Effect of rescanning with reduced interlayer times on microstructure and properties of titanium alloy via selective laser melting

Abstract

In the process of selective laser melting, laser rescanning technology is often used to optimize the residual stress and other properties of the formed parts. In order to improve the performance of parts and reduce the manufacturing time concurrently, this paper proposed a combined rescanning strategy. Based on finite element analysis, molten pool solidification behavior was simulated and studied. Ti6Al4V alloy samples were fabricated and analyzed by changing the rescanning strategies and process parameters. The microstructure, relative density, size of pore defects and residual stress were investigated under different rescanning strategies. It can be seen that the average cooling rate ranked in a descending order of SLM > re-SLM+IL1 > re-SLM, and samples formed by rescanning without layer interval had the best relative density and residual stress optimization effect, while the microstructure of each scanning strategy was all acicular α′ phase. When the number of rescanning interlayers and laser power was “one-layer” and 140 W respectively, the residual stress went down from 353 to 294 MPa. Finally, a simplified model was proposed to calculate the time cost for fabrication of rescanning with reduced interlayer times.

Keywords

Laser rescanning selective laser melting titanium alloy scanning strategy optimization microstructure

Introduction

Ti6Al4V alloy has shown promising applications in aerospace field for high specific strength. However, the low plasticity and high hardness lead to the difficulty of Ti6Al4V alloy mechanical processing. As a key process in the field of metal 3D printing, SLM (selective laser melting) technology has been widely used in personalized creative manufacturing of Ti6Al4V alloys due to its high efficiency, low cost, high degree of integration, and strong adaptability to the processing of complex structural parts in recent years.^1,2 During the forming process of SLM, each layer of the component is fused by selectively melting the appropriate area of powder with a laser beam. During the fusion of a layer, the temperature goes up above the melting point temperature of titanium alloy and rapidly drops down with almost 10⁸ K/s cooling rate which leads to excessive internal temperature gradient, resulting in the imbalance of metal thermal deformation at different parts and higher internal residual stress.³ Residual stress will cause defects such as warpage, cracks, and pores.⁴ Powder bed preheating can reduce temperature gradient and residual stress, but the temperature of the powder bed cannot reach higher preheating temperature.⁵ The subsequent heat treatment can effectively reduce the residual stress of the SLM forming part, but it increases excess cost and time.⁶ In this context, recent development in laser rescanning open new perspectives for the improvement of the components performance, especially for the residual stress and volume porosity.

Rescanning is a technique to solidify surface which is rescanned by heat source during the SLM process.⁷ Rescanning remelts the unmelted defects forming at the first scanning, at the same time, it conducts secondary distribution of temperature gradient to restrict the residual stress inside the forming SLM parts, resulting in the improvement of the relative density and mechanical properties.^8,9 Therefore, rescanning has been supposed to be a promising method to optimize the performance of SLM forming part in recent years.

Kruth et al. studied the influence of different process parameters on the residual stress by bridge curvature method. They found that short scan vectors and powder bed preheating decreased high temperature gradient and resulted in the reduction of residual stress.¹⁰ Thermal stress in particular directions was decreased by the optimization of scan vectors. Duan et al.¹¹ investigated the effect of rescanning strategy on the volume porosity and wear resistance and found that the samples formed by square-framed rescanning strategy hold an excellent bio-tribological properties. Griffiths et al.¹² rescanned SLMed Al-Mg-Zr alloy parts and found that the laser energy that inflicted on solidified surface reduced during rescanning, leading to the remelting of columnar grain area at the top of the original molten pool. New equiaxed grains and columnar grains were precipitated, and the equiaxed grains in the deep side of the molten pool were retained. Therefore, rescanning can reduce the number of columnar grains, increasing the number of equiaxed grains, and refined the grains size, but it increased the processing time of forming. Shiomi et al.¹³ used heat treatment and rescanning to reduce residual stress of SLMed steel model. He found that the residual stress of parts treated at 700°C for 1 h reduced by 70%, while residual stress of parts processed using rescanning technology decreased by 55%. Yu et al.¹⁴ studied the influence of the rescanning direction of laser on the AlSi10Mg parts. They found that rescanning in the same and opposite directions as the first scanning direction reduced the surface roughness of parts from 20.67 to 11.67 and 10.8 μm respectively. The remelting resulted in the escape of under-melting tiny voids from the molten pool, so it decreased the porosity of the printed parts. Qiu et al.¹⁵ formed 316L samples by selective laser melting, and used optical microscope and scanning electron microscopy to characterize the surface structure of samples, the results showed that the remelting of the previous layer determined the development of the porosity of the interlayer interface. Demir and Previtali¹⁶ discussed the three different kinds of strategies to manufacture the 18Ni300 maraging steel by SLM, and found that the remelting strategy was effective in enhancing parts density and caused a slight grain phase. Due to the thermal conductivity of metal solidified surface is much higher than that of metal powder, the cooling rate increases during the solidification process of rescanning. As the rescanning cycle increased, residual stress decreased when the cooling rate decreased. Miao et al.¹⁷ used ANSYS to simulate the SLM forming process of Ti6Al4V, indicated the temperature field and stress field during the forming process, and proposed a rescanning strategy to reduce residual stress. The rescanning process parameters were optimized for the melted aviation nozzle rings to reduce the average residual stress from 322 to 254 MPa.

However, most researchers only paid attention to rescanning all the layers of parts. There were few literatures related to the effect of rescanning with reduced interlayer times. Xiao et al.¹⁸ found that the yield strength, microhardness, and residual stress of SLMed Ti6Al4V parts all increased with the increasing of the rescanning cycle from 0 to 3, but decreased after rescanning four times, the rescanning had little effect on the microstructure of Ti6Al4V. Ali et al. established a Ti6Al4V-SLM simulation model through ABAQUS and proposed that high temperature gradient was due to the alternating cooling and heating cycles during the SLM process, resulting to large residual stress in the forming parts. They also rescanned SLMed Ti6Al4V parts with 150% energy density and found that the residual stress decreased but the build time increased.¹⁹ Karimi et al.²⁰ investigated the effects of the number of remelting on microstructure and mechanical properties of the SLM-built materials, he found that although the continuous remelting layer increased the production time and the processing cost of the parts, the sample mechanical strength and degree of homogenization could make up for the cost increased. However, the influence of the rescanning with reduced interlayer times on the characteristics of SLMed Ti6Al4V was still absent. In this work, the influence of the different rescanning strategies on the molten pool solidification behavior, relative density and microstructure was analyzed by experiments of Ti6Al4V samples printed using different scanning strategies of once scanning, rescanning, and combined scanning strategies. On this basis, the influence of the rescanning interlayers times, laser power, and hatch spacing on residual stress was investigated. The purpose of this study is to evaluate the effect of rescanning in SLM and to provide a new idea for the balance between performance optimization and time saving.

Materials and methods

The Ti6Al4V powder with the particle size of 15–53 μm was utilized in this study, provided by Wuxi Falcontech Co., Ltd. This powder was manufactured through an atomization comminuting process. The chemical composition of powder was shown in Table 1.

Table 1.

Composition of Ti6Al4V alloy powder (wt%).

Element	Al	Fe	C	O	N	H	Ti
Content	6.0	0.12	0.02	0.09	0.01	0.002	Bal.

Experiments in this study were conducted on YLM-120 selective laser melting equipment (Yongnian Laser, Suzhou, China). This equipment used SPI red POWER Fiber Laser with laser wavelength from 1060 to 1070 nm. In order to avoid high laser energy at the corner of the rescanned layers, cuboid samples with a dimension of 20 × 10 × 6 mm³ were fabricated by S-shaped scanning strategy and the powder bed was preheating in 100°C. Argon gas was used during the fabrication of samples to restrict the oxygen under 0.2%. As shown in Figure 1, the rescanning direction and combined scanning direction were same as the once scanning direction. Samples with once scanning were defined as “SLM,” samples rescanned each layer were defined as “re-SLM,” samples rescanned with one-layer and two-layer interval were defined as “reSLM+IL1” and “re-SLM+IL2.” The first scanning parameters, such as laser power, scanning speed, hatch spacing, and layer thickness, were set to an optimized condition according to our preliminary experiment. The process parameters of each scanning strategies are shown in Table 2.

Table 2.

Process parameters of different scanning strategies.

Scanningstrategy	Laserpower(W)	Hatchspacing(mm)	Scanningspeed(mm/s)	Layerthickness(μm)
SLM	120	0.1	1200	30
re-SLM;re-SLM+IL1;re-SLM+IL2	80	0.1
	100
	120
	140
	160
	120	0.05
		0.075
		0.1
		0.125
		0.15

Figure 1.

Details of experiments and simulation: (a) sketch of once scanning strategy, (b) sketch of rescanning strategy, and (c) building direction of combined rescanning strategy.

The relative density was measured by the drainage method. The surface morphology and distribution of pores were observed by RTEC White light interference profilometry. Kroll etchant reagent (HF:HNO₃: H₂O = 2:1:15) was used to erode samples for 30 s after grinding and polishing. Then the microstructure was characterized using a DM-2700M optical microscope (OM), SU1510 scanning electron microscope (SEM), and D2 PHASER X-ray diffractometer (XRD). The residual stress was measured with the hole-drilling method by JHMK-L Residual Stress Measurement instrument. The Vickers-hardness was measured by HVS1000ZCM-XY microhardness tester. The applied load was 500 g and the holding time was 15 s. Measuring points were selected separated by 0.3 mm on the polished surface.

The material properties of solid and powder Ti6Al4V used in this study were set according to the work done by Bovineau et al.²¹ and material properties of melted Ti6Al4V was set according to the work by Parry et al.²² Element Birth and Death of ANSYS was used to build the simulation model to indicate the influence of different rescanning strategies on the temperature field. The Gaussian laser heat source was applied on the powder bed in the form of heat flux, and laser heat source was loaded in APDL. The room temperature was assumed as T₀ = 25°C before the laser heat source was loaded. The convective heat transfer coefficient was applied to the surface of the powder bed, and the boundary conditions can be calculated by the following equation²³:

- k_{n} \frac{\partial T}{\partial n} - q + h (T - T_{0}) + σ ε_{θ} (T^{4} + T_{0}^{4}) = 0

(1)

Where k_n is the effective thermal conductivity of powder, T is the temperature of powder, h is the heat transfer coefficient, q is the rate of the heat input from the beam, σ is the Stefan–Boltzmann constant and ε_θ is the emissivity. As shown in Figure 2, a 1.44 × 0.96 × 0.03 mm³ powder bed model and 3.84 × 1.92 × 0.36 mm³ substrate model were built in order to improve the simulation efficiency and calculation accuracy. The height of each layer of powder was 0.03 mm, and each layer was printed by S-shape scanning strategy. The monitoring points were located on the upper surface of each printing layer, and the temperature monitoring points of the upper and lower layers were separated by 0.03 mm. Cooling time for 2 s was set to simulate the diffusion of the upper powder bed after the calculation of the heat source loading of the lower Ti6Al4V was completed.

Figure 2.

(a) Established finite element model and the temperature monitoring point and (b) result of temperature field simulation.

Results and discussions

Molten pool solidification behavior

The heating and cooling curves of the upper and lower layers with laser applied to the upper layer were shown in Figure 3(a). From the simulation data, it can be seen that molten pool maximum temperature of upper layers under different scanning strategies was 2683°C, 2800°C, 2870°C, respectively, while the data of lower layers was 1694°C, 1712°C, 1775°C, separately. It can be clearly found that all the molten pool maximum temperature was higher than the melting point of Ti6Al4V, which meant that the upper and lower layers under different scanning strategies were all melted and the remelting of upper layers caused by rescanning can melt the lower layers. The maximum molten pool temperature ranked in a descending order of SLM > re-SLM+IL1 > re-SLM. The effect of laser on the material during rescanning was much weaker than that during once scanning. A large amount of laser power was reflected from the surface of the metal, resulted in lower heat absorbance by the metal material than that of the once scanning, so the molten pool formed by rescanning was smaller than that of once scanning.^24,25 However, the thermal conductivity in solidified metal during the rescanning was much more efficient than that in powder.²⁶ Thereby, the thermal effect on the substrate during rescanning was significantly higher than that during once scanning. The average cooling rate and the maximum temperature of the molten pool under different scanning strategies were shown in Figure 3(b). It can be observed that the average cooling rate ranked in a descending order of SLM > reSLM+IL1 > re-SLM, indicating that rescanning can reduce the temperature gradient, and played a similar role like powder bed preheating.²⁷ The increasing trend of the maximum temperature of the molten pool of upper and lower layers was corresponded to Figure 3(a).

Figure 3.

Curve of (a) cooling rate and the melting point of Ti6Al4V and (b) the average cooling rate, the maximum temperature of the molten pool of upper layers and the maximum temperature of the molten pool of lower layers under different rescanning strategies.

Microstructure

The OM and SEM images of samples obtained with the side surface and the upper surface as the shooting direction were shown in Figure 4. As was shown in Figure 4(f), it can be clearly seen that the acicular α′ phase was wrapped in the columnar primary α phase from the SEM images with 500× magnification. The OM images with 100× magnification indicated that light and dark grains grew alternately and the main axis of the columnar grains was parallel to the building direction. As illustrated by Figure 4(b) and (c), the columnar crystal grains overlapped each other. Although Ti6Al4V was an alloy composed of α + β, the XRD image of Ti6Al4V samples was illustrated in Figure 5, indicating that β phase diffraction peaks did not exist and only α and α′ phase diffraction peaks existed. Combined with the simulation of molten pool, the cooling rate of samples manufactured by each scanning strategy reached 10⁶–10⁷°C/s, which far exceeded the grain cooling forming rate of β phase of 410°C/s.²⁸ When the titanium alloy was rapidly cooling from high temperature. The β phase was too late to transform into the α phase caused by the extremely high cooling rate, so it transformed into the α′ phase with the same composition as the parent phase and different grain structure.²⁹ Then acicular martensite structure was precipitated inside the grains. When the laser beam was applied on the upper layers of powder, the top of the solidified columnar grains of the lower layers remelted, and the powder of the upper layers were melted to form a new columnar top. As is shown in the molten pool simulation, the remelting of the lower layers is mainly owing to the remelting of upper layers caused by rescanning. The columnar grains grew epitaxially in the building direction. From Figure 4(d), it can be clearly seen that the distance between the grain boundaries was about 130 μm. According to Figure 4(e), the boundary between grains was not clear, and it can be observed that the acicular α′ phase wrapped in two different grains grown in a cross-bending way. The boundary of the melt tracks became indistinguishable, and the martensite phase inside the melt tracks showed a cross-melt growth morphology. The powder melting process under the S-shaped scanning strategy advanced from one side to the other side of the part. At this time, the powder in the unmelted area had poor thermal conductivity, thus the heat transfer of the molten pool was mainly through the heat conduction of solidification zone and air convection.³⁰ A low heat dissipation rate resulted in large temperature gradient and high solidification speed of the liquid phase. When laser applied on the solidified melt tracks again, prior-β grains grew due to the larger temperature gradient at the same melt tracks, so the grain structure grew across the melt tracks along the building direction.³¹ Columnar grains that overlap each other were formed, and acicular martensite was formed after cooling. It can also be seen from the OM diagram that the thickness of acicular martensite was related to the rescanning strategy. As mentioned before, the temperature of the molten pool gradually decreased from SLM to reSLM+IL1 to re-SLM. It can be observed that the thickness of acicular martensite ranked in a descending order of SLM > re-SLM+IL1 > re-SLM. Therefore, smaller martensite grains formed due to the smaller driving force.

Figure 4.

Images of the microstructures of the Ti6Al4V samples under different rescanning strategies: (a) SLM sample in upper direction, (b) re-SLM sample in upper direction, (c) re-SLM+IL1 sample in upper direction, (d) SLM sample in side direction,(e) re-SLM+IL1 sample in side direction, and (f) SEM image of α′ phase of the re-SLM sample.

Figure 5.

X-ray diffraction patterns of samples under different rescanning strategies.

Relative density

In this study, drainage method was used to measure the relative density. The measured data was shown in Table 3. It can be seen that the value ranked in a descending order of re-SLM > re-SLM+IL1 > SLM, which were in good agreement with the OM images (100× magnification) from Figure 6(a) to (c), suggesting that the rescanning strategy decreased the pore distributed over the samples surface. Surface morphology was also captured using Scanning electron microscope (SEM) from Figure 6(d) to (f), showing the surface of samples formed by different scanning strategies with 500× magnification. Part of the metal powder in the sample in once scanning was not fully melted, which was easier to cause defects.³² When the next scanning was performed, the previous molten pool was fully cooled, resulted in the high temperature gradient. It was difficult for melt materials of once scanning to flow across the melt tracks, so the lack of metal caused by the rapid solidification of melt materials cannot be replenished in time.³³ Compared with only once scanning, the temperature gradient of the adjacent melt tracks under the rescanning and combined scanning was lower, the probability of the melt flowing across the melt tracks increased, so the probability of the microscopic voids in the adjacent melt tracks declined. Samples re-SLM+IL1 were rescanned every two layers, so the surface defects remelting behavior of the middle layer between any two layers still depends on the laser applied on the upper layer. However, the laser energy is consumed after passing through the powder layer of the upper layer, so remelting effect on the lower layer of the upper layer heat transfer is weaker than that of directly rescanning applied on the lower layer. During the SLM forming process, a little spherical pore formed near the molten pool due to Marangoni-convection.³⁴ The rescanning strategy remelted the solidified meta. The pores formed in the once scanning strategy were also remelted in rescanning strategy which resulted to smoother undulations on the surface. The spread powder of upper layers was more uniformly distributed, which reduced the probability of unmelted defects. It was possible for gas in pores at bottom of molten pool to float up and escape, so the pores at the top of the new layers also reduced after rescanning.³⁵

Table 3.

The relative density, void size, and hardness of different samples.

Samples	SLM	re-SLM+IL1	re-SLM
Relative density (%)	95.7 ± 1.1	97.8 ± 1.2	98.7 ± 0.7
Void width (μm)	77.8 ± 5.0	42.4 ± 3.7	34.7 ± 2.7
Void depth (μm)	5.2 ± 0.5	4.3 ± 0.6	4.8 ± 0.7
Hardness (HV_0.5)	404.2 ± 7.6	412.8 ± 5.9	423.2 ± 6.7

Figure 6.

OM and SEM images of the microsurface of the Ti6Al4V samples under different rescanning strategies: (a) OM image of the SLM sample, (b) OM image of the re-SLM sample, (c) OM image of the re-SLM+IL1 sample, (d) SEM image of the SLM sample, (e) SEM image of the re-SLM+IL1 sample, and (f) SEM image of the re-SLM sample.

In this study, white light interference profilometry was used to measure the void width and void depth, using light interference fringes on different surfaces to measure the size of pores. As can be seen in Figures 6(d) and 7(a), an unmelted powder particle with a width of about 50 μm existed at the bottom of the sample “SLM.” Compared to the sample “SLM,” the average void width of sample “re-SLM+IL1” and “re-SLM” went down to 42.4 and 34.7 μm from 77.8 μm, respectively, while the void depth kept almost the same. It can be concluded that the rescanning strategy and combined scanning strategy had a positive effect on reducing defects width.

Figure 7.

Microscopic three-dimensional topography of voids obtained from samples: (a) SLM, (b) re-SLM+IL1, and (c) re-SLM.

Table 3 also showed the hardness of the different samples. The hardness of “SLM,”“re-SLM,” and “re-SLM+IL1” samples was 404.2, 412.8, and 432.2 HV_0.5, respectively. When the bearing surface was subjected to pressure, the Ti6Al4V alloy with defects deformed along a plane perpendicular to the load. This was mainly caused by the lack of carrying capacity of the pores. The pores of samples released the load and expanded the indentation. As mentioned before, the relative density value ranked in a descending order of re-SLM > re-SLM+IL1 > SLM. Therefore, the microstructure became dense due to the remelting of pores. This is the reason why the hardness of samples printed by rescanning strategy was higher than that of samples without rescanning.

Residual stress

Residual stress of samples was measured using the hole-drilling method. First, the three-dimensional strain rosette was attached to the center of the measurement area. Then, a hole was drilled in the center area of the strain rosette. At last, the strain rosette was penetrated, and a hole with a depth of 1–3 mm was punched on the surface of the sample. The residual stress testing data of samples rescanned with zero, one, and two layers interval was shown in Figure 8, the value (353 MPa) of residual stress of samples without rescanning was marked in the image with a dashed line. It can be observed that rescanning applied on each layer had the best effect to release residual stress of the sample. The residual stress of the sample increased with the increasing of the rescanning interlayers times. The rescanning release of residual stress comes from the micro-molten pool formed from the upper layer to lower layer, the temperature gradient of the remelted molten pool was lower than gradient of the primary formed molten pool, so the residual stress will be released. The enhancement of laser energy caused the increasing of the micro-molten pool area, and the remelting effect of releasing residual stress also enhanced. If rescanning strategy melted all the melting tracks formed in the first scanning, the residual stress will all be affected by the temperature gradient of remelting. According to the principle of solidification and crystallization, the energy entering the molten pool increased, the molten pool maximum temperature and the life of the liquid phase also increase, so the temperature gradient of the molten pool decreases, resulting in the decrease of the average cooling rate and the decrease of metal subcooling degree, making the microstructure transformation more uniform which led to the reduction of the microstructure stress. With the extension of the high temperature time on the printing layer, the thermal deformation of the unevenly heated parts of the printing layer reduced, which was beneficial to the release of residual stress after cooling.³⁶ Compared with the sample without rescanning, the residual stress of the remelted sample reduced. Rescanning strategy applied on each layer had a relative optimal effect on reducing the residual stress. As the times of rescanning interlayers increased, the improvement effect of rescanning on residual stress showed a gradual decreasing trend.

Figure 8.

Residual stresses of samples under different rescanning strategies and different process parameters: (a) hatching space and (b) laser power.

As can be seen from Figure 8(a), the value of residual stress under rescanning and combined rescanning strategy all reached minimum when the hatch spacing was 0.1 mm. When the hatch spacing was 0.05 and 0.075 mm respectively, the interval between the melt tracks of the printing layer was so short that the laser beam of rescanning crossed two or three tracks to remelt the solidified surface, which means that the hatch spacing was reset during rescanning, and the residual stress did not reduce after redistribution. When the hatch spacing was 0.125 and 0.15 mm, the interval between the melt tracks of the printing layer was so long that rescanning did not fully remelt the melt tracks formed in the first scanning, so the residual stress was not fully released. When the hatch spacing was 0.1 mm, the laser beam of rescanning remelted the printing layer along the melt tracks formed in the first scanning, the residual stress was released during the remelting process.

The influence of rescanning laser power on residual stress of parts was shown in Figure 8(b). The residual stress of sample under 80 W was higher than sample without rescanning, it can be inferred that the laser energy was insufficient to remelt the solidified metal surface and the residual stress generated in the first scanning strategy was not released but slightly increased. When the laser power was increased to 100 W, laser rescanned the solidified surface to form the micro-molten pool without any residual stress inside. Rescanning released the residual stress around micro-molten pool, so the residual stress of parts declined. When the laser power was increased to 120 W, the area of the micro-molten pool was enlarged, and the ability to release the residual stress was improved. The residual stress of this sample was lower than that of the 100 W sample. The residual stress of the combined rescanning under the laser power of 140 W was minimum in the “re-SLM+IL1” curve. The once scanning layer cannot optimize the residual stress of the rescanning layer by the remelting with a laser power of 120 W. Therefore, laser energy with a laser power of 140 W was required to reduce residual stress by combined rescanning. The residual stress rose from when the laser power increased from 140 to 160 W which was considered to the fact that laser energy of rescanning can fully remelt the melt tracks in the first scanning. The residual stress depended on the remelting temperature gradient rather than the first scanning temperature gradient. The rising of laser power means the rising of laser energy applied in the molten pool, so the temperature gradient around the molten pool increases, resulting to an increasing in residual stress of parts. However, the change of the molten pool caused by the scanning strategy of rescanning every three layers was almost negligible for the formed part. Therefore, the rescanning with two-layer interval almost had no optimization effect on residual stress.

This decrease of residual stress is not free and an additional time for the fabrication is still necessary. In the process of forming parts via selective laser melting, the time for manufacturing parts includes powder spreading time and laser scanning time. The time saved is not fixed, due to different powder spreading devices (speed of roller), sample size and path planning. The total parts manufacturing time can be calculated by the following formula:

T_{m} = T_{p} + [it / (n + 1) + it]

(2)

Where T_m is the total parts manufacturing time, T_p is the total powder spreading time, t is the laser scanning time for once scanning of single layer, i is the number of total sliced layers, and n is the rescanning interlayers times. Compared with rescanning (layer by layer), rescanning with one-layer interval can save 25% of theoretical laser scanning time.

Conclusions

In this study, SLMed Ti6Al4V samples rescanned with zero, one, and two interval layers were fabricated under different laser power and hatch spacing, the influence of rescanning with interval layers on the molten pool solidification behavior, relative density, and microstructure were systematically investigated. In addition, the residual stress of Ti6Al4V alloy under different rescanning parameters were compared and studied. Compared with rescanning strategy without layer interval, combined rescanning strategy can improve the processing efficiency and shorten the forming time. The detail conclusions can be drawn as follows:

According to the simulation results, average cooling rate and temperature of the molten pool under different rescanning strategies were calculated and studied. The simulation results suggested that rescanning strategy and combined rescanning strategy can both decrease the cooling rate of molten pool.

In terms of microstructure, the acicular α′ phase of samples by remelting turned finer, and grains grew across the melt tracks. The relative density of Ti6Al4V alloy samples under rescanning strategy and combined rescanning strategy was higher than that of samples under once scanning strategy. Additional remelting can reduce the width of the pores and improve the relative density of the samples by rescanning or combined rescanning strategies.

In combined rescanning strategies, residual stress of parts can be released from 353 to 294 MPa with the process parameters of laser power 140 W, scanning speed 1200 mm/s, hatch spacing 0.1 mm, and layer thickness 0.03 mm. However, the mechanism of combined scanning on releasing residual stress of upper and lower layers should be explored in the future.

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Project of Team of National Defense Science and Technology Innovation of China (No. 2016300TS009).

ORCID iDs

Yiqing Ma

Weipeng Duan

Xiaojin Miao

Jitai Han

References

Yang

Wang

, et al. Rapid fabrication of metallic mechanism joints by selective laser melting. Proc IMechE, Part B: J Engineering Manufacture 2011; 225(12): 2249–2256.

Behera

Dougherty

Singamneni

Selective laser melting of stainless steel and silicon nitride fibre metal matrix composites. Proc IMechE, Part B: J Engineering Manufacture 2020; 234(12): 1513–1525.

Tang

Chen

, et al. Numerical simulation of selective laser melting temperature conduction behavior of H13 steel in different models. Optik 2020; 201: 163336.

Denlinger

Heigel

Michaleris

Residual stress and distortion modeling of electron beam direct manufacturing Ti-6Al-4V. Proc IMechE, Part B: J Engineering Manufacture 2015; 229(10): 1803–1813.

Polozov

Sufiiarov

Kantyukov

, et al. Microstructure, densification, and mechanical properties of titanium intermetallic alloy manufactured by laser powder bed fusion additive manufacturing with high-temperature preheating using gas atomized and mechanically alloyed plasma spheroidized powders. Addit Manuf 2020; 34: 101374.

Fang

Huang

, et al. Review on residual stress in selective laser melting additive manufacturing of alloy parts. Opt Laser Technol 2020; 129: 106283.

Keller

Mokhtari

Vieille

, et al. Influence of a rescanning strategy with different laser powers on the microstructure and mechanical properties of Hastelloy X elaborated by powder bed fusion. Mater Sci Eng A 2021; 803:140474.

Hosseini Yekta

Sadough Vanini

Pouyafar

. Feasibility of partial remelting process in globular microstructure preparation for semi-solid processing of M2 high-speed steel. Proc IMechE, Part B: J Engineering Manufacture 2015; 229(6): 1002–1009.

Pal

Lojen

Hudak

, et al. As-fabricated surface morphologies of Ti-6Al-4V samples fabricated by different laser processing parameters in selective laser melting. Addit Manuf 2020; 33: 101147.

10.

Kruth

Deckers

Yasa

, et al. Assessing and comparing influencing factors of residual stresses in selective laser melting using a novel analysis method. Proc IMechE, Part B: J Engineering Manufacture 2012; 226(6): 980–991.

11.

Duan

Han

Effect of laser rescanning on Ti6Al4V microstructure during selective laser melting. Proc IMechE, Part B: J Engineering Manufacture 2021; 235(5): 763–771.

12.

Griffiths

Rossell

Croteau

, et al. Effect of laser rescanning on the grain microstructure of a selective laser melted Al-Mg-Zr alloy. Mater Charact 2018; 143: 34–42.

13.

Shiomi

Osakada

Nakamura

, et al. Residual stress within metallic model made by selective laser melting process. CIRP Ann Manuf Technol 2004; 53(1): 195–198.

14.

Sing

Chua

, et al. Influence of re-melting on surface roughness and porosity of AlSi10Mg parts fabricated by selective laser melting. J Alloys Comp 2019; 792: 574–581.

15.

Qiu

Wang

Aladawi

, et al. Influence of laser processing strategy and remelting on surface structure and porosity development during selective laser melting of a metallic material. Metallurgical Mater Trans A 2019; 50(9): 4423–4434.

16.

Demir

Previtali

Investigation of remelting and preheating in SLM of 18Ni300 maraging steel as corrective and preventive measures for porosity reduction. Int J Adv Manuf Technol 2017; 93(5–8): 2697–2709.

17.

Miao

Han

, et al. Effect of laser rescanning on the characteristics and residual stress of selective laser melted Titanium Ti6Al4V alloy. Materials 2020; 13(18): 3940.

18.

Xiao

Chen

, et al. Effect of rescanning cycles on the characteristics of selective laser melting of Ti6Al4V. Opt Laser Technol 2020; 122: 105890.

19.

Ali

Ghadbeigi

Mumtaz

Effect of scanning strategies on residual stress and mechanical properties of selective laser melted Ti6Al4V. Mater Sci Eng A 2018; 712: 175–187.

20.

Karimi

Suryanarayana

Okulov

, et al. Selective laser melting of Ti6Al4V: effect of laser re-melting. Mater Sci Eng A 2021; 805:140558.

21.

Boivineau

Cagran

Doytier

, et al. Thermophysical properties of solid and liquid Ti-6Al-4V (TA6V) alloy. Int J Thermophys 2006; 27(2): 507–529.

22.

Parry

Ashcroft

Wildman

RD.

Understanding the effect of laser scan strategy on residual stress in selective laser melting through thermo-mechanical simulation. Addit Manuf 2016; 12: 1–15.

23.

Schoinochoritis

Chantzis

Salonitis

Simulation of metallic powder bed additive manufacturing processes with the finite element method: a critical review. Proc IMechE, Part B: J Engineering Manufacture 2017; 231(1): 96–117.

24.

Liu

Selective laser remelting of an additive layer manufacturing process on AlSi10Mg. Results Phys 2019; 12: 982–988.

25.

Pal

Gubeljak

Hudák

, et al. Evolution of the metallurgical properties of Ti-6Al-4V, produced with different laser processing parameters, at constant energy density in selective laser melting. Results Phys 2020; 17: 103186.

26.

Majeed

Khan

Rasheed

Finite element analysis of melt pool thermal characteristics with passing laser in SLM process. Optik 2019; 194: 163068.

27.

Kenel

Dasargyri

Bauer

, et al. Selective laser melting of an oxide dispersion strengthened (ODS) γ-TiAl alloy towards production of complex structures. Mater Des 2017; 134: 81–90.

28.

Ahmed

Rack

HJ.

Phase transformations during cooling in α+β titanium alloys. Mater Sci Eng A 1998; 243(1–2): 206–211.

29.

Sallica-Leva

Caram

Jardini

, et al. Ductility improvement due to martensite α' decomposition in porous Ti-6Al-4V parts produced by selective laser melting for orthopedic implants. J Mech Behav Biomed Mater 2016; 54: 149–158.

30.

Xia

Liu

, et al. Modeling and numerical studies of selective laser melting: multiphase flow, solidification and heat transfer. Mater Des 2020; 196: 109115.

31.

Thijs

Verhaeghe

Craeghs

, et al. A study of the microstructural evolution during selective laser melting of Ti–6Al–4V. Acta Mater 2010; 58(9): 3303–3312.

32.

Song

, et al. Micro selective laser melting of NiTi shape memory alloy: defects, microstructures and thermal/mechanical properties. Opt Laser Technol 2020; 131: 106374.

33.

Kumar Mishra

Kumar

. Effect of surface morphology on the melt pool geometry in single track selective laser melting. Mater Today Proc 2020; 27(2): 816–823.

34.

Vaziri

Shahverdi

Shabestari

, et al. Effect of re-scanning on tribological characterization of laser surface alloyed layers. Mater Des 2010; 31(8): 3875–3879.

35.

Yadroitsava

Els

Booysen

, et al. Peculiarities of single track formation from TI6AL4V alloy at different laser power densities by selective laser melting. S Afr J Ind Eng 2015; 26(3): 86–95.

36.

Chen

Yin

Zhu

, et al. Effect of overlap rate and pattern on residual stress in selective laser melting. Int J Mach Tools Manuf 2019; 145: 103433.