An investigation on graphite behavior and coating properties in the molten pool based on different powder particle sizes

Abstract

This paper aims to explore the influence mechanism of powder particle sizes on the microstructure and properties of coatings and identify the effect of powder particle size difference on the coating graphite phase. NbC-reinforced Ni-based coatings were in-situ synthesized by laser cladding to investigate the impact of powder particle sizes on the morphology, structure, and properties of coatings. The results indicate that increasing powder particle size enlarges the coating area and decreases the coating width and dilution ratio. Meanwhile, the defect ratio first increases and then decreases. The XRD test suggests that the coating mainly consists of NbC, solid solution (Fe, Ni), B₄C, Cr₂C, and CrB_2. Different powder particle sizes do not change the phase composition of coatings but affect the graphite phase morphology. The morphology transforms from spherical to flocculent as the powder size varies from micrometer to nanometer. The hardness of coatings gradually increases, and the friction and wear properties decrease with the growth of powder particle size. The dispersed graphite phase in the nano coating plays a self-lubricating role in the friction and wear process. This research provides a reference and theoretical basis for selecting powder particle size in laser cladding.

1. Introduction

Laser cladding is a surface modification technology [1, 2, 3, 4], which has strengths of material saving, small heat-affected zone, and high efficiency and reliability [5, 6, 7] compared with traditional technology. In laser cladding, powder materials are critical in changing the properties of cladding layers. Powder materials usually include carbide (NbC, WC, and SiC), oxide (Al₂O₃), and boride (B₄C) [[8], [9], [10]]. Among them, B₄C is a common boride reinforcement phase featuring high hardness, low density, and high chemical inertia. This powder material has been widely used in aerospace and medical fields. B₄C can be decomposed to produce B and C atoms to react with Ti and Nb, generating hard or wear-resistant phases (such as TiC, NbC, and TiB₂). Zhao et al. [11] prepared TiC/B₄C/Ni204-based coatings by laser cladding. They found that the TiB₂ phase was successfully synthesized in situ in the composite coating with 30% B₄C, 5% TiC, and 65% Ni204. Meanwhile, graphite was formed in the coating, improving significantly the average coating hardness and wear resistance. Zhang et al. [9] fabricated FeCoCrCrNiMnTi_x (B₄C) _y high entropy alloy (HEAs) coatings on Q355 steel using laser cladding. The addition of B₄C created composite borides and carbides and reduced Ti solution in the fcc and bcc phases. As B₄C content grew, the F FeCoCrCrNiMnTi_x (B₄C) _y HEAs coating was improved in terms of hardness, bcc content, and wear resistance. Wang et al. [12] adopted laser cladding to prepare Fe-matrix composite layers on a steel substrate by adding Nb and B₄C powders. The phases in the composite layer mainly included a-Fe, Fe₂B, Cr₇C₃, and Cr₂₃C₆. The addition of Nb and B₄C led to the in-situ formation of NbC phase and the increase in the amount of Fe₂B phase, refining the microstructure in the Fe-matrix composite layer.

Studies of powder materials have found that not only the choice of materials but also powder sizes significantly impact coating properties. Due to the strengths of Nanomaterials [13,14], such as large specific surface area and high chemical activity, the coating performance can be improved by adding nano-particles. For example, Li et al. [15] prepared micro- and nano-structured WC reinforced Co-based cladding layers on Ti6Al4V substrates. They found that reducing particles to nano size significantly improved the wear resistance of the cladding layer, indicating a positive impact of particle size on the performance of the cladding layer. Perrin et al. [16] produced tungsten carbide-doped NiCrBSi coatings on steel by laser cladding. The coating hardness increased with increasing particle size of carbide powder. The minimum particle size of carbide powder, ranging from 20 to 53 μm, led to the lowest tip wear and mass loss. So, the wear resistance decreased with the increase in particle size. Li et al. [17] fabricated Fe-based coatings with different-sized WC particles on the surface of 16Mn steel. A change in particle size of WC had an effect not on the phase composition but on the microhardness and wear resistance. Xu et al. [18] used B₄C nanoparticles to strengthen IN625 powder. The reinforced powder refined the dendritic structure, causing the hardness of reinforced layers to increase by 64% and the high-temperature wear loss to decrease by 86%. The result indicated that B₄C nanoparticles enhanced the properties of the cladding layer. Carbon-rich areas were formed in the cladding layer.

Currently, few studies have focused on the influence of different particle sizes on the coating performance. The mechanism for the morphology of carbon-rich areas in coatings is still unknown. Therefore, this paper uses different-sized B₄C and Nb powders to synthesize NbC in situ. The paper explores the influence of different-sized powders on the microstructure evolution and morphology of carbon-rich areas in coatings, and reveals the effect of powder particle sizes on the coating properties. The research results provide a reference and theoretical support for selecting powder particle sizes in laser cladding in-situ synthesis of NbC.

2. Method

2.1. Experimental design and materials

Powders used in the experiment include Ni45 powder with particle size from 80 to 110 μm, Nb powder with 99.9% purity, and B₄C powder with 99.9% purity. Table 1 provides the chemical composition of Ni45 powder. These powders were mixed and milled in the planetary ball mill for 2 h. The powder ratio of B₄C (Fig. 1(b)) and Nb (Fig. 1(a))is 1–1.3 (Atom ratio). Ni45 (Fig. 1(c)) powder accounts for 80 wt% of the total volume of the mixed powders. Then, a five-percent PVA solution was used to pre-place the mixed powders in the 45# steel substrate with dimensions of 40 mm × 20 mm × 10 mm. Finally, the preset substrate was put into a dryer at 80 °C to dry for 2 h to remove moisture from the powders. The powder morphology and particle size diagram are shown in Fig. 1(a–c).

Table 1.

Chemical composition of Ni45 powder (mass fraction, %).

Elements	C	Si	O	Fe	Cr	B	Ni
Ni45	0.32	3.35	<0.05	2.75	7.75	1.65	Balance

Fig. 1.

Powder morphology and particle size diagram (a) Nb (b) B₄C (c) Ni45.

This experiment adopted the single-factor method. The preliminary experiment was conducted to identify three parameters, including laser power, scanning speed, and gas flow. Meanwhile, the particle sizes of Nb and B₄C powders were varied. There were five groups of tests in this research. Table 2 lists the levels of the experimental factors.

Table 2.

Experimental factors and levels.

No.	Laser power kW	Scanning speed mm/s	Powder particle sizes	Gas flow NL/min
S1	1.8	5	50 nm	6
S2	1.8	5	500 nm	6
S3	1.8	5	1 μm	6
S4	1.8	5	50 μm	6
S5	1.8	5	100 μm	6

2.2. Equipment

Fig. 2 illustrates the laser cladding system, including Fiber laser (YLS-3000, IPG, Germany), Pneumatic powder feeder (GZ-DPSF-2, China), Laser water cooler (TFLW-4000WED-01-3385, China), Laser cladding head (FDH0273/f = 300, lasermech, USA), PLC machine controller (Mitsubishi, Japan), and multi-degree-of-freedom industrial manipulator (M-710Ic/50, Fanuc, Japan). Ar was selected as the shielding gas during cladding.

Fig. 2.

Laser cladding system.

The cladding samples were corroded in 4% nitric acid for 30 s after wire-electrode cutting, mounting, and polishing. Optical fiber measuring equipment (Hirox Co Ltd, Tokyo, Japan) was used to measure the geometric characteristics of the coating, including the area (A), width (W), dilution ratio (η), and defect rate (D). The dilution ratio and defect rate were calculated by Equations (1), (2), respectively.

$$\eta =\frac{A_2}{A_1+A_2}\times 100\%$$ (1)

$$D=\frac{P}{A_1+A_2}\times 100\%$$ (2)

Where A₁ is the coating area, A₂ is the dilution area, and P is the pore area (see Fig. 3). The experiment used a scanning electron microscope (TM3030Plus, Hitachi, Japan) to measure the coating morphology and adopted the energy spectrometer (Model 550i, IXRF, America) to test the element composition and distribution. The system parameters of X-ray diffraction analysis (X-Pert Pro MPD, Netherlands) included Cu Kα radiation at 400 kV and 200 mA (λ = 0.15418 nm), 2θ ranging from 10° to 90°, a scanning step at 0.05°, and measurement time at 10 s per step. A friction and wear tester (UMT-2, Bruker, America) was used to measure the wear resistance in a reciprocating wear testing with the Tungsten steel as friction pair at 30 N loading force. The porosity defects were avoided in the performance test to establish the accuracy of the test results.

Fig. 3.

Morphology of the cladding layer.

3. Results and analysis

3.1. Macro morphology analysis of coatings

Fig. 4(a) displays the geometric morphology of the coating under different powder particle sizes. Defects such as pores and cracks were observed in the coating. The varying powder particle sizes affect the area, width, dilution ratio, and defect rate of the coating. The coating area((Fig. 4(b)) is relatively large at powder particle sizes of 50 μm and 100 μm. Equation (3) indicates the influence of powder particle sizes on laser absorptivity.

$$E_{\mathrm{a}}=\pi {\mathrm{r}}^{2}ZJ$$ (3)

Where, E_a refers to the powder particles' energy absorbed by laser radiation; r is the particle size of powders; Z is the energy absorption rate of the powder particles by laser radiation; J is the energy density of the laser beam spot. When E_a is unchanged, the energy absorption rate increases with the decrease in the powder particle size. In other words, small-sized powders have a higher energy absorption rate than large-sized ones [17]. Consequently, when the nanopowder melts to form a molten pool, more energy is absorbed to lengthen the life of the molten pool. The liquid molten pool gradually flattens. Accordingly, the molten pool of nanopowder has a larger width than that of the micron powder, see in Fig. 4(c).

Fig. 4.

Morphological analysis of S1–S5: (a) coating morphology; (b) coating area histogram; (c) melt width histogram; (d) dilution rate histogram; (e) defect rate histogram.

The dilution rate of small-sized powders is significantly higher than that of large-sized powders due to higher laser absorptivity, see in Fig. 4(d). Equation (4) provides the relationship between powder surface area and powder particle size.

$$S_{\mathrm{a}}=\frac{S}{V}=\frac{4\pi \mathrm{r}}{4\pi {\mathrm{r}}^3/3}=\frac{3}{\mathrm{r}}$$ (4)

Where S_a is the powder surface area; S is the surface area of a particle; V is the particle volume, and r is the particle radius. The surface area of the powder increases with the decrease in the powder particle size. Powders melt mainly through laser energy absorbed by the particle surface. Equation (4) indicates that owing to a larger surface area, the small-sized powder can absorb more laser energy, which causes the molten pool to increase temperature and decrease liquid surface tension. However, reducing the surface tension of the cladding material can effectively improve the dilution ratio between the cladding layer and the substrate. Therefore, small-sized powders have a higher dilution ratio than large-sized powders.

The coating has the highest defect rate at a powder particle size of 500 nm, see in Fig. 4(e). Pores emerge mainly because the C atoms produced by decomposing B₄C are oxidized to generate gas during laser cladding, leaving pores in the cladding layer. Due to the large surface area, nano B₄C has a strong absorbability during ball milling. Accordingly, the surrounding nanopowder is absorbed to form blocky nano B₄C particles. Melting these particles consumes more heat, causing insufficient fluidity in the molten pool. The pores in the molten pool cannot escape in time. So, small-sized powders increase pores in the coating [19]. When the powder sizes are 1 μm and 50 μm, cracks are generated in the coating and mostly spread along the vertical direction. The formation of cracks is related to the difference in thermophysical properties between carbide and substrate. High internal stress leads to the generation and propagation of cracks.

3.2. Structure analysis

Two groups with the largest difference in powder particle size were selected to explore the influence of powder particle size on coating phase composition. Fig. 5 provides the XRD diffraction spectrums of coatings, as the particle sizes of Nb and B₄C are 50 nm and 100 μm. Based on the figure, the coating mainly consists of NbC, solid solution (Fe, Ni), B₄C, Cr₂C, and CrB₂. Ni45 powder is decomposed to produce the Cr element and B₄C is decomposed to generate the C element. These decomposed elements react with each other to form Cr₂C and CrB₂. Fe and Ni in Ni45 powder react with 1045 steel to produce solid solution (Fe, Ni). Nb reacts with the C element to form NbC [8,20]. The phase composition of the cladding layer does not have an evident change, indicating varying the powder particle size does not affect the phase composition. The PDF card are shown in Table 3.

Fig. 5.

XRD spectra of test S1and S5.

Table 3.

PDF card of physical phase.

Number	Phase	PDF card
1	CrB2	34–0369
2	B4C	35–0798
3	NbC	38–1364
4	Cr2C	14–0519
5	[Fe,Ni]	26–0790

The EDS test was adopted to analyze the coating to clarify the morphology of each phase structure. Fig. 6 displays the EDS energy spectrum at 100 μm particle size. The figure indicates three different color particles in the cladding layer, dotted for the EDS test. The silvery white petal-shaped particles consist of Nb and C elements with an atomic ratio close to 1 : 1 (see Fig. 6(a)). Combined with the XRD results, these particles are identified as the reinforcement phase of in-situ synthesized NbC. Dark grey massive particles are mostly precipitated near NbC and present evident segregation and accumulation of Cr after dotting (Fig. 6(b)). The particles are the chromite reinforcement phase, generated by the reaction of Cr and C elements in the molten pool. NbC preferentially precipitates in the molten pool. When the liquid metal crystallizes, heterogeneous nucleation has less nucleation energy than homogeneous one [21], and the subsequently precipitated nuclei tend to attach to the surface of the first precipitated NbC to form composite phases. Fig. 6(c) provides the light grey structure, dominated by the Ni element. Some alloy elements, such as Fe and Nb, are dissolved in the Ni-based structure, playing a role of solution strengthening.

Fig. 6.

Coating microstructure and element distribution diagram of test S5.

The formation mechanism of the molten pool requires to be clarified to explore the generation of the physical phase structure, as shown in Fig. 7. First, Ni45, Nb, and B₄C powders are uniformly mixed and pre-placed on the substrate. The molten pool gradually forms by heating the laser beam. Ni45 and B₄C powders decompose to produce Cr, C, and B atoms. B and Si in Ni45 powder enable B₄C more easily dissolved in the molten pool, creating a C-rich environment [11]. Further stirring in the molten pool rearranges atoms to form multiple phases such as NbC and CrB₂. In solidification of the molten pool, NbC with low Gibbs free energy is first precipitated. Subsequently, the chrominide compounds generate since B has a higher mole fraction than C [22]. This process consumes a lot of B and Cr atoms, decreasing the elements that can react with C. So, there are unbound C atoms in the molten pool, providing a carbon source to grow the graphite phase [19]. Finally, structures, including petal shape, flocculent carbide, spherical carbide, and grey cube, are formed in the coating.

Fig. 7.

Formation mechanism of the molten pool.

3.3. Mechanism of graphite-phase changing

The observation of the coating structure indicates that the carbon-rich areas in the molten pool are usually spherical and slightly flocculent when the powder size is micron scale. Most of them are converted into flocculent structures at the nanometer scale. The two groups with the largest difference in particle size were selected for line scanning to better analyze the components. Fig. 8 presents the results of scanning the 100-μm powder. The scan starts with the dark grey Cr compound and ends with the light grey (Fe, Ni) mixture. The figure demonstrates that the carbon-rich areas contain C elements and a small amount of Fe elements, almost without Ni and Cr elements.

Fig. 8.

EDS line scan of test S5.

Fig. 9 shows the line scanning results of 50 nm powder. The scan starts from the light grey (Fe, Ni) mixture to the black carbon-rich area. The black area consists of C atoms with a slight decrease in the Fe element content. Combined with Fig. 8, changing powder particle size does not impact the composition of the carbon-rich area, except for the structure from spherical to flocculent.

Fig. 9.

EDS line scan of test S1.

Besides, the formation mechanism of the graphite phase in the molten pool was explored to identify the reason for the graphite-phase change from spherical to flocculent. The morphology of graphite precipitates differs owing to the composition segregation and temperature difference during solidification, including nodular, flake, and irregular graphites. Spherical graphite in the molten pool has a similar formation mechanism as graphite spheroidization in nodular cast iron. According to the interfacial energy theory in metal solidification, carbon atoms preferentially diffuse and combine with the sections with lower interfacial energy in the grains. However, the final morphology of graphite growth depends on the interfacial energy of the graphite edge and base planes. Spherical graphite is formed as interfacial energy is higher at the edge plane than at the base plane. The appearance of flocculent graphite is similar to the “graphite degradation” in nodular cast iron. The composition fluctuation in the molten pool causes the composition difference in regions, and some regions are not conducive to the spheroidization of the graphite phase. Simultaneously, the cooling conditions deviating from the equilibrium state in laser cladding cause the non-spheroidization of graphite [19] and finally form the flocculent graphite.

Nanopowders have a longer liquid phase life than micron powders in molten pool formation. The composition fluctuation in the molten pool brings composition differences in regions, unfavourable for the spheroidization of the graphite phase. Moreover, nanopowder can be kept in liquid for a long time due to a large surface area, increasing the decomposition of B₄C. The growing anti-spheroidizing elements, such as B, accumulate at the grain boundary, forming non-spherical graphite when carbon crystallizes in the later period. In solidification, the oxygen in the molten pool reacts with the spheroidizing elements in the liquid phase, causing the attenuation of spheroidizing elements. When the residual amount of spheroidizing elements is lower than the critical value required for graphite spheroidization, spheroidization decaying occurs [23] with the graphite phase transferring from spherical to flocculent.

3.4. Influence of different powder particle sizes on the hardness

Fig. 10 provides the coating hardness of different powder sizes. The figure indicates that varying powder particle sizes affects the coating hardness. The hardness increases with the growth of powder particle sizes, minimizing at the powder size of 50 nm.

Fig. 10.

Histogram of coating hardness for tests S1 to S5.

The coating hardness highly correlates to the morphology dimension and content distribution of the reinforcement phase. Fig. 11 exhibits the coating microstructure. In Fig. 11(a), the grain morphology is dominated by fine dendrite and flocculent graphite phases. The dendrite has no obvious directivity, and the dispersed graphite phase reduces the coating hardness. As powder particle sizes increase, the precipitated grains are mainly fine equiaxed crystals and spherical graphite phases (see Fig. 11(b)). The rounder graphite sphere leads to fewer cracks and damages in coatings. Spherical graphite reduces alloy element segregation, improving the microstructure uniformity and the coating hardness [22]. According to Fig. 11(c), the grain morphology is inclined to petal-like equiaxed crystal and dendrite with less precipitated graphite phase. More carbon elements are dissolved in the Ni-based structure to form the reinforcement phase, thus increasing the hardness. In Fig. 11(d), the coating mainly contains equiaxed crystal particles with fine grain size and without obvious directivity. This morphology is an ideal form of generating phase, according to the Hall-Petch relationship (as shown in Equation (5)) [24]:

$$\Delta {\sigma }_{\mathrm{G}\mathrm{B}}=k{d}^{-\frac{1}{2}}$$ (5)

Where k is the strengthening coefficient and d is the average grain size. Δσ_GB refers to the strengthening capacity of grain size. In the equation, d is inversely proportional to Δσ_GB, which indicates that finer grain sizes have a higher grain-strengthening capacity.

Fig. 11.

Microstructure of the coating: (a) S1; (b) S2; (c) S3; (d) S4; (e) S5.

Finer grain sizes in the coating lead to an increase in the grain and subgrain boundaries that hinder the sliding. More dislocations stack at the grain boundaries during the plastic deformation, improving the resistance to dislocations and strengthening fine grains. Based on Fig. 11(e), the precipitated grains are mainly equiaxed grains and dendrites. The varying grain morphology is an important factor for the change in coating hardness.

3.5. Influence of powder particle sizes on tribological characteristics

Fig. 12(a) shows the varying coefficient of friction (COF) of coatings in Samples 1–5 with time. COF is a significant indicator of the coating wear resistance. Generally, a higher COF value indicates a decrease in wear resistance. Fig. 12(b) presents the wear amount of coatings in Samples 1–5. The wear amount of the cladding layer increases with increasing powder particle sizes.

Fig. 12.

Coating friction and wear performance diagram of test S1 to S5 (a) COF (b) wear volume.

Fig. 13(a), (b), and (c) display the wear morphology of S1, S3, and S5. Fig. 13(a) indicates that the growth of powder particle sizes gradually increases the wear amount of the coating, with the wear mechanism shifting from adhesive to abrasive wear. The coating surface of nanopowder has an evident black carbon-rich area, which acts as a self-lubricating additive in the friction and wear process. The graphite phase is extruded and coated on the coating surface, forming a solid lubricating film. The lubricating film leads to a smoother coating surface, significantly reducing wear [25,26].

Fig. 13.

Coating wear morphology: (a) S1; (b) S3; (c) S5.

The growth of powder particle sizes gradually decreases the graphite phase (see Fig. 13(b)), weakening the self-lubricating effect. The hard phase in the coating mainly consists of in-situ synthesized NbC, providing support in the friction and wear process to reduce the contact area between the coating and the friction pair. High-speed wear causes a small part of the hard phase to peel off and act as wear debris to intensify wear, resulting in a large amount of wear debris and grooves in the coating. According to Fig. 13(b) and (c), the scratch surface is covered with a black film. The friction heat generated in the friction creates a layer of oxide film on the coating surface. The oxide film protects the worn surface from damage by the friction pair to some extent [27].

Fig. 14 exhibits the friction and wear mechanism of nanopowder coating (a) and micron-powder coating (b). Based on Fig. 14(a), the friction pair is pressed into the coating surface under the normal load. The friction movement shears the junction between the friction pair and the coating, resulting in the transfer and loss of the coating material. Here, the wear mechanism in the coating is adhesive wear. According to Fig. 14(b), the hard phase synthesized in situ in the coating plays a supporting role. Without the lubrication of the graphite phase, the intensified wear causes part of the hard phase to peel off to produce chips. The chips as grinding generate scratches on the coating surface during the wear process. At this time, the wear mechanism in the coating is abrasive wear.

Fig. 14.

Wear mechanism of powder particle sizes on coatings (a) Nanopowder coating; (b) Micron powder coating.

4. Conclusion

This paper varies the particle size of Nb and B₄C to in-situ synthesize NbC by laser cladding to investigate the influence of powder particle size on the microstructure and properties of the coating. The conclusions are drawn as follows:

• (1)The powder particle size has a significant impact on the coating morphology. As the powder particle size reduces, the melting width and dilution ratio decrease. Meanwhile, the defect ratio first increases and then decreases. This is because the small-size powder has a larger surface area to absorb more laser energy, leading to a longer-life molten pool, a larger melting width, and a higher dilution ratio of the coating.
• (2)The results of the XRD test indicate that the coating mainly contains NbC, solid solution (Fe, Ni), NbC, B₄C, Cr₂C, and CrB₂. The change in powder particle sizes does not affect the phase composition of the coating.
• (3)The change in powder particle sizes affects the morphology of the coating carbon-rich region. The carbon-rich region is spherical in micron powder and flocculent in nanopowder. As Anti-spheroidized elements aggregate in the nanopowder molten pool and the composition in the pool fluctuates, the graphite phase transforms from spherical to flocculent.
• (4)The wear volume of the coating minimizes at the powder particle size of 50 nm and maximizes at the powder particle size of 100 μm. The wear resistance decreases gradually with the increase of powder particle sizes. The wear mechanism of coatings is adhesive at the nanometer scale and abrasive at the micron scale. The carbon-rich area in the nano-powder coating forms a solid lubricating film during the friction and wear process to generate a smoother coating surface. The film plays a good anti-wear effect and increases the wear resistance of the coating.
• (5)The hardness of the substrate is 40HRC and the wear amount is 820 μm³. The hardness and the wear amount after cladding increase to about 70HRC and 260 μm³, respectively. Compared with the substrate, the hardness of the coating is improved by 1.75 times, and the wear amount is increased by 3.15 times. This result indicates that the performance of the laser cladding in-situ synthesized NbC coating is effectively improved.

Author contribution statement

Kun Yue: Performed the experiments; Wrote the paper. Guofu Lian; Jiayi Zeng: Conceived and designed the experiments. Changrong Chen; Ruqing Lan: Analyzed and interpreted the data. Linghua Kong: Contributed reagents, materials, analysis tools or data.

Funding statement

Professor Guofu Lian was supported by Science and Technology Innovation Key Project of Fujian Province [2021G02009], Fujian Industrial Industry-University Collaboration Project [2020H6028].

Data availability statement

No data was used for the research described in the article.

Declaration of interest's statement

The authors declare no competing interests.