Surface Friction and Interfacial Wear Mechanisms in CeO₂-Ni/WC Cladding Layers on 45 Steel

Abstract

This study investigates the insufficient surface hardness of medium-carbon 45 steel and the drawbacks associated with conventional surface modification techniques (e.g., cracking in laser cladding, weak bonding in thermal spraying, restricted thickness in chemical deposition). Series CeO₂-Ni/WC composite claddings (with CeO₂ content ranging from 0.5 to 2.0 wt %) were fabricated via vacuum cladding. The cladding with 0.5 wt % CeO₂ (NWC5) exhibited the lowest porosity (0.0673%) and finest grain size (12.06 nm) and demonstrated the highest microhardness (1042.74 HV0.2) and elastic modulus (269.06 GPa), respectively. The interfacial friction coefficient (0.343–0.444) was significantly reduced compared to the surface friction coefficient (0.562–0.617). Wear track analysis revealed that the width in the cladding layer-to-substrate transition zone (CTSZ) was 22.1–43.2% wider than that in the substrate-to-cladding layer transition zone (STCZ). This disparity is attributed to stress concentration induced by the abrupt hardness change across the CTSZ, promoting the formation of a three-tiered step structure (with a step height difference of 2.1–4.1 µm). In contrast, the progressive hardness series in the STCZ facilitated a smoother wear surface. The dominant wear mechanism was identified as a combination of abrasive and oxidative wear. This study provides a theoretical foundation for optimizing such high-reliability components.

1. Introduction

AISI 1045 steel, a medium-carbon structural steel, is widely utilized in mechanical manufacturing due to its balanced mechanical properties and cost-effectiveness. This material is commonly employed in critical components under high-load conditions, such as automotive transmission gears, machine tool spindles, and connecting shafts in heavy machinery. However, its insufficient surface hardness severely limits its application in complex environments. In high-friction scenarios, 45 steel is prone to surface abrasion, leading to increased failure rates and shortened maintenance intervals. Consequently, surface modification technologies to enhance surface hardness and wear resistance have emerged as effective strategies for optimizing its performance, which can significantly extend component service life and reduce lifecycle costs.

Metal matrix composite coatings/claddings have been widely implemented through laser cladding [1,2,3,4,5,6], thermal spraying [7,8,9,10], and chemical deposition [11,12,13] to enhance substrate hardness and wear resistance. While these surface engineering techniques demonstrate significant mechanical enhancements, inherent limitations persist: Laser cladding induces substrate distortion and residual stress due to instantaneous thermal input (peak temperature > 2000 °C), increasing interfacial microcrack initiation probability. Thermal-sprayed coatings exhibit weak interfacial bonding strength owing to mechanical adhesion, leading to delamination risks. Chemical deposition achieves nanocrystalline structures but suffers from low deposition rates (<5 μm/h) and thickness limitations (<100 μm), failing under heavy-load conditions. In contrast, vacuum cladding technology creates oxide-free interfaces (<10⁻³ Pa vacuum) with optimized wettability, achieving metallurgical bonding.

The CeO₂-Ni/WC composite cladding layer effectively addresses the limitations of conventional materials [14,15,16,17]. As a hard phase material, tungsten carbide (WC) demonstrates superior characteristics including a high melting point (2870 °C), remarkable hardness (22–24 GPa), and excellent wettability with nickel-based alloys, providing a stable structural foundation for the composite layer [18,19,20,21]. Research indicates that rare earth oxides (e.g., Y₂O₃ [22,23,24,25], La₂O₃ [26,27]) as reinforcement phases can significantly enhance the microstructure and mechanical properties of nickel-based composites. Notably, CeO₂ exhibits prominent advantages in improving wear resistance of metal matrix composites due to its unique lattice matching characteristics.

Current research on CeO₂ reinforced Ni-based composite claddings predominantly focuses on surface tribological characterization, while leaving a critical knowledge gap regarding multiscale friction mechanisms in the substrate-cladding layer interface. Additionally, interfacial friction testing can also serve as a method for assessing the interfacial bonding strength and debonding resistance. To address this, the present study fabricates series-structured CeO₂-Ni/WC cladding layers (CeO₂ content: 0.5–2.0 wt %) on 45 steel substrates via vacuum cladding technology, systematically investigating the synergistic influence mechanisms of CeO₂ doping on interfacial friction performance. Through the innovative establishment of a cross-scale evaluation framework for surface-interfacial tribological performance, this study has revealed the stepwise effect in the interface transition zone, providing an important theoretical foundation for the interface optimization design of high-reliability engineering components.

2. Materials and Methods

2.1. Materials Preparation

In this study, all AISI 1045 steel substrates were subjected to grinding and polishing to achieve a surface roughness of Ra = 0.5 μm, ensuring consistency in experimental conditions. The microstructures of the Ni-based alloy and WC are shown in Figure 1a,b, respectively. The main chemical composition of the nickel-based alloy can be found in our previous work [28]. The chemical composition (wt %) of the Ni-based alloy is as follows: C 0.7–1.1, B 3.0–4.0, Si 3.5–5.0, Cr 15.0–17.0, and Fe ≤ 5.0, with the balance being Ni. The CeO₂ particles (average particle size: 32 nm; supplied by Shanghai Buwei Applied Materials Technology Co., Ltd. (Shanghai, China)) were used. The TEM image in Figure 1c shows the distribution and morphology of these particles.

The cladding layer preparation method in this study was consistent with that in our previous work [28]. However, this paper provides more detailed process parameters: planetary ball milling was performed in a zirconia jar with a ball-to-powder ratio of 3:1 at 300 rpm for 2 h, using an intermittent cycle (10 min milling/5 min pause). The cladding material was deposited onto the surface of the 45 steel substrate using a quantitative coating method to ensure uniform cladding layer thickness. The sample contained in a graphite crucible was placed in a vacuum furnace, where the initial vacuum level was better than 6.67 × 10⁻² Pa, as measured by a hot-cathode ionization gauge. The cladding process was conducted under the following thermal schedule: the temperature was first raised to 200 °C within 30 min and held for 30 min; it was then increased to 1060 °C at a heating rate of 10 °C/min and held for 10 min. After the isothermal holding, the heating was terminated, and the sample was furnace-cooled to below 100 °C before removal. Subsequent analysis and characterization were performed after the sample cooled down to room temperature. All samples contained 30 wt % WC and varying amounts of CeO₂ (0.5, 1.0, 1.5, and 2.0 wt % for samples NWC5, NWC10, NWC15, and NWC20, respectively), with the balance being a Ni-based alloy powder.

2.2. Materials Characterization

Following vacuum cladding, 10 mm × 10 mm × 8 mm specimens were sectioned from cladding layer cross-sections. These were progressively ground, polished, and chemically etched in aqua regia (3 s). The microstructure of the cladding layer and the morphology and composition of the wear tracks were analyzed using a QUANTA FEG-450 (FEI Company, Hillsboro, OR, USA) field-emission scanning electron microscope (FEI Company, Hillsboro, OR, USA), equipped with an energy-dispersive X-ray spectroscopy (EDS (EDAX LLC, Mahwah, NJ, USA)) system. The analyses were conducted at an accelerating voltage of 18 kV, with the X-ray source being the field-emission gun (FEG FEI Company, Hillsboro, OR, USA). Image-Pro Plus 6.0 software (Media Cybernetics, Inc., Rockville, MD, USA) quantified porosity via threshold segmentation. At least 8 metallographic images were acquired from different sample areas using a scanning electron microscope at 500 × magnification. Images were converted to binary images through threshold segmentation using Image-Pro Plus 6.0 software, with pores displayed as black and the matrix as white via a combination of automated and manual threshold adjustment. After error correction, porosity was determined by calculating the pore area percentage. XRD (D/max-2400) scanned phases from 10° to 90°. CeO₂ underwent TGA (TGA 4000, PerkinElmer, Waltham, MA, USA) in N₂ from RT to 1100 °C at 10 °C/min using about 10 mg.

2.3. Mechanical Properties

Metallographic specimens were sectioned from the CeO₂-Ni/WC cladding layers using wire electrical discharge machining. After grinding and polishing, microhardness testing was performed using a W1102D37 micro-Vickers (Akashi Corporation, Kawasaki, Kanagawa, Japan) hardness tester under a load of 200 gf with a dwell time of 10 s and an indent spacing of 300 μm. The elastic modulus of the cladding layer surface was measured by a KLA G200 nanoindenter (KLA Corporation, Milpitas, CA, USA), equipped with a Berkovich tip (tip radius: 150 nm). Tests were conducted at a peak load of 10 mN and a loading rate of 0.5 mN/s, with a 4 × 8 array of points on the surface. The nanoindentation data were analyzed using the Oliver–Pharr method. During the analysis, the Poisson’s ratio of the nickel-based coating (ν) was assigned a value of 0.3. All mechanical tests were repeated at least three times to ensure reliability.

2.4. Friction Test

Dry sliding tests were conducted in ball-on-disk configuration using an HT-1000 tribometer (Lanzhou Zhongke Kaihua Science and Technology Development Co., Ltd., Lanzhou, Gansu Province, China). A commercial Si₃N₄ ball (diameter: 6 mm, Hardness: 1800 HV) served as the counter body. The test conditions were set with an initial maximum contact stress of 1.5 GPa and a sliding speed of 0.235 m/s. For surface friction tests, the wear track radius was 4 mm, giving a sliding distance of 423.0 m. For interfacial friction tests, metallographic specimens were hot-mounted in a 25 mm diameter resin and polished at cross-sections, with a wear track radius of 2 mm and sliding velocity of 0.1175 m/s; the total sliding distance was measured to be 211.5 m. Friction tests were performed on specimens with a surface roughness (Ra) of 0.05 μm under ambient conditions of (25 ± 2) °C and a relative humidity of (45 ± 5)%. Schematic diagrams of both test configurations are shown in Figure 2. Following tribological tests, the 3D topography of wear tracks was measured and wear volumes were calculated using an Rtec UP 2000 (Rtec Instruments, San Jose, CA, USA) a white light interferometer, while phase composition of tribo-oxides were analyzed with an ATR3110 Raman spectrometer (B&W Tek (Shanghai) Co., Ltd., Shanghai, China) (532 nm laser, 1 μm spot diameter) and a micro-focus X-ray diffractometer (D8 DISCOVER, Bruker AXS GmbH, Karlsruhe, Germany) (50 μm beam diameter, 2θ range: 10°–110°). The wear rate W_s in surface friction tests was calculated using Equation (1) based on the Archard wear model:

$$W_s= {\displaystyle \frac{V}{Fs}},$$

(1)

where V is the wear volume (mm³), F is the load (N), and s is the total sliding distance (mm).

3. Result and Discussion

3.1. Phase Composition and Distribution of Elements

Figure 3a shows the XRD patterns of the cladding layers with varying CeO₂ contents, revealing that the primary phases include γ-Ni, WC, Cr₇C₃, Cr₂₃C₆, W₂C, Ni₃Si, Ni₃Fe, and Ni₃B. The formation mechanisms of dominant phases in this cladding layer have been comprehensively elucidated in our prior work [28]. We adopt these established conclusions to provide the foundation for subsequent analysis of microstructure–property relationships. Additionally, weak diffraction peaks corresponding to CeO₂ were observed at 2θ values of 28.5° and 47.5°. Figure 3b presents the TGA curve of CeO₂, showing a mass loss of only 1.7% at 1060 °C. This low mass loss, indicating good thermal stability, is consistent with the detection of CeO₂ diffraction peaks in the XRD analysis, together confirming the presence of CeO₂ in the cladding layer. Figure 3c presents the measurement results of the porosity in the cladding layer, which shows that the porosity increases with the increasing CeO₂ content. When the CeO₂ content exceeds 1.5 wt %, the agglomeration of nanoparticles induces a diffusion blocking effect and an imbalance in solidification shrinkage compensation, ultimately leading to an increase in the overall material porosity [29,30,31]. As the CeO₂ content increases, the particles are more prone to aggregation, which in turn leads to a gradual increase in porosity.

The average grain size of the samples was calculated from XRD diffraction data using the Scherrer equation. Details of the calculation procedure can be found in our previous work [28]. As shown in Figure 3d, the calculated average grain sizes for the NWC5, NWC10, NWC15, and NWC20 samples were 12.06 nm, 12.93 nm, 16.21 nm, and 15.64 nm, respectively. Grain size increases with CeO₂ content, peaking at NWC15, which is a similar trend to porosity. The unmodified Ni/30 wt % WC cladding layer (0 wt % CeO₂) fabricated under identical conditions exhibited an average grain size of 25 nm. CeO₂ addition induced significant grain refinement, reducing the minimum grain size to 12.06 nm with a 51.76% decrease. The addition of a trace amount of CeO₂ (0.5 wt %) acts as heterogeneous nucleation sites, promoting grain refinement through heterogeneous nucleation. Simultaneously, it improves interfacial wettability and enhances melt fluidity, thereby promoting complementary shrinkage and reducing porosity. However, excessive CeO₂ (2.0 wt %) forms agglomerates, losing its effectiveness as heterogeneous nucleation sites, consequently impeding solute transport and leading to grain coarsening and increased porosity [29].

Figure 4 shows the SEM and EDS mapping images on the metallographic surface and interface of cladding layers with different CeO₂ contents. In order to identify the predominant phases in the cladding layer, EDS spot sweeps were performed for different color shapes. Based on EDS point analysis data, combined with XRD patterns and elemental mapping results, it is concluded that the white particulate phase in Figure 4a is composed predominantly of WC and W₂C (Point 1); the light-gray blocky phase corresponds to Ni₃Si (Point 2). The black acicular phase in Figure 4b is composed predominantly of Cr₇C₃ and Cr₂₃C₆ (Point 3), while the gray matrix phase contains γ-Ni, Ni₃B, and Ni₃Fe.

Figure 4c exhibits the cross-sectional morphology of the NWC5 cladding layer, with an average thickness of 3.28 ± 0.68 mm. Figure 4(c1–c4) shows EDS mapping images within the rectangular box in Figure 4c, where the diffusion bonding zone is primarily composed of Ni and Fe. An EDS line scan (Figure 4c, line 1) from the cladding layer to the substrate (Figure 4(c5)) shows a interdiffusion zone depth of approximately 210 μm. Compared to the cladding system without CeO₂ [28], CeO₂ reduces the surface tension during the melting process, thereby enhancing the wettability of the molten metal. Additionally, CeO₂ decomposes at high temperatures to produce Ce³⁺, which accumulates at the grain boundaries, reducing the activation energy of diffusion and facilitating the rapid migration of elements like Ni and Fe [30]. TGA analysis of CeO₂ shows that CeO₂ decomposes at high temperatures. Finally, under the same cladding process conditions, the depth of the interdiffusion zone significantly increased (from 40 μm to 210 μm).

3.2. Microhardness and Nanoindentation Analysis

Figure 5a illustrates the interfacial microhardness distribution of CeO₂-Ni/WC cladding layers. The results demonstrate that the substrate exhibits a microhardness of approximately 255 HV0.2, while the interdiffusion zone achieves the maximum microhardness value of 394.6 HV0.2 (NWC5) due to interdiffusion of Ni and Cr. The fusion cladding layer shows a progressive increase in microhardness followed by stabilization. Figure 5b presents the surface microhardness distribution, with measured values of 1042.74 ± 34.58 HV0.2, 931.41 ± 54.12 HV0.2, 963.14 ± 66.49 HV0.2, and 872.36 ± 93.51 HV0.2 for NWC5, NWC10, NWC15, and NWC20, respectively (error bars represent standard deviations). NWC5 exhibits the highest average surface microhardness (1042.74 HV0.2), which decreases with increasing CeO₂ content, accompanied by expanding standard deviations (34.58→54.12→66.49→93.51). The coarsened grains and elevated porosity in the cladding layers induced by excessive CeO₂ additions are identified as primary contributors to the enhanced microhardness variability.

The elastic modulus of cladding layers was systematically characterized using nanoindentation testing. As illustrated in Figure 6, a 4 × 8 grid array with 20 μm spacing between adjacent points was established on the sample surface for comprehensive measurements. Notably, the NWC5 specimen exhibited the maximum elastic modulus, with a significant decreasing trend observed as the CeO₂ content increased.

Table 1. Mechanical properties of cladding layers.

Sample	Nanohardness	Elasticity Modulus	H/E	H3/E2
	H (GPa)	E (GPa)	/	GPa
NWC5	12.34 ± 1.43	269.06 ± 6.61	0.0459	0.0260
NWC10	10.94 ± 1.06	240.47 ± 5.94	0.0455	0.0226
NWC15	10.17 ± 1.39	237.09 ± 7.16	0.0429	0.0187
NWC20	9.98 ± 1.63	222.91 ± 8.09	0.0434	0.0188

summarizes the nanohardness (H), elastic modulus (E), H/E ratio, and H³/E² ratio for each coating layer. The H/E ratio serves as an indicator of fracture toughness, and a higher value suggests a greater resistance to crack initiation during friction. The H³/E² ratio is a measure of resistance to plastic deformation, and a higher value implies a stronger ability to resist penetration by the Si₃N₄ indenter during the friction process. Among the coatings with different CeO₂ contents, NWC5 exhibits the highest H/E ratio (0.0459) and H³/E² ratio (0.0260), indicating its superior tribological performance [28]. It is noteworthy that while lattice refinement induces solid solution strengthening effects that improve microhardness; an excessive CeO₂ addition ultimately results in an elastic modulus reduction.

3.3. Surface Friction Analysis

As illustrated in Figure 7a, the friction coefficient evolution of CeO₂-Ni/WC cladding layers exhibits two distinct characteristic stages: an initial run-in period followed by a steady-state wear phase. During the run-in stage (about 0–10 min), the point contact configuration between the Si₃N₄ counter ball and cladding layer surface induces significant localized stress concentration. The interlocking effect of surface asperities results in pronounced oscillations in the friction coefficient (±0.22 amplitude). Continuous debris generation and accumulation during this phase drive dynamic surface topographical reconstruction. Following this initial period (>10 min), the friction coefficient stabilizes (±0.07 amplitude). This transition is attributed to the formation of a third-body layer from wear debris, which effectively reduces shear strength at the contact interface and establishes a new tribological equilibrium. Figure 7b,c present the average friction coefficients and wear rates of CeO₂-Ni/WC cladding layers during the steady-state wear phase. The measured values for NWC5, NWC10, NWC15, and NWC20 specimens are 0.618, 0.670, 0.613, and 0.579 for friction coefficients, and 1.46 ± 0.18 × 10⁻⁸, 1.89 ± 0.11 × 10⁻⁸, 2.68 ± 0.31 × 10⁻⁸, and 15.43 ± 0.98 × 10⁻⁸ mm³/(N·m) for wear rates, respectively. Notably, NWC20 exhibits an inverse correlation between the highest wear rate (476% increase compared to NWC15) and the lowest friction coefficient. This phenomenon is primarily attributed to its characteristic 0.253% porosity: The porous architecture functions as prefabricated debris reservoirs, facilitating the transition from sliding to rolling contact through third-body lubrication, thereby reducing the friction coefficient. Coarse grains (average size 15.64 nm) induce stress concentration at grain boundaries, which promotes microcrack nucleation/propagation and accelerates debris generation, consequently elevating wear rate.

Micro-area XRD analysis of the wear track on NWC20 specimen (Figure 7d) reveals the coexistence of NiO, Cr₂O₃, CeO₂, and Fe₃O₄ phases. Although the friction test was conducted at ambient temperature, localized transient heating (400–600 °C) generated by plastic deformation and adhesive effects between asperities exceeds the oxidation threshold of Ni (400 °C). This thermomechanical coupling induces selective oxidation of surface Ni atoms, forming NiO. Chromium demonstrates higher oxidation priority due to its more negative Gibbs free energy of oxidation (ΔG° = –1058 kJ/mol) compared to Ni (ΔG° = –489 kJ/mol), leading to preferential formation of Cr₂O₃. The oxidation pathway of Fe involves sequential reactions: initial generation of metastable FeO followed by its subsequent oxidation to thermodynamically stable Fe₃O₄ [31]. It should be noted that the CeO₂ peaks originate from the intentionally added reinforcing phase during cladding preparation, with no detectable participation in tribo-oxidation processes.

To investigate the tribological mechanism of CeO₂-Ni/WC cladding layers, systematic characterization of worn surfaces was conducted using SEM. Figure 8 presents the typical wear morphology of CeO₂-Ni/WC cladding layers after friction tests, with the wear mechanism elucidated through combined micro-area X-ray diffraction and EDS analyses. Microstructural observations reveal that the worn surface consists of a gray continuous matrix and discrete white hard phase particles (Figure 8(c1)). EDS results (

Table 2. Chemical compositions (at %) of different points in Figure 8(c1).

Point	Ni	Cr	B	Si	Fe	C	O	W	Ce
5	4.3	2.6	3.2	5.6	3.1	26.8	22.2	32.2	0
6	20.5	18.9	0.6	6.2	3.2	23.5	25.3	1.7	0.1
7	27.5	2.9	1.2	12.6	6.3	15.8	31.5	2.2	0

) indicate that the gray debris (spot 6 in Figure 8(c1)) primarily contains metallic oxides (Fe₃O₄, NiO, Cr₂O₃) and carbides (Cr₇C₃, Cr₂₃C₆), while the white particles (spots 5 and 7 in Figure 8(c1)) are predominantly composed of hard phases including WC, W₂C, and Ni₃Si. Characteristic spalling pits, microcracks, and plowing grooves are observed on the worn surface. The formation mechanism of spalling pits involves microcrack initiation (Figure 8(b1)) induced by stress concentration and subsequent propagation leading to material delamination (Figure 8(a1)), predominantly occurring during the initial run-in period with a significant fluctuation in friction coefficient. The detached alloy fragments are progressively crushed by silicon nitride counter balls and deposited into surface defects (pores and spalling pits). As friction progresses, continuous microcrack generation (arrows in Figure 8b) and subsequent delamination within the debris layer establish a dynamic equilibrium during the steady-state wear stage. Notably, the hard phase particles (white particles) induce prominent plowing effects, forming characteristic grooves on the cladding layer surface (Figure 8(d1)), which correlates with their superior hardness properties. The wear behavior of CeO₂-Ni/WC cladding layers is governed by the synergistic interaction between abrasive wear and oxidative wear mechanisms [32].

Figure 9 presents the quantitative three-dimensional morphological analysis of wear tracks. Systematic characterization through three representative cross-sections of 3D wear profiles reveals key tribological parameters for NWC5, NWC10, NWC15, and NWC20 specimens: the average wear track widths measure 0.34, 0.39, 0.27, and 0.57 mm, respectively; maximum wear depths reach 3.53, 3.77, 4.02, and 7.29 μm; and surface roughness average (Ra) values are quantified as 522.1, 509.7, 691.2, and 1213.6 nm. The NWC5 specimen demonstrates the minimum average wear width (0.34 mm) and maximum depth (3.53 μm), attributed to its superior microhardness (1042.74 HV0.2) and elastic modulus (269.06 GPa). In contrast, the NWC20 specimen exhibits the highest Ra value (1213.6 nm), resulting from stress concentration and hard phase delamination induced by its elevated porosity (0.253%). Notably, distinctive lip-like protrusions formed by debris accumulation are observed at wear track edges, which can be mechanistically linked to plastic flow in contact zones and debris redeposition processes.

3.4. Interface Friction Analysis

Figure 10a illustrates the friction coefficient evolution at the interface of CeO₂-Ni/WC cladding layers. Similar to surface friction behavior, the friction coefficient progression exhibits distinct running-in and steady-state wear stages, yet with notable interfacial characteristics: the running-in duration decreases from 10 min (surface friction) to 6 min, accompanied by a significant reduction in steady-state average friction coefficients (Figure 10b). Quantitative analysis reveals that the steady-state friction coefficients for NWC5, NWC10, NWC15, and NWC20 specimens decrease from 0.601, 0.617, 0.589, and 0.562 (surface friction) to 0.343, 0.359, 0.444, and 0.418 (interface friction), respectively. A negative correlation was observed between the stable friction coefficient at the interface and both porosity and grain size, where reduced porosity and smaller crystallite size led to a lower friction coefficient. Raman spectroscopy analysis of wear tracks (Figure 10c) indicates a consistent phase composition between interface and surface friction regions, predominantly comprising NiO, Cr₂O₃, CeO₂, and Fe₃O₄. To systematically investigate spatial heterogeneity in tribological behavior, the wear track is divided into four characteristic zones based on Si₃N₄ counter ball trajectories (schematic in Figure 10d): cladding layer-to-substrate transition zone (CTSZ), substrate zone (SZ), substrate-to-cladding layer transition zone (STCZ), and cladding layer zone (CZ). This zoning strategy establishes a spatially resolved framework for analyzing material transfer and interfacial response mechanisms during friction processes.

To investigate the spatial heterogeneity of tribological characteristics across wear track regions during interface friction, systematic SEM characterization and EDS analysis were conducted on the NWC5 specimen. Figure 11a shows the wear track morphology, revealing significant width variations: the SZ exhibits the maximum width (874 μm), while the CZ shows the minimum (474 μm). The CTSZ (799 μm) is 43.2% wider than the STCZ (558 μm). Elemental mapping in Figure 11(a1) demonstrates gradient distributions: O content follows SZ > CTSZ > CZ > STCZ, Fe decreases as SZ > STCZ > CTSZ > CZ, and Ni inversely distributes with CZ > CTSZ > STCZ > SZ. Figure 11b shows the STCZ region, where distinct accumulation of hard phases can be observed. Figure 11c–e display microstructural features of STCZ, SZ, CTSZ, and CZ, respectively, showing characteristic wear morphologies including black debris, plowing grooves (Figure 11c), microcracks, and spalling pits (Figure 11e). The black debris containing gray/white particles was analyzed by area-spot EDS (

Table 3. Chemical compositions (at %) of different positions in Figure 11c,d.

Area	C	O	Si	Cr	Fe	Ni	Ce	W
1	14.87	4.87	0.57	0.03	79.16	0.38	0.03	0.07
2	11.24	50.49	5.19	2.87	20.07	8.20	0.05	1.88
3	9.67	49.97	6.70	5.11	11.48	14.31	0.06	2.69
4	17.77	24.20	2.72	8.10	22.45	22.49	0.12	2.16

) and Raman spectroscopy, confirming identical composition to surface friction debris: black metallic oxides (Fe₃O₄, NiO, Cr₂O₃), gray carbides (Cr₇C₃, Cr₂₃C₆), and white hard phases (WC, W₂C). Notably, the compositional similarity between the debris and original cladding layer (CeO₂-Ni/WC), combined with spatial O-Fe-Ni distribution patterns, suggests debris primarily originates from CZ. Tribological material transfer leads to debris accumulation in CTSZ and SZ, while its content drastically reduces in STCZ due to interfacial barrier effects.

The formation mechanisms of plowing grooves, microcracks, and spalling pits remain consistent with those observed during surface friction (see Figure 8 for mechanistic details). The wear mechanisms at interface friction maintain substantial consistency with surface friction, predominantly governed by the synergistic interaction between abrasive wear and oxidative wear. Specifically, cyclic detachment of hard phase particles (e.g., WC, W₂C) generates third-body abrasives, producing characteristic plowing grooves (as shown in Figure 11c). Concurrently, frictionally activated oxidation reactions induce in situ formation of metallic oxides (Fe₃O₄, NiO) through oxygen interaction with metallic matrices (Fe, Ni), as evidenced by Raman analysis. These oxide layers undergo dynamic fracture–reformation cycles under alternating shear stresses. Notably, significant plastic deformation was observed on the inner surface of the CTSZ wear track (Figure 11d). No microcracks induced by cyclic stress were observed at the CTSZ, suggesting good interfacial bonding strength. Microhardness gradient analysis (Substrate: 255 HV0.2 → Interdiffusion zone: 394.6 HV0.2 → Cladding layer: 1042.74 HV0.2) reveals that stress mismatch induced by abrupt hardness variations governs the deformation behavior when Si₃N₄ counter balls slide from the high-hardness cladding layer (1042.74 HV0.2) through the medium-hardness interdiffusion zone (394.6 HV0.2) to the low-hardness substrate (255 HV0.2). This hardness transition triggers shear localization in CTSZ, where accumulated lattice distortion energy from the impeded dislocation slip is ultimately released via plastic flow, forming the wave-like deformation morphology shown in Figure 11d [33]. The CZ matrix phase (e.g., γ-Ni phase) shown in Figure 11e also exhibited plastic deformation on its surface.

Figure 12a–c present the interfacial wear track morphologies of NWC10, NWC15, and NWC20 specimens. Across different regions of the wear track, the variation in wear track width followed a pattern similar to that of NWC5, in the order SZ > CTSZ > STCZ > CZ. Distinct plastic deformation zones were observed on the inner side of wear tracks in both the CTSZ (Figure 12(a3)) and STCZ (Figure 12(a4)) regions of NWC10, with these deformation zones primarily concentrated on the 45 steel substrate side. This plastic deformation originated from stress concentration induced by abrupt changes in material hardness, causing coordinated deformation of the soft 45 steel substrate under frictional shear stress. Additionally, cyclic loading induced a plastic flow at weak interfacial bonding sites in the transition zone, forming strain accumulation zones.

High-magnification images of CTSZ and STCZ for NWC15 (Figure 12(b3,b4)) are shown in Figure 12(b5,b6). In CTSZ (Figure 12(b5)), microstructural characterization reveals that the dark debris matrix contains a substantial dispersion of white hard phase particles. Hard phase particles progressively generate grooves along the sliding direction of Si₃N₄ counter balls, exhibiting depth/width gradients from shallow/narrow to deep/wide, with final particle embedding at groove termini (inside the yellow box of CTSZ). Conversely, in SZ, hard phases directly embed into the 45 steel substrate without forming extended grooves (inside the yellow box of SZ). This divergence stems from material hardness gradients: the cladding layer surface hardness (963.14 HV0.2), interdiffusion zone (CTSZ, 307.35 HV0.2), and substrate (257.76 HV0.2). The reduced hardness gap between CTSZ (307.3 HV0.2) and WC particles (~2000 HV) enhances shear resistance against particle penetration, thereby prolonging groove propagation in CTSZ [34]. Two evolutionary characteristics are observed in the STCZ (Figure 12(b6)): (1) Substantial depletion of white hard phase particles within the dark debris matrix, showing a 76% reduction in volume fraction compared to CTSZ (statistical analysis across three fields by Image-Pro Plus 6.0 software). (2) Narrow elongated shallow grooves (blue box, aspect ratio > 15:1) predominantly formed in the SZ by hard phase particles. This provides evidence of a size-selection mechanism during particle migration: initial particles generated in CZ undergo mechanochemical degradation (tribo-oxidation + fragmentation) in CTSZ and plastic dissipation in SZ (45 steel substrate hardness 257 HV0.2), resulting in submicron residues at STCZ that exhibit diminished groove-forming capacity. Distinct plastic deformation zones were also observed on the inner side of wear tracks in the CTSZ and STCZ regions of the NWC20 specimen. Quantitative analysis demonstrated that all specimens exhibited more pronounced wear characteristics in the CTSZ compared to the STCZ during interfacial friction testing (detailed data are shown in

Table 4. The width of wear track in different areas of the cladding layer (μm).

Sample	SZ	STCZ	CZ	CTSZ
NWC5	874 ± 106	558 ± 65	474 ± 46	799 ± 98
NWC10	747 ± 95	464 ± 42	432 ± 39	598 ± 52
NWC15	811 ± 89	562 ± 34	553 ± 45	686 ± 56
NWC20	695 ± 72	437 ± 37	411 ± 39	622 ± 43

). Specifically, the average wear track width in CTSZ regions increased by 34.13% relative to STCZ (data range: 22.1%–43.2% for NWC5-NWC20 series specimens). Mechanistic analysis revealed two predominant factors: Firstly, the stepwise hardness transition from the high-hardness cladding layer to medium-hardness interdiffusion zone in CTSZ induced localized stress concentration through stress gradient effects during Si₃N₄ counter ball sliding. Secondly, wear debris generated in the CZ directly interacted with CTSZ, intensifying abrasive wear. In contrast, the reduced wear in STCZ originated from (1) a progressive hardness transition from the low-hardness substrate to high-hardness cladding layer that mitigated stress concentration; (2) approximately 76% less wear debris accumulation compared to CTSZ. These differential mechanisms ultimately resulted in significantly greater wear track width in CTSZ than in STCZ.

To elucidate the interfacial friction-wear mechanisms, this study developed schematic diagrams of friction processes for both the CTSZ (Figure 13a) and STCZ (Figure 13b). Figure 13 (a1 and a2, respectively) displays the 3D surface topography and cross-sectional profile of the wear track in the CTSZ region of the NWC15 specimen. Experimental data from Figure 11 and Figure 12 confirm that CTSZ exhibits a 34.13% greater average wear track width than STCZ, resulting from dynamic stress impacts caused by hardness gradients and synergistic debris effects. Detailed analysis of Figure 13(a2) reveals distinct stepped morphologies at both CZ-CTSZ and CTSZ-SZ interfaces. The formation mechanism involves two critical aspects: (1) Wear debris generated in the cladding zone (CZ) preferentially accumulates in CTSZ, where the mixture of hard phase particles and metal oxides intensifies three-body abrasion. (2) Cyclic stress impacts induced by Si₃N₄ counter balls crossing abrupt hardness transition interfaces ultimately develop the three-tiered stepped morphology (step height differential: 2.1–4.1 μm).

Figure 13(b1,b2) systematically characterizes the three-dimensional morphology and cross-sectional profile of wear tracks in the STCZ of the NWC15 specimen. Analysis of the 2D wear track profile in Figure 13(b2) reveals continuous gradient transitions at both SZ-STCZ and STCZ-CZ interfaces, devoid of distinct stepped morphologies, with wear debris predominantly accumulating at the STCZ-CZ interface. Unlike the CTSZ, where abrupt hardness transitions generate dynamic stress impacts and stepped structures, the progressive hardness gradient from the low-hardness substrate to high-hardness cladding layer in STCZ prevents significant stress concentration, resulting in a stabilized stress distribution during Si₃N₄ counter ball sliding. Crucially, the sustained three-body abrasion induced by debris-laden counter balls promotes uniform material removal, ultimately forming a smooth transitional morphology in the STCZ wear track [35,36].

3.5. Discussion

CeO₂-Ni/WC composite cladding layers were fabricated on 45 steel substrates via vacuum cladding technology. Microstructural characterization revealed defect-free dense structures with metallurgical bonding at the interface (diffusion layer depth: approximately 210 μm). Increasing the CeO₂ content from NWC5 to NWC20 induced monotonic porosity growth (0.0673%→0.253%) and grain coarsening up to 16.21 nm in NWC15 (34.4% increase vs. NWC5). These structural evolutions led to mechanical property degradation: microhardness decreased from 1042.74 ± 34.58 HV0.2 (NWC5) to 872.36 ± 93.51 HV0.2 (NWC20), accompanied by a 17.2% reduction in elastic modulus (269.06→222.91 GPa). In surface friction tests, NWC20 exhibited the lowest friction coefficient (0.579) and highest wear rate (15.43 ± 0.98 × 10^-8 mm³/(N·m)), attributed to a reduced load-bearing capacity from high porosity (0.253%) and coarse grains (15.64 nm). During interfacial friction, the average friction coefficient decreased significantly compared to surface friction, with this coefficient exhibiting a similar negative correlation with both porosity and grain size. Additionally, stress concentration induced by abrupt hardness changes and cumulative plastic deformation caused by cyclic loading resulted in distinct plastic deformation zones on the inner side of wear scars adjacent to the 45 steel substrate. The CTSZ developed a characteristic three-tier stepped morphology due to cyclic stress impacts at abrupt hardness transitions. Conversely, the STCZ showed a smooth wear morphology resulting from progressive hardness gradients. No microcracks were observed in the interdiffusion zone, indicating good interfacial bonding. Micro-area XRD and Raman analysis of wear debris confirmed competing mechanisms: three-body abrasion and oxidative wear (NiO, Cr₂O₃, Fe₃O₄).

4. Conclusions

In this paper, CeO₂ (0.5,1.0,1.5 and 2.0 wt %)-Ni/WC composite cladding layers on 45 steel were fabricated by vacuum cladding. The detailed phase composition, mechanical properties, and surface and interface tribological characteristics of composite cladding layers were investigated. The conclusions can be drawn as follows.

(1)

The phases present in the CeO₂-Ni/WC composite cladding layer include γ-Ni, Ni₃Si, Ni₃Fe, Ni₃B, WC, W₂C, Cr₇C₃, Cr₂₃C₆, and CeO₂. Among all cladding layers, NWC5 exhibited the lowest porosity (0.0673%) and the smallest average grain size (12.06 nm).

(2)

The microhardness and elastic modulus of NWC5 are 1042.74 ± 34.58 HV0.2 and 269.06 GPa, respectively. The combined effects of reduced porosity and grain refinement enhance the mechanical properties of NWC5.

(3)

NWC20 demonstrated the lowest average friction coefficient (0.579) and the highest wear rate (15.43 × 10⁻⁸ mm³/(N·m)). This correlation originates from the synergistic effects of structural defects in NWC20, characterized by high porosity (0.253%) and coarse grains (15.64 nm), where porous architecture facilitates crack propagation while grain coarsening diminishes grain boundary strengthening. NWC5 achieved the minimum wear rate of 1.46 × 10^-8 mm³/(N·m), representing a 90.54% reduction compared to NWC20. The wear mechanism was predominantly governed by the three-body abrasion from hard phase particles, coupled with cyclic formation/delamination of tribo-oxidation layers.

(4)

The interfacial steady-state friction coefficient of CeO₂-Ni/WC composite cladding layers ranged from 0.343 to 0.444, with wear mechanisms dominated by abrasive wear and oxidative wear. Quantitative analysis revealed a 22.1%–43.2% increase in wear track width at the CTSZ compared to the STCZ. Mechanistic investigations demonstrated that cyclic stress impacts generated by silicon nitride counter balls traversing abrupt hardness transition interfaces in CTSZ induced a characteristic three-tiered stepped morphology (step height differential: 2.1–4.1 μm). In contrast, STCZ exhibited progressive hardness gradients and homogenized stress distribution, resulting in smooth transitional morphology in wear track.