Microstructure effect on sliding wear of 316L stainless steel selectively laser melted

Abstract

Due to varying thermal cycles, the resulting microstructure of metal additive manufacturing differs from the conventionally processed counterpart alloys. Since the mechanical properties depend on the microstructure, the wear resistance of components manufactured by laser powder bed fusion (LPBF) is determined by the processing parameters. This work focuses on microhardness and sliding wear of 316L stainless steel, evaluated nanoindentation and pin-on-disc, respectively, analysed through optical microscopy, scanning electron microscopy (SEM), X-ray diffraction (XRD) and glow discharge emission spectrometry (GDOES). The results show that the LPBF-processed specimens have about 40% higher microhardness and ca. 30% lower wear rate than the wrought counterpart. The enhanced sliding wear resistance is associated with the higher density of dislocations at the cellular subgrain boundaries.

Keywords

laser powder bed fusion microstructure stainless steel wear dislocations

Introduction

Austenitic stainless steel (SS) is one of the most used materials in additive manufacturing (AM) mainly due to its high laser absorption capacity (>30%),¹ and its high tensile strength, excellent corrosion resistance, biocompatibility and recyclability, making it high-applicable in mechanical engineering and biomedical application.² As the demand for high-performance materials increases, it is necessary to subject additively manufactured samples to tribological tests where they are expected to perform better than those obtained by conventional processes such as casting or forging.

From a tribological perspective, hardness measurements can be used to estimate wear resistance. However, its greater hardness because of the reduction in grain size is not a guarantee that it is more resistant to wear compared to conventionally processed SS.^3,4 Zhu et al.⁵ corroborated that the wear rate of a range of alloys processed by laser powder bed fusion (LPBF) in materials is inversely proportional to the material's hardness, as proposed originally by Archad.⁶ However, the presence of pores on the surface and the residual stresses negatively affect the performance of LPBF-processed samples. Several recent investigations have been carried out focusing on the tribological response of AM materials.⁷ Still, no consensus exists that these materials have greater wear resistance than those conventionally processed.

Sun et al.⁸ reported that the wear resistance of LPBF-processed 316L SS specimens is adversely affected by porosity, which in turn is determined by the processing parameters. Chen and Gu⁹ analysed the wear rate of additively manufactured 4Cr5MoSiV steel. They reported that wear rate correlates with volumetric energy density conveyed by the laser and the resulting relative density and hardness. Yan et al.¹⁰ studied the microstructure and tribological properties of LPBF-processed Fe–16Mn–10Al–1.5C alloy, determining that microhardness increases with the decrease of the laser scanning speed and that the lowest speed produces the highest wear resistance. The beneficial effect of precipitates on wear performance has been explored by intentionally adding reinforcing particles in the laser-assisted process.^11,12

The effects of build-up strategies have also been found relevant. Rathod et al.¹³ studied the effect of scanning strategy and heat treatment on the tribological properties of Al–12Si fabricated by LPBF, reporting that the chess-board strategy leads to low porosity and thus lower wear rate and that posterior heat treatment improves the wear resistance over the as-prepared LPBF material. Li et al.¹⁴ investigated the effects of build-up directions on the tribological behaviour of 316L SS samples produced by LPBF through a sliding wear test at different temperatures. The results showed that the effects of build direction on wear rate are not significant; however, at higher test temperatures, the wear rate reached the maximum at 200 °C and then decreased at temperatures further increased due to activation of the oxide layers as working lubricants, protecting the samples from further material loss. Yang et al.¹⁵ sought high wear resistance 316L SS by analysing the effect of remelting on a zigzag strategy. They reported a maximum enhancement of wear resistance in the zigzag with a 90° rotation sample. The remelting with small hatch spacing was found to cause the grains to rotate, resulting in improved resistance to slipping.

Different testing techniques are reportedly used to assess the wear resistance of LPBF-processed materials: pin-on-disk, ball-on-ring, fretting or scratch testing.^14–17 In addition, indirect evaluation through nanoindentation has also been reported for additively manufactured materials, for instance, titanium alloy manufactured by LPBF.^18–20 Bahshwan et al.²¹ employed a micro scratch tribometer to assess the effect of build orientation on the anisotropy of 316L SS fabricated by LPBF, demonstrating two dominant wear mechanisms: (1) abrasive-oxidative wear and (2) delamination wear. However, adhesive mode of wear, as well as two- and three-body abrasion, fatigue and tribo-oxidative wear, were also reported by García-León et al.²² on borided AISI 316L SS under a ball-on-flat test.

Lastly, the differing microstructure of AM-processed materials is considered a key factor in wear performance. Additively manufactured 316L SS presents a hierarchical microstructure. At the larger scale, there are micro-sized molten pools, within which there are dendritic subgrains (e.g. columnar or cellular, depending on the observation orientation). Subgrains are defined as grains oriented at a <15° angle with respect to the nearest grain boundary, corresponding to a low-angle grain boundary. The formation of subgrains has been attributed to the high solidification rate (∼10⁶ K/s) and micro-segregation.^23–26 However, the relation between these substructures and their effect on wear performance has not been straightforwardly reported. Therefore, this work provides an experimental demonstration of the effect of the AM-specific microstructure on wear resistance tested in sliding. Furthermore, demonstrate that 316L SS processed by LPBF possesses superior wear resistance than conventionally processed SS.

Materials and methods

LPBF processing

Austenitic SS, grade AISI 316L, was used to fabricate prismatic samples with dimensions of 15 × 10 × 6 mm³ (Figure 1(a)). The prime material was metallic powder provided by General Electric Additive, with an average particle size of 32 µm (range from 19 to 52 µm). For the LPBF deposition, a direct metal laser melting system of Concept Laser (MLAB 200R) was used. The set-up featured a 200-W (Nb:YAG) fibre laser, wavelength 1064 nm, with a laser beam focused of 75 μm diameter. The sealed working chamber was fed high-purity nitrogen, reducing the oxygen content to less than 1000 ppm. The powders were deposited on a 16-mm thick SS support plate.

A series of preliminary experiments were implemented to identify the following processing parameters as suitable: laser power of 160 W, scanning speed of 700 mm/s, hatch spacing of 60 μm and layer thickness of 30 µm. A meander scan pattern was implemented, rotating the scan direction by 67° in each subsequent layer (Figure 1(b)).

Microstructure characterisation

The elemental composition of the printed samples was verified by glow discharge emission spectrometry (GDOES), bulk analysis mode, using GDA 750 HR (Spectruma Analytik GmbH). The average of three spots and the corresponding deviations are reported in Table 1.

Table 1.

Elemental composition of the as-built AISI 316L determined by GDOES.

Element (wt-%)
Fe	Cr	Ni	Mo	Si	Mn	Co	C	P
Bal.	18.9 ± 1.6	12.5 ± 0.9	2.7 ± 0.2	0.9 ± 0.2	0.5 ± 0.03	0.1 ± 0.01	0.02 ± 0.01	0.06 ± 0.01

For metallographic analysis, the samples were ground using SiC paper, starting from 120 down to 1500 grit, then finely polished using 0.05 µm alumina paste followed by 0.01 µm diamond paste. The microstructure was revealed by chemical etching immersion in Marble's reagent etchant (10 g CuSO₄, 50 ml HCl and 50 ml water) for 50 s. Surface morphology was inspected by optical microscopy (OM) (Nikon Eclipse MA 200) equipped with a darkfield filter and scanning electron microscopy (SEM) (TESCAN – MIRA 3) equipped with an energy-dispersive spectrometer (EDS) to investigate the microstructure and elemental chemical composition. The OM and SEM micrographs were processed and analysed using ImageJ software (National Institutes of Health, USA) to characterise the microstructural features and evaluate porosity. Moreover, the relative density was determined using Archimedes’ principle.

X-ray diffraction (XRD) (Panalytical – Empyrean) was utilised for phase characterisation, with a copper anode wavelength (λ) of 1.5406 (Å) for K_α1 emission equipped with a Ni filter. The interplanar spacing was calculated using Bragg's law and the lattice parameter identified by Miller indexing the peaks. The crystallite size (D) and microstrain (ε) were calculated by Williamson–Hall (W–H) method (Equation (1)), where θ is the diffraction angle, K corresponds to the Scherrer constant (here 0.94), and β_T is the full width at half maximum (FWHM) of the corresponding peak:

β_{T} = \frac{K λ}{D \cos θ} + 4 ε \tan θ .

(1)

Dislocation density (δ), defined as the length of dislocation lines per unit volume, was calculated as the inverse of D.²⁷ The W–H method can be rearranged as a linear equation where D is estimated from the intercept with the ordinate axis and ε from the slope of a linear fit.

The uniform deformation energy density model (UDEDM), derived from the W–H method, allows determining the strain energy density (U), considering the anisotropy of the crystal Equation (2):

U = \frac{ε^{2}}{2} E_{h k l},

(2)

where Young's modulus (E_hkl) can be calculated for the FCC crystal (austenite) by Equation (3):

\frac{1}{E_{h k l}} = S_{11} + (2 S_{12} - 2 S_{11} + S_{44}) A_{h k l},

(3)

where S₁₁, S₁₂ and S₄₄ are the components of the compliance matrix of the material. For the 316L steel, the values of 10.7 × 10⁻³, −4.25 × 10⁻³ and 8.6 × 10⁻³ [GPa⁻¹], respectively, are adopted.²⁷ The A_ℎkl factor is used to describe anisotropy values using Equation (4).

A_{h k l} = \frac{(h^{2} k^{2} + k^{2} l^{2} + l^{2} h^{2})}{{(h^{2} + k^{2} + l^{2})}^{2}} .

(4)

Wear testing

The wear resistance of the samples was evaluated using a pin-on-disc tribometer (CSM Instruments) under dry conditions, at room temperature (20 °C), in air with a relative humidity of approximately 50%. Samples were tested under a normal force of 10 N at a velocity of 20 mm/s for a sliding distance of 24 m. The counterpart was a sphere with a radius of 2.54 mm made of silicon nitride (Si₃N₄) with an average hardness of 1600 HV. The surface finish prior to the wear test was determined by a roughness tester (MarSurf PS 10) with a vertical resolution of 0.008 μm.

The wear track (scar depth and width) and the volume loss were evaluated using a 3D optical profiler (Zygo, ZeGage). In addition, mass loss was measured using an analytical balance with a precision accuracy of 0.1 mg. Each of the reported results is an average of over three measurements. Conventional wrought SS 316L was also tested as reference material for comparison purposes. In addition, the friction coefficient of the specimens was recorded during the wear tests. The wear rate ( $ω$ ) was calculated by Equation (5):

ω = \frac{V}{F_{N} \times d},

(5)

where ω is quantified in [mm³ N⁻¹ m⁻¹], V is the volume loss in [mm³] estimated either directly or by mass loss, F_N is the applied normal force [N] and d − sliding distance [mm].

Microhardness was measured using a Vickers hardness tester (TUKON 2500), using 500 g force and 10 s dwell time, following the ASTM E384-17 standard. Mean values were determined from five measurements on the top surface.

Nanoindentation tests were carried out on the top surface with a force of 150 mN and a holding time of 10 s using a nanoindentation tester (NHT3) equipped with a standard Berkovich indenter. The nanohardness (H) and reduced elastic modulus (E_r) of the samples were calculated from the loading–unloading curve using Oliver and Pharr analysis²⁸ by applying Equations (6) and (7), respectively:

H = \frac{F_{\max}}{A_{c}} = \frac{F_{\max}}{26.43 h_{c}^{2}},

(6)

where

F_{\max}

is the maximum indentation load, h_c is the contact depth under the maximum indentation load,

v_{s}

and

v_{i}

are the Poisson's ratio of the sample and the indenter, respectively and

E_{s}

and

E_{i}

are the Young's modulus of the samples and the indenter. The reduced elastic modulus was evaluated by Equation (7):

\frac{1}{E_{r}} = \frac{1 - v_{s}^{2}}{E_{s}} + \frac{1 - v_{i}^{2}}{E_{i}} .

(7)

Results and discussion

Microstructure analysis

Figure 2(a) shows a micrograph of 316L SS processed by LPBF in a plane parallel to the build direction, where it is possible to observe the stacking of the molten pools, and the dark areas represent the grain boundaries. It is worth noting that a relative density of 99.1% has been obtained due to the proper selection of processing parameters. Figure 2(b) shows the typical microstructure of the 316L SS conventionally processed where polygonal grains with frequent large recrystallisation twins are observed, with an average grain size of about 25 μm.

Figure 3(a) shows a heterogeneous grain size distribution in the selective laser-melted samples. The average grain size is around 17.5 µm. Figure 3(b) presents the grain distribution in more detail, where it is possible to observe that there are cellular subgrains within each grain with a cell size of around 0.6 µm. It is worth noting that the orientation of the subgrains influences the grain size, which in turn depends on the processing parameters.

A more detailed appearance of the microstructure inside the molten pool is presented in Figure 4. The SEM image shows the solidification behaviour in the plane parallel to the build direction, but it should also be considered that the solidification proceeds in all three dimensions. The arrow indicates the epitaxial growth inheriting crystallographic orientation from parental grains. The dashed lines represent the grain boundaries formed by subgrains ordered in different crystal orientations or high-angle grain boundaries.

The layer-by-layer nature of the LPBF process leads to a repeated thermal cycle, which can initiate diffusional processes in the solid state, leading to the precipitation of nanoparticles. Temperature gradient and solidification rate are the key parameters for understanding solidification. In LPBF, high thermal gradients (10⁶ K/m) and ultra-fast cooling rates (10⁶ K/s) result in the cellular/dendritic microstructure.^23–25

Due to the high crystallisation rate, a solute-rich boundary layer is built up in front of the solid–liquid interface, thus approaching conditions for constitutional supercooling. In LPBF, the solid–liquid interface in the melt pool has a complex shape due to a changing temperature gradient along the interface. The highest temperature gradient is expected at the bottom of the melt pool and gradually decreases along the interface towards the top and sides of the pool. The solidification rate is thus expected to be the lowest at the bottom and the highest at the top and sides of the pool.²⁹ A steep temperature gradient is typically associated with crystallisation in a cellular manner, whereas a mild slope of temperature gradient promotes solidification with dendrites of very characteristic shape.

Figure 5 shows a close-up of the melting pool with characteristics of competitive growth across the melting pool. In the case of a preferred crystal orientation aligned with the direction of heat flow (vertical arrow), the grains grow along the underlying direction, consistent with a high solidification rate. Whereas the grains with other orientations (see inclined arrow as an example) growth may be gradually arrested due to the competitive growth of different grains.³⁰ The competitive growth mechanism leads to the unidirectional columnar grain structure during solidification.³¹

Figure 6 shows the dendritic subgrains forming a cellular structure in the direction perpendicular to the columnar orientation, along with EDS mapping of a representative area to identify the elements around and inside the cellular subgrains. The corresponding EDS spectra are shown in Figure 7. The cellular structure is featured by element segregation boundary. In particular, the amount of Cr and Mo is higher on grain boundaries on 316L SS.³² The results show no overall difference in chemical composition concerning the values determined by GDOES. The formation of the cellular colonies results from constitutional undercooling, which is related to the diffusion of elements in liquid and solid phases. Enrichment of cell walls with Cr and Mo segregation has been observed in austenitic SS.³³ Figure 6 illustrates the cellular structure with a cell size of around 600 nm. According to Hong et al.,³⁴ the boundaries of the cellular subgrains are enriched with a high density of dislocations.

Improved cell connections between molten pools are achieved by successive rotation of the deposited layers.³⁵ In addition, the average cell size varies from the scanning strategy. According to Salman et al.,³ the stripe strategy leads to a lower average cell size, which in turn is linked to a smaller grain size. An additional feature of the intercellular boundaries in LPBF materials is the high density of dislocations forming characteristic dislocation structures.

X-ray diffraction analysis

The alloy is found to be austenite with no other phases, in particular, no martensite. The lattice spacing values differ slightly, 3.58 nm for the additively processed and 3.59 nm for the conventional SS. Compared to the wrought counterpart, the peak positions are slightly shifted in the 2θ space, with differing relative intensities of the respective peaks (Figure 8). The first observation indicates deformation of the unit cell typically associated with residual stresses. The shift to the right in the LPBF peaks indicates compressive stress, consistent with the vertiginous heating and cooling cycles produced by the laser-material interactions. The difference in relative peak heights is indicative of crystallographic texture.

Table 2 summarises the results of the XRD pattern analysis by applying the W–H method for each evaluation plane. The samples produced by LPBF show a reduced crystallite size (values of D between 52 and 66 nm) compared with the 316L SS conventional processed (value of D = 207 nm). The peak position and the peak broadening are consistent for the LPBF samples; the dislocation density is one order of magnitude higher than that of the wrought sample.

Table 2.

Results of the Williamson–Hall method for crystallite size, microstrain and dislocation density evaluation considering anisotropy.

PLANE	Peak position	FWHM	Crystallite size	Dislocation density	Energy density	Microstrain
	2θ (°)	β (°)	D (nm)	δ (10¹⁴ m⁻²)	U (MPa)	ε (10⁻³)
TOP	43.593	0.260	52.09	3.69	0.1277	0.96
	50.684	0.320				1.65
	74.604	0.446				1.17
SIDE	43.848	0.238	65.53	2.33	0.1690	1.10
	50.979	0.324				1.90
	74.839	0.423				1.35
FRONT	43.766	0.262	51.81	3.72	0.3349	1.55
	50.850	0.371				2.68
	74.782	0.519				1.90
WROUGHT	43.697	0.333	206.68	0.234	1.3838	3.16
	50.719	0.612				5.44
	74.698	0.750				3.86

Hardness testing

Microhardness

The highest microhardness was found at the top surface (271.48 ± 23.68 HV_0.5), while Vickers microhardness practically does not change at the front and side surfaces, with a value of 245.84 ± 9.18 and 245.38 ± 8.79, respectively. These values are similar to those reported in the literature^26,36,37 and are generally consistent with the microstructure-property relationship reported for anisotropic 316L.³⁸ In the specific case of grain size, yield strength, and hardness are directly linked following the well-known Hall–Petch relationship.

The anisotropy is evidenced in the values of hardness differing for the faces, which is not the case of the conventionally manufactured SS shows isotropy, with a practically constant microhardness (190 ± 4 HV_0.5). On average, the samples manufactured by LPBF have microhardness increased by 42% compared to the wrought specimens.

Nanohardness

According to classical wear theory, hardness measurements are used as an approximate indicator of wear resistance. The wear resistance of a material is also related to its ability to resist elastic deformation until failure (H/E_r) under some conditions.³⁹ A higher H/E_r ratio approximately reflects better wear resistance of a material. Also, the $H^{3} / E_{r}^{2}$ ratio is proportional to the material's resistance to plastic deformation.¹⁸ Therefore, in Figure 9, the loading–unloading curve is analysed to determine whether the top and side manufacturing drawings differ.

The obtained results are summarised in Table 3, showing that the top plane offers greater resistance to elastic deformation (H/E_r) and the highest resistance to plastic deformation ( $H^{3} / E_{r}^{2}$ ) compared to the side plane (Figure 10). These results are consistent with those obtained by microhardness and are related to the smaller grain size. They also agree with the results reported in the literature.⁴⁰ Therefore, it can be concluded that the upper plane presents higher resistance to indentation deformation than the lateral plane.

Table 3.

Nanoindentation on the different faces on 316L SS processed by LPBF.

	H (GPa)	E_r (GPa)	H/E_r	$H^{3} / E_{r}^{2}$
Top	2.84 ± 0.039	141.18 ± 2.48	0.0206	0.00121
Side	2.74 ± 0.019	153.19 ± 6.22	0.0179	0.00088

Wear performance

The as-built roughness of the samples was 9.85 ± 1.78 μm; prior to evaluating the tribology response, the average surface roughness at the tested surfaces was approximately 0.54 μm.

The friction coefficient and wear resistance are not intrinsic material properties. The results are specific to a tribo-system composed of Si₃N₄ indenter and the base material (test material). Figure 11 shows the variation of the friction coefficient (COF) during the pin-on-disc test as a function of the sliding distance. Wrought and laser-processed SS show a running-in response characterised by a steady increase of the COF. In contrast, the conventionally processed SS transits faster to the steady state characterised by a stable COF. After ca. 5 m, the COF reaches a maximum stabilising thoroughly after 8 m. Whereas, for the additively processed SS, the ascent slope is lower, and the steady state is reached after ca. 9 m. Once in the steady-state, the COF of the conventional SS stabilises at the value of 0.1409 ± 0.0064, whereas for LPBF-processed SS at the value of 0.1426 ± 0.0058.

The differences in the COF are associated with the presence of small porosities on the surface of the LPBF-processed samples. Fluctuations in the COF can be attributed to wear mechanisms and to the appearance of the stick-slip phenomenon, which generates entrapment and expulsion of wear particles.

Figure 12 illustrates the average wear volume and mass loss for wrought and LPBF 316L SS after a sliding distance of 24 m. This value is representative because volume loss is considered independent of the sliding distance once the steady state has been reached.⁴¹ The highest wear resistance is associated with the highest hardness of the additively manufactured specimen, which is consistent with Archard's law.

With an average volume loss of 21.09 ± 4.24 × 10⁻³ mm³ for the additively manufactured SS and 31.37 ± 2.40 × 10⁻³ mm³ for the conventionally processed steel, the wear rate of 8.79 × 10⁻⁵ and 1.307 × 10⁻⁴ mm³/Nm, was obtained respectively. The additively manufactured specimen shows a 32.8% lower wear rate than the wrought specimens.

Figure 13 presents the surface morphology of the wear track of wrought and additively manufactured 316L SS. Although both materials were tested against the same counter body, the following differences are observed: (1) the diameter of the wear track for the conventional sample has an extension of 3.375 µm, while the LPBF sample has a smaller diameter of 2885 µm; (2) the width of the wear track for the conventional specimen is around 604 µm, while for the LPBF the width was 548 µm; (3) wrought samples show a depth around 28 µm and the LPBF show less depth 19 µm. These observations validate the lower volume loss in the SS processed by LPBF.

Figure 14 shows the morphology of the wear tracks after the pin-on-disc test, exposing relevant features of the LPBF-processed material. Figure 14(a) and (b) displays the surface as observed under OM, and Figure 14(c) and (d) shows the 3D profile. In Figure 14, it is possible to observe a plowing groove with shallow depth on the surface of the wear track. Figure 15 shows the wear track of the conventionally processed SS, analogous to Figure 14.

In the OM images of the wear tracks, the dark zones are identified as smearing (Figure 15); such material agglomerations have been associated with a tribo-film, a compact oxidised layer on the surface, which in some cases can protect against material removal.²²

Effect of microstructure

The LPBF-processed 316L steel has shown a superior wear resistance as compared to the conventional 316L SS, which, on the micrometer scale, is shown to be related to grain size and morphology. On the smaller length scale, the formation of small domains (cells) inside large grains and subgrain structures explains this effect by providing more barriers to dislocation sliding, which is the primary mechanism of plastic deformation in austenitic steel and responsible for the high yield strength of 316L LPBF samples.⁴² The high dislocation density deduced from the XRD analysis is attributed to the formation of the cellular microstructure, which is correlated to spatial variations in the local Cr concentration and nanoinclusions formation.⁴³ The higher density of dislocations is associated with higher strength and hardness because the grain and subgrain boundaries hinder the movement of dislocations, thus strengthening the material.⁴⁴ The more grain boundaries, the more difficult for the dislocations to slide across, resulting in a harder material. According to dislocation theory,⁴⁵ the shear stress is proportional to the dislocation density, giving rise to a relation similar to that of Hall–Petch, Equation (8).

τ = τ_{o} + α G b \sqrt{δ},

(8)

where

τ_{o}

is the intrinsic strength of the material (without dislocations), α is a correction factor (0.1) and b is the Burgers vector. Additively manufactured materials typically have a high content of dislocations (>10¹⁴ m⁻²)^43,46–48; as a result, the materials exhibit high wear resistance.

Figure 16 conceptualises the effect of LPBF-specific microstructure, in particular the grain sizes relative to wrought 316L SS, during indentation for hardness measurement. Whereas the microstructure of the wrought alloy is considered isotropic, for the LPBF-processed alloy, two cases are considered depending on the orientation of the columnar structure with respect to the indenter tip. Two extreme cases are considered: (i) indenter oriented perpendicular to the long axis of the cellular columns (middle scheme) and (ii) pin oriented parallel to the long axis of the cellular columns (bottom scheme). Since grain size influences the number of dislocations piled up at the grain boundaries, there is indeed a higher number of barriers to dislocation sliding in both cases of the LPBF material compared to the wrought alloy, explaining the overall higher microhardness and greater wear resistance.

The microstructure's relative orientation also explains the differing load–displacement curves observed in the nanoindentation of top and lateral faces (Figure 9). On the one hand, the results are consistent with the relative size of the domain available for dislocation sliding.⁴⁹ On the other hand, the top face is composed of several cell colonies of different orientations (Figure 10(a)). In contrast, the lateral face has a honeycomb microstructure (Figure 10(b)), which could help transfer load and absorb more strain energy. The columnar structures may also buckle instead of transferring the load, leading to lowered force-bearing capability.

The effect of microstructure during sliding wear testing is conceptualised in Figure 17, considering the same extreme cases as in Figure 16. In addition, grain boundaries are colour coded to represent those in direct contact with the test pin (red) and those affected or receiving stress transferred from the pin (yellow). In conventional SS with large grains and a smaller number of grain boundaries, dislocations during plastic deformation can move over a longer distance before being blocked by a grain boundary. Whereas, for additively manufactured SS, a larger number of cells in the cellular or columnar arrangement pose more barriers to the dislocation movement and thus increase the wear resistance with elevated hardness.

During sliding, the force acting on the sample can be divided into tangential ( $F_{t}$ ) and normal ( $F_{n}$ ) components. The normal component causes the material to deform downward under the load (compression), while $F_{t}$ is the tangential force between materials in contact.⁵⁰ The normal force $F_{n}$ is influenced by the roughness angle ( $θ_{Ra}$ ) and the applied load (P), as expressed by Equation (9); and $F_{t}$ can be described as the product between shear stress (τ) and the contact area (A) given by Equation (10).

F_{n} = \frac{2 P \tan θ_{Ra}}{π},

(9)

F_{t} = A τ,

(10)

where τ is related to the shear modulus (G) and the plastic shear strain, depending on Young's modulus and dislocation density, respectively. As the indenter advances in sliding, it activates slipping; consequently, dislocations pile up at the grain boundaries. At the same time,

F_{n}

promotes the formation of grooves in the wear track. Since the columnar structures are more likely to slip than the cellular structures,⁵¹ the cellular grains can bear more force in the plastic deformation range. Then, the refined subgrains enhance the microhardness and wear resistance of the material in the subsequent passage of the pin, which is not the case in conventionally processed SS with larger grains. In addition, at certain levels of applied force, the piled-up dislocations affect the internal stresses. In the conventionally processed SS, the accumulation of dislocations is found only at the grain boundaries, while in the SS processed by LPBF, dislocations are also found in the subgrains in addition to the vicinity of cellular walls.⁴⁶ Consequently, the microstructure-enhanced strain hardening strongly affects the wear behaviour of the austenitic SS 316L processed by LPBF.

Once the elastic limit is exceeded, dislocations accumulate, and the hardened grains eventually peel off. The wear debris generated in the process leads to abrasive wear in the three-body mode. The abrasive particles plow along the sliding direction, forming the plowing groove with shallow depth on the surface of the wear track. In the conventionally processed SS, an adherent and strain-hardened tribo-layer is smeared on the worn surface, promoting the adhesive wear mode.

Conclusions

Austenitic SS 316L processed by LPBF has ca. 30% higher sliding wear resistance as compared with a wrought counterpart, which is mainly explained by the formation of dendritic cellular subgrains resulting from the ultra-fast solidification rate. The increased number of barriers to dislocation sliding is consistent with increased hardness, which correlates with higher wear resistance.

The LPBF-processed alloy's microstructure is composed of micro-sized melt pools comprising dendritic cellular subgrains of a diameter smaller than 600 nm and extending up to 1 mm in bundles of various orientations. The cellular walls are enriched in Cr and Mo due to micro-segregation consistent with competitive grain growth favouring the alignment in the build direction.

The higher density of dislocations is consistent with the dendritic cellular subgrains that differ in cell parameter and crystallite size, the LPBF-processed material being characterised by cell parameter of 3.58 nm and crystalline size of ca. 60 nm, as compared with respectively 3.59 and 206 nm of the wrought counterpart.

Tribological performance of the LPBF-processed material is also favoured by forming a compact oxidised layer, that is tribo-film.

Anisotropy of the microstructure is relevant to the response to sliding wear, with the upper plane presenting greater resistance to indentation deformation. The strain hardening plays an essential role in affecting the wear behaviour of the austenitic SS 316L processed by LPBF.

Footnotes

Acknowledgements

The authors are grateful for financial support from SENESCYT, FONDEQUIP EQM grant number 180081, by Agencia Nacional de Investigación y Desarrollo (ANID), and Natural Science and Engineering Research Council of Canada.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the SENESCYT [grant number ARSEQ-BEC-000329-2017]; Agencia Nacional de Investigación y Desarrollo (ANID) project: FONDEQUIP EQM 180081.

ORCID iDs

Germán Omar Barrionuevo

Magdalena Walczak

Jorge Ramos-Grez

Alexis Debut

References

Bergström

. The absorptance of metallic alloys to Nd:YAG and Nd:YLF laser light . Thesis, 2005, p.149.

Haghdadi

Laleh

Moyle

, et al. Additive manufacturing of steels: a review of achievements and challenges. J Mater Sci 2021; 56: 64–107.

Salman

, Brenne F, Niendorf T, et al. Impact of the scanning strategy on the mechanical behavior of 316L steel synthesized by selective laser melting. J Manuf Process 2019; 45: 255–261.

Gupta

, Singla AK, Ji H, et al. Impact of layer rotation on micro-structure, grain size, surface integrity and mechanical behaviour of SLM Al–Si–10Mg alloy. J Mater Res Technol 2020; 9: 9506–9522.

Zhu

Zou

Yang

. Wear performance of metal parts fabricated by selective laser melting: a literature review选区激光熔化成形件磨损特性综述. J Zhejiang Univ Sci A 2018; 19: 95–110.

Archard

. Contact and rubbing of flat surfaces. J Appl Phys 1953; 24: 981–988.

Barrionuevo

Walczak

Ramos-grez

, et al. Microhardness and wear resistance in materials manufactured by laser powder bed fusion: machine learning approach for property prediction. CIRP J Manuf Sci Technol 2023; 43: 106–114.

Sun

Moroz

Alrbaey

. Sliding wear characteristics and corrosion behaviour of selective laser melted 316L stainless steel. J Mater Eng Perform 2014; 23: 518–526.

Chen

. Effect of metallurgical defect and phase transition on geometric accuracy and wear resistance of iron-based parts fabricated by selective laser melting. J Mater Res 2016; 31: 1477–1490.

10.

Yan

Gao

Chang

, et al. Microstructure and tribological property of selective laser melted Fe-Mn-Al-C alloy. Mater Lett 2020; 270: 127699.

11.

Almangour

Grzesiak

Yang

. Rapid fabrication of bulk-form TiB2/316L stainless steel nanocomposites with novel reinforcement architecture and improved performance by selective laser melting. J Alloys Compd 2016; 680: 480–493.

12.

Wang

Zhang

, et al. Effect of Cr3C2 content on 316L stainless steel fabricated by laser melting deposition. Vacuum 2018; 147: 92–98.

13.

Rathod

Nagaraju

Prashanth

, et al. Tribological properties of selective laser melted Al–12Si alloy. Tribol Int 2019; 137: 94–101.

14.

Ramezani

, et al. Tribological performance of selective laser melted 316L stainless steel. Tribol Int 2018; 128: 121–129.

15.

Yang

Khonsari

, et al. On enhancing surface wear resistance via rotating grains during selective laser melting. Addit Manuf 2020; 36. DOI: https://doi.org/10.1016/j.addma.2020.101583.

16.

Tang

Zhang

. Microstructural evolution, mechanical and tribological properties of TiC/Ti6Al4 V composites with unique microstructure prepared by SLM. Mater Sci Eng A 2021; 814: 141187.

17.

Huang Y, Yang S, Gu J, et al. Microstructure and wear properties of selective laser melting 316L. Mater Chem Phys 2020; 254: 123487.

18.

Zhou

Yuan

, et al. Two ways of evaluating the wear property of Ti–13Nb–13Zr fabricated by selective laser melting. Mater Lett 2019; 242: 9–12.

19.

Kumar

Joshi

Pruncu

, et al. Microstructure and tribological response of selective laser melted AISI 316L stainless steel: the role of severe surface deformation. J Mater Eng Perform 2021; 30: 5170–5183.

20.

Efremenko

Zurnadzhy

Chabak

, et al. A comparison study on the effect of counter ball material on sliding wear response of SLM-printed biomedical 316L steel. Mater Today Proc 2022; 66: 2587–2593.

21.

Bahshwan

Gee

Nunn

, et al. In situ observation of anisotropic tribological contact evolution in 316L steel formed by selective laser melting. Wear 2022; 490–491. DOI: https://doi.org/10.1016/j.wear.2021.204193.

22.

García-León RA, Martínez-Trinidad J, Zepeda-Bautista R, et al. Dry sliding wear test on borided AISI 316L stainless steel under ball-on-flat configuration: a statistical analysis. Tribol Int 2021; 157. DOI: https://doi.org/10.1016/j.triboint.2021.106885.

23.

Chadwick

Voorhees

. The development of grain structure during additive manufacturing. Acta Mater 2021; 211: 39–53.

24.

Liu

, et al. Tuning strength-ductility combination on selective laser melted 316L stainless steel through gradient heterogeneous structure. Addit Manuf 2021; 48. DOI: https://doi.org/10.1016/j.addma.2021.102373.

25.

Barrionuevo

Ramos-Grez

Walczak

, et al. Numerical analysis of the effect of processing parameters on the microstructure of stainless steel 316L manufactured by laser-based powder bed fusion. Mater Today Proc 2021; 59: 93–100.

26.

Zhang

Chu

Wang

, et al. Comprehensive effects of placement orientation and scanning angle on mechanical properties and behavior of 316L stainless steel based on the selective laser melting process. J Alloys Compd 2019; 791: 166–175.

27.

Stinville

Tromas

Villechaise

, et al. Anisotropy changes in hardness and indentation modulus induced by plasma nitriding of 316L polycrystalline stainless steel. Scr Mater 2011; 64: 37–40.

28.

Oliver

Pharr

. An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. J Mater Res 1992; 7: 1564–1583.

29.

Lee

Nordin

Babu

, et al. Effect of fluid convection on dendrite arm spacing in laser deposition. Metall Mater Trans B 2014; 45: 1520–1529.

30.

Zhang

Yocom

Mao

, et al. Microstructure evolution during selective laser melting of metallic materials: a review. J Laser Appl 2019; 31: 031201.

31.

Barrionuevo

, et al. Microstructure simulation and experimental evaluation of the anisotropy of 316 L stainless steel manufactured by laser powder bed fusion. Rapid Prototyp J 2022. DOI: https://doi.org/10.1108/RPJ-04-2022-0127.

32.

Saeidi

Gao

Zhong

, et al. Hardened austenite steel with columnar sub-grain structure formed by laser melting. Mater Sci Eng A 2015; 625: 221–229.

33.

Yadroitsev

Yadroitsava

Du Plessis

, et al. Fundamentals of laser powder bed fusion of metals. Amsterdam: Elsevier, 2021.

34.

Hong

Zhou

Zheng

, et al. The cellular boundary with high density of dislocations governed the strengthening mechanism in selective laser melted 316L stainless steel. Mater Sci Eng A 2021; 799: 140279.

35.

Yan

, et al. Effect of building directions on the surface roughness, microstructure, and tribological properties of selective laser melted Inconel 625. J Mater Process Technol 2021; 288: 116878.

36.

Kovacı

Seçer

. Improved tribological performance of AISI 316L stainless steel by a combined surface treatment: surface texturing by selective laser melting and plasma nitriding. Surf Coat Technol 2020; 400: 126178.

37.

Larimian

Kannan

Grzesiak

, et al. Effect of energy density and scanning strategy on densification, microstructure and mechanical properties of 316L stainless steel processed via selective laser melting. Mater Sci Eng A 2020; 770: 138455.

38.

Pitrmuc

, et al. Mechanical and microstructural anisotropy of laser powder bed fusion 316L stainless steel. Materials (Basel) 2022; 15. DOI: https://doi.org/10.3390/ma15020551.

39.

Ehtemam-Haghighi

Cao

Zhang

. Nanoindentation study of mechanical properties of Ti based alloys with Fe and Ta additions. J Alloys Compd 2017; 692: 892–897.

40.

Upadhyay

Kumar

. Scratch and wear resistance of additive manufactured 316L stainless steel sample fabricated by laser powder bed fusion technique. Wear 2020; 458–459. DOI: https://doi.org/10.1016/j.wear.2020.203437.

41.

Chen

, et al. Effect of Ti addition on the sliding wear behavior of AlCrFeCoNi high-entropy alloy. Wear 2020; 462–463. DOI: https://doi.org/10.1016/j.wear.2020.203493.

42.

Man

, et al. The effect of sub-grain structure on intergranular corrosion of 316L stainless steel fabricated via selective laser melting. Mater Lett 2019; 243: 157–160.

43.

Sprouster

, et al. Dislocation microstructure and its influence on corrosion behavior in laser additively manufactured 316L stainless steel. Addit Manuf 2021; 47: 102263.

44.

Lin

, et al. Selective laser melting processing of 316L stainless steel: effect of microstructural differences along building direction on corrosion behavior. Int J Adv Manuf Technol 2019; 104: 2669–2679.

45.

Hull

Bacon

. Introduction to Dislocations, vol. Oxford, 25. 2011.

46.

Kurian

Mirzaeifar

. Deformation mechanisms of the subgranular cellular structures in selective laser melted 316L stainless steel. Mech Mater 2020; 148: 103478.

47.

Kang

Dong

Wang

, et al. Dislocation evolution in 316L stainless steel subjected to uniaxial ratchetting deformation. Mater Sci Eng A 2010; 527: 5952–5961.

48.

Bertsch

Meric de Bellefon

Kuehl

, et al. Origin of dislocation structures in an additively manufactured austenitic stainless steel 316L. Acta Mater 2020; 199: 19–33.

49.

Javaid

Pouriayevali

Durst

. Dislocation–grain boundary interactions: recent advances on the underlying mechanisms studied via nanoindentation testing. J Mater Res 2021; 36: 2545–2557.

50.

Wang

Zhang

Cui

, et al. Novel gradient alloy steel with quasi-continuous ratios fabricated by SLM: material microstructure and wear mechanism. Mater Charact 2021; 174: 111020.

51.

Yang

Zhu

Khonsari

, et al. Wear anisotropy of selective laser melted 316L stainless steel. Wear 2019; 428–429: 376–386.