Enhancement of Scan Strategy for Improving Surface Characteristics of the SLMed Inconel 718 Alloy Component

Abstract

The technique of selective laser melting has garnered significant interest due to its capacity to fabricate metal parts of any shape using a single printing process. In this study, a scan strategy was proposed for printing the inner structure part based on the varying performance of the Inconel 718 alloy part produced using different scan techniques. The test findings indicated that the surface quality of the printed material along the scan line exhibited significantly superior performance in comparison to that produced in a vertical direction to the scan line. Regarding the scan strategy, it is evident that the zigzag shape scan strategy exhibited superior performance in terms of the microstructure of the printed part. This is attributed to the more consistent cooling rate on each scan track. On the other hand, the square-framed scan strategy demonstrated a relatively optimal condition for the bending deformation of the printed part compared to other scan strategies. This is due to the more appropriate behavior of the molten pool during the outer edge printing process. After considering all of these criteria, a novel scan approach was created that included various scan strategies. The test findings indicated that the component produced using this novel scanning technique exhibited the combined benefits of both the square-shaped and zigzag-shaped scanning strategies.

1. Introduction

The rapid development of additive manufacturing technology in recent years can be attributed to its distinct forming mechanism, which sets it apart from traditional manufacturing processes. By sequentially constructing the designated components layer by layer, based on the information provided for each layer, this approach has the potential to immediately produce the required shape of the part without requiring additional processing [1]. This technology is garnering significant attention from several fields due to this very reason. Additive manufacturing encompasses various specific technologies, including fused deposition modeling (FDM), stereolithography (SLA), and others. Fused deposition modeling (FDM) primarily utilizes a heating process to melt the plastic wire material within the extrusion head, allowing it to take the required shape on the workplate [2,3]. Stereolithography appearance (SLA) utilizes photosensitive resin as its raw material and employs ultraviolet light to sculpt the desired form. Selective laser melting, a very promising method in additive manufacturing, offers distinct benefits over other AM technologies due to its capacity to fabricate metal components. The metal powder was melted layer by layer using laser energy until the entire object was produced [4]. Figure 1a displays the schematic diagram of SLM. Figure 1b depicted the entire manufacturing process.

Due to the numerous benefits of Selective Laser Melting (SLM), researchers from various disciplines are currently directing their efforts towards enhancing the properties of printed components to expand their potential applications. Optimizing the processing settings during the printing process was undoubtedly crucial for enhancing the overall performance of the printed pieces. The scan technique, which refers to the trajectory of the laser energy throughout the printing process, has received significant attention from numerous studies owing to its critical significance. Mertens et al. [5] modified the scan technique in the printing process and discovered that by adjusting the tilt and process parameters, the quality of the downward-facing surface can be enhanced. Rashid et al. [6] employed several scanning procedures to fabricate samples of 17-4PH stainless steel. They discovered that the hardness of the sample produced using the double scan strategy was greater than that of the sample produced using the single scan strategy. Su et al. [7] examined the degree of overlap in the printing process. The researchers discovered that using an intra-layer type scan method resulted in a relatively ideal deposition efficiency in the selective laser melting (SLM) process. Carter et al. [8] discovered that the scanning technique had a substantial influence on the grain structure of the nickel-based superalloy CM247LC component, mostly because of the varying sizes of the repeating pattern. In their study, Han et al. [9] introduced a novel scan method that was found to be more appropriate for printing metal parts with superior surface quality compared to the other scan strategies employed in their research. In their study, Jhabvala et al. [10] examined the impact of the scan strategy on the temperature distribution of the molten pool throughout the forming process. They investigated the effects of several scan strategies on the homogenous heating of the part using simulation methods. In their study, Qian et al. [11] employed a helix scan approach and observed a substantial decrease in the distortion of the printed layer. Ali et al. [12] conducted a study to examine the correlation between scan approach and residual stress in the printed component. Based on the testing results, it was concluded that the TC4 part printed using a 90° alternating scan approach exhibited the lowest residual stress. Parry et al. [13] employed a simulation method to investigate the temperature distribution of the molten pool during the forming process, specifically under various scan strategies. Based on the simulation findings, they put up a novel scan strategy for printing TC4 components. In their study, Dai et al. [14] observed that the choice of scan approach led to variations in the behavior of the molten pool boundaries. Additionally, they noted that grain formation exhibited distinct characteristics during this process. In their study, Song et al. [15] employed both finite element analysis and experimentation to investigate the impact of the scan approach on the forming process. The researchers discovered that variations in the temperature distribution of the molten pool led to distinct changes in the residual stress behavior. Based on the modeling results, a new scan strategy was presented that involved rotating 15° on each layer. The experimental results showed that the part produced using this new scan strategy had a relatively ideal residual stress. In their study, Geiger et al. [16] employed various scan procedures to fabricate an IN 738LC part. They observed that the creation of grains was significantly influenced by the scanning direction, leading to variations in the properties of the printed component. Yan et al. [17] employed the LCR method to assess the residual stress of the 316L stainless steel component produced with varying residual stress levels. They discovered that the chess board scan approach outperformed the meander and stripe methods. Koutny et al. [18] conducted a comparison of the properties of a printed item using different scan techniques. The researchers discovered that the relative density exhibited an increase while employing the chessboard scan approach, although there was only minimal benefit in reducing faults and cracks. Cheng et al. [19] employed finite element analysis to investigate the correlation between scan techniques and residual stress in the printed component. It is evident that employing a 45° inclined line scanning technique can effectively decrease residual stress in both horizontal and vertical directions.

The testing findings clearly demonstrate that the scan method has a significant influence on the behavior of the molten pool during the printing process, leading to variations in the performance of the printed samples. Furthermore, the scan approach has an impact on the surface property, but less research has been conducted on this particular issue. This study focused on researching the scan strategy and its impact on the surface quality of the printed part. The scan strategy was then optimized to obtain a greater surface quality in the printed sample.

2. Materials and Instrument

2.1. Material

The Inconel 718 powder utilized in this study was supplied by Falcon Tech. Co. Ltd., Wuxi, China, and was produced by the gas atomization technique.

Table 1. Information about the powder used in this work.

Chemical Composition /wt.%	Cr	Fe	Nb	Mo	Ti	Al	Co
	18.93	17.32	5.18	3.08	0.92	0.47	0.041
	Cu	C	Si	Mn	P	S	Ni
	0.058	0.05	0.056	0.13	0.0058	0.0038	Bal
Particle size distribution/μm	D10		D50			D90
	22.3		33.2			50.4

displayed the chemical composition and diameter of the powder utilized in this study.

2.2. Instrument

The Inconel 718 powder was fabricated utilizing the YLM-120 Selective Laser Melting (SLM) process, manufactured by Jiangsu Yongnian Laser Forming Technology Co., Ltd., (Suzhou, China), as depicted in Figure 2. The process parameters employed in this study were documented in

Table 2. Process parameters used in this work.

Process Parameters	Laser Power	Scan Speed	Layer Thickness	Defocusing Amount	Hatch Spacing	Protective Gas	Building Direction
Value	200 W	1000 mm/s	70 μm	0.0 mm	0.08 mm	Argon	Y-axis

, while the scan tactics utilized may be observed in Figure 3.

The Mitutoyo Roughometer, Surftest SJ-410 (Mitutoyo, Kawasaki, Japan), a device provided by Mitutoyo in Japan, was used to measure surface roughness. Given the compact size of the design sketch, both the acceleration and deceleration lengths used for measurement were 0.45 mm, while the total length for each measurement was 0.9 mm. In order to minimize measurement errors due to position, five locations were measured on a single surface, with a spacing of 1μm between each measurement. In this study, the surface flatness of the printed part was also examined to provide a more thorough comprehension of its surface quality. The measurement was mostly conducted using the RA-7525SEI-4, a Trilinear Coordinates Measuring Instrument manufactured by Hexagon Metrocogy, RA-7525SEI-4 (Hexagon, Stockholm, Sweden). The scanning approach utilized a triangle laser measurement technique to obtain highly precise findings. The scanner head utilized in this case was the HP-L-20.8, with a working distance of 180 ± 40 mm. The shape inaccuracy had an approximate magnitude of 9 μm, occurring at a frequency of 100 Hz. The scanning mode employed was ultrafine quality, ensuring high precision in the measurement procedure. The point cloud data obtained from the scanning process were loaded into the Polyworks program, where it was aligned with the design drawing. The flatness value was determined by extracting the specified surface, and all the computations were performed using Polyworks 2016.

In order to elucidate the previously reported phenomenon, the surface morphology of the printed component was examined using a white light interferometer (WLI) provided by RTEC, MFD-D (RTEC, San Jose, CA, USA). The device had a magnification of 20×, a working distance of 4.7 mm, and an aperture of 0.4 mm. The apparatus had a field of view, spatial sampling, and resolution of 860 × 650 μm², 0.34 μm, and 0.35 μm, respectively. In this study, a scanning electron microscope (SEM) provided by Carl Zeiss, Sigma 360, (Carl Zeiss, Jena, Germany) was also utilized. The voltage applied was 20 kilovolts (20 kV), and the equipment utilized a SE2 type detector. The distance between the object being observed and the lens, known as the working distance, was measured to be 8.7 mm. The magnification rate, which indicates how much larger the object appears compared to its actual size, was determined to be 42X. The measuring instruments utilized in this study were depicted in Figure 4.

3. Results and Discussion

The study focused on analyzing the surface roughness of the samples printed using various scan methodologies. Specifically, the side surfaces and positive surface were examined independently. Since the negative surface was attached to the workplate, a slow wire cutting machine was used to separate the sample from the workplate, which was irrelevant for the investigation. The precise measurement of the surface area was presented in Figure 5. The surface roughness value was documented in

Table 3. The measured value of the surface roughness.

Measured Position	Positive Surface/μm	Side Surface 1/μm	Side Surface 2/μm	Side Surface 3/μm	Side Surface 4/μm
Sample 1	6.003	9.745	12.785	10.021	14.277
Sample 2	5.286	12.533	9.213	12.319	11.882
Sample 3	5.229	10.257	13.522	10.503	13.963
Sample 4	5.471	8.160	8.305	9.007	10.452

.

To give a more intuitive view of the whole variation trend of the surface roughness caused by different scan strategies, a line chart was generated accordingly using Origin Lab V8.0, and the line chart is shown in Figure 6.

Figure 6 illustrates that the Ra value of side surface roughness was greater than that of positive surface roughness. The primary cause was primarily attributed to the additional powder adhering to the lateral surface during the printing procedure. Since the powder was already present on the side surface during the printing process, the laser spot can generate more heat near the edge, causing the extra powder to melt during the forming process. The powder adhered to the side surface. However, during the formation process of the positive surface, the powder intended for the subsequent layer printing was not spread out, resulting in the inability of the powder to bond during the forming process. Figure 7 displayed the morphology of both the positive and side surfaces.

Figure 7 demonstrates that the side surface exhibited substantial powder bonding, while the positive surface had a noticeable fluctuation wave. The distinct morphology found on the positive and side surfaces corroborated the aforementioned argument.

Regarding the various lateral surfaces, distinct disparities are also evident. For instance, the second side surface of Sample 2 exhibited a notably lower elevation in comparison to the first side surface of the same sample. Based on the comprehensive testing findings, it can be inferred that the side surface roughness is reduced when printed along the scan path compared to when printed vertically to the scan path. The power fluctuation primarily caused the surface route to turn. In order to prevent the occurrence of failure in the printed part, a time compensation of approximately 0.3 s was consistently implemented during the turning process. While the powder can be effectively blended in the turning position, more powder was more prone to being bonded in this process, leading to a decrease in surface quality. In summary, in order to enhance the roughness of the side surface, it was found that printing along the scan path was more effective than printing perpendicular to the scan path, as it reduced the influence of power fluctuations. The morphologies of the side surface 2 were documented using WLI, as depicted in Figure 8. This provided additional confirmation for the explanations provided in this context.

Surface flatness was also studied in this work. The detailed information on the testing data was shown in

Table 4. The measured value of the surface flatness.

Measured Position	Positive Surface/mm	Side Surface 1/mm	Side Surface 2/mm	Side Surface 3/mm
Sample 1	0.088	0.083	0.075	0.087
Sample 2	0.096	0.082	0.071	0.088
Sample 3	0.081	0.086	0.073	0.084
Sample 4	0.099	0.081	0.077	0.082

. A line chart was also given in Figure 9.

The testing data reveals that the printed output on different side surfaces showed minimal variation under varied scan strategies. However, the difference on the positive surface was more pronounced compared to the side surfaces. The main cause of this was mostly due to the contrasting attributes of the positive and side surfaces. The positive surface was created as a single plane using laser-induced printing, whereas the side surface was made by the gradual accumulation of each layer. This indicates that the positive surface exhibited a much greater width in comparison to the side surface. The dimension of the surface has a strong influence on bending deformation, which is a major factor affecting surface flatness. The correlation between the dimension and deformation sketch was shown in the following manner, accompanied by a schematic diagram and a fitting image of an actual printing part displayed in Figure 10.

In summary, the square-framed scan method was better suited for printing samples with lower side surface roughness, whereas the chessboard scan strategy resulted in samples with relatively ideal positive surface flatness. The combination of two scan techniques is seen in Figure 11. The measured data and the line charts were presented in

Table 5. The surface roughness and flatness of the parts printing using new proposed scan strategies.

Roughness	Positive Surface/μm	Side Surface 1/μm	Side Surface 2/μm	Side Surface 3/μm	Side Surface 4/μm
Sample 5	5.351	8.035	7.993	7.645	9.225
Sample 6	5.422	8.116	8.179	8.331	9.550
Sample 7	5.627	7.967	8.003	8.074	9.961
Flatness	Positive surface/mm	Side surface 1/mm	Side surface 2/mm	Side surface 3/mm
Sample 5	0.083	0.072	0.070	0.072
Sample 6	0.085	0.071	0.071	0.075
Sample 7	0.088	0.073	0.071	0.073

and Figure 12, respectively.

The Suoyan HVS-1000 Digital Micro Vickers Hardness Tester was used to do microhardness analysis. The Vickers hardness test was conducted on the XOY surface after it was polished. The test involves applying a pressure of 200 g for a duration of 12 s on a randomly selected flat area on the XOY surface. Five more flat areas are also picked for the test. Any measurement values that deviate significantly from the largest value are discarded, and the average is calculated using the remaining measurement values. Indentation diagram can be seen in Figure 13. Detailed testing results can be seen in

Table 6. Microhardness measurements conducted on printed items using different scan strategies.

Scan Strategy	Figure 11a	Figure 11b	Figure 11c
Microhardness (HV)	348	345	355

.

The findings above clearly demonstrate that the implementation of the new scan algorithms led to a substantial enhancement in the overall surface quality of the printed item. Due to variations in the number of circles employed in square-framed scan techniques, a slight discrepancy in positive surface flatness can be observed in this study. It is evident that as the number of cycles increases, the positive surface flatness exhibits a declining tendency. The correlation between the number of cycles used by the square-framed scan approach and the positive surface flatness is such that a smaller number of cycles results in superior flatness. Considering all the findings from this study, it is evident that the sample produced by combining the one-circle square-framed scan strategy and the chess board scan strategy resulted in a significantly higher overall surface quality compared to samples printed using alternative scan strategies. As for the microhardness, few differences can be seen on this index printed under these three different scan strategies.

4. Conclusions

This study introduces a novel scan approach for printing Inconel 718 parts, combining a square-framed scan strategy with a chess board scan strategy. The specific conclusions were listed as follows:

(1)

The side surface roughness exhibited superior performance when printed along the scan path, as opposed to when printed vertically to the scan path, due to the fluctuation of laser power at the turning location. The flatness value exhibited minimal variation when printed using different scan paths.

(2)

The primary cause of variations in surface flatness is bending deformation, with the positive surface being the most affected due to the accumulation of deformation and its larger area compared to the side surface. The findings indicated that the chess board scan strategy exhibited a relatively ideal condition as a result of the printing process having a more uniform temperature distribution.

(3)

The newly proposed scan strategy combines the benefits of both the square-framed scan strategy and the chess board scan strategy. The flatness value of the positive surface was nearly identical to that printed using the chess board scan strategy, while the roughness value of the side surface was even lower compared to that printed using the square-framed scan method.