Using Coatings Based on the ZrN System to Improve the Corrosion Resistance of Stainless Steel Products

Abstract

This article investigates the anticorrosive properties of Zr-ZrN coatings, including Zr-(Zr,Hf)N, Zr-(Zr,Ti)N, Zr,Hf-(Zr,Hf,Nb)N, and Zr,Nb-(Zr,Nb)N, deposited on AISI 321 stainless steel substrates. The hardness and elasticity modulus of these coatings, as well as their scratch test strength, were measured. Corrosion current densities were calculated using the polarisation resistance method and by extrapolating the linear sections of the cathodic and anodic curves under electrode polarisation. The structure and composition of the sample surfaces were analysed by transmission electron microscopy. Notably, the nitride coatings reduced the corrosion current density in a 3% aqueous NaCl solution at 25 °C by more than 10 times, from 6.96 for the uncoated substrate to 0.17 μA/cm² for the Zr-(Zr,Ti)N-coated sample. The addition of Ti nitride to Zr-ZrN led to the most significant decrease in the corrosion current density. However, the introduction of Nb caused an increase in the corrosion rate and a decrease in the polarisation resistance, and Hf did not affect the corrosion-protective properties of the studied nitride coatings.

1. Introduction

Nitride coatings have been employed to improve the corrosion resistance of stainless steel since at least the early 1990s. In particular, the corrosion resistance of austenitic steel was markedly increased by depositing TiN and ZrN coatings [1]. In the case of ZrN, the highest corrosion resistance was achieved using alternating layers of Zr and ZrN. The corrosion resistance of TiN and ZrN coatings on an AISI 304 stainless steel substrate in a 0.5 N NaCl solution has also been investigated [2], showing that TiN is inert, whereas ZrN reacts with an aggressive environment. The authors suggested that the nitrogen (N) in ZrN is replaced by oxygen (O), forming a zirconium oxide (ZrO₂) film, which has good corrosion resistance and improves corrosion protection. ZrO₂ forms spontaneously upon contact with an O-containing environment, occurring naturally during the coating deposition process. The inclusion of a ZrO₂ layer in the coating structure reduces the corrosion current density in 3.5 wt.% NaCl and 20 wt.% HCl solutions compared with a ZrN coating without an oxide layer [3]. The anticorrosive properties of the multilayer coating are attributed to a combined effect, including increased strength and reduced stress levels owing to the multilayer structure and high corrosion resistance provided by the ZrO₂ sublayer. Compared with the uncoated sample, the corrosion current density (i_corr) and polarisation current density (i_p) values for samples with ZrN/Zr bilayers were significantly lower (at least one order of magnitude) than those measured for samples with a single Zr or ZrN coating [4]. This result was ascribed to a decrease in the density of point defects in coatings with a multilayer structure. The interlayer interface inhibits the corrosion channel from passing through the coatings into the metal substrate, minimising the area of impact and, thus, improving corrosion resistance.

TiN and ZrN coatings were shown to significantly enhance the corrosion resistance of 316L stainless steel [5]. Electrochemical impedance spectroscopy and polarisation curves showed that the TiN coating had higher corrosion resistance than the ZrN coating. At the same time, a comparison of the anticorrosion properties among TiN, CrN, and ZrN coatings showed that ZrN provided the strongest corrosion protection, whereas TiN only outperformed the uncoated samples [6]. To further reduce corrosion, coatings combining TiN and ZrN have also been employed. TiN/ZrN coating with a nanocomposite structure improved the corrosion resistance by approximately 6 times compared with the uncoated sample [7]. This coating also increased the corrosion potential from −0.085 to 0.080 V, which significantly exceeds the results obtained for monolithic CrN and TiN coatings. The corrosion properties of nanocrystalline (Ti,Zr)N coatings on an AISI 304 stainless steel substrate were also studied [8]. The steady-state potential during the corrosion process (E_corr) only slightly differs for the coated and uncoated samples. This may indicate that the corrosion of the TiN- or (Ti,Zr)N-coated samples was mainly caused by the dissolution of the metal substrate, and not directly by the TiN or (Ti,Zr)N coatings. At the same time, TiN and (Ti,Zr)N coatings significantly reduce the i_corr values compared with uncoated samples.

The coating structure also has a significant effect on its corrosion resistance. Single-layer, multilayer, and gradient ZrN coatings were shown to reduce i_corr by 83.0%, 99.8%, and 95.8%, respectively [9]. Thus, the use of gradient and multilayer coating structures can enhance corrosion resistance by increasing the crack resistance and mitigating the formation of penetration channels. The coating structure can be modified by introducing additional elements, such as silicon (Si). A nanocomposite coating (Zr,Si)N, in which the nanoscale crystallised ZrN phase is embedded in an amorphous silicon nitride matrix, was used to improve the corrosion resistance of stainless steel in a 3.5% NaCl solution [10]. The results showed that (Zr,Si)N coatings are more uniform than ZrN and exhibit higher corrosion resistance than ZrN and the uncoated sample. ZrN-coated steel exhibited localised corrosion through pores in the coating, whereas (Zr,Si)N-coated steel exhibited uniform corrosion at the coating–steel interface [11].

An additional increase in corrosion resistance can be achieved by optimising the deposition parameters. For instance, the nitrogen gas (N₂) pressure has a significant effect on corrosion resistance because of the changes in coating defects [12]. The ZrN coating deposited on a 316 L steel substrate at an optimal N₂ pressure (3.5 Pa) demonstrated increased corrosion resistance (E_corr = 0.307 V, i_corr = 2.085 × 10⁻⁷ A/cm²), and the coating structure exhibited minimal defects. The effect of substrate bias voltage on the corrosion properties of the ZrN coating was also studied [13]. Additional post-treatment of the coated samples, such as vacuum annealing, also enables further improvements in corrosion resistance. For example, the corrosion resistance of vacuum-annealed 304SS steel with a ZrN coating in a NaCl solution increased by at least 100 times compared with that of uncoated samples [14]. Multilayer nanostructured ZrN/CrN coatings provided a significant increase in resistance to aggressive environments, owing to the enhanced barrier properties of multilayer structures [15,16]. Similarly, multilayer (Cr,Zr)N/ZrN coatings exhibited more effective anticorrosion properties than single-layer (Cr,Zr)N and ZrN coatings [17]. For multilayer (Cr,Zr)N/ZrN coatings, i_corr was only 2.078 × 10⁻⁷ A/cm², 2 times lower than that of the single-layer (Cr,Zr)N coating. The multilayer coating with local coherent interfaces appears to be beneficial. The ratio of the nanolayer coating thicknesses also affects the corrosion resistance. Potentiodynamic polarisation tests showed that multilayer CrN/ZrN and CrN/(Cr,Al)N coatings significantly improve the corrosion resistance of stainless steel [18]. Moreover, the CrN/ZrN coating demonstrated the lowest corrosion current density because it had a lower defect density and higher chemical stability than the CrN/(Cr,Al)N coating. Multilayer CrN/ZrN coatings (with different modulation periods of 1, 8, 15, and 30) were applied to AISI 420 steel substrates [19].

The corrosion properties of NbN-based coatings have also been actively studied. In particular, multilayer TiN/NbN coatings provided more corrosion protection than single-layer TiN and NbN coatings [20,21,22]. The corrosion resistance of multilayer coatings improved with an increase in the total number of interfaces in the coatings. Furthermore, effective corrosion protection was provided using CrN/NbN coatings with a nanolayer structure [23]. Electrochemical corrosion testing revealed that the corrosion resistance of the NbN coating tended to increase with increasing spray power [24]. Multilayer CrN/NbN coatings demonstrated a low corrosion rate in 3.5% NaCl solutions and effectively protected the steel substrate from corrosion [25,26,27]. Nanolayer CrN/NbN coatings showed distinct passivation behaviour, with pitting potentials ranging from 230 to 400 mV, depending on their chemical composition and residual stress level [28]. The defect rate in CrN/NbN coatings decreased, and the corrosion resistance increased, with increasing bias voltage and substrate surface temperature [29].

An investigation of (Nb,Al)N coatings with different Al contents showed that the addition of Al negatively affects the corrosion resistance [30,31]. Substrates with a CrN/NbN coating showed higher corrosion resistance than substrates with a (Cr,Al,Ti)N coating [32]. Moreover, the introduction of 40 or 60 at.% Al increased the coating resistance to corrosion in a 3.5% NaCl solution, owing to the formation of Al(OH)₃ and Nb₂O₅ films.

HfN-based coatings are also effective in corrosion protection [33]. A multilayer Hf/HfN and HfN/VN coating exhibited low corrosion current density and a small number of corrosion pits after salt spray test [34,35]. The corrosion current density of samples with (Hf,Y)N coatings (9.8 to 18.5 at.% Y) decreased by two orders of magnitude compared with that of the substrate, reaching a minimum of 4.21 × 10⁻⁸ A/cm² [36].

Overall, it can be concluded that ZrN-, NbN-, and HfN-based coatings can improve the corrosion resistance of stainless steel surfaces. The introduction of additional elements and the use of multilayer coating structures are generally effective for further improving the corrosion resistance. However, the introduction of some elements, such as Al, into the coating composition can negatively impact its corrosion resistance.

The present article considers the corrosion properties of stainless steel specimens with (Zr,Nb)N, (Zr,Hf)N, and (Zr,Nb,Hf)N coatings. Samples with and without ZrN and (Zr,Ti)N coatings are also considered for comparison. A previous comparison of Ti6Al-4V alloy specimens with the aforementioned coatings showed the particular effectiveness of the (Zr,Hf)N coating in reducing corrosion [37]. The introduction of Hf into ZrN suppressed the corrosion rate in 3% NaCl to 0.118 μA/cm², whereas the addition of Nb to the ZrN system accelerated corrosion (3.440 μA/cm²).

It is necessary to account for the significant differences in the corrosion processes that are characteristic of titanium alloys and stainless steel. Titanium alloys, such as Ti6Al-4V, are characterised by inherently high corrosion resistance in 0.9% and 3.0% NaCl solutions, which is due to the formation of a dense non-conducting oxide film on the titanium alloy surface [38,39,40,41]. Conversely, the corrosion resistance of stainless steels in these environments is related to the surface self-passivation that occurs upon contact with a corrosive environment. In this case, the transition of the stainless steel surface to a passive state is accompanied by etching of the iron (Fe) component, and the remaining chromium (Cr; more than 12 at.%) and nickel (Ni; more than 9 at.%) quickly form a dense layer that prevents the onset and spread of corrosion [42]. Hence, the corrosion resistance of stainless steel is determined not by the chemical resistance of the main elemental component, as in titanium alloys, but by the content of alloying components (Cr and Ni). Moreover, unlike titanium alloys, many stainless steels require special treatment in oxidising environments to impart their passive state [43].

The objective of this work is to increase the resistance to corrosion in sea water and wear resistance of stainless steel products operating under conditions of simultaneous exposure to corrosion and mechanical wear.

2. Materials and Methods

2.1. Coating Deposition

Samples are cylindrical in shape, 25 mm in diameter and 3 mm in thickness.

Coating deposition was performed using a VIT-2 vacuum plasma system (IDTI RAS—MSUT, STANKIN, Moscow, Russia), as previously described [44,45,46,47,48,49]. Evaporators implementing controlled accelerated arc technology [50,51,52] were used. This technology allows for reduced energy consumption and minimises the number of microparticles in the plasma flow.

Before coating deposition, the samples were prepared as follows:

• Rinsing in an alkaline solution at 80 °C under ultrasonication.
• Rinsing in specially purified water (filtration on a carbon filter).
• Blowing with a stream of purified air (filtration).

During deposition, the arc currents of the titanium (Ti) and zirconium (Zr) cathodes were 75 and 80 A, respectively. The arc current of the Zr (99.98%), Ti (99.97%), Zr-Nb (1:1), and Zr-Hf (1:1) cathodes was 65, 70, 85, and 90 A, respectively. The remaining parameters were the same for all processes: N₂ pressure of 0.42 Pa, substrate voltage of 150 V, tool rotation speed of 0.7 rpm, and substrate temperature maintained in the range of 400–450 °C.

2.2. Coating Properties

The surface morphology and structure of the coatings were examined by scanning electron microscopy (SEM; EVO 50, Carl Zeiss, Oberkochen, Germany), energy-dispersive X-ray spectroscopy (EDX; X-Max 80 mm², OXFORD Instruments, High Wycombe, UK), and transmission electron microscopy (TEM; JEM 2100, JEOL, Akishima, Japan). The elemental composition was evaluated by TEM-EDX (INCA Energy, OXFORD Instruments, High Wycombe, UK). Samples were prepared for TEM analysis by focused ion beam milling using a Strata 205 device (FEI, Hillsboro, OR, USA). Phase analysis was performed on an X-ray diffractometer (XRD)–D2 Phaser 2 Gen. (Bruker, Billerica, MA, USA). Hardness and elastic modulus were measured using an SV-500 nanotester (Nanovea, Irvine, CA, USA) and a Berkovich indenter at a maximum load of 200 mN. Average values were determined based on 20 measurements. Scratch testing was conducted using an SV-500 tester (Nanovea, Irvine, CA, USA) according to the ASTM C1624-05 standard [53].

Corrosion studies were conducted using an AUTOLAB PGSTAT302 N potentiostat-galvanostat (Metrohm Autolab B.V., Herisau, Switzerland) in a 3% NaCl solution at 25 °C, employing a three-electrode electrochemical cell with a platinum auxiliary electrode and a saturated silver chloride reference electrode. The intended operating environment is seawater. The salinity of the World Ocean averages 3.47%; in fresher seas (such as the Black Sea or the Sea of Azov), salinity is lower (approximately 1.5–2.0%), meaning 3% NaCl solution simulates seawater well. Potential values were converted to the standard hydrogen electrode scale. The working electrode area was 1 cm². The open circuit potential (ECP) was assessed for 1 h before potentiodynamic studies were performed. The susceptibility of samples to corrosion was assessed based on the potentiodynamic curves; the potential sweep rate was V_p = 1 mV/s, performed at an electrode polarisation (η) of ±800 mV [38], which was determined using the following formula:

$$\eta = E - E_{corr},$$

(1)

where $E$ is the electrode potential under an applied current, and $E_{corr}$ is the steady-state potential for the corrosion process. The sample was held in 3% NaCl at 25 °C for 1 h before measurement. Corrosion current densities were calculated using Nova 2.0 software (Netherlands) and the polarisation resistance method, for experimental data obtained at low polarisations (η = ±20 mV):

$$\eta \approx {\displaystyle \frac{RT}{nF}} {\displaystyle \frac{i}{i_0}}=R_{0}i$$

(2)

where $R_0={\displaystyle \frac{RT}{i_{0}nF}}$ is the charge transfer resistance, Ω [42].

Additionally, all obtained potentiodynamic curves were analysed using Nova 2.0 (Netherlands) by extrapolating the linear sections of the cathodic and anodic curves with an electrode polarisation of ±100 mV from the Tafel equation:

$$\eta =a+b lg{i}_{corr}$$

(3)

where $a={\displaystyle \frac{nF}{RT}} \lg i_o$n and $b={\displaystyle \frac{nF}{RT}}$ are Tafel coefficients, B and B/decade, respectively; $i_{corr}$ is the corrosion current density, A/cm²; R is the universal gas constant, 8.314 J/(mol∙K); T is the absolute temperature, K; F is the Faraday constant, 96.485.33 C/mol; i_o is the exchange current density, A/cm²; and n is the number of electrons participating in the reaction [54].

Corrosion tests were carried out in a 3.0 wt.% aqueous NaCl solution at a temperature of 25 °C for 7 days. Three independent experiments were conducted for each coating type. The spread of values for certain parameters was within 17%. The final tables show the average values from the three tests.

3. Results

3.1. Substrate Properties

AISI 321 is an austenitic-ferritic stainless steel. The elemental composition of the steel is presented in

Table 1. Elemental composition of AISI 321 steel.

Element	Mass Fraction (%)
Si	Max 1.0
C	Max 0.08
P	Max 0.045
Mn	Max 2.0
Cr	17–19
S	Max 0.03
Ni	9–11
Ti	0.5
Fe	Rest

.

AISI 321 is an austenitic-ferritic stainless steel. X-ray diffraction (XRD) reveals that uncoated samples contain two phases: austenite (Fm3m, A1) and ferrite or alpha-iron (Im3m, A2), in approximately equal amounts. This is consistent with the selected area electron diffraction (SAED) data (Figure 1). Austenite is stabilised by Ni because it is a high-temperature phase. Judging by the intensities of the XRD peaks, the original sample contains slightly more alpha-iron than the sample observed after corrosion testing.

3.2. Mechanical Properties of the Coatings

The general appearance of the samples with coatings is shown in Figure 2. The samples have a colour that is characteristic of these types of coatings.

The coating thickness measured on the transverse section is 4.8–5.5 µm; a certain variation in thickness is typical for coatings deposited by the PVD method.

Table 2. Main characteristics of the studied coatings.

Coating	Elemental Composition [at.%]				Hardness	Adhesion Strength to the Substrate, LC2 (N)
	Zr	Nb	Hf	Ti
(Zr,Nb)N	50.5	49.5	−	−	HV 2336 ± 115	15 ± 2
ZrN	100	−	−	−	HV 2993 ± 145	12 ± 2
(Zr,Hf)N	76.12	−	23.88	−	HV 3350 ± 120	7 ± 1
(Zr,Ti)N	46.58	−	−	53.42	HV 2727 ± 86	6 ± 2
(Zr,Nb,Hf)N	72.07	24.87	3.05	−	HV 2860 ± 95	6 ± 1

presents the main characteristics of the coatings. In terms of composition, deposition using two cathodes of single-element composition (Zr and Ti) and a two-component Zr-Nb cathode (1:1) results in coatings with nearly equivalent (1:1) compositions, but in the case of a Zr-Hf cathode, although the elemental ratio in the cathode composition is 1:1, the coating composition is dominated by Zr with a significantly lower content of hafnium (Hf). This likely occurs because the atomic weight of Hf is significantly higher than that of Zr (178.49 and 91.224 g/mol, respectively [55]). Hf atoms require more energy to reach the substrate surface, so some of them are deposited on the chamber walls and the tooling, whereas lighter Zr atoms reach the substrate in greater quantities. A similar process occurs for the (Zr,Nb,Hf)N coating deposited using Zr-Hf (1:1) and Zr-Nb (1:1) cathodes. The (Zr,Hf)N coating exhibits the highest hardness, and the (Zr,Nb)N coating exhibits the lowest hardness. The (Zr,Nb)N coating, however, demonstrated the highest adhesion strength to the substrate (i.e., the highest total coating failure rate, L_C2), and the ZrN coating also demonstrated relatively high L_C2 values. The other coatings demonstrated relatively low adhesion strengths. Note that this relatively low adhesion strength (measured using a scratch test) is due to both the large difference in the hardness of the coating and the substrate, as well as the use of a “realistic” substrate (without special surface treatment, such as polishing). These substrate properties were selected to accurately represent the operating conditions of real products in a seawater environment.

A previous study on steel samples with similar coatings showed that the highest wear resistance obtained during ball-on-disk testing was provided by a Zr,Hf-(Zr,Hf,Nb)N coating (wear was 1.5 times lower than that of the uncoated sample) [56]. This coating also provided the greatest reduction in the coefficient of friction.

3.3. Corrosion Behaviour of the Samples with Nitride Coatings

Polarisation studies of the corrosion behaviour of nitride coatings on the surface of AISI 321 steel were carried out by recording the anodic and cathodic polarisation curves for each sample in 3% NaCl at 25 °C (Figure 3). The general appearance of the anodic curve of an uncoated AISI 321 steel sample in 3% NaCl at 25 °C suggests the self-passivation of this steel in an oxygen-containing environment [40]. The anodic curve for AISI 321 steel in 3% NaCl displays a dissolution region from −0.77 to −0.29 V and a passivation region (−0.29 to 0.00 V). With a further increase in potential, a sharp increase in current density is observed on the anodic curve (Figure 3), which is characteristic of oxygen evolution and substrate dissolution [57].

The application of Zr-ZrN, Zr,Nb-(Zr,Nb)N, Zr-(Zr,Hf)N, Zr,Hf-(Zr,Hf,Nb)N, and Zr-(Zr,Ti)N coatings with a thickness of ~3 μm shifts the corrosion potential by 0.20–0.25 V to a more electropositive region (Figure 3). This indicates an active interaction between the coatings and the corrosive environment, involving the formation of oxonitrile layers with increased electrical resistance [39,44]. Moreover, the passivation region that is characteristic of uncoated AISI 321 steel disappears in the anodic polarisation curves of all coated samples, which further indicates the blocking of the corrosive dissolution process of AISI 321 steel.

A comparison of the values of open circuit potentials (E_oc, V) for samples of AISI 321 steel with Zr-ZrN, Zr,Nb-(Zr,Nb)N, Zr,Hf-(Zr,Hf,Nb)N coatings with their corrosion potentials (E_corr) (

Table 3. Corrosion properties extrapolated from the linear sections of the anodic and cathodic polarisation curves, obtained for AISI 321 steel samples with different coatings in 3% NaCl at 25 °C.

№	Coating	Eoc, V	Ecorr, V	Tafel Plot Extrapolation Method
				icorr, μA/cm2	P, μm/year	βc, V/dec	βa, V/dec	Ecorr calc, V	R0, kΩ
1	uncoated	−0.492	−0.759	6.96	80.8	0.15	0.45	−0.767	6.9
3	Zr-ZrN	−0.529	−0.516	0.25	2.9	0.11	0.11	−0.519	98.0
4	Zr,Nb-(Zr,Nb)N	−0.599	−0.571	0.78	9.1	0.19	0.11	−0.565	39.6
8	Zr-(Zr,Hf)N	−0.474	−0.523	0.31	3.6	0.15	0.12	−0.517	94.6
9	Zr,Hf-(Zr,Hf,Nb)N	−0.578	−0.559	0.45	5.2	0.16	0.10	−0.562	61.9
10	Zr-(Zr,Ti)N	−0.484	−0.563	0.16	1.9	0.15	0.16	−0.563	207.0

and

Table 4. Polarisation resistance determined from the corrosion potential region, obtained for AISI 321 steel samples with different coatings in 3% NaCl at 25 °C.

№	Coating	Ecorr, V	Polarisation Resistance Method
			icorr, μA/cm2	P, μm/year	R0, kΩ cm2
1	uncoated	−0.759	4.77	55.4	5.5
3	Zr-ZrN	−0.516	0.34	3.9	77.8
4	Zr,Nb-(Zr,Nb)N	−0.571	0.84	9.7	31.1
8	Zr-(Zr,Hf)N	−0.523	0.34	3.9	76.9
9	Zr,Hf-(Zr,Hf,Nb)N	−0.559	0.57	6.6	45.8
10	Zr-(Zr,Ti)N	−0.563	0.17	2.0	154.4

) demonstrates a shift to a more electropositive region of E_oc in relation to the E_corr value by 0.013–0.021 V, which indicates the occurrence of self-hazardization processes of the surface during the exposure of these coatings to 3% NaCl for 1 h at 25 °C. For the samples with Zr-(Zr,Hf)N and Zr-(Zr,Ti)N coatings, the value of E_oc is lower than E_corr by 0.049 and 0.079 V, respectively, which is explained by the absence of oxide films before the start of testing after 1 h of exposure in 3% NaCl at 25 °C. It should be noted that E_oc for the uncoated sample of AISI 321 steel is shifted by 0.267 toward a more electropositive region with respect to E_corr, which indicates self-passivation of this alloy in 3% NaCl at 25 °C.

Taking into account the minimum corrosion currents observed at the corrosion potential in a 3% aqueous NaCl solution at 25 °C for both the coated uncoated AISI 321 steel samples, the polarisation resistance was determined from the corrosion potential region (Table 3), and the corrosion current density was estimated by extrapolating the linear sections of the anodic and cathodic polarisation curves (Table 4).

Note that the current density values for AISI 321 steel coated with Zr-ZrN, Zr,Nb-(Zr,Nb)N, Zr-(Zr,Hf)N, Zr,Hf-(Zr,Hf,Nb)N, and Zr-(Zr,Ti)N, as calculated by extrapolating the Tafel sections of the anodic and cathodic polarisation curves (Table 3), are within the range of 0.17–0.84 μA/cm², which are somewhat higher than the current density values calculated based on the polarisation resistance in the corrosion potential region (Table 4). This indicates that the method of determining the polarisation resistance for nitride coatings is more adequate for assessing the corrosion rate [38,57].

The polarisation resistance values of the coated samples during the corrosion process are in the range of 31.1–154.4 kΩ cm². A comparison of the current density and polarisation resistance (Figure 4) reveals a discrepancy of 3–27%, which largely reflects the general error of electrochemical methods for assessing corrosion resistance [38,57].

The structure of this error includes the initially different degree of surface oxidation for two identical samples and, consequently, different primary oxidation rates of coatings during the studies, as well as the non-equilibrium of the processes caused by the flow of corrosion current in the system [38,57].

For subsequent evaluations of the corrosion resistance, the data obtained based on the polarisation resistance in the corrosion potential region (Table 3) were used. However, the corrosion behaviour of uncoated AISI 321 steel in a 3% aqueous NaCl solution at 25 °C is well-characterised using the Tafel equation, with a fairly high corrosion current of 6.96 μA/cm².

The high corrosion resistance of AISI 321 steel samples with Zr-(Zr,Ti)N, Zr-ZrN, and Zr-(Zr,Hf)N coatings in 3% NaCl at 25 °C is characterised by relatively low corrosion current densities (0.17–0.34 μA/cm²) and high polarisation resistances (76.9–154.4 kΩ cm²), which is over an order of magnitude higher than the polarisation resistance of uncoated AISI 321 steel (6.9 kΩ cm²). The performance of the Zr-(Zr,Ti)N, Zr-ZrN, Zr-(Zr,Hf)N coatings is primarily ascribed to the presence of a Zr O₂ transition layer, which blocks the spread of corrosion deep into the coating. Samples with a Zr-(Zr,Ti)N coating demonstrate the lowest corrosion currents, namely 0.17 μA/cm², owing to the faster dissolution of TiN and the formation of a poorly soluble titanium oxide of non-stoichiometric composition on the surface [38,58], expressed as follows:

2TiN + 2yH₂O ⟶ 2TiNxOy + (1 − x) N₂ + 4yH⁺ + 4ye⁻

(4)

2TiN_xO_y + 2H₂O ⟶ 2Ti³⁺ + (y + 1) O₂ + xN₂+ 4H⁺ + 10e⁻

(5)

Ti³⁺ + xH₂O ⟶ TiO_x + 2xH⁺ + (2x − 3) e⁻

(6)

The introduction of Hf into the coating does not lead to a significant change in corrosion resistance because the ZrO₂ and HfO₂ oxides that form during corrosion have similar chemical properties [56].

The introduction of niobium (Nb) into the coating slightly increased the corrosion current densities to 0.84 and 0.57 μA/cm² for Zr,Nb-(Zr,Nb)N and Zr,Hf-(Zr,Hf,Nb)N, respectively. This is due to the dissolution of NbN and the formation of a complex oxide (NbO_x) with a non-stoichiometric composition, which has increased solubility in water because of the acid–base interaction [59,60]. In this case, partial destruction of the NbO_x corrosion product layer and the development of a corrosion process within the transition layer of Zr and Nb are expected.

3.4. Elemental Distribution on the Surface of Coated Samples After Corrosion Exposure

Elemental mapping of the coated specimen surfaces after corrosion testing (Figure 5) reveals some differences in O intensity. Considering that all the coatings contain Zr, this element was chosen as an indicator of the presence of a coating in the 25 µm × 25 µm sample surface area under observation. Accordingly, the O content primarily indicates oxidation processes on the coating surface. The highest O intensity is observed for the Zr,Nb-(Zr,Nb)N and Zr,Hf-(Zr,Hf,Nb)N coatings (i.e., coatings containing Nb). These results correlate well with the data obtained above.

3.4.1. Coating ZrN

XRD shows that peaks from the fcc Fe phase (A1) are more prominent in the substrate (Figure 6). SAED shows that the structure closer to the substrate edge is more dispersed. This is also evident in the images. The number of phases, according to SAED, is approximately equal.

3.4.2. Coating Zr,Hf-(Zr,Hf)N

XRD analysis of the Zr,Hf-(Zr,Hf)N coating shows (Figure 7) that the substrate has more intense peaks from the bcc Fe phase (A1). SAED leads to the same conclusion, indicating that the A1 phase is predominant. The structural dispersion, as seen from both the images and SAED, is significantly lower on the surface of the sample than in the bulk.

From the EDX data, a decrease in Cr content is observed closer to the substrate surface (Figure 8). On the coating surface, the presence of iron oxide particles is noticeable (based on the O and Fe content), but there are no signs of significant oxidation in the coating. At the same time, in some measurement areas (area 2 in Figure 8a–c), both O and the coating elements are present, which indicates the potential formation of zirconium oxides, oxynitrides, and hafnium oxides.

3.4.3. Coating Zr-(Ti,Zr)N

XRD data show that the substrate exhibits more noticeable peaks from the bcc Fe phase (A1) (Figure 9). From the SAED results, it is evident that the structure closer to the edge of the substrate is more dispersed. This is even more noticeable in the images. Consistent with XRD, the bcc Fe phase (A1) is predominant, according to SAED.

3.4.4. Coating Zr,Nb-(Zr,Nb,Hf)N

XRD reveals that the amount of bcc Fe (A1) is significantly greater than that of the fcc phase (Figure 10). Based on the SAED reflections, the A2 phase is more abundant than A1 closer to the edge of the substrate, but at greater depth, the A1 phase dominates. The structural dispersion does not change within the visible range of the substrate, as confirmed by SAED. Despite signs of surface oxidation in the coating in the elemental mapping, TEM studies did not reveal an oxide layer. This layer may be too thin, or it may have been destroyed during lamella preparation.

3.4.5. Coating Zr,Nb-(Zr,Nb)N

XRD data show that the amount of bcc Fe phase (A1) is higher than that of the fcc phase (Figure 11). This is consistent with the SAED reflections, showing that there is more bcc A1 phase closer to the surface. The structure’s dispersion changes towards the edge of the substrate.

4. Discussion

The anticorrosive properties of coated samples are largely related to the adhesion strength between the coating and the substrate. Additionally, the features of the interface between the steel substrate and the coatings with different compositions must be considered. Previous studies on the properties of coatings deposited on a steel substrate show the presence of a transition zone between the coating and the substrate, formed as a result of diffusion and the action of high-energy ions, inducing the mixing of elements in the interfacial zone [61,62,63,64]. When the coating peels off from the substrate, an aggressive environment can penetrate the resulting cavities, and active corrosion of the substrate begins [65,66]. In turn, the strength of the adhesive bond between the coating and the substrate is largely determined by the properties of the substrate (e.g., grain size, phases, hardness, and surface roughness) [67,68,69,70,71]. In particular, a decrease in the surface roughness of the stainless steel substrate leads to increased coating adhesion [69] and increased resistance to corrosion in alkaline environments [65]. The residual stress gradient at the coating–substrate interface and defects on the substrate surface significantly affect the performance and failure mode of the coating. The coating wear rate can increase by up to 72% owing to an increase in interface roughness [72]. Laser-modified interfaces significantly reduce the grain size of deposited coatings, weaken their crystallographic texture, and reduce residual stresses, which leads not only to increased adhesion but also to increased corrosion resistance [70]. A similar effect is achieved by plasma [73] and plasma-laser [71] texturing of the substrate before coating deposition. The adhesion strength at the coating–substrate interface also depends on the type of interatomic bonds and the load-bearing capacity of the substrate (hardness), as well as the strength of the outer layers of the substrate [74]. The presence of coating defects is associated with substrate roughness values, with the number of surface defects being a key parameter influencing the rate of coating wear [72]. Furthermore, surface coating defects serve as sites of stress concentration, leading to premature failure of the coating during operation.

In the samples studied herein, a sufficiently tight interface between the coatings and the substrate is observed, with no signs of delamination (Figure 12). The Zr,Nb-(Zr,Nb)N coating exhibits the highest resistance to destruction in the scratch test, whereas the Zr-(Ti,Zr)N, Zr,Hf-(Zr,Hf)N, and Zr,Nb-(Zr,Nb,Hf)N coatings exhibit relatively low fracture strength. At the same time, the Zr,Hf-(Zr,Hf,Nb)N coating provides the best wear resistance and the lowest friction coefficient in the ball-on-disk test, which has previously been established [56]. Thus, the Zr-(Ti,Zr)N coating, although not providing the highest adhesion to the substrate, provides the highest resistance to corrosion. In this case, it can be assumed that the key to corrosion resistance is the non-porous structure of the applied multilayer coatings, in which dense surface films of zirconium oxides and oxonitrides play the primary corrosion-protective role. The addition of Ti nitride to Zr-ZrN significantly decreases the corrosion current density. The introduction of Nb, however, increases the corrosion rate and reduces the polarisation resistance to corrosion, whereas Hf has no effect on the corrosion-protective properties of the nitride coatings. Although the surface oxide layer can only be directly observed in the sample coated with Zr,Hf-(Zr,Hf)N, it may also be present in other samples, based on the elemental mapping data. Further investigation is needed to achieve a more pronounced oxide layer, potentially by increasing the sample soak time in the saline solution and raising the experimental temperature, thereby allowing for a more detailed study.

5. Conclusions

The anticorrosive properties of coatings based on the ZrN system were studied herein, specifically for Zr-(Zr,Hf)N, Zr-(Zr,Ti)N, Zr,Hf-(Zr,Hf,Nb)N, and Zr,Nb-(Zr,Nb)N deposited on AISI 321 stainless steel substrates. The experiments revealed the following findings:

• The (Zr,Hf)N coating had the highest hardness, and the (Zr,Nb)N coating had the lowest hardness. The (Zr,Nb)N coating demonstrated the highest adhesion strength to the substrate (based on the total coating failure rate, L_C2), and the ZrN coating also demonstrated a high L_C2 value, whereas the other coatings demonstrated relatively low values.
• The Zr-ZrN, Zr,Nb-(Zr,Nb)N, Zr-(Zr,Hf)N, Zr,Hf-(Zr,Hf,Nb)N, and Zr-(Zr,Ti)N coatings provided a 10-fold reduction in the corrosion current density in a 3% aqueous NaCl solution at 25 °C, decreasing from 6.96 for the uncoated AISI 321 steel to 0.17 μA/cm² for the Zr-(Zr,Ti)N-coated sample.
• The addition of titanium to ZrN resulted in a significant decrease in corrosion current densities from 0.34 to 0.17 μA/cm², which may be due to the formation of poorly soluble titanium oxide TiO₂ on the surface. However, the addition of Nb increased the corrosion rate and reduced the polarisation resistance to corrosion. This may be due to the formation during corrosion of a non-stoichiometric composition of complex oxides NbO_x with increased solubility in water due to the acid–base interaction. Hf did not affect the corrosion-protective properties of the nitride coatings.
• Although the presence of a surface oxide layer was directly observed only in the Zr,Hf-(Zr,Hf)N-coated sample, it was likely present in the other samples. Further research is necessary to achieve a more pronounced oxide layer, allowing for more detailed investigation.
• Although the Zr,Hf-(Zr,Hf,Nb)N coating provided the best wear resistance and the lowest friction coefficient in the ball-on-disk test, consistent with previous work [56], this coating is not preferred for corrosion protection. In terms of corrosion resistance, the highest performance was provided by the Zr-(Zr,Ti)N, Zr-ZrN, and Zr-(Zr,Hf)N coatings. Thus, in terms of both high corrosion resistance and high wear resistance, the Zr-(Zr,Hf)N coating is recommended. In the absence of noticeable mechanical wear, the Zr-(Zr,Ti)N coating is the most appropriate for corrosion protection in NaCl solutions (e.g., seawater).