Influence of Laser Energy Variation on the Composition and Properties of Gradient-Structured Cemented Carbide Layers Produced by LP-DED

Abstract

In this study, graded cemented carbide layers were fabricated using Laser Powder-Directed Energy Deposition (LP-DED) to investigate the effects of laser input energy and WC content on crack formation, compositional distribution, and hardness. Two-layer structures were formed, with the first layer containing either 30.5 wt.% or 42.9 wt.% WC and the second layer containing 63.7 wt.% WC. Crack formation was evaluated in situ using acoustic emission (AE) sensors, and elemental composition and Vickers hardness were measured across the cross-section of the deposited layers. The results showed that crack formation increased with higher laser power and higher WC content in the first layer. Elemental analysis revealed that higher laser input led to greater Co enrichment and reduced W content near the surface. Additionally, the formation of brittle structures was observed under high-energy conditions, contributing to increased hardness but decreased toughness. These findings indicate that both WC content and laser energy strongly influence the microstructural evolution and mechanical properties of graded cemented carbide layers. Optimizing the balance between WC content and laser parameters is essential for improving the crack resistance and performance of cemented carbide layers in additive manufacturing applications.

1. Introduction

Additive manufacturing (AM) has emerged as a transformative approach in modern manufacturing, enabling the fabrication of complex geometries and functionally optimized components with high material efficiency and reduced lead times [1,2,3]. Among the various metal AM technologies, Directed Energy Deposition (DED) is particularly suitable for applications requiring material reinforcement, repair, and the deposition of multi-material or gradient structures due to its ability to selectively deposit material in a layer-by-layer manner [4,5]. Laser Powder-Directed Energy Deposition (LP-DED), a subclass of DED, utilizes a high-power laser to melt metal powders as they are delivered into the melt pool. This technique provides precise control over the geometry and composition of the deposited layers, making it well-suited for creating composite overlays containing hard ceramic phases, such as cemented carbides reinforced with tungsten carbide (WC) particles [6]. These overlays are increasingly used in wear-resistant tooling applications, such as molds, dyes, and cutting tools, where extended service life and cost efficiency are critical.

To achieve high performance while maintaining low production costs, a widely adopted strategy is to deposit cemented carbide layers only on the surface of inexpensive steel substrates [7]. However, a major challenge associated with such dissimilar material combinations lies in the significant mismatch of thermophysical properties, such as the coefficients of thermal expansion and thermal conductivity. These mismatches often lead to the formation of residual stresses and, consequently, to cracking during or after deposition [8,9,10]. To mitigate these issues, the concept of functionally graded materials (FGMs) has been proposed. In FGMs, the composition of the deposited material is gradually varied across the thickness of the layer, reducing the abrupt transition in properties at the material interface. This gradation helps to alleviate residual stress concentration and enhances interfacial bonding, thus reducing crack formation and delamination [11,12]. Recent research has shown that WC-based gradient cemented carbide layers manufactured using LP-DED can significantly improve adhesion and mechanical performance while suppressing cracks [13]. Despite these advancements, the formation of high-quality FGMs using LP-DED remains a complex task, influenced by multiple process parameters. Among these, laser input energy plays a crucial role, as it governs the size and stability of the melt pool, the extent of powder dilution, and the solidification dynamics—all of which affect the final microstructure, composition uniformity, and defect formation [14,15]. Additionally, the WC content in each layer directly influences the hardness, wear resistance, and thermal stress behavior.

Although previous studies have investigated the influence of process parameters on residual stress, tribological behavior, and microstructural evolution in LP-DED-processed materials [16,17,18], the interplay between WC content and laser energy input has not been fully elucidated in literature, especially in the context of producing crack-free graded cemented carbide overlays. In particular, the effect of laser energy input on melt penetration during the fabrication of graded cemented carbides and its subsequent influence on material properties and microstructural characteristics have not been clearly understood.

Therefore, in this study, functionally graded cemented carbide layers with systematically varied WC contents were fabricated on steel substrates using LP-DED. The influence of laser input energy on crack formation, microstructural evolution, compositional distribution, and hardness was thoroughly investigated. The results provided valuable insights into optimizing LP-DED process parameters for the fabrication of robust, crack-resistant graded composites suitable for tooling applications.

2. Materials and Methods

To fabricate two-layer graded cemented carbide layers with suppressed crack formation, it is first necessary to select an appropriate WC content for the first layer that does not induce cracking. Yamashita et al. conducted a 10 mm cladding bead formation on SKH51 (ISO HS-6-5-2 [19]) steel substrates using seven types of WC-Co granulated powders with different WC contents and found that cracks did not occur when the WC content was 42.9 wt.% or lower [13].

Based on these findings, two WC content levels, 30.5 wt.% and 42.9 wt.%, were selected for the first layer in this study to form a graded composite structure. Additionally, a WC content of 63.7 wt.% was selected for the second layer to mitigate rapid changes in material properties.

A multi-beam LP-DED system capable of emitting six laser beams simultaneously was used in the experiment (ALPION Series, Muratani Machine Inc., Ishikawa, Japan), as shown in Figure 1. The laser wavelength was 975 nm, and the focused laser spot diameter was 0.3 mm. To investigate the effect of laser input energy on the formation of the graded composite layer, experiments were conducted under different laser power conditions. The experimental conditions are summarized in

Table 1. Process parameters used in LP-DED experiments.

Parameter	Unit	Value
Laser power	W	80	100	120	140
Scan Speed	mm/s	20
Powder Feed Rate	mg/s	20
Laser wavelength	nm	975
Focused spot diameter	mm	0.3
Layer Thickness	mm	0.1

. The laser output powers were set at 80 W, 100 W, 120 W, and 140 W, with a constant feed rate of 20 mm/s.

SKH51 (ISO HS-6-5-2), with dimensions of 20 mm × 40 mm × 60 mm, was used as the substrate material. As shown in Figure 1, a 10 mm long cladding bead was deposited onto the substrate, and two layers were stacked with a pitch of 0.1 mm to form the graded cemented carbide layer.

Cracks formed during cladding bead deposition were monitored using acoustic emission (AE) sensors, which were mounted on the side of the substrate. Crack events were identified by the occurrence of burst-type AE signals corresponding to crack initiation.

Cross-sectional elemental analyses of the graded cemented carbide layers were conducted using a tabletop SEM (scanning electron microscope) (JCM-6000Plus, JEOL Ltd., Tokyo, Japan) coupled with an EDS (energy dispersive X-ray spectroscopy) system, allowing the determination of compositional distributions across the layers. Vickers hardness (MMT-X series, Matsuzawa Co., Ltd., Tokyo, Japan) measurements were also performed on the cross-sections to assess variations in material properties associated with changes in the metallurgical microstructure.

3. Results and Discussion

3.1. Crack Evaluation of Graded Composition Cemented Carbide Layers

Figure 2 shows the cross-sectional observation of the graded cemented carbide layer formed under the conditions of 42.9 wt.% WC in the first layer, a laser power of 80 W, and a scanning speed of 20 mm/s, as observed using a digital microscope. The line extending from the surface of the layer toward the substrate represents the crack, while the spherical black regions indicate pores.

Figure 3 presents the results of evaluating the number of crack initiation events based on burst-type waveforms detected by the AE sensor during the formation of graded cemented carbide layers under each experimental condition, with error bars indicating the standard deviation. In the first layer, no crack initiation was detected for the 30.5 wt.% WC condition at a laser power of 80 W.

Experimental results showed that the number of cracks was lower under the 30.5 wt.% WC condition than under the 42.9 wt.% WC condition. For both WC contents, the number of cracks increased with higher laser power, likely due to the increased energy input affecting the metallurgical behavior of the deposited layers.

3.2. Evaluation of Elemental Composition and Hardness Characteristics

To understand the cause of the variation in crack formation, the elemental composition across the cross-section of the graded cemented carbide layers was analyzed. Figure 4 shows a cross-sectional SEM image of the compositionally graded cemented carbide layer under the condition of 30.5 wt.% WC in the first layer. The cladding was performed under a laser power of 80 W and a scan speed of 20 mm/s. Elemental analysis was conducted at 50 μm intervals at the marked cross positions in the image. Figure 5 presents elemental analysis results obtained at 50 μm intervals from the interface between the substrate and the deposited layer, under the condition of 30.5 wt.% WC in the first layer. The data represent average values measured with a resolution of 512 × 384 pixels over an analysis area of 37 μm × 50 μm.

With increasing laser power, the W content near the surface decreased, whereas the Co content increased. At lower laser power, the W gradient was more gradual, while at higher power levels, the W content rose sharply from the interface. Notably, the point at which the W content surpassed that of Co shifted closer to the substrate as the laser power increased.

Figure 6 presents elemental analysis results obtained at 50 μm intervals from the interface between the substrate and the deposited layer, for the condition in which the first layer contained 42.9 wt.% WC. The data represented average values measured with a resolution of 512 × 384 pixels over an analysis area of 37 μm × 50 μm. Consistent with the 30.5 wt.% WC case, W content near the surface decreased, and Co content increased as the laser power increased.

In both cases, Co content remained lower than the W content from the interface to the surface, though the gap between them narrowed with increasing laser power.

The differences in composition ratios at each position under varying laser powers were attributed to the significant influence of melting depth induced by laser heating.

When a circular and uniformly distributed heat source was applied to the surface of a semi-infinite solid during laser thermal processing, the temperature rise from the material surface to a certain depth along the center of the heat source can be expressed by the following equation:

$$T\left(z,t\right)={\displaystyle \frac{4AP}{\pi K{D}^2}}\sqrt{4\alpha t}\left(ierfc\sqrt{{\displaystyle \frac{z^2}{4\alpha t}}}-ierfc\sqrt{{\displaystyle \frac{z^2+{\left(\frac{D}{2}\right)}^2}{4\alpha t}}}\right),$$

(1)

where A is the absorption coefficient at the material surface, P (W) is the laser beam power, K (W/(m·K)) is the thermal conductivity, D (mm) is the spot diameter of the heat source, α (m²/s) is the thermal diffusivity, z (mm) is the distance from the material surface into the material, and t (s) is time. Therefore, assuming that the melting depth increased with the temperature, it can be considered proportional to the absorption rate and inversely proportional to the thermal conductivity. The laser absorption coefficients at a wavelength of 1 µm were 0.82 for WC and 0.38 for Co [20,21], whereas the thermal conductivities were 29.3 W/(m·K) for WC and 99.2 W/(m·K) for Co [22,23]. Accordingly, it was inferred that an increased WC ratio results in a higher temperature within the overlay cladding, thereby leading to a greater melt-in depth.

To evaluate the effect of varying composition percentages on material properties, the hardness of the deposited layer was measured. Hardness values at distances of 30 μm and 150 μm from the interface between the substrate and the deposited layer, with a WC content of 30.5 wt.% in the first layer, are presented in Figure 7. At a distance of 30 μm from the interface, the Vickers hardness increased with increasing laser power. This behavior may be attributed to the greater melting depth associated with higher laser power. At a distance of 150 μm from the interface, no significant change in hardness was observed with varying laser powers. Figure 8 presents the Vickers hardness measurements at 30 μm and 150 μm from the interface between the substrate and the deposited layer, with a WC content of 42.9 wt.% in the first layer. At 30 μm, the hardness exhibited an increasing trend with higher laser power, and a similar tendency was observed at 150 μm.

The elemental analysis results at a distance of 150 μm, shown in Figure 5 and Figure 6, indicated a slight decreasing trend in the W content with increasing laser power. Given that the hardness of WC was 2440 HV [24], it was expected that the hardness would decrease with increasing laser power; however, the Vickers hardness measurements exhibited the opposite trend.

Cemented carbides are known to form brittle metallographic phases, such as W₂C and W₃Co₃C, with increasing melting temperature [25,26]. To clarify the influence of these phases, SEM observations were carried out. Figure 9 shows the SEM image acquired using the BED (backscattered electron detector) and the EDS elemental mapping results for the first layer with 30.5 wt.% WC. The area shown in the image corresponds to the location of the Vickers indentation made during the hardness measurement. Elemental analysis revealed that the white regions consisted solely of W, indicating that they corresponded to WC. According to elemental analysis, the Co-rich regions exhibited dark areas and regions of intermediate contrast relative to the surrounding light and dark areas. The dark regions were attributed to metallic Co, whereas the intermediate contrast regions were identified as dendritic structures. Fine dendritic structures were observed in the samples fabricated at 100 W and 120 W, whereas the sample processed at 140 W exhibited a coarser morphology with more pronounced intermediate contrast regions. These regions, associated with dendritic structures, are believed to correspond to the formation of the brittle structures M₆C(W₃Co₃C, W₄Co₂C) and M₁₂C(W₆Co₆C) [27]. The increased heat input at 140 W is likely to induce enhanced constitutional supercooling and solute segregation, particularly favoring the enrichment of Co and W in localized regions. These conditions are conducive to the formation of brittle structure intermetallic compounds, which typically precipitate when the local composition and thermal gradients satisfy the critical conditions for nucleation. The observed coarsening of dendritic structures further supports the hypothesis of slower solidification kinetics at higher laser powers.

Figure 10 shows the SEM image acquired using the BED (backscattered electron detector) and the EDS elemental mapping results for the first layer with 42.9 wt.% WC. The area shown in the image corresponds to the location of the Vickers indentation made during the hardness measurement. Unlike the layer with 30.5 wt.% WC content, the samples processed at 120 W and 140 W did not exhibit dendritic structures. Instead, the intermediate contrast regions observed between the light and dark areas were composed of large crystals. Elemental mapping confirmed the presence of both W and Co, indicating the formation of the brittle structures. The high WC content at 42.9 wt.% is expected to significantly modify the thermophysical properties of the melt, such as increasing its viscosity and altering thermal gradients. These changes may stabilize the solid–liquid interface and suppress the constitutional supercooling typically required for dendritic growth. Moreover, the enrichment of solid particles within the melt could act as heterogeneous nucleation sites for equiaxed crystals while limiting solute segregation and promoting the formation of intermetallic brittle structure compounds.

These observations suggest that extensive precipitation of the brittle structures occurred in regions beyond the original WC and Co phases and that the transformation of Co-rich areas into the hard brittle structures contributed to the observed increase in hardness. Moreover, because the formation of such microstructures is typically associated with reduced toughness, it is likely that the increased laser input energy led to a higher incidence of crack formation.

It is well known that an increase in WC content and temperature significantly affects the precipitation of brittle phases [28,29]. In the present experiment, the first layer with a higher WC content of 42.9 wt.% exhibited a greater amount of brittle phase formation. Additionally, an increase in laser power led to a larger area of brittle phase formation.

Therefore, increasing the WC content in the graded composition contributed to a greater amount of brittle phase formation, and higher laser power not only influenced the melting depth but also promoted the formation of brittle phases. To reduce brittle phase formation during the fabrication of graded cemented carbide layers, it is important to limit the input laser energy. However, since sufficient bonding with the underlying layers must be ensured, further investigation is required to determine the appropriate level of laser energy input.

4. Summary

This study investigated the fabrication of functionally graded cemented carbide layers using Laser Powder-Directed Energy Deposition (LP-DED), focusing on the effects of WC content and laser input energy on crack formation, microstructure, and material properties. It was revealed that using a lower WC content of 30.5 wt.% in the first layer significantly suppressed crack formation, compared to a higher content of 42.9 wt.%. In both cases, increasing the laser power led to a higher frequency of crack initiation events, likely due to increased thermal input affecting melt pool behavior and solidification dynamics, thereby promoting microstructural instability and residual stress accumulation.

Elemental analysis showed that higher laser power decreased W content and increased Co content near the surface of the deposited layers. This was attributed to differences in laser absorptivity and thermal conductivity between WC and Co, with higher WC content resulting in deeper melt penetration. Hardness measurements indicated that Vickers hardness increased with higher laser power, particularly near the interface between the substrate and the deposited layer. This behavior was considered to result from the formation of brittle intermetallic compounds such as W₂C and M₆C at elevated temperatures. SEM observations confirmed the presence of these brittle phases, with their morphology and distribution varying significantly depending on WC content and laser power. In particular, higher WC content and greater laser input promoted the precipitation of brittle structures, contributing to increased hardness but also reducing toughness and increasing the susceptibility to cracking.

To achieve reliable fabrication of functionally graded cemented carbide layers, it is essential to carefully control laser input energy and appropriately select WC content. The findings of this study provide valuable insights for optimizing LP-DED process parameters to produce crack-resistant graded composites suitable for tooling applications. Future work should aim to further optimize process conditions to suppress the formation of brittle phases while maintaining the desired hardness and wear resistance, potentially through the implementation of real-time process monitoring and advanced quality control strategies.