Interfacial Gradient Optimization and Friction-Wear Response of Three Architectures of Ni-Based Cold Metal Transfer Overlays on L415QS Pipeline Steel

Abstract

Pipeline steels under cyclic loading in corrosive environments are prone to wear and corrosion–wear synergy. Low-dilution, high-reliability Ni-based Cold Metal Transfer (CMT) overlays are therefore required to ensure structural integrity. In this work, three overlay architectures were deposited on L415QS pipeline steel: a single-layer ERNiFeCr-1 coating, a double-layer ERNiFeCr-1/ERNiFeCr-1 coating, and an ERNiCrMo-3 interlayer plus ERNiFeCr-1 working layer. The microstructure, interfacial composition gradients, and dry sliding wear behavior were systematically characterized to clarify the role of interlayer design. The single-layer ERNiFeCr-1 coating shows a graded transition from epitaxial columnar grains to cellular/dendritic and fine equiaxed grains, with smooth Fe dilution, Ni–Cr enrichment, and a high fraction of high-angle grain boundaries, resulting in sound metallurgical bonding and good crack resistance. The double-layer ERNiFeCr-1 coating contains coarse, strongly textured columnar grains and pronounced interdendritic segregation in the upper layer, which promotes adhesive fatigue and brittle spalling and degrades wear resistance and friction stability. The ERNiCrMo-3 interlayer introduces continuous Fe-decreasing and Ni-Cr/Mo-increasing gradients, refines grains, suppresses continuous brittle phases, and generates dispersed second phases that assist crack deflection and load redistribution. Under dry sliding, the tribological performance ranks as follows: interlayer + overlay > single-layer > double-layer. The ERNiCrMo-3 interlayer system maintains the lowest and most stable friction coefficient due to the formation of a dense tribo-oxidative glaze layer. These results demonstrate an effective hierarchical alloy-process design strategy for optimizing Ni-based CMT overlays on pipeline steels.

1. Introduction

Pipeline steels in long-term service are exposed to high pressures, Cl⁻/S-containing media, and cyclic loading, where surface wear and corrosion-wear synergy have become critical factors limiting service life and operational safety [1,2,3]. Depositing functional overlays (e.g., Ni-based alloys) on steel substrates can significantly improve corrosion and wear resistance as well as reparability [4,5]. Among various processes, Cold Metal Transfer (CMT) offers low heat input, low spatter and stable short-circuit transfer, enabling high-quality formation while effectively reducing dilution and the width of the heat-affected zone (HAZ). CMT is a modified Gas Metal Arc Welding (GMAW) process tailored for low-heat-input cladding/overlay applications. Its core mechanism involves digitized current regulation to achieve spatter-free short-circuit transfer: during deposition, the welding current is actively interrupted at the moment the filler wire droplet contacts the molten pool, minimizing heat input (30%–50% lower than conventional GMAW) while ensuring stable metal deposition. Key characteristics include low dilution rates (<10%), a narrow heat-affected zone (HAZ), and excellent metallurgical bonding—distinguishing it from non-welding surface-treatment technologies (e.g., physical vapor deposition, thermal spraying) or high-heat-input welding processes. This clarification ensures readers contextualize the Ni-based “overlays” as weld-deposited functional coatings, not alternative surface layers. It has shown clear advantages in hardfacing and corrosion-resistant alloy overlays. Previous studies have demonstrated that CMT can produce low-dilution coatings thicker than 2.5 mm, with lower dilution and more uniform penetration compared with GMAW-RE and related processes [6,7,8]. From a solidification metallurgy perspective, the temperature gradient (G) and interface growth rate (R) determine morphology and microstructural scale, which underpins the typical transition from interfacial epitaxial columnar grains to cellular/dendritic and fine equiaxed grains, as well as the development of strong textures in overlays [9]. In Ni-based overlay systems, ERNiFeCr-1 and ERNiCrMo-3 fillers have been widely applied in steel/Ni-based dissimilar joints and surface overlays, and datasheets confirm their good resistance to localized corrosion and compatibility in dissimilar welding [10,11,12].

Nevertheless, directly depositing a single Ni-based layer on steel still presents serious challenges. First, the steep gradients in chemical composition and thermophysical properties at the steel/overlay interface promote elemental interdiffusion and the formation of continuous brittle phases and carbides. Second, solidification segregation and strong textures in the upper layer tend to generate long columnar grains and anisotropy, which can trigger adhesion–fatigue coupling and sudden increases in friction coefficient during sliding. Third, repeated remelting and reheating in multi-pass/multi-layer deposition amplify interdendritic segregation and interlayer stress concentration [13,14,15]. Whang et al. reported that introducing a Ni-based interlayer in stainless steel-Ni/WC composite overlays reduced internal stress, suppressed cracking and improved interfacial uniformity [16]. Liu et al. demonstrated that a two-layer design (Ni-WC top layer with an Inconel 625 interlayer) can effectively replace hard chrome plating and enhance combined wear and corrosion resistance [17]. Consistently, Du et al. showed that increasing Ni content in the interlayer mitigates thermal expansion mismatch, reduces cracking susceptibility and improves interfacial toughness and structural continuity across different systems [18,19]. Thus, interlayer design has emerged as an effective process-material coupling strategy. High-Ni or Mo-containing interlayers can provide a graded buffer in terms of chemistry, thermophysical properties and solidification conditions, thereby lowering interfacial dilution/embrittlement risks and adjusting G/R in the upper layers to weaken competitive columnar growth [20,21,22]. Recent comparative studies and reviews indicate that Ni/Cr/Mo-enriched interlayers (such as Alloy 625-type ERNiCrMo-3) can reduce dilution, smooth the Fe→Ni gradient, suppress cracking, refine the microstructure and enhance the stability of corrosion and wear resistance (in X65 steel/625 overlays and Ni interlayers mitigating cracking and thermal mismatch) [23,24]. In addition, Mo-containing systems tend to form low-shear-strength composite oxide films or “glaze layers” (e.g., Cr₂O₃/NiO with Mo oxides) during sliding, which can markedly reduce friction and wear in the intermediate temperature range until film fracture and spallation occur; this self-lubricating tribo-film mechanism has been confirmed in various Ni-based composite and self-lubricating coatings [25,26,27]. Despite the proven effectiveness of Ni-Mo interlayers in mitigating interfacial embrittlement and improving mechanical properties, critical research gaps remain: (1) No systematic comparison has been conducted on the microstructural evolution and tribological mechanisms of three typical CMT overlay architectures (single-layer, double-layer homogeneous alloy, and ERNiCrMo-3 interlayer + ERNiFeCr-1 working layer) on L415QS low-alloy high-strength pipeline steel, a material widely used in oil and gas transportation but rarely studied in Ni-based CMT overlays. (2) Previous studies focused on single-structure optimization, lacking in-depth analysis of how interlayer design (layer number and alloy type) synergistically regulates solidification behavior, interfacial composition gradients, and tribo-oxide film formation. (3) The correlation between high-angle grain boundary (HAGB) distribution, secondary dendritic arm spacing (SDAS), and friction stability in Ni-based CMT overlays has not been clarified [28,29,30]. This work fills these gaps by systematically characterizing the three architectures, revealing the role of interlayer design in optimizing interfacial gradients and wear resistance, and providing a hierarchical material-process design strategy for pipeline steel applications.

Against this background, L415QS pipeline steel is selected as the substrate, and three CMT overlays are designed: (i) single-layer ERNiFeCr-1; (ii) double-layer ERNiFeCr-1/ERNiFeCr-1; and (iii) ERNiCrMo-3 interlayer + ERNiFeCr-1 working layer. Geometrical formation, solidification morphologies, elemental gradients and phase distributions are characterized by SEM/EDS/EBSD, and dry sliding friction–wear tests are conducted. The effects of layer number and interlayer design on epitaxial-to-cellular/dendritic transitions, interlayer remelting/refinement zones, interdendritic precipitation and grain boundary networks are clarified, together with their impact on friction coefficient evolution and wear mechanisms (abrasive, adhesive and oxidative–glazing coupling). Based on these results, a hierarchical material–process design concept and key parameter window (heat input, interpass temperature, overlap ratio and alloy retention) are proposed for Ni-based CMT overlays tailored to pipeline steel service conditions.

2. Experimental Procedure

2.1. Materials and Equipment

The substrate was L415QS pipeline steel (Baoshan Iron & Steel Co., Ltd. (Baosteel), Shanghai, China), a typical low-cost, high-strength structural steel widely used in oil and gas transportation, which requires low heat input during overlay welding to minimize distortion and performance degradation. The dimensions of the pipeline steel plates were 300 mm × 150 mm × 20 mm. The chemical compositions of the L415QS steel and nickel-based alloys are listed in

Table 1. Chemical composition of the L415QS steel and nickel-based alloys (wt.%).

Materials	C	Si	Mn	P	S	Cr	Mo	Ni	Ti	Cu	Al	Nb	V	Fe
L415QS	0.18	0.45	1.25	0.025	0.015	0.42	0.46	0.39	0.015	0.33	-	0.05	0.05	Bal.
ERNiFeCr-1	0.003	0.15	0.92	0.003	0.002	23.17	3.38	43.63	1.12	2.65	0.09	-	-	Bal.
ERNiCrMo-3	0.016	0.09	0.21	0.004	0.002	21.48	8.86	64.70	0.16	0.01	0.15	3.74	-	Bal.

. ERNiFeCr-1 wire (Beijing Jinwei Welding Materials Co., Ltd., Beijing, China) was selected as the overlay material because it provides excellent resistance to general corrosion, pitting, and stress corrosion cracking, while offering a thermal expansion coefficient and other physical properties better matched to the steel substrate. ERNiCrMo-3 wire was employed as the interlayer material; its high Ni content improves compatibility with the Fe-based substrate compared with Fe-richer overlays, and the Mo addition enhances pitting resistance, thereby suppressing excessive Fe diffusion into the overlay and reducing the tendency for brittle interfacial phase formation. To clarify the influence of overlay pass number and interlayer design on the CMT coating microstructure and properties, three comparative configurations were prepared: (i) single-layer overlay; (ii) double-layer overlay; and (iii) ERNiCrMo-3 interlayer plus one ERNiFeCr-1 overlay layer.

Prior to deposition, the substrate plates were cleaned and ground to remove surface contaminants. The plates were then rigidly clamped on the welding platform to avoid vibration during cladding, and the leading end of the filler wire was cleaned to prevent arc instability and inhomogeneous bead formation. The clamped plates were aligned with good perpendicularity and positioned to match the programmed torch path. The CMT system consisted of a KUKA welding robot, a CMT Advanced 4000 pulsed power source, a Fronius torch, a flexible welding fixture table, and a fume extraction unit. Before welding, a dry-run was conducted to verify the torch trajectory and avoid collision with the workpiece. Cladding was performed using single-pass, unidirectional tracks. For multi-layer specimens, the deposition direction of each subsequent layer was parallel to and opposite to that of the previous layer. The welding parameters were set to a current of 178 A, voltage of 25.4 V, wire feed speed of 8.0 m/min, and an overlap ratio of 40%. To avoid excessive heat accumulation and its adverse effects on the microstructure and properties, each pass was followed by air cooling, and the inter℃ before the next pass. The shielding gas was high-purity argon-helium mixture (70% Ar/30% He, each 99.99% pure) with a flow rate of 20 L/min, and the welding travel speed was 8.0 mm/s. To avoid excessive heat accumulation, the interpass temperature was strictly controlled below 100 °C (monitored by an infrared thermometer) before depositing the next layer.

2.2. Microstructure Characterization and Properties Test

Cross-sectional specimens of the CMT overlays were cut by electrical discharge machining using a Mo wire. The cross-sections were ground sequentially with 80#, 400#, 800#, 1500#, 2000#, and 3000# SiC papers, followed by mechanical polishing with a nano-SiO₂ suspension to obtain a mirror finish, and subsequently etched using a 4 vol.% nitric acid-ethanol solution. Surface and cross-sectional morphologies were examined using a ZEISS Sigma 300 field-emission scanning electron microscope (FE-SEM). An Oxford energy-dispersive X-ray spectrometer (EDS) was employed for elemental analysis. An Oxford Nordlys Nano EBSD detector combined with AZtecCrystal software (v6.0) was used for phase identification, phase fraction statistics, grain size and grain orientation analysis, texture analysis, and dislocation density evaluation. For EBSD, the sample surfaces were ground up to 3000#, mechanically polished to a mirror finish, and then vibratory polished for 8 h using a VibroMet 2 (Buehler) with a nano-SiO₂ suspension to remove surface deformation. During SEM and EDS analyses, the accelerating voltage was set to 10–15 kV with a working distance of 6–8 mm. The EBSD step size was selected as approximately one-tenth of the average grain size.

To evaluate and compare the wear performance and wear mechanisms of the overlays, linear reciprocating dry sliding tests were carried out on planar specimens taken from the near-surface functional region of the coatings. A Si₃N₄ ball with a diameter of 6 mm was used as the counter body. The stroke length was 7 mm, the normal load was 30 N, and the sliding speed was 120 mm/min, with a total test duration of 30 min. A fresh Si₃N₄ ball was used for each test to avoid cross-contamination and environmental history effects. All tests were conducted at room temperature. After wear testing, the specimens were ultrasonically cleaned in alcohol, dried thoroughly, and subsequently examined by SEM to characterize the worn surface morphologies.

3. Results and Discussion

3.1. Microstructure

ERNiFeCr-1 was deposited on L415QS pipeline steel by the CMT process to obtain a geometrically uniform single-layer multi-pass overlay, as shown in Figure 1. The individual bead thickness is approximately 3.8 mm, with a bead width of about 5.2 mm. The cross-section is dense, and no lack of fusion or macro-cracks is observed, as shown in Figure 1a–c. From the fusion line to the surface, the microstructure exhibits a continuous transition from epitaxial columnar grains to cellular–dendritic grains and locally fine equiaxed grains. Near the fusion line, grains grow preferentially along the heat flow direction, showing a typical epitaxial columnar/cellular–columnar morphology (Figure 1d). In the middle region of the layer, the reduced thermal gradient and enhanced constitutional supercooling promote cellular–dendritic growth, with fine secondary phases and a few pores observed between dendrites, as shown in Figure 1e. In the multi-pass overlap zones, banded remelting/re-solidification features appear, indicating secondary refinement and orientation adjustment induced by inter-pass thermal cycling. Overall, sound metallurgical bonding is achieved, and only a narrow dilution zone is formed at the interface, as shown in Figure 1f.

The elemental distribution across the substrate/coating interface is shown in Figure 2. Fe exhibits the highest intensity on the substrate side and gradually decreases across the fusion line into the overlay. Ni and Cr are significantly enriched within the overlay and decrease toward the substrate, while Mn and Si are nearly uniformly distributed in the coating. Refractory elements W and Mo appear as weak but homogeneous background signals, indicating their segregation into interdendritic regions during solidification. C shows localized band-like enrichment near the fusion line and along interdendritic areas, in accordance with secondary phases, implying the preferential formation of local carbides. No pronounced spike-like segregation of a single element or continuous brittle interlayer is observed, confirming that the low heat input of the CMT process promotes stable chemical dilution and sound metallurgical bonding.

The EBSD results of the interfacial region are shown in Figure 3. The band-contrast and orientation maps indicate that grains in the overlay near the interface grow epitaxially along the build normal, forming a pronounced texture, whereas grains in the heat-affected zone (HAZ) of the substrate are refined with dispersed orientations, as shown in Figure 3a. The phase distribution map in Figure 3c shows that the overlay is mainly composed of fcc Ni, while the substrate and HAZ consist of bcc α-Fe. Small amounts of Fe₃C and Fe₂Nb are detected along the fusion line and in interdendritic regions. Their volume fractions are listed in

Table 2. Volume fractions of phases and corresponding colors in the phase map.

Phase	α-Fe	Ni	TiC	Fe2Mo	Cr7C3	Fe3C	Fe2Ni
Vol. %	34.3%	60.1%	0.1%	0.3%	0.7%	0.5%	0.1%
color

, and these phases appear predominantly as fine granular or short stripe-like particles. Grain boundary statistics, as shown in Figure 3d reveal that high-angle grain boundaries (HAGBs, >15°) account for approximately 52.3%, while low- and medium-angle boundaries represent about 47.7%. The relatively high HAGB fraction is beneficial for crack blunting and toughness enhancement of the overlay and is attributed to grain refinement induced by multi-pass overlap and the associated remelting-recrystallization process.

Using the same CMT parameters, a geometrically uniform double-layer multi-pass ERNiFeCr-1 overlay was deposited on the L415QS substrate, as shown in Figure 4a. Macroscopically, the first (Clad layer 1) and second layer (Clad layer 2) exhibit clear boundaries without lack-of-fusion defects. The total thickness is approximately 5.0–5.5 mm, and the bead width of the second layer is about 6.15 mm. Compared with the single-layer overlay, the double-layer configuration shows a fuller surface profile and more pronounced remelting bands in the overlap regions, indicating a stronger interpass thermal cycle. Near the fusion line between the substrate and the first layer, grains grow epitaxially along the heat flow direction, forming columnar to cellular–columnar structures, as shown in Figure 4b,c. In the middle of the first layer, the microstructure transforms into cellular–dendritic morphology. At the interface between the first and second layers, as shown in Figure 4d–f, a continuous and dense remelted band is observed, with sound metallurgical bonding. The dendrite arm spacing in the second layer is clearly smaller than that in the first layer, and fine equiaxed grain islands appear near the interface. This is attributed to the heat flow directed into Clad layer 1 during deposition of the second layer, which promotes grain coarsening in the underlying layer while refining the newly solidified upper layer. Overall, the upper columnar grains in the double-layer coating exhibit a higher aspect ratio and stronger preferred orientation than those in the single-layer coating.

Elemental mapping of the region between Clad layer 1 and Clad layer 2 is shown in Figure 5. Fe, Ni, Cr, Mn, Si, W, and Mo display an approximately uniform distribution across the interlayer region, without continuous enrichment or depletion bands. This is consistent with the graded “Fe-decreasing into the overlay/Ni,Cr-decreasing toward the substrate” behavior observed at the substrate/coating interface in the single-layer specimen, and indicates that the second layer induces sufficient remelting and mixing of the upper part of the first layer, eliminating potential compositional jumps. Carbon shows only slight interdendritic enrichment, and no continuous carbide layer is detected, which reduces the risk of brittle interlayer formation. In general, the chemical homogeneity at the interlayer interface in the double-layer coating is superior to the graded transition at the substrate/coating interface of the single-layer coating, which is beneficial for interlayer strength and resistance to cracking.

EBSD analysis of the substrate/Clad layer 1 interface, as shown in Figure 6a–d shows pronounced epitaxial growth and strong texture in the first layer near the fusion line. The substrate is mainly α-Fe (bcc), whereas Clad layer 1 is dominated by fcc Ni. Discrete Fe₃C and Fe₂Nb particles, together with minor Fe₂Mo/Cr₇C₃, are observed along the fusion line and interdendritic regions, mostly as granular or fine stripe-like phases decorating interdendritic areas or local interfaces. In the interfacial region between Clad layer 1 and Clad layer 2, as shown in Figure 6e–h, large columnar grains with competitive growth appear on both sides of the curved remelted band, indicating enhanced preferential growth of the second layer under relatively higher heat input and lower thermal gradient. The phase constitution remains as an fcc-Ni matrix with a small fraction of interdendritic carbides/intermetallics, and no continuous brittle layer is observed. Grain boundary misorientation maps show that HAGBs (>15°) form a dense network within the recrystallized refinement band at the interlayer region, while they are relatively sparse inside the upper columnar grains. Compared with the single-layer coating (HAGBs ≈ 52%), the double-layer coating exhibits a slightly higher HAGB fraction within the refined interlayer band and a slightly lower fraction in the coarse columnar zone, reflecting a cooperative “crack-arresting refined band and load-bearing columnar region” behavior.

ERNiCrMo-3 was first deposited as a transition layer on L415QS pipeline steel, followed by an ERNiFeCr-1 overlay, forming a three-layer “substrate/interlayer/overlay” structure, as shown in Figure 7a. As shown in Figure 7b,c, the substrate/interlayer interface exhibits epitaxial columnar to cellular–columnar growth along the heat flow direction, with a dense fusion line and no lack-of-fusion or cracking. Within the interlayer, as shown in Figure 7d, the grains are finer and the orientations are more uniformly distributed than in the single-layer coating. At the interlayer/overlay interface, as shown in Figure 7e,f, a continuous remelted band is observed. Above this band, the ERNiFeCr-1 layer is dominated by cellular–dendritic grains with locally distributed equiaxed grains as shown in Figure 7g,h. Compared with the double-layer same-alloy configuration (ERNiFeCr-1/ERNiFeCr-1), the upper columnar grains show a slightly lower aspect ratio and weaker preferred orientation, and the interlayer refinement band is more pronounced. This indicates that the transition layer effectively modifies the G/R ratio and solute undercooling conditions in the upper layer and suppresses excessive competitive columnar growth.

The elemental distributions at the substrate/interlayer and interlayer/overlay interfaces are shown in Figure 8 and Figure 9. From the substrate to the interlayer, Fe decreases smoothly, while Ni and Cr gradually increase; Mo is clearly enriched on the interlayer side, consistent with the high-Mo composition of ERNiCrMo-3. C, Mn, Si, W show no continuous enrichment bands and only slight fluctuations in interdendritic regions, as shown in Figure 8. Compared with the steep chemical gradient at the substrate/coating interface in the single-layer specimen, the introduction of the interlayer redistributes the gradient into a controlled “substrate ↔ interlayer” region, thereby reducing the stress concentration and embrittlement risk associated with directly depositing a Ni-based coating on steel. From the interlayer to the overlay, as shown in Figure 9, the overall distribution remains uniform. Ni and Cr are maintained at high levels in both layers, Mo decreases from the interlayer into the ERNiFeCr-1 overlay which is a low-Mo system, and Fe increases mildly, without spike-like enrichment or continuous depletion bands. Relative to the “double-layer same-alloy” coating, this system achieves comparable interlayer chemical homogeneity while introducing a beneficial Mo gradient, which enhances resistance to localized corrosion.

The EBSD results are shown in Figure 10. At the substrate/interlayer interface, as shown in Figure 10a–d, pronounced epitaxial columnar growth and strong texture are observed in the interlayer adjacent to the fusion line, while the heat-affected zone in the substrate is refined with dispersed orientations. Phase maps confirm α-Fe with a bcc structure in the substrate and fcc Ni in the interlayer. Within the interlayer, as shown in Figure 10e–h, clusters of columnar and lath-like grains intersect, and a high density of high-angle grain boundaries (HAGBs) is present, indicating that the combination of low heat input and high alloy content promotes recrystallization and orientation redistribution, which is beneficial for crack path deflection. At the interlayer/overlay interface, as shown in Figure 10i–l, fan-shaped competitive columnar grains form on both sides of the curved remelted band, but their size and texture intensity are weaker than those of the upper columnar zone in the “double-layer same-alloy” coating, demonstrating a “softening” effect of the transition layer on the solidification conditions of the overlay. Both the interlayer and overlay are mainly composed of fcc Ni, with a small amount of discrete granular or fine stripe-like second phases (Fe₂Nb, Fe₂Mo, Cr₇C₃, Fe₃C, etc.) distributed in interdendritic regions. No continuous brittle networks are observed. Compared with the single-layer direct Ni-based deposition, these phases are more discretely dispersed; compared with the double-layer same-alloy coating, Mo-containing precipitates are more concentrated in the interlayer and significantly reduced in the overlay, consistent with the designed elemental gradient.

In summary, the ERNiCrMo-3 interlayer combined with the ERNiFeCr-1 overlay produces a multiscale hierarchical structure characterized by epitaxial refinement at the substrate, chemical buffering in the interlayer, interfacial remelting-induced refinement, and moderated competitive growth in the overlay. The beads exhibit regular geometry (single-pass width ~6.15 mm) and dense metallurgical bonding. Chemically, a controlled Fe-decreasing and Ni/Cr-increasing gradient with Mo enrichment confined to the interlayer effectively suppresses local embrittlement. Microstructurally, the system is dominated by an fcc-Ni matrix with dispersed carbides/intermetallics, without continuous brittle networks. Compared with single-layer direct Ni-based deposition, this design alleviates the interfacial compositional jump and excessive columnar growth; compared with the double-layer same-alloy configuration, it maintains similar interlayer homogeneity while optimizing the corrosion and cracking resistance through tailored Mo partitioning. This provides a clear processing route for tuning texture and grain boundary networks via coupled “interlayer alloying + interpass thermal control.”

To further clarify the solidification characteristics of the three architectures, the EBSD grain boundary data and etched cross-sectional micrographs were quantitatively analyzed. The area fraction of high-angle grain boundaries (HAGBs, misorientation > 15°) was obtained from EBSD boundary maps, and the secondary dendritic arm spacing (SDAS) was measured on longitudinal sections by the linear-intercept method. For each coating, three characteristic regions along the build direction were considered: (i) the fusion-line/HAZ region adjacent to the L415QS substrate, (ii) the interfacial refined or remelted band, and (iii) the near-surface functional region that directly participates in sliding wear. In the single-layer ERNiFeCr-1 overlay, the HAGB fraction increases from approximately 47.5% at the fusion line to 53.2% in the near-surface cellular–dendritic zone, while the SDAS decreases, reflecting progressive refinement during multi-pass solidification. In the double-layer same-alloy coating, the remelted interlayer band exhibits a relatively high HAGB fraction of around 62.4% and a fine SDAS, whereas the upper columnar zone of Clad layer 2 shows a lower HAGB fraction and a coarser SDAS, which is consistent with its strong texture and pronounced interdendritic segregation. For the ERNiCrMo-3 interlayer + ERNiFeCr-1 overlay, both the interlayer refinement zone and the overlay surface show comparatively high HAGB fractions (≈68.2%) together with the smallest SDAS values among the three designs, indicating a dense HAGB network combined with refined dendrite arms that is expected to facilitate crack deflection and homogenize plastic deformation during sliding.

3.2. Friction-Wear Behavior

The evolution of friction coefficient as a function of time for the three coatings is shown in Figure 11. Sample 3 (interlayer + overlay) exhibits the lowest and most stable friction coefficient in the initial stage, with values consistently lower than those of Samples 1 and 2. A slight decrease and plateau appear in the intermediate stage, followed by a marked increase and pronounced fluctuations after approximately 10³·s. Sample 1 (single-layer overlay) maintains an intermediate friction level with small fluctuations over the whole test duration, indicating relatively stable tribological behavior. In contrast, Sample 2 (double-layer same-alloy overlay) shows a significant increase in friction in the mid-to-late stage, accompanied by large oscillations, suggesting unstable transitions in the sliding process. Worn surface observations confirm these tendencies: Sample 1 is characterized by continuous parallel grooves with limited adhesion and fine spallation, as shown in Figure 12a,a1. Sample 2 shows pronounced lips at groove edges, smeared transfer layers, and localized delamination and fatigue cracks along the sliding direction, as shown in Figure 12b,b1. Sample 3 exhibits a compact glaze-like layer composed of dense flaky/blocky oxide–transfer films, intersected by through-thickness cracks and local spalled regions, indicating brittle fracture and re-exposure after prolonged sliding, as shown in Figure 12c,c1.

These features correlate well with the underlying microstructures. The near-surface region of the single-layer coating consists of a hierarchical “epitaxial columnar→cellular–dendritic→locally fine equiaxed” structure. Dispersed Fe₃C/Fe₂Nb particles in the Ni-Cr matrix and a relatively high fraction of high-angle grain boundaries (HAGBs ≈ 52%) enhance resistance to plowing and promote crack deflection. As a result, the friction coefficient remains moderate with limited fluctuation, and the dominant wear mechanisms are mild abrasive wear with local adhesive wear. In the double-layer coating, the upper layer contains long, strongly textured columnar grains. Although the interlayer remelted band refines the microstructure locally, the pronounced orientation anisotropy and interdendritic segregation (chain-like carbides along dendrite arms) in the top layer facilitate adhesion–fatigue coupling under tangential loading. The wear process evolves from initial abrasive grooves to rapid formation and periodic delamination of adhesive transfer layers, leading to an increased and highly fluctuating friction coefficient, as well as large-scale spalling and lip formation. This trend is consistent with the higher near-surface HAGB fractions and smaller SDAS measured in the interlayer-containing coating compared with the single- and double-layer overlays.

For Sample 3, the ERNiCrMo-3 interlayer redistributes the substrate-to-coating chemical gradient into a smooth Fe-decreasing/Ni,Cr-increasing profile with Mo enrichment in the interlayer. This reduces interfacial embrittlement and suppresses excessive columnar growth in the ERNiFeCr-1 top layer. The near-surface microstructure consists of cellular–dendritic grains superimposed with an interfacial refinement band. During sliding, Mo promotes the formation of a dense tribo-oxide/glaze layer containing Cr₂O₃/NiO and MoO₃. This layer exhibits low shear strength, which explains the low and stable friction coefficient in the early and intermediate stages. With increasing sliding time and cyclic thermo-mechanical loading, the glaze thickens, develops through-cracks, and undergoes brittle spallation (blocky delamination in Figure 12(c1)), causing the abrupt rise and fluctuation of friction at later stages. Accordingly, the dominant wear mechanism for Sample 3 is oxidative/glazing wear in the early-to-middle stage, followed by adhesion-spallation triggered by glaze fracture. The total wear volume remains the lowest among the three coatings, though control of glaze stability is required.

The overall wear resistance ranking is: Sample 3 (interlayer + overlay) > Sample 1 (single-layer) > Sample 2 (double-layer same-alloy). The superior performance of Sample 3 over an extended sliding duration arises from three key factors: (i) the interlayer provides chemical and thermophysical decoupling, mitigating substrate interfacial embrittlement and excessive columnar growth in the overlay; (ii) Mo enrichment facilitates the formation of a stable, low-shear composite tribo-oxide glaze; and (iii) the interfacial remelting refinement band and enhanced HAGB network improve crack deflection and energy dissipation. The single-layer coating benefits from dispersed second phases and HAGBs against plowing, but lacks a Mo-driven self-lubricating oxide film. The double-layer same-alloy coating is penalized by coarse columnar grains and interdendritic segregation, which accelerate adhesion-fatigue-type failure.

These findings suggest the following process guidelines: (i) prioritize the ERNiCrMo-3 interlayer+ERNiFeCr-1 overlay configuration; (ii) carefully control heat input and interpass temperature for the top layer to form a distinct but not excessive remelted band, promoting near-surface cellular–dendritic refinement and a high HAGB fraction while suppressing high-aspect-ratio columnar grains; (iii) retain sufficient Mo in the interlayer (avoid over-dilution) to support the formation of a stable MoO₃/spinel-based glaze; and (iv) employ a reasonably high overlap ratio and surface finishing to reduce initial groove defects and extend the stable-glaze regime. Following these principles enables low and stable friction while mitigating late-stage glaze instability and spallation, thereby achieving an improved balance between wear resistance and resistance to cracking in service.

During sliding, the ERNiCrMo-3 interlayer facilitated the formation of a dense Mo-enriched tribo-oxide layer (composed of MoO₃, Cr₂O₃, and NiO). Since direct characterization via XPS or Raman spectroscopy was not performed, the composition attribution is supported by relevant literature: Ni-Cr-Mo alloys typically form composite oxide films containing low-shear-strength MoO₃, wear-resistant Cr₂O₃, and dense NiO during dry sliding, which collectively reduce friction and improve wear resistance [31]. Specifically, MoO₃ acts as a self-lubricating phase, while Cr₂O₃ and NiO enhance the film’s compactness and adhesion. This Mo-enriched tribo-oxide layer explains the low and stable friction coefficient in the early-to-middle stages of the interlayer-containing sample.

4. Conclusions

This study employed the Cold Metal Transfer (CMT) process to deposit ERNiFeCr-1 coatings on L415QS pipeline steel and designed three structural configurations—single-layer, double-layer same-alloy, and ERNiCrMo-3 interlayer+ERNiFeCr-1 overlay—to systematically evaluate the effects of interlayer architecture and compositional gradients on microstructure, interfacial phase constitution, and tribological behavior.

(1) CMT enables the formation of dense Ni-based overlays with sound metallurgical bonding on L415QS. In the single-layer coating, the microstructure evolves hierarchically from epitaxial columnar grains near the fusion line to cellular–dendritic and fine equiaxed grains toward the surface. In the double-layer same-alloy system, a remelted band is formed at the interlayer, but the upper layer exhibits pronounced competitive columnar growth. With the introduction of an ERNiCrMo-3 interlayer, a controlled Fe-decreasing and Ni-Cr-increasing gradient is established from substrate to interlayer to ERNiFeCr-1 overlay, Mo is mainly retained in the interlayer, the columnar texture in the top layer is alleviated, and interdendritic second phases become discretely dispersed, avoiding continuous brittle networks and spike-like segregation.

(2) Tribological comparison shows that wear resistance and frictional stability follow the following order: interlayer + overlay > single-layer overlay > double-layer same-alloy overlay. The single-layer coating is governed by mild abrasive wear with limited local adhesion. The double-layer same-alloy coating suffers from increased friction and strong fluctuations at later stages due to anisotropic columnar grains and adhesion–fatigue coupling. In contrast, the ERNiCrMo-3 interlayer system maintains the lowest and most stable friction coefficient in the early and middle stages, attributed to the formation of a dense Mo-containing tribo-oxide/glaze layer, and only exhibits an increase after glaze cracking at prolonged sliding. Accordingly, an ERNiCrMo-3 interlayer + ERNiFeCr-1 overlay configuration is recommended, with controlled heat input and interpass temperature to form a distinct but not excessive remelted band and to preserve Mo in the interlayer, thereby achieving an optimal balance between structural stability and wear performance.