cr, Y, and N atoms, decreasing the bond strength between the elements. The nitrogen atoms can be placed in positions previously occupied by the Cr atoms. Figure 5b shows the spectra of N1s, and there is a single broad peak with significant variations in peak intensity. The N 2 flow rate of 30 sccm leads to maximum nitrogen content, while 10 sccm promotes a minimum and 50 sccm, an intermediate nitrogen concentration. The N1s peaks lie at about 397.4 eV for all nitrogen flow rates, i.e., 10, 30, and 50 sccm. This binding energy is associated with the transition metal nitride CrN, exhibiting the theoretical peak at 396.7 eV. The lower binding energies for 10 and 50 sccm can be attributed to the increased distance between the Cr, Y, and N atoms and the decreasing atomic bonding. The N atoms can be placed in positions previously occupied by the Cr atoms [6,29]. For the Y3d spectra (Figure 5c), two peaks, 3d 3/2 and 3d 5/2 , slightly shifted. The 3d 3/2 and 3d 5/2 binding energies lie at approximately 158.8 and 156.6 eV, respectively, compared to the theoretical values of 160 and 158 eV. This small energy shift is due to the (Cr, Y)N x coatings bond [30,31].
From the spectra of atomic concentrations of Cr, Y, and N as a function of nitrogen flux (see Figure 5d), we analyzed that nitrogen concentrations increase in (Cr, Y)N x films from 10 to 50 sccm. However, the nitrogen flow rate of 30 sccm leads to a maximum atomic concentration of nitrogen. This can be attributed to increasing the nitrogen flow. For the (Cr, Y)N x coatings deposited by HiPIMS, the flux of the ionized species increases as the negative peak voltage of the HiPIMS increases (from −718 to −757 V) [30]. Figure 6 displays X-ray diffractograms of (Cr, Y)Nx films deposited by HiPIMS under different N2 flow rates (10,20,30,40, and 50 sccm). For the film grown with an N2 flow rate of 10 sccm, one can mainly note the presence of the hexagonal phase Cr2N with a (100) preferential orientation in the normal surface direction (ND). This agrees with the XPS analyses that verified higher binding energies for the Cr2p photoelectrons as a result of Cr2N occurrence for the 10 sccm nitrogen flow rate. With a nitrogen flow rate of 20 sccm, the (300) peak of Cr2N remains broad but becomes small, thus indicating a low volume fraction of the nanostructured hexagonal phase. According to previous works [27,29], it was analyzed that the nitride peaks tend to widen as the flow of N2 gas is reduced. A facecentered cubic phase (fcc-CrN) also occurs with five diffraction lines, i.e., (111), (200), (220), (311), and (440). Fcc-CrN is encountered for the different nitrogen flow rates and is the major phase for all flow rates above 10 sccm. The preferred growth orientation is (100) for all N2 flow rates. This is desired for (Cr, Y)Nx coatings under situations of high atomic mobility and a low percentage of yttrium (<2 at.%) [27,29]. The increase in nitrogen flow rate of the N2 ratio according to Reference [29] was also analyzed for favoring preferential growth orientation in (100) caused by an increase in ion density. In addition, the diffractograms revealed peaks from the austenitic stainless-steel substrate (Figure 6), and Table 1 displays the XRD data of the (Cr, Y)Nx films, such as domain size, planes, and orientation. Table 1 (10,20,30,40, and 50 sccm). For the film grown with an N 2 flow rate of 10 sccm, one can mainly note the presence of the hexagonal phase Cr 2 N with a (100) preferential orientation in the normal surface direction (ND). This agrees with the XPS analyses that verified higher binding energies for the Cr2p photoelectrons as a result of Cr 2 N occurrence for the 10 sccm nitrogen flow rate. With a nitrogen flow rate of 20 sccm, the (300) peak of Cr 2 N remains broad but becomes small, thus indicating a low volume fraction of the nanostructured hexagonal phase. According to previous works [27,29], it was analyzed that the nitride peaks tend to widen as the flow of N2 gas is reduced. A face-centered cubic phase (fcc-CrN) also occurs with five diffraction lines, i.e., (111), (200), (220), (311), and (440). Fcc-CrN is encountered for the different nitrogen flow rates and is the major phase for all flow rates above 10 sccm. The preferred growth orientation is (100) for all N 2 flow rates. This is desired for (Cr, Y)N x coatings under situations of high atomic mobility and a low percentage of yttrium (<2 at.%) [27,29]. The increase in nitrogen flow rate of the N 2 ratio according to Reference [29] was also analyzed for favoring preferential growth orientation in (100) caused by an increase in ion density. In addition, the diffractograms revealed peaks from the austenitic stainless-steel substrate (Figure 6), and Table 1 displays the XRD data of the (Cr, Y)N x films, such as domain size, planes, and orientation. Table 1     The High-Angle Annular Dark-Field Scanning Transmission Electron Microscopy (HAADF-STEM) image of the cross-section shows the morphology of the Cr base layer Nanomaterials 2022, 12, 2410 9 of 17 and the (Cr, Y)N x films with a highly segmented columnar structure (Figure 7a). Figure 7b shows (Cr, Y)Nx multilayer nanostructure with N 2 flux of 50 sccm. So that Figure 7b shows that columnar grains exhibit a zig-zag grain growth morphology (black arrows) that is consistent across multiple layers. Bright-Field Transmission Electron Microscopy (BF-TEM) image of the nanostructured film grown with a nitrogen flow rate of 50 sccm and oscillatory motion of the substrate also shows the columnar feature of the deposition (Figure 7c). The Selected Area Electron Diffraction (SAED) patterns of the (Cr, Y)N x films confirm the formation of nanosized grains that generate the rings in the Diffraction Patterns (DPs), as shown in Figure 7d. Figure 8b-e shows schematic illustrations of the crystal structures of the Cr 2 N, CrN, Cr, YN, and Y phases. The CrN based coatings are characterized by their fine granulation and dense structure, and the thickness of the coatings is related to peak voltage, which corroborates the (Cr,Y)Nx coatings in that increasing the nitrogen flux increases the thickness of the coating by increasing the peak voltage [20]. The High-Angle Annular Dark-Field Scanning Transmission Electron Microscopy (HAADF-STEM) image of the cross-section shows the morphology of the Cr base layer and the (Cr, Y)Nx films with a highly segmented columnar structure (Figure 7a). Figure 7b shows (Cr, Y)Nx multilayer nanostructure with N2 flux of 50 sccm. So that Figure 7b shows that columnar grains exhibit a zig-zag grain growth morphology (black arrows) that is consistent across multiple layers. Bright-Field Transmission Electron Microscopy (BF-TEM) image of the nanostructured film grown with a nitrogen flow rate of 50 sccm and oscillatory motion of the substrate also shows the columnar feature of the deposition (Figure 7c). The Selected Area Electron Diffraction (SAED) patterns of the (Cr, Y)Nx films confirm the formation of nanosized grains that generate the rings in the Diffraction Patterns (DPs), as shown in Figure  7d. Figure 8b-e shows schematic illustrations of the crystal structures of the Cr2N, CrN, Cr, YN, and Y phases. The CrN based coatings are characterized by their fine granulation and dense structure, and the thickness of the coatings is related to peak voltage, which corroborates the (Cr,Y)Nx coatings in that increasing the nitrogen flux increases the thickness of the coating by increasing the peak voltage [20].   Figure 9a. The coating surface for all N2 flow rates exhibited cone-shaped growth-defect particles that showed the smallest amount for the 50 sccm conditions. Generally, these growth defects come from impurity particles still located on the substrate or the surface of the growing multilayer-like (Cr, Y)Nx coatings during the onset of deposition. The impact of the N2 flow rate on the arithmetic mean of surface roughness (Ra) and the corresponding effect on peak voltage HiPIMS is seen in Figure 9b. As the nitrogen flow rate increases from 10 to 50 sccm, the surface roughness decreases, while the negative peak voltage increases from 718 to 757 V during HiPIMS pulses.   Figure 9a. The coating surface for all N 2 flow rates exhibited cone-shaped growth-defect particles that showed the smallest amount for the 50 sccm conditions. Generally, these growth defects come from impurity particles still located on the substrate or the surface of the growing multilayer-like (Cr, Y)N x coatings during the onset of deposition. The impact of the N 2 flow rate on the arithmetic mean of surface roughness (Ra) and the corresponding effect on peak voltage HiPIMS is seen in Figure 9b. As the nitrogen flow rate increases from 10 to 50 sccm, the surface roughness decreases, while the negative peak voltage increases from 718 to 757 V during HiPIMS pulses. The surface maps obtained by tapping-AFM of the (Cr, Y)Nx films deposited by HiPIMS under substrate oscillation are presented in Figure 9a. The coating surface for all N2 flow rates exhibited cone-shaped growth-defect particles that showed the smallest amount for the 50 sccm conditions. Generally, these growth defects come from impurity particles still located on the substrate or the surface of the growing multilayer-like (Cr, Y)Nx coatings during the onset of deposition. The impact of the N2 flow rate on the arithmetic mean of surface roughness (Ra) and the corresponding effect on peak voltage HiPIMS is seen in Figure 9b. As the nitrogen flow rate increases from 10 to 50 sccm, the surface roughness decreases, while the negative peak voltage increases from 718 to 757 V during HiPIMS pulses.   Figure 10 shows the mechanical and tribological properties from which it is possible to observe an evolution of the residual stresses in the multilayer-like (Cr, Y)N x coatings deposited by HiPIMS as a function of N 2 flux. All coatings exhibited compressive residual stresses. By increasing the N 2 flow rate from 10 to 50 sccm, residual stresses increased (from −162 to −911 MPa). Residual stresses in magnetron-sprayed coatings often arise due to intrinsic growth and thermal stresses [27]. The compressive stresses gradually grew with increasing nitrogen content because of rising negative peak voltages. Hence, incorporating nitrogen in the coatings at a higher flow rate can effectively reduce surface roughness and cone-shaped defects and increase the compressive stresses. Nanomaterials 2022, 12, x FOR PEER REVIEW 11 of 17 Figure 10 shows the mechanical and tribological properties from which it is possible to observe an evolution of the residual stresses in the multilayer-like (Cr, Y)Nx coatings deposited by HiPIMS as a function of N2 flux. All coatings exhibited compressive residual stresses. By increasing the N2 flow rate from 10 to 50 sccm, residual stresses increased (from −162 to −911 MPa). Residual stresses in magnetron-sprayed coatings often arise due to intrinsic growth and thermal stresses [27]. The compressive stresses gradually grew with increasing nitrogen content because of rising negative peak voltages. Hence, incorporating nitrogen in the coatings at a higher flow rate can effectively reduce surface roughness and cone-shaped defects and increase the compressive stresses. When the N2 flow rate increases, the hardness values also increase, see Figure 10. The highest hardness value (32.6 ± 2.9 GPa) was measured to the N2 flow rate equal to 50 sccm. This is caused by the highest peak voltage during sputtering and the maximum compressive stresses, despite a nitrogen content lower than 30 sccm. There was a positive correlation between the rising N2 flow ratio (from 10 to 40 sccm) and the hardness values, which increased from 19.0 ± 1.7 to 28.1 ± 2.4 GPa, and the elastic modulus, which increased from 266.4 ± 59.0 to 362.8 ± 54.3 GPa due to the replacement of Me-Me bonds by Me-N bonds in the (Cr, Y)Nx film [32][33][34][35]. The hardness values obtained are higher than those of pure multilayer-like CrN coating in Reference [20] that was grown with 500 Hz and exhibited a maximum hardness of 26 GPa. This increase was due to the positive influence of yttrium on the microstructure of the fcc-CrN phase, which was considerably refined.
To evaluate the suitability of (Cr, Y)Nx coatings for tribological applications, the resistance to plastic indentation (H 3 /Er 2 ) was calculated with the data obtained from the instrumented nanohardness [36]. The H 3 /Er 2 ratio [37] is defined by the transition from elastic to plastic contact, proportional to the coatings type resistance to plastic deformation, and correlates well with wear resistance. Thus, the higher the H 3 /Er 2 ratio, the greater the capacity of the coating to dissipate energy through plastic deformation during loading. As nitrogen flow rates increase, the H 3 /Er 2 ratio increases, as seen in Figure  10.
The highest wear rate (1.92 × 10 −6 mm3 N −1 m −1 ) and wear volume (0.0096 ± 0.0015 mm³) were observed in (Cr, Y)Nx coatings formed under nitrogen flux below 20 sccm due to the lower hardness (19.0 ± 1.7 GPa) and larger granulometry of these coatings, as shown in Figure 10. Studies in Reference [38] reported a reduction in wear rate when reducing When the N 2 flow rate increases, the hardness values also increase, see Figure 10. The highest hardness value (32.6 ± 2.9 GPa) was measured to the N 2 flow rate equal to 50 sccm. This is caused by the highest peak voltage during sputtering and the maximum compressive stresses, despite a nitrogen content lower than 30 sccm. There was a positive correlation between the rising N 2 flow ratio (from 10 to 40 sccm) and the hardness values, which increased from 19.0 ± 1.7 to 28.1 ± 2.4 GPa, and the elastic modulus, which increased from 266.4 ± 59.0 to 362.8 ± 54.3 GPa due to the replacement of Me-Me bonds by Me-N bonds in the (Cr, Y)N x film [32][33][34][35]. The hardness values obtained are higher than those of pure multilayer-like CrN coating in Reference [20] that was grown with 500 Hz and exhibited a maximum hardness of 26 GPa. This increase was due to the positive influence of yttrium on the microstructure of the fcc-CrN phase, which was considerably refined.
To evaluate the suitability of (Cr, Y)N x coatings for tribological applications, the resistance to plastic indentation (H 3 /E r 2 ) was calculated with the data obtained from the instrumented nanohardness [36]. The H 3 /E r 2 ratio [37] is defined by the transition from elastic to plastic contact, proportional to the coatings type resistance to plastic deformation, and correlates well with wear resistance. Thus, the higher the H 3 /E r 2 ratio, the greater the capacity of the coating to dissipate energy through plastic deformation during loading. As nitrogen flow rates increase, the H 3 /E r 2 ratio increases, as seen in Figure 10. The highest wear rate (1.92 × 10 −6 mm3 N −1 m −1 ) and wear volume (0.0096 ± 0.0015 mm 3 ) were observed in (Cr, Y)N x coatings formed under nitrogen flux below 20 sccm due to the lower hardness (19.0 ± 1.7 GPa) and larger granulometry of these coatings, as shown in Figure 10. Studies in Reference [38] reported a reduction in wear rate when reducing the grain size. Moreover, increasing nitrogen flow rates above 30 sccm decrease the wear rate from 0.513 × 10 −6 to 0.130 × 10 −6 mm 3 N −1 m −1 and the wear rate from 2.6 × 10 −3 ± 0.9 × 10 −3 to 0.6 × 10 −3 ± 0.9 × 10 −3 mm 3 , due to improvements in coating hardness (Cr,Y)N x [39].
(Cr, Y)N x coatings with higher N 2 concentrations (30, 40, and 50 sccm) showed better performance in wear resistance (Figure 11), as well as the most increased coating toughness given by the H 3 /E 2 ratios ( Figure 10) [40]. Hence, we can observe the positive effects of the N 2 flux on the phase composition, crystalline structure, and tribological and mechanical properties of multilayer-like (Cr, Y)N x films produced by HiPIMS. The best mechanical and tribological properties were measured to thin films manufactured at the N 2 flow equal to 50 sccm. According to Reference [41], increasing hardness was also verified for rising nitrogen contents in pure CrN films.   Figure 11 displays the wear tracks of (Cr, Y)N x coatings generated by the dry linear reciprocating wear test, as measured by non-contact optical profilometry. All (Cr, Y)N x coatings grown with different nitrogen flow rates showed material accumulation at the edges. The most significant wear track depth is for 10 sccm flow. As the flow of nitrogen increases, the maximum depths tend to decrease. The lowest maximum depth is for (Cr, Y)N x films manufactured with a flow rate of 50 sccm [41]. Moreover, the grooves in the lower region of the indicative abrasive wear ranges are also notable. Figure 12 displays the worn surface of the Al 2 O 3 balls, presenting traces of adhered debris coming from the counter body. For Al 2 O 3 balls in Figure 12, it is possible to observe material adhering to the surface, and for the 10 sccm flow rate, it is possible to follow wear marks, as expected from the abrasive wear mechanism. Note that the surface of the sphere with nitrogen fluxes of 10 and 20 sccm has the most material adhered [41,42].   The friction coefficient was measured during the dry reciprocating wear tests. Table 2 shows the maximum, and minimum peak friction coefficients (µ), as well as their root, mean square (RMS) values (µ rms ), which are a statistical measure of the magnitude of friction along the tests. The multilayer-like (Cr, Y)N x films manufactured with an N 2 flux of 50 sccm exhibits a minimum µ rms corresponding to the lowest wear rate [42]. This is possibly related to the reduced wear rate observed, which can be correlated with the obtained nanohardness values. When analyzing the friction coefficient of the (Cr, Y)N x films, it is notable that the nitrogen flow of 30 sccm showed a rapid and sharp increase in the initial cycles, as shown in Figure 13. Coatings produced with a nitrogen flow of 10, 20, 40, and 50 sccm initially show a similar trajectory, with the coefficient of friction rising slower. However, for 10 sccm, there is a significant decrease in the friction coefficient. For flows of 20 and 40 sccm, the friction reaches a steady-state, while the coefficient of friction for the 50 sccm condition reaches a maximum steady-state value after a second gradual increase at the end of the cycle.

Conclusions
Regarding the application of different nitrogen flow rates (10,20,30,40, and 50 sccm) and the addition of 2 at.% yttrium on the microstructure, the following conclusions can be drawn: 1. The thickness and deposition rate increase as the nitrogen flow rate increases, consequent to the rise in HiPIMS peak voltages applied to the Cr-Y target. 2. The addition of 2 at.% Y increases the hardness of the fcc-CrN phase by almost 30% due to considerable refinement of the film architecture. 3. Phase formation at low nitrogen flux 10 to 30 sccm has a microstructure formed by fcc-CrN and hexagonal Cr2N; at higher nitrogen fluxes, there is only fcc-CrN formation.

Conclusions
Regarding the application of different nitrogen flow rates (10,20,30,40, and 50 sccm) and the addition of 2 at.% yttrium on the microstructure, the following conclusions can be drawn:

1.
The thickness and deposition rate increase as the nitrogen flow rate increases, consequent to the rise in HiPIMS peak voltages applied to the Cr-Y target.

2.
The addition of 2 at.% Y increases the hardness of the fcc-CrN phase by almost 30% due to considerable refinement of the film architecture.