Next Article in Journal
Research on Low-Frequency Sound Absorption Based on the Combined Array of Hybrid Digital–Analog Shunt Loudspeakers
Previous Article in Journal
Development of a Coastal Erosion Monitoring Plan Using In Situ Measurements and Satellite Images
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Effect of Multilayer Cr/CrN Coatings Deposited by Magnetron Sputtering on the Thermal Stability and High-Temperature Steam Oxidation Resistance of Zirconium Alloy E110

by
Nazerke Muktanova
1,2,
Madina Kamet
3 and
Bauyrzhan Rakhadilov
1,*
1
PlasmaScience LLP, Ust-Kamenogorsk 070018, Kazakhstan
2
LLP Institute of Materials Science and Metallurgy, Ust-Kamenogorsk 070018, Kazakhstan
3
School of Sciences and Humanities, Nazarbayev University, Astana 010000, Kazakhstan
*
Author to whom correspondence should be addressed.
Appl. Sci. 2025, 15(23), 12773; https://doi.org/10.3390/app152312773
Submission received: 18 October 2025 / Revised: 15 November 2025 / Accepted: 25 November 2025 / Published: 2 December 2025
(This article belongs to the Section Materials Science and Engineering)

Abstract

The effect of the multilayer structure of Cr/CrN coatings deposited by reactive magnetron sputtering on zirconium alloy E110 (Zr–1Nb) on their thermal stability and resistance to steam oxidation at 1100 °C was studied. Coatings with different numbers of alternating Cr and CrN sublayers (1, 2, 4, and 6) were fabricated using experimental methods of X-ray diffraction (XRD) and scanning electron microscopy with energy-dispersive analysis (SEM/EDS). It was shown that an increase in the number of alternating Cr and CrN sublayers leads to the preservation of the cubic phase of CrN, the formation of a dense structure, and a decrease in Cr–Zr interdiffusion. After testing, multilayer coatings retaining the internal structure and a sufficiently structurally dense Cr2O3 layer effectively ensured air penetration. The best thermal stability was demonstrated by a six-layer coating, ensuring minimal oxidation and preservation of the E110 substrate.

1. Introduction

After major nuclear accidents, such as the Fukushima-1 nuclear power plant disaster, the development of accident-tolerant fuel (ATF) systems has become one of the priority areas for improving the safety of nuclear reactors in the event of a loss-of-coolant accident (LOCA). Traditional zirconium alloys used as fuel element cladding materials react intensively with water vapor at temperatures above 1000–1200 °C, leading to high-temperature steam oxidation (HTSO). As a result, an exothermic reaction of zirconium oxidation to zirconium dioxide (Zr → ZrO2) occurs with the release of hydrogen, which exacerbates the thermal regime of the core and contributes to the degradation of the cladding structure [1,2].
One of the most effective ways to increase the thermal and chemical resistance of cladding is to apply chromium-based coatings that, when heated, form a dense and adhesive-strong layer of Cr2O3 oxide, effectively preventing oxygen diffusion and slowing down the oxidation process of zirconium alloys in a steam environment up to temperatures of 1300 °C [3,4]. However, under prolonged and cyclic thermal exposure, homogeneous chromium coatings are subject to CrZr interdiffusion, the formation of brittle intermetallic phases (e.g., ZrCr2), as well as cracking and recrystallization, which reduces their protective properties [5,6].
A number of researchers note that during oxidation, chromium coatings form a multilayer structure consisting of successive layers of Cr2O3, residual Cr, ZrCr2 intermetallic compound, and ZrO2 oxide from the surface to the substrate [7,8]. However, the mechanisms of their formation are interpreted differently. According to Brachet et al. [9], a surface layer of Cr2O3 and an interphase layer of ZrCr2 are formed at the initial stage of oxidation, and as the test time increases, oxygen diffuses along the boundaries of Cr grains, interacts with zirconium, forming ZrO2 and causing the loss of the protective properties of the coating. Han et al. [10] showed that chromium is completely oxidized at the early stage of vapor oxidation, after which the reduction of Cr2O3 by the zirconium substrate leads to the formation of subsurface layers of ZrO2 and Cr, and further diffusion of Zr contributes to the formation of ZrCr2 at the interface.
The oxidation resistance of chromium coatings is determined by their structure, thickness, adhesion, and level of internal stresses [11,12,13,14]. In recent years, to increase the durability and stability of the protective layer, multilayer Cr/CrN systems consisting of alternating plastic metal and hard nitride sublayers have been actively researched. This architecture helps reduce internal stresses, slow down crack propagation, create a barrier to oxygen and chromium diffusion, and limit Cr–Zr interdiffusion during vapor oxidation. According to Sidelev et al. [15,16], multilayer Cr/CrN coatings obtained by magnetron sputtering on zirconium alloys (e.g., E110) are characterized by higher thermal and interphase stability at temperatures up to 1400 °C compared to single-layer chromium.
However, chromium nitride (CrN) has limited thermal stability: at temperatures of around 900–1100 °C, it decomposes according to the scheme CrN → Cr2N → Cr + N2, which can impair the barrier properties of the coating at HTSO in the case of suboptimal parameters of the multilayer structure—sublayer thickness, period, and texture [17]. Thus, it is the coating architecture—the number of layers, the multilayer period, and its total thickness—that plays a decisive role in ensuring structural stability and thermal stability. In addition, an important factor determining the stability of the system is the physical integrity of the layers—the absence of pores, cracks, and disruption of interlayer boundaries—which helps maintain the structural and thermal stability of the substrate under high-temperature conditions. It has been previously reported that multilayer Cr/CrN coatings effectively improve the mechanical and protective properties of various metal substrates. For example, Cai et al. [18] studied Cr/CrN films deposited by multi-arc ion plating on Ti-6Al-4V titanium alloys. It was shown that the introduction of soft Cr interlayers disrupts the columnar growth of nitride layers, promotes the formation of CrN(200) texture, and improves wear and erosion resistance due to crack dispersion at the Cr/CrN boundaries. However, this work only considered mechanical aspects at room temperatures and did not investigate the behavior of the coatings under high-temperature oxidation conditions.
In a later study using HiPIMS technology, Li Z. et al. [19], multilayer (CrN/Cr)ₙ coatings were produced on Zr-4 alloy, and it was found that increasing the number of layers increases the hardness (up to ~12 GPa) and resistance to steam oxidation at 1200 °C due to the formation of a fine-grained structure and a dense Cr2O3 oxide barrier. However, the study was limited to Zr-4 alloy and did not take into account the effect of Nb alloying, typical for operational claddings of E110 (Zr-1Nb) alloy, nor did it analyze the evolution of phases at lower temperatures typical for long-term operation in a reactor environment.
This work expands on these results and makes a new scientific contribution. For thefirst time, the influence of the architecture of multilayer Cr/CrN coatings with different numbers of layers (1, 2, 4, and 6) deposited by reactive magnetron sputtering on their structural evolution, phase stability, and thermal-oxidative behavior at 1100 °C was systematically studied for the E110 (Zr-1Nb) alloy.
In contrast to [19,20], the presented work is aimed at studying the architectural optimization and phase stability of multilayer Cr/CrN coatings specifically for the industrial alloy E110, which allows for a deeper understanding of the patterns of formation of barrier layers and ensures the practical application of such systems under conditions of high-temperature steam exposure.
In this regard, the aim of this work is to study the influence of the multilayer architecture of Cr/CrN coatings obtained by magnetron sputtering on the thermal stability and resistance to high-temperature vapor oxidation of zirconium alloy E110. The main focus is on investigating the influence of the number of layers on the kinetics of vapor oxidation, the morphology and evolution of the oxide film, as well as the role of multilayer architecture in suppressing interdiffusion and defects at the coating-substrate interface.

2. Materials and Methods

Plates made of zirconium alloy E110 (Zr–1% Nb), widely used as a material for nuclear reactor fuel element cladding, were used as substrates. Plates made of the E110 zirconium alloy (Zr–1Nb), compliant with GOST 25278.10–82 and TU 95.166–83, were used as substrates. The material was supplied by the A.A. Bochvar High-Technology Research Institute of Inorganic Materials (Moscow, Russia) in the as-received condition. The starting material was in the form of rods with a diameter of 50 mm. Plates measuring 10 × 10 × 2 mm were mechanically manufactured under laboratory conditions. The E110 alloy is characterized by high corrosion resistance and thermal stability due to the presence of a solid solution of niobium in an α-zirconium matrix. The main physical, mechanical, and chemical properties are density—6.5 g/cm3, elastic modulus—≈ 100 GPa, tensile strength—400–480 MPa, hardness—≈ 150 HV, melting point—approximately 1850 °C. The chemical composition of the alloy met the regulatory requirements (Zr—main, Nb—1.0 ± 0.1 wt.%).
The surface of the substrates was subjected to step-by-step mechanical grinding with abrasive paper with a gradual decrease in grain size (from P400 to P4000), after which it was polished to a mirror finish using an aluminum oxide suspension (Al2O3, 1 μm). Before coating, the samples were cleaned in an ultrasonic bath in acetone and ethanol (10 min in each solution) and dried in a stream of high-purity nitrogen. Multilayer Cr/CrN coatings were produced using reactive magnetron sputtering. Metallic chromium (99.9% purity, 75 mm diameter) served as the target. Deposition was performed at a constant power of 200 W in an argon/nitrogen mixture at a total pressure of ~0.2 Pa. The pressure was controlled using a vacuum gauge. The working gas flow ratio (Ar/N2) was 80/40 sccm, and the distance between the target and the substrate was 100 mm. The substrate temperature was maintained at 250 °C. Uncoated E110 alloy plates, as well as samples with a single-layer CrN coating and multilayer Cr/CrN coatings with different numbers of layers, were used as substrates. Before deposition, the substrate surface was ion-blasted with an Ar+ stream for 15 min to enhance coating adhesion. The deposition time for each layer was 1 h, and the total number of layers varied from 2 to 6. The deposition was carried out in an alternating regime: the chamber sequentially operated in two gas environments—in an atmosphere of pure argon (Ar) to form a metallic layer of Cr and in a mixture of Ar + N2 to deposit a nitride layer of CrN. High-temperature steam oxidation (HTSO) resistance tests were performed in a Zhengzhou CY Scientific CY-CVD1200-50-200×200-3TH-Q (Zhengzhou, China) tube furnace equipped with a programmable temperature control system. The samples were heated to 1100 °C at a rate of 10 °C/min. Before testing, a flow of high-purity argon (99.9%) was supplied to the working chamber, removing residual air and stabilizing the gas environment during the experiment. After reaching a temperature of approximately 400 °C, water vapor generated by an SG-20 S 5008 steam generator (PromService LLC, St. Petersburg, Russia) was supplied to the chamber at a minimum excess pressure necessary to maintain stable saturated steam formation. Water vapor was supplied using an SG-20 S 5008 steam generator at a rate of 10 sccm, while the partial pressure of water vapor was 5 kPa. During heating, the working chamber was purged with a flow of high-purity argon (99.9%) at a rate of 300 sccm, and during cooling—150 sccm, which ensured gas environment stability and prevented secondary oxidation of the samples. The isothermal holding time was 2 h, after which cooling was accomplished by moving the sample into the cold zone of the channel, where the temperature did not exceed 250 °C. The change in sample weight before and after testing was determined using a VK-300 analytical balance (accuracy ±0.005 g).
The surface morphology and microstructure of the cross-sections of the coatings were examined using a CIQTEK SEM3200 scanning electron microscope (Hefei, China), equipped with an XFlash Detector 730M-300 (Bruker, Billerica, MA, USA) energy-dispersive X-ray microanalysis (EDS) system. This made it possible to additionally determine the distribution of chemical elements in different areas of the coating and at the interface with the substrate. The phase composition of the coatings was analyzed by X-ray diffraction (XRD) on an X’Pert PRO (Philips, Amsterdam, The Netherlands) using Cu Kα radiation (λ = 1.5406 Å) at a voltage of 40 kV and a current of 30 mA. The measurements were performed in the angle range 2θ = 20–90°, with a step of 0.05° and an exposure time of 1 s.

3. Results

Figure 1 shows the cross-sections of Cr/CrN coatings deposited by magnetron sputtering on an E110 alloy substrate with different numbers of layers, before high-temperature steam oxidation (HTSO) testing. All coatings form a tight, continuous contact with the substrate without cracks, pores, or signs of delamination. The interlayer boundaries are smooth and clearly defined, and the light and dark bands correspond to alternating Cr and CrN sublayers. The single-layer coating (Figure 1a) exhibits a uniform, dense structure without pronounced layered alternation. The total coating thickness is approximately 1.15–1.20 μm. The interface between the coating and the substrate is clearly defined, and the structure is uniform across the thickness. The two-layer coating (Figure 1b) exhibits a distinct architecture consisting of a lower metallic Cr sublayer and an upper nitride CrN layer. The overall thickness is approximately 1.4 μm, with the top layer being approximately 0.9 μm thick and the Cr sublayer being approximately 0.4–0.5 μm. The coating adheres tightly to the substrate, and the layer structure is uniform, without pores or defects. In the four-layer coating (Figure 1c), alternating Cr/CrN layers of uniform thickness are clearly visible. The overall thickness reaches approximately 2.7 μm, with the thickness of individual layers varying within the range of 0.46–1.07 μm. The structure is dense and uniform, without through defects. The six-layer coating (Figure 1d) demonstrates the most developed multilayer architecture with regularly alternating Cr and CrN sublayers 0.4–1.0 μm thick. The overall coating thickness is approximately 4.18 μm. The microstructure is dense, the boundaries between the layers are clearly visible, and there are no pores.
Next, the phase composition of the coatings with different numbers of layers was studied before high-temperature steam oxidation (HTSO) testing. Figure 2 shows the X-ray diffraction (XRD) patterns of the original E110 (Zr-1Nb) substrate and the Cr/CrN coatings produced by magnetron sputtering with different numbers of layers.
All coatings exhibit diffraction peaks corresponding to the Cr (BCC), CrN (cubic, Fm-3m), and ε-Cr2N (hexagonal, P312) phases, as well as reflections of metallic zirconium (Zr) on the substrate side. With increasing layer count, an increase in the intensity of the CrN peaks and a decrease in the intensity of the Cr2N reflections are observed, indicating more complete nitridation processes and stabilization of the cubic CrN phase in the multilayer coatings. Four- and six-layer samples typically developed a more uniform nanocrystalline structure, as evidenced by a decrease in peak width and an increase in peak intensity.
The samples were then subjected to high-temperature steam oxidation (HTSO) testing. The change in sample weight before and after HTSO testing was determined by direct weighing. Before weighing, the samples were pre-dried at 120 °C for 1 h to remove residual moisture and volatile products. Each sample was weighed three times before and after HTSO testing, after which the average weight was calculated.
The results of the change in sample weight before and after high-temperature steam oxidation testing are shown in Table 1.
The obtained data show that the greatest weight change is observed for the uncoated E110 alloy, which is due to active oxidation of its surface. The application of CrN coatings and multilayer Cr/CrN systems significantly reduces the weight change (to +0.119–0.299 g), indicating improved oxide resistance. However, for the six-layer coating, a slight increase in Δm (+0.241 g) is observed, likely due to partial oxidation of the interlayer regions, which does not affect the integrity of the coating but does result in a slight weight gain.
Thermal exposure resulted in changes in their phase composition, as shown in the X-ray diffraction patterns (Figure 3).
Figure 3 shows the diffractograms of the initial sample made of the E110 alloy (Zr-1Nb) and Cr/CrN coatings with different multilayer architectures (1-, 2-, 4-, and 6-layer) after testing for high-temperature steam oxidation at 1100 °C.
The analysis shows a significant change in the phase state of the surface depending on the number of coating layers, which reflects differences in the thermal stability and protective mechanisms of multilayer structures.
The diffractogram of the initial sample shows that the sample is characterized by a typical phase structure represented by tetragonal (t-ZrO2) and monoclinic (m-ZrO2) modifications of zirconium oxide, which are formed as a result of surface oxidation of the alloy under the influence of atmospheric oxygen. The presence of two polymorphic modifications is explained by the partial thermal transformation of the tetragonal structure into a monoclinic one during cooling after thermomechanical treatment. The absence of metallic Zr peaks indicates complete oxidation of the surface layer and the formation of a continuous oxide film several micrometers thick. After high-temperature vapor oxidation, the single-layer CrN coating is completely oxidized to form the corundum phase Cr2O3, with no CrN reflections. The formation of the corundum phase alone confirms that high-temperature vapor exposure caused complete decomposition of chromium nitride according to the reaction:
2 C r N + 3 2 O 2 C r 2 O 3 + N 2
As a result, a structurally dense Cr2O3 oxide layer was formed, exhibiting high thermodynamic stability. In addition to chromium oxides, the X-ray diffraction pattern shows weak peaks corresponding to the tetragonal (t-ZrO2) and monoclinic (m-ZrO2) modifications of zirconium oxide. Their appearance is associated with partial oxidation of the E110 alloy substrate through microdefects in the coating and local areas of discontinuity in the oxide layer. The joint presence of Cr2O3, m-ZrO2, and t-ZrO2 indicates the formation of a composite oxide structure on the surface of the sample, where the outer layer is represented by a structurally dense Cr2O3 film, and near the substrate there are zones of zirconium oxides arising from limited oxygen diffusion through the coating. The results obtained indicate the complete decomposition of the nitride layer and the formation of a structurally dense but not completely sealed protective oxide coating that performs a protective function. However, the presence of weak ZrO2 peaks indicates the insufficient barrier efficiency of such a single-layer system.
In the case of a two-layer coating, thermal exposure results in the formation of a multiphase oxide structure consisting mainly of the Cr2O3 oxide phase, as well as traces of zirconium oxides (m-ZrO2 and t-ZrO2) formed due to limited oxygen diffusion to the substrate. The diffractogram shows intense peaks corresponding to Cr2O3 (corundum structure, space group R-3c), with characteristic reflections (104), (110), (113), (024), (116), (214), and (300) located in the regions 2θ ≈ 33.6°, 36.2°, 54.8°, 63.0°, 65.0°, and 69.1°. Among them, the most intense is the (110) reflection, which indicates the presence of a preferential orientation of Cr2O3 crystallites along this plane. The formation of the oxide structure is explained by the partial oxidation of the upper CrN nitride layer and the Cr metal sublayer, occurring according to the reactions:
2 C r + 3 2 O 2 C r 2 O 3   a n d   2 C r N + 3 2 O 2 C r 2 O 3 + N 2
The preferred orientation of Cr2O3 crystallites along the (110) plane is due to the directed growth of oxide grains during coating cooling after spraying. The formation of the Cr2O3 phase occurs from metastable CrOx oxide compounds under thermal influence, and the (110) orientation is the most thermodynamically stable for a corundum-like structure. The proposed chemical transformations (1) and (2), describing the sequential oxidation of chromium nitride, are consistent with previously published data on the behavior of Cr and CrN coatings exposed to high temperatures in a vapor or oxidizing environment. According to literature sources [20,21,22,23,24], at temperatures above 900–1000 °C, the phase evolution of CrN can proceed according to the scheme: CrN → Cr2N → Cr2O3 + N2, which is caused by the gradual release of nitrogen and the oxidation of chromium with the formation of stable oxide Cr2O3. These results confirm the thermodynamic validity of the proposed reactions and their correspondence to experimentally observed processes in Cr/CrN systems during high-temperature steam oxidation.
The presence of weak m-ZrO2 and t-ZrO2 peaks indicates limited oxidation of the substrate, which is significantly less than in the case of a single-layer coating. The formation of a stable Cr2O3 phase indicates high thermodynamic stability of the coating under HTSO conditions and ensures the formation of a structurally dense passive film, which plays a key role in protecting the E110 substrate from intense oxidation. The four-layer Cr/CrN coating is characterized by the formation of a predominantly Cr2O3 oxide phase with a corundum structure, accompanied by weak reflections of Cr2N and Cr. The presence of these phases indicates partial oxidation of the upper layers while maintaining the internal multilayer architecture. This indicates that oxygen diffusion into the coating was limited and that the oxidation process was completed mainly in the upper part of the multilayer structure. The appearance of diffraction reflections of the ε-Cr2N phase in a 4-layer coating is associated with incomplete nitridation of individual sublayers during sequential deposition. Weak m-ZrO2 and t-ZrO2 peaks confirm limited substrate oxidation. The presence of Cr(110) (~44.4°) and Cr(211) (~74.2°) confirms that some areas of the Cr metal layer remained largely unchanged. This indicates high stability of the interlayer distribution and the effectiveness of alternating nitride and metal layers in preventing through-oxidation.
The six-layer coatings demonstrate that the multilayer coating architecture provides the most stable phase state and the highest thermal stability among the samples studied. The diffractogram is dominated by intense peaks corresponding to the Cr2O3 phase (corundum structure, space group R-3c), indicating the predominance of a well-ordered oxide structure with preferential crystallite orientation. The presence of multiple high-intensity Cr2O3 reflections indicates the formation of a dense and continuous protective oxide film that prevents further oxygen penetration. In addition to the dominant oxide, weak reflections of CrN (111) (~37.5°) and (220) (~63.4°) remain, indicating partial preservation of the original nitride phase in the inner layers of the coating. The retention of the CrN phase after thermal treatment indicates limited oxygen diffusion into the structure and confirms the effective protective role of the external oxide barrier. The Cr2N phase is retained only in small quantities, primarily in the transition zones between individual layers, and does not significantly affect the overall phase composition of the coating. A decrease in the intensity of its diffraction peaks indicates increased uniformity of nitrogen distribution and more complete formation of the cubic CrN phase (Fm-3m). Weak Cr (200) (~44.4°) peaks are also observed, indicating the preservation of individual metal layers that have not undergone complete oxidation. The absence of ZrO2 reflections indicates effective insulation of the substrate from the oxidizing environment. Thus, the six-layer coating demonstrates the most effective phase state, in which the upper part of the structure is represented by a stable Cr2O3 phase, and the lower layers retain a combination of CrN and Cr, ensuring mechanical integrity and thermal protection.
Figure 4a–e shows the morphological features of the surface of the initial E110 alloy sample, as well as single-layer and multilayer Cr/CrN coatings after testing for resistance to high-temperature steam oxidation at 1100 °C for 2 h.
The surface of the initial zirconium alloy (Figure 4a) is characterized by pronounced roughness and the presence of numerous depressions and pores of various shapes and sizes. The dark areas correspond to zones of intense oxidation and local destruction of the oxide layer, where corrosion products have peeled off or flaked off. The light areas, on the contrary, represent areas of a denser and more stable oxide layer, presumably consisting of ZrO2. The observed network of microcracks was formed as a result of thermal stresses arising from differences in the coefficients of thermal expansion between the substrate and the forming oxide film. Taken together, this indicates low thermal stability of the material without a protective coating and intense oxidation processes when exposed to water vapor at 1100 °C. The single-layer CrN coating partially protects the substrate, as evidenced by the zone of partial coating delamination from the Zr-1Nb alloy observed in Figure 4b. The left side of the image corresponds to a Zr-rich substrate, while the right side corresponds to a coating characterized by elevated Cr and O contents. The difference in microstructure on either side of the crack indicates the boundary between the coating and the substrate and confirms partial coating delamination with localized oxidation of the surface layer to Cr2O3. The transition zone is clearly defined, indicating a reliable interface between the coating and the base material. Further, on the surface of the two-layer coating (Figure 4c), a relatively smooth morphology is observed with a single delamination line at the boundary between the layers. The number of cracks is significantly lower than in the single-layer coating. This indicates that the introduction of a Cr metal sublayer improves thermoelastic compatibility and adhesion, reducing the likelihood of cracking during thermal cycling. However, individual defects (a single diagonally oriented thermal crack) indicate that the two-layer structure only partially provides long-term protection. The surface of the four-layer coating (Figure 4d) exhibits a morphology that differs from the other samples.
To clarify their nature, additional EDS analysis was conducted. Figure 5 shows the surface mapping results of the four-layer coating after HTSO.
The observed rounded objects represent swellings of the surface oxide layer of Cr2O3, partially exposed in places, and therefore appear as dark pores/craters in the SEM images. EDS mapping shows elevated O content along the rims of these objects with low N content, which is consistent with the model of localized gas accumulation (O2, N2, H2) under the oxide film during HTSO and its subsequent swelling and rupture.
Cracks are observed between the bubble-like areas, caused by thermal stresses and local expansion of the oxide phase. At the same time, the coating retains its overall integrity: no continuous destruction or delamination is observed. Finally, the surface of the six-layer coating (Figure 4e) is characterized by the densest and most uniform structure. No microcracks or pores were found. The coating has a fine-grained, uniform morphology, which indicates the formation of a stable and continuous oxide film, consisting mainly of Cr2O3. This structure ensures the integrity of the coating and indicates its high thermal stability and strong adhesion to the substrate after high-temperature steam oxidation (HTSO) testing.
To establish the relationship between surface morphology and the structural integrity of the coatings, their cross-sections were examined. Figure 6 shows the microstructures of the cross-sections of Cr/CrN coatings after high-temperature steam oxidation (HTSO) testing at 1100 °C.
The images show: (a) single-layer CrN coating; (b) 2-layer Cr/CrN; (c) 4-layer Cr/CrN; (d) 6-layer Cr/CrN coating. The measured total coating thicknesses are approximately 1.6 ± 0.23 μm, 2.85 ± 0.32 μm, 5.8 ± 0.36 μm, and 9.62 ± 0.23 μm, respectively. After exposure to high-temperature water vapor, the overall integrity of the coatings is maintained, and a clear distinction is observed between the coating and the E110 alloy substrate. However, in multilayer systems, with an increase in the number of layers, partial blurring of the interlayer boundaries is observed due to diffusion and oxidation, which indicates thermal interaction between the Cr and CrN sublayers.
Figure 7 further presents the mapping results (a, b, c, d, e) performed on the coating cross-sections, as well as the element distribution along the selected analysis line (a.1, b.1, c.1, d.1, e.1). The resulting EDS maps allow us to visualize the spatial distribution of the main elements (Cr, N, Zr, etc.) and evaluate the degree of their interdiffusion between the coating layers and the substrate after high-temperature steam oxidation testing.
The presented micrographs and elemental distribution maps obtained by EDS clearly show the boundary between the E110 alloy substrate and the deposited coatings. On the uncoated sample (Figure 7a), significant oxygen enrichment is observed near the surface of the E110 alloy substrate after testing, indicating the formation of a continuous ZrO2 oxide film. The elemental distribution (Figure 7(a.1)) shows a high Zr signal across the thickness, while the oxygen gradually increases from the inner layers to the surface. For the single-layer CrN coating (Figure 7b) after HTSO, the formation of a surface oxide layer consisting predominantly of Cr2O3 is observed. The elemental distribution profile (Figure 7(b.1)) shows an increase in the O concentration and a decrease in N in the upper zone of the coating, which corresponds to partial oxidation of chromium nitride. The coating-substrate interface is clearly defined, and Zr diffusion into the coating is absent. However, oxygen enrichment is observed in the near-surface zone of the substrate, indicating the formation of a ZrO2 oxide layer. The two-layer Cr/CrN coating (Figure 7c) exhibits a uniform distribution of Cr and N across the coating thickness, with a distinct oxygen-enriched zone near the surface. The linear profile (Figure 7(c.1)) confirms the presence of a diffusion barrier at the Cr–CrN interface, which limits the penetration of O and Zr. The structure retains a layered morphology, which contributes to increased thermal stability. The four-layer Cr/CrN architecture (Figure 7d) exhibits a more uniform distribution of elements and a decrease in the O signal intensity compared to the two-layer version. The profile (Figure 7(d.1)) shows that Cr and N are uniformly distributed, and Zr does not exhibit noticeable diffusion from the substrate. The presence of additional interfaces helps inhibit oxygen interpenetration, forming a stable protective structure. The six-layer Cr/CrN coating (Figure 7e) maintains clear layer separation and high coating density after HTSO. The oxygen zone is limited to the upper surface layer, indicating the formation of a thin passive oxide film of Cr2O3. According to the linear profile (Figure 7(e.1)), Cr and N are distributed stably, while Zr maintains a sharp transition at the substrate-coating interface. This structure ensures superior thermal stability and minimal diffusion of elements between the substrate and coating. Thus, the results of EDS mapping and linear analysis confirm the high thermal stability of multilayer Cr/CrN coatings and the effectiveness of their barrier properties when exposed to saturated water vapor at 1100 °C.
Comparison of the structural and phase characteristics showed that multilayer Cr/CrN coatings retain high thermal stability after exposure to water vapor at 1100 °C. Chromium nitride oxidizes to form a stable oxide Cr2O3 according to the reaction: 2CrN + 1.5O2 → Cr2O3 + N2 ↑. The Cr2O3 phase has a lower density (ρ ≈ 5.2 g/cm3) compared to CrN (ρ ≈ 6.0 g/cm3), which causes volume expansion (~10–15%) and a visual thickening of the upper zone [25]. The actual thickness of the metallic nitride layer remains virtually unchanged. The resulting Cr2O3 layer functions as a diffusion barrier, preventing the penetration of oxygen and zirconium. In multilayer systems, residual Cr and CrN phases are retained, whereas in single- and two-layer coatings, complete conversion to Cr2O3 is observed. To investigate the resistance of various coatings to high-temperature steam oxidation, the evolution of single- and multilayer CrN and (CrN/Cr)ₙ systems after HTSO at 1100 °C for 2 h was assessed. The interpretation of the results was compared with the oxidation model schematically presented in Figure 8.
When exposed to saturated water vapor on an uncoated Zr-1Nb substrate, intense oxidation of the zirconium occurs, forming a thick, porous, and cracked ZrO2 layer. Deep oxygen saturation of the base metal is observed, accompanied by a significant increase in mass and pronounced surface degradation. In the case of a single-layer CrN coating, the protective capacity is insufficient. The thin CrN layer completely or almost completely transforms into loose Cr2O3, in which defects and diffusion channels form. Through these defects, oxygen and water vapor freely reach the substrate, causing intensive growth of ZrO2 beneath the Cr2O3. The barrier function of the coating is lost in the early stages of testing. In the two-layer (Cr/CrN)2 system, the top layer of the coating also oxidizes to Cr2O3, but the combined thickness and number of interfaces remain insufficient to effectively suppress oxygen diffusion. The inner layers are partially preserved, but localized formation of ZrO2 at the interface with the substrate indicates a breakdown of the protective barrier. This structure provides only moderate resistance to HTSO. In the four-layer (Cr/CrN)4 coating, a dense, continuous Cr2O3 layer forms, significantly reducing the penetration of oxygen and water vapor. The inner Cr/CrN sublayers are partially oxidized but retain a pronounced multilayer morphology and integrity, providing significantly higher resistance compared to single- and two-layer versions. Finally, in the six-layer (Cr/CrN)6 structure, the maximum number of Cr/CrN interfaces forms the most effective diffusion barrier. After HTSO, a clearly defined multilayer structure is maintained, the upper Cr2O3 layer remains dense and stable, and the formation of ZrO2 is virtually absent. This coating demonstrates the highest resistance and provides almost complete protection of the substrate from vapor oxidation.

4. Conclusions

Cr/CrN coatings deposited by reactive magnetron sputtering on E110 zirconium alloy (Zr–1Nb) demonstrated high thermal stability and resistance to steam oxidation at 1100 °C. Studies have shown that increasing the number of alternating Cr and CrN sublayers helps stabilize the cubic CrN phase and suppresses the formation of ε-Cr2N, resulting in the formation of a fine-grained multilayer structure with distinct interlayer boundaries.
High-temperature steam oxidation tests confirmed that all coatings form a protective Cr2O3 oxide layer. However, it is the multilayer systems that preserve the internal architecture and effectively limit the interdiffusion of Cr and Zr. The best results were obtained for a six-layer coating, characterized by minimal weight gain, the absence of ZrO2 phases, and the preservation of clear boundaries between the sublayers. Microstructure and phase composition analysis revealed that during oxidation, the outer layers of the coating partially transform into Cr2O3 oxide, while the inner sublayers retain Cr and CrN phases, acting as an effective diffusion barrier. The observed formation of a thin α-Zr(N) sublayer at the substrate interface further prevents oxygen penetration and stabilizes the coating structure.
Thus, multilayer Cr/CrN coatings provide reliable protection of E110 zirconium alloy from high-temperature steam oxidation due to the combined effects of interlayer architecture, a dense oxide barrier, and limited diffusion of elements.

Author Contributions

B.R. and N.M. formal analysis; supervision; writing—review and editing; M.K. and N.M. investigation; methodology; B.R. and N.M. resources; data curation, writing—original draft preparation. All authors have read and agreed to the published version of the manuscript.

Funding

This study was funded by the Committee of Science of the Ministry of Science and Higher Education of the Republic of Kazakhstan (grant No. AP19680681).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding author.

Conflicts of Interest

Authors Nazerke Muktanova and Bauyrzhan Rakhadilov were employed by the company PlasmaScience LLP. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Terrani, K.A. Accident Tolerant Fuel Cladding Development: Promise, Status, and Challenges. J. Nucl. Mater. 2018, 501, 13–30. [Google Scholar] [CrossRef]
  2. Ma, H.-B.; Yan, J.; Zhao, Y.-H.; Liu, T.; Ren, Q.-S.; Liao, Y.-H.; Zuo, J.-D.; Liu, G.; Yao, M.-Y. Oxidation Behavior of Cr-Coated Zirconium Alloy Cladding in High-Temperature Steam above 1200 °C. npj Mater. Degrad. 2021, 5, 7. [Google Scholar] [CrossRef]
  3. Deng, J.; Geng, D.; Sun, Q.; Song, Z.; Sun, J. Steam Oxidation of Cr-Coated Zirconium Alloy Claddings at 1200 °C: Kinetics Transition and Failure Mechanism of Cr Coatings. J. Nucl. Mater. 2023, 586, 154684. [Google Scholar] [CrossRef]
  4. Wang, D.; Zhong, R.; Zhang, Y.; Chen, P.; Yu, J.; Lan, Y.; Su, G.H.; Qiu, S.; Tian, W.; Liao, Y.; et al. Oxidation Behavior and Kinetics of Magnetron-Sputtered Cr-Coated Zr Alloy Cladding in 1000–1300 °C Steam Environment. Corros. Sci. 2023, 218, 111215. [Google Scholar] [CrossRef]
  5. Yeom, H.; Johnson, G.; Maier, B.; Dabney, T.; Sridharan, K. High Temperature Oxidation of Cold Spray Cr-Coated Accident Tolerant Zirconium-Alloy Cladding with Nb Diffusion Barrier Layer. J. Nucl. Mater. 2024, 588, 154822. [Google Scholar] [CrossRef]
  6. Interdiffusion Behavior Between Cr and Zr and Its Effect on the Microcracking Behavior in the Cr-Coated Zr-4 Alloy|Nuclear Science and Techniques. Available online: https://link.springer.com/article/10.1007/s41365-021-00977-w (accessed on 17 October 2025).
  7. Liu, J.; Tang, C.; Steinbrück, M.; Yang, J.; Stegmaier, U.; Große, M.; Yun, D.; Seifert, H.J. Transient Experiments on Oxidation and Degradation of Cr-Coated Zircaloy in Steam up to 1600 °C. Corros. Sci. 2021, 192, 109805. [Google Scholar] [CrossRef]
  8. Chen, H.; Wang, X.; Zhang, R.; Chen, H.; Wang, X.; Zhang, R. Application and Development Progress of Cr-Based Surface Coating in Nuclear Fuel Elements: II. Current Status and Shortcomings of Performance Studies. Coatings 2020, 10, 835. [Google Scholar] [CrossRef]
  9. Brachet, J.-C.; Rouesne, E.; Ribis, J.; Guilbert, T.; Urvoy, S.; Nony, G.; Toffolon-Masclet, C.; Le Saux, M.; Chaabane, N.; Palancher, H.; et al. High Temperature Steam Oxidation of Chromium-Coated Zirconium-Based Alloys: Kinetics and Process. Corros. Sci. 2020, 167, 108537. [Google Scholar] [CrossRef]
  10. Han, X.; Chen, C.; Tan, Y.; Feng, W.; Peng, S.; Zhang, H. A Systematic Study of the Oxidation Behavior of Cr Coatings on Zry4 Substrates in High Temperature Steam Environment. Corros. Sci. 2020, 174, 108826. [Google Scholar] [CrossRef]
  11. Rakhadilov, B.; Magazov, N.; Kakimzhanov, D.; Apsezhanova, A.; Molbossynov, Y.; Kengesbekov, A.; Rakhadilov, B.; Magazov, N.; Kakimzhanov, D.; Apsezhanova, A.; et al. Influence of Spraying Process Parameters on the Characteristics of Steel Coatings Produced by Arc Spraying Method. Coatings 2024, 14, 1145. [Google Scholar] [CrossRef]
  12. Zhassulan, A.; Rakhadilov, B.; Baizhan, D.; Kengesbekov, A.; Kakimzhanov, D.; Musataeva, N.; Zhassulan, A.; Rakhadilov, B.; Baizhan, D.; Kengesbekov, A.; et al. Influence of TiO2 Nanoparticle Concentration on Micro-Arc Oxidized Calcium–Phosphate Coatings: Corrosion Resistance and Biological Response. Coatings 2025, 15, 1142. [Google Scholar] [CrossRef]
  13. Rakhadilov, B.; Kakimzhanov, D.; Mural, A.; Turabekov, Y.; Rakhadilov, B.; Kakimzhanov, D.; Mural, A.; Turabekov, Y. Effect of Technological Parameters of Detonation Spraying on the Properties of Cr2O3 Coatings. Coatings 2025, 15, 943. [Google Scholar] [CrossRef]
  14. Sagdoldina, Z.; Rakhadilov, B.; Maulet, M.; Sulyubayeva, L.; Drenda, C.; Bolatov, S.; Sagdoldina, Z.; Rakhadilov, B.; Maulet, M.; Sulyubayeva, L.; et al. Comprehensive Study of the Mechanical and Tribological Properties of NiCr-Al Detonation Coatings. Appl. Sci. 2025, 15, 7513. [Google Scholar] [CrossRef]
  15. Sidelev, D.V.; Ruchkin, S.E.; Syrtanov, M.S.; Kashkarov, E.B.; Shelepov, I.A.; Malgin, A.G.; Polunin, K.K.; Stoykov, K.V.; Mokrushin, A.A. Protective Cr Coatings with CrN/Cr Multilayers for Zirconium Fuel Claddings. Surf. Coat. Technol. 2022, 433, 128131. [Google Scholar] [CrossRef]
  16. Sidelev, D.V.; Syrtanov, M.S.; Ruchkin, S.E.; Pirozhkov, A.V.; Kashkarov, E.B.; Sidelev, D.V.; Syrtanov, M.S.; Ruchkin, S.E.; Pirozhkov, A.V.; Kashkarov, E.B. Protection of Zr Alloy under High-Temperature Air Oxidation: A Multilayer Coating Approach. Coatings 2021, 11, 227. [Google Scholar] [CrossRef]
  17. Liu, Z.R.; Du, J.W.; Chen, L. Influence of Oxygen Content on Structure, Thermal Stability, Oxidation Resistance, and Corrosion Resistance of Arc Evaporated (Cr, Al)N Coatings. Surf. Coat. Technol. 2022, 432, 128057. [Google Scholar] [CrossRef]
  18. Cai, X.; Gao, Y.; Cai, F.; Zhang, L.; Zhang, S. Effects of Multi-Layer Structure on Microstructure, Wear and Erosion Performance of the Cr/CrN Films on Ti Alloy Substrate. Appl. Surf. Sci. 2019, 483, 661–669. [Google Scholar] [CrossRef]
  19. Li, Z.; Liu, C.; Chen, Q.; Yang, J.; Liu, J.; Yang, H.; Zhang, W.; Zhang, R.; He, L.; Long, J.; et al. Microstructure, High-Temperature Corrosion and Steam Oxidation Properties of Cr/CrN Multilayer Coatings Prepared by Magnetron Sputtering. Corros. Sci. 2021, 191, 109755. [Google Scholar] [CrossRef]
  20. Mammeri, F.Z.; Ayad, A.; Riah, B.; Chekour, L.; Djouadi, M.A.; Rouag, N. Morphology and Crystallography of Cr, CrN and CrAlN Films—Effect of Annealing Temperature. Surf. Coat. Technol. 2019, 357, 823–832. [Google Scholar] [CrossRef]
  21. Milošev, I.; Abels, J.M.; Strehblow, H.H.; Navinšek, B.; Metikoš-Huković, M. High-Temperature Oxidation of Thin CrN Coatings Deposited on Steel. J. Vac. Sci. Technol. A Vac. Surf. Film. 1996, 14, 2527–2534. [Google Scholar] [CrossRef]
  22. Chen, H.Y.; Lu, F.H. Oxidation Behavior of Chromium Nitride Films. Thin Solid Film. 2006, 515, 2179–2184. [Google Scholar] [CrossRef]
  23. Qi, Z.B.; Liu, B.; Wu, Z.T.; Zhu, F.P.; Wang, Z.C.; Wu, C.H. A Comparative Study of the Oxidation Behavior of Cr2N and CrN Coatings. Thin Solid Film. 2013, 544, 515–520. [Google Scholar] [CrossRef]
  24. Meng, C.; Jiang, J.; Ma, J.; Zhang, T.; Liu, W.; Xie, P.; He, X. Improving the High-Temperature Oxidation Resistance of CrN Coating by Gradient Deposition Considering Internal Stress Effects. Nucl. Mater. Energy 2023, 37, 101531. [Google Scholar] [CrossRef]
  25. Rumble, J. (Ed.) CRC Handbook of Chemistry and Physics, 102nd ed.; CRC Press: Boca Raton, FL, USA, 2021. [Google Scholar]
Figure 1. Cross-sections of Cr/CrN coatings produced by magnetron sputtering with different numbers of layers: (a) single-layer CrN; (b) two-layer Cr/CrN; (c) four-layer Cr/CrN; (d) six-layer Cr/CrN.
Figure 1. Cross-sections of Cr/CrN coatings produced by magnetron sputtering with different numbers of layers: (a) single-layer CrN; (b) two-layer Cr/CrN; (c) four-layer Cr/CrN; (d) six-layer Cr/CrN.
Applsci 15 12773 g001aApplsci 15 12773 g001b
Figure 2. X-ray diffraction patterns of the (a) E110 substrate and Cr/CrN coatings deposited by magnetron sputtering with different numbers of layers: (b) single-layer, (c) two-layer, (d) four-layer, and (e) six-layer.
Figure 2. X-ray diffraction patterns of the (a) E110 substrate and Cr/CrN coatings deposited by magnetron sputtering with different numbers of layers: (b) single-layer, (c) two-layer, (d) four-layer, and (e) six-layer.
Applsci 15 12773 g002
Figure 3. X-ray diffraction patterns of the initial E110 sample and Cr/CrN coatings with different multilayer architectures (1-, 2-, 4-, and 6-layer) after high-temperature steam oxidation testing at 1100 °C.
Figure 3. X-ray diffraction patterns of the initial E110 sample and Cr/CrN coatings with different multilayer architectures (1-, 2-, 4-, and 6-layer) after high-temperature steam oxidation testing at 1100 °C.
Applsci 15 12773 g003
Figure 4. Surface morphology of the sample without coating (a) and with Cr/CrN coatings of different multilayer architectures—single-layer (b), two-layer (c), four-layer (d), and six-layer (e)—after testing for high-temperature steam oxidation at 1100 °C for 2 h.
Figure 4. Surface morphology of the sample without coating (a) and with Cr/CrN coatings of different multilayer architectures—single-layer (b), two-layer (c), four-layer (d), and six-layer (e)—after testing for high-temperature steam oxidation at 1100 °C for 2 h.
Applsci 15 12773 g004aApplsci 15 12773 g004b
Figure 5. SEM image and EDS maps of element distribution (Cr, O, N) on the surface of a four-layer Cr/CrN coating after high-temperature steam oxidation.
Figure 5. SEM image and EDS maps of element distribution (Cr, O, N) on the surface of a four-layer Cr/CrN coating after high-temperature steam oxidation.
Applsci 15 12773 g005aApplsci 15 12773 g005b
Figure 6. Microstructures of cross-sections of Cr/CrN coatings after high-temperature steam oxidation (HTSO): (a) single-layer, (b) two-layer, (c) four-layer and (d) six-layer coating.
Figure 6. Microstructures of cross-sections of Cr/CrN coatings after high-temperature steam oxidation (HTSO): (a) single-layer, (b) two-layer, (c) four-layer and (d) six-layer coating.
Applsci 15 12773 g006aApplsci 15 12773 g006b
Figure 7. Cross-sectional mapping of the (a) uncoated and (be) multilayer Cr/CrN coating samples after high-temperature steam oxidation and the (b.1,c.1,d.1,e.1) linear EDS profile across the coating thickness and (a.1) without coating.
Figure 7. Cross-sectional mapping of the (a) uncoated and (be) multilayer Cr/CrN coating samples after high-temperature steam oxidation and the (b.1,c.1,d.1,e.1) linear EDS profile across the coating thickness and (a.1) without coating.
Applsci 15 12773 g007aApplsci 15 12773 g007bApplsci 15 12773 g007cApplsci 15 12773 g007dApplsci 15 12773 g007eApplsci 15 12773 g007fApplsci 15 12773 g007g
Figure 8. Schematic diagrams illustrating the behavior of single-layer (a), bilayer (b), four-layer (c), and six-layer (d) Cr/CrN coatings during high-temperature steam oxidation.
Figure 8. Schematic diagrams illustrating the behavior of single-layer (a), bilayer (b), four-layer (c), and six-layer (d) Cr/CrN coatings during high-temperature steam oxidation.
Applsci 15 12773 g008
Table 1. Change in sample mass before and after high-temperature steam oxidation (HTSO) testing.
Table 1. Change in sample mass before and after high-temperature steam oxidation (HTSO) testing.
Sample No.Coating TypeMass Before Testing, gMass After Testing, gChange in Mass Δm, g
1E110 (uncoated)10.87511.255+0.380
2CrN (single layer)10.43210.731+0.299
3Cr/CrN (2-layer)9.2239.342+0.119
4Cr/CrN (4-layer)10.82311.017+0.194
5Cr/CrN (6 layers)11.90112,142+0.241
Note: relative mass increase was calculated using the formula: m % = m a f t e r m b e f o r e m b e f o r e × 100 .
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Muktanova, N.; Kamet, M.; Rakhadilov, B. Effect of Multilayer Cr/CrN Coatings Deposited by Magnetron Sputtering on the Thermal Stability and High-Temperature Steam Oxidation Resistance of Zirconium Alloy E110. Appl. Sci. 2025, 15, 12773. https://doi.org/10.3390/app152312773

AMA Style

Muktanova N, Kamet M, Rakhadilov B. Effect of Multilayer Cr/CrN Coatings Deposited by Magnetron Sputtering on the Thermal Stability and High-Temperature Steam Oxidation Resistance of Zirconium Alloy E110. Applied Sciences. 2025; 15(23):12773. https://doi.org/10.3390/app152312773

Chicago/Turabian Style

Muktanova, Nazerke, Madina Kamet, and Bauyrzhan Rakhadilov. 2025. "Effect of Multilayer Cr/CrN Coatings Deposited by Magnetron Sputtering on the Thermal Stability and High-Temperature Steam Oxidation Resistance of Zirconium Alloy E110" Applied Sciences 15, no. 23: 12773. https://doi.org/10.3390/app152312773

APA Style

Muktanova, N., Kamet, M., & Rakhadilov, B. (2025). Effect of Multilayer Cr/CrN Coatings Deposited by Magnetron Sputtering on the Thermal Stability and High-Temperature Steam Oxidation Resistance of Zirconium Alloy E110. Applied Sciences, 15(23), 12773. https://doi.org/10.3390/app152312773

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop