1 Introduction

Laser metal deposition (LMD) is the most common type of directed energy deposition (DED) technology that uses a laser to fabricate high-strength, short-duration metal components directly from a computer-aided design (CAD) model without the use of molds [1]. As shown in Fig. 1, LMD uses laser power as a heat source to melt the substrate, and the raw materials are fed into the molten pool in the form of metal wire or powder. A single track is formed with the movement of the laser beam, and then a 3D part is formed as this movement is continued layer upon layer [2, 3]. LMD is superior to traditional technologies in terms of geometric freedom, production flexibility, and low thermal input; thus, it has attracted considerable attention in the fields of aerospace, aviation, and mold as an effective and efficient process to fabricate metal and cladding parts and repair and refurbish damaged parts [4, 5].

Fig. 1
figure 1

Schematic of LMD: a wire feeding and b powder feeding

Full size image

Although LMD is an important technique for manufacturing, repairing, and refurbishing damaged components, its lack of consistent and reliable quality limits its applications. Several scholars have analyzed the influence of LMD process parameters, including laser power, scanning speed, and powder flow rate, on the properties of the fabricated or remanufactured parts to increase the quality and reliability of these parts. Riquelme et al. [6] have investigated the influence of laser power and scanning speed on the microstructure of Al/SiCp composites during LMD and evaluated the wear behavior and microhardness of the fabricated parts. Eo et al. [2] have studied the effects of the melt pool oxidation and inclusion characteristics of LMD 316L stainless steel on the mechanical properties and their relation to laser power, scanning speed, and powder chemistry. Xiang et al. [7] compared the microstructures and mechanical properties of CrMnFeCoNi prepared using LMD and casting methods. The results showed that the samples prepared using LMD contain columnar and equiaxed grains and that the proportion of these grain structures can be modified by changing the laser power, whereas those prepared through casting possess a coarse dendritic structure. In addition, the mechanical properties of the 1400-W samples prepared using LMD are better than those prepared using casting. Xue et al. [8] investigated the mechanism by which various laser energy densities affect the microstructure and microhardness of fabricated parts during LMD. They found that a decrease in laser energy density (LED) can alter the microstructure of the parts, decrease their sub-grain size, and increase their microhardness.

Numerous studies have focused on improving the properties of fabricated parts during LMD by controlling the process parameters in a continuous-wave laser. However, the effects of pulsed laser parameters on these properties remain unclear despite their significant effect on the microstructure and mechanical properties of parts prepared using LMD [9,10,11,12,13]. The scanning pattern also considerably influences the fine structure and mechanical properties of parts manufactured via LMD using continuous-wave lasers [14,15,16,17,18,19,20], but its effects when pulse wave lasers are used remain to be determined. Gharbi et al. [9] investigated the effects of pulsed laser parameters on the surface finish caused by LMD on titanium alloy (Ti6Al4V). They demonstrated that high mean power enhances surface quality, and the usage of a pulsed mode with several duty cycles provides clear smoothening effects. Li et al. [10] found that the microstructure during pulsed laser processing shows finer columnar dendrites than that during continuous laser processing. In addition, the dendrite development direction is more oriented toward the laser scanning direction than the continuous laser power sample. Pinkerton et al. [11] found that pulsed laser may reduce the porosity of fabricated parts more during LMD than during continuous laser processing. Ravi et al. [12] found that laser mode and laser power significantly affect the grain structure. The continuous-wave mode usually leads to large columnar crystals, whereas the pulse wave mode produces finer equiaxed crystals. The higher the laser power, the larger the grain size and the coarser the microstructure. Moat et al. [13] investigated the influence of laser pulse length and duty cycle on multi-track residual stress distributions during laser direct metal deposition. The results demonstrate that the tensile stress gradient increases while the residual stresses decrease with the duty cycle.

During LMD, the scanning strategy plays an important role in improving the accuracy and performance of fabricated parts because of the variation in thermal behavior when different path strategies are used. Zhou et al. [14] studied the microstructure and mechanical properties of TC4 samples fabricated through LMD using different scanning strategies. They found that the one-way scanning method provides finer grains and a stronger texture index than the cross-scanning method. Moreover, the one-way scanning method has higher ultimate tensile strength and plasticity than the cross-scanning method. Wan et al. [15] found that the crystalline structure, tensile strength, and fatigue strength change with varying scanning directions. Woo et al. [19] investigated the effect of scanning patterns on different microstructural properties of ferritic and austenitic steel. In both parts, epitaxial grain expansion along the construction direction produces an equiaxed grain structure. The ferritic steel parts exhibit a preferentially tilted grain pattern in the bidirectional scanning method.

Despite the tremendous efforts of many researchers, several aspects of LMD remain poorly understood, which affects the efficiency of the products and limits the application and reliability of this technology. Therefore, further studies should be conducted to determine the potential advantages of LMD. Within this context, consideration of the influence of pulsed laser parameters and different scanning strategies on the microstructure and the mechanical and geometric properties of additively manufactured walls using LMD is a relatively new addition to this field. In this study, the LMD was utilized to fabricate 316L stainless thin-walled parts. Two strategies were used to investigate the capability of LMD: (1) changing pulsed laser parameters according to (a) different frequencies, same duty cycle, and (b) different duty cycles, same frequency; and (2) changing the path strategy (unidirectional and bidirectional scanning strategies) during pulsed laser. The microstructure and the microhardness profile of the macro cross section of the samples were characterized. This study also investigated the influence of pulsed laser parameters and scanning strategies on the geometric stability of the printed wall of the fabricated samples.

2 Experimental Procedures

2.1 Materials

AISI316L stainless steel was used as the substrate and powder materials in the experiments. The chemical composition and the mechanical and thermal properties are shown in Tables 1 and 2, respectively. The powder had a size of 45–120 µm and a spherical shape. The substrates were machined to dimensions of 120 mm × 15 mm with a thickness of 10 mm, polished, and then cleaned carefully before LMD.

Table 1 Chemical compositions of 316L stainless steel
Full size table
Table 2 Mechanical and thermal properties
Full size table

2.2 LMD Production

AISI316L LMD thin-walled parts were produced using a laser material deposition system, as illustrated in Fig. 2. The system consisted of a LaserLine diode laser generator with a 4000-W maximum power supply (Manual LDF 4000-100 VGP),a six-axis manipulator (KUKA-ZH 30/60III) that can reach 2233 mm with a total load of 65 kg and a payload of 30 kg, a Precitec laser head (YC52), and a metal powder feeder (Raychem RC-PGF-D). The metal powder was conveyed through four coaxial nozzles into the deposition point by a pressurized carrier gas (argon). Argon was also used as a shielding gas during LMD. The standoff distance of the laser head was 15 mm above the substrate, and the beam spot diameter was 2 mm.

Fig. 2
figure 2

Schematic of the LMD system

Full size image

LMD experimental work was carried out in two groups (D and F), as presented in Table 3. The process parameters, such as the laser power, scanning speed, and powder flow rate, were fixed during all experiments at 1400 W, 6 mm/s, and 8 g/min, respectively. In all experiments, the Z-increment after adding each layer was 2 mm as the manipulator robot moved up to add the next layer. Each group consisted of two sections that differed in scanning strategy (i.e., unidirectional or bidirectional), as shown in Fig. 3. In group D, the effects of the duty cycle of the pulsed laser on AISI316L LMD thin-walled parts in both scanning patterns were investigated. The frequency was fixed, but the duty cycles were varied in each experiment. The values of duty cycles were started from the pulse width equal to two-thirds of the pulse period (2:1), then three-quarters of the pulse period (3:1), and the last one was four-fifths of the pulse period (4:1). In group F, the effects of pulse frequency on LMD thin-walled parts in both scanning patterns were analyzed. The duty cycle was fixed, but the frequencies varied from low to high in each experiment.

Table 3 Experiments groups
Full size table
Fig. 3
figure 3

Scanning pattern used in the research: a unidirectional (section D1 and F1 experiments), b bidirectional (section D2 and F2 experiments)

Full size image

The specific energy of a laser beam can be expressed as the LED, as shown in Eq. (1), which is defined as the energy input per unit area of the laser beam on the substrate (J/mm2) [21, 22]. The laser power mode in this research is pulsed waves, and the average power can be measured by using Eq. (2). The peak power is 1400 W in all experiments. However, the average power was distinguished in group D experiments as 933.8 W (d11 and d21), 1050 W (d12 and d22), and 1120 W (d13 and d23) because of the variations in the duty cycle. The average power of group F experiments was 700 W in all experiments because the duty cycle was fixed in all experiments. Changing the pulse laser parameters can change the amount of energy distribution within the manufacturing process, as shown in Table 4, which can change the melting behavior of the material [23].

$${\text{LED}} = \frac{{{\text{Laser}}\,{\text{Power}} }}{{{\text{Scan}}\,{\text{speed}}\, \times \,{\text{Beam}}\,{\text{Diameter}}}}$$
(1)
$${\text{Average}}\,{\text{Power}} = \frac{{{\text{Duty}}\,{\text{Cycle }}}}{100}\, \times \,{\text{Peak}}\,{\text{Power}}$$
(2)
Table 4 Laser energy density during manufacturing
Full size table

2.3 Characterization Techniques

AISI316L parts were fabricated using LMD as shown in Fig. 4. They were cut from the middle and then mounted in resin. The samples were ground down from 240 to 2000 grit using SiC papers, polished, and then ultrasonically cleaned with water and alcohol. Then, the samples were etched in a saturated oxalic acid solution for microstructure analysis. The optical microscopic (OM) images were obtained under a LEICA DMi8 optical microscope at a magnification of 50 µm. The 3D measuring laser microscope OLS4000 was used to take cross-sectional images of the samples at a magnification of 800 µm. The microhardness was measured using a Vickers microhardness machine (HVS-1000Z) along with the height of samples, with a dwell period of 10 s and a test load of 200 g. The geometric properties (height, width, and depth) and grain size of the LMD-fabricated parts were measured using ImageJ software 1.53k.

Fig. 4
figure 4

LMD thin-walled samples: a unidirectional pattern fabricated part, b bidirectional pattern fabricated part

Full size image

3 Results and Discussion

3.1 Geometric Properties of LMD-Fabricated Samples

Figure 5 shows the macro cross-section images of LMD thin-walled samples taken under a 3D measuring laser microscope at a magnification of 800 µm. The geometric properties (height, width, and depth) of the samples are shown in Fig. 6.

Fig. 5
figure 5

Macro cross-section images of LMD thin-walled samples from a group D and b group F

Full size image
Fig. 6
figure 6

Geometric properties (height, width, and depth) measured using ImageJ software

Full size image

3.1.1 Impact of Pulsed Laser Parameters on the Geometric Properties of LMD-Fabricated Samples

The cross-section height, width, and depth of the LMD thin-walled parts processed by pulse laser with different parameters are listed in Table 5 and Fig. 7. These properties have been affected during the part fabrication because of the periodic characteristics of the pulse laser. As shown in Fig. 7a, b, the height and width of the parts from group D increase as the duty cycle is increased from 66.7% to 80%. This result can be ascribed to the fact that a higher duty cycle corresponds to higher heat input, and the molten pool temperature can melt more metal material. Figure 7d, e show the height and width of the parts from group F with different frequencies and the same duty cycle (50%). The deposition height and width are the smallest at 13.33 Hz and the highest at 50 Hz. The height and width of the samples increase with increasing frequency. Although the heat input of the pulsed laser is the same due to the same duty cycle, the heat accumulation is different in each experiment.

Table 5 Geometric properties of LMD samples
Full size table
Fig. 7
figure 7

Geometric properties of samples: a, b, and c are the height, width, and depth of group D samples, respectively; d, e, and f are the height, width, and depth of group F samples, respectively

Full size image

Geometric stability cannot be thoroughly evaluated only by the height and width of the deposition layers. The depth is also an important index because it reflects the quality of fabricated parts and directly affects the interlayer bonding strength of formed parts. During LMD, the metal powder and substrate are heated simultaneously to form a molten pool on the substrate surface, allowing the deposited materials to mix with the substrate and create an excellent metallurgical bond [24]. Figure 7 also presents the influence of different laser parameters on melting depth; when the duty cycle is reduced at the same frequency, the laser penetration to the substrate is reduced, as shown in Fig. 7c, because of the reduction in heat input. In contrast to the height and width of the parts from group F, the depth increases with decreasing frequency. This result is because when the frequency decreases, the laser duration in a single pulse cycle is prolonged, and the penetration into the substrate increases, as shown in Fig. 7f. The effects of the pulsed parameters, including duty cycle (D1 and D2 samples) and frequency (F1 and F2 samples), on the geometric properties of the parts are the same in the two scanning patterns (unidirectional and bidirectional). These results suggest that the geometric stability increases with increasing duty cycle but decreases with increasing frequency.

3.1.2 Impact of Scanning Pattern on the Geometric Properties of LMD-Fabricated Samples

The thin-walled parts were fabricated in this study through LMD by moving a laser beam source layer by layer to melt the feed material powder onto a substrate. The design of scanning paths is significant for optimizing the performance of manufactured samples because it changes the thermal history of each layer [25]. The samples from D1 (d11, d12, and d13) and D2 (d21, d22, and d23) only differ in scanning strategy, and other experimental parameters are unchanged. Similarly, the samples from sections F1 and F2 only differ in scanning strategy, and other experimental parameters are unchanged. Given that the unidirectional scanning pattern is used to fabricate the samples from sections D1 and F1, the deposition of the new layer starts from the starting point of the preceding one. Thus, the cooling rate of the produced layer is uniform. In addition, this scanning pattern prolongs the cooling time of the samples. The deposition pattern also affects the shape of the molten pool and temperature gradient during LMD [26]. Thus, in a unidirectional scanning pattern, the subsequent layer is deposited on the previous one at a lower temperature, driving the thermal gradient higher. By comparison, the samples from sections D2 and F2 are produced with a bidirectional pattern, indicating that the preceding layer has a higher temperature, resulting in a low thermal gradient. Given their longer cooling time, higher cooling rate, and higher thermal gradient, the samples from sections D1 and F1 have better surface quality than the samples from sections D2 and F2.

Moreover, the samples from sections D1 and F1 are more geometrically stable than those from sections D2 and F2 in terms of height, width, and penetration depth, as shown in Figs. 5 and 7.

3.2 Microstructure of LMD-Fabricated Samples

Metallographic images of the central region of the samples from six different zones (beginning, center, and end) are shown in Figs. 8 and 9. The laser energy is absorbed when the laser beam interacts with powder particles during LMD, causing the laser energy to attenuate. The residual laser energy reacts with the substrate surface, melting it if the laser beam energy is sufficient. Otherwise, the substrate is not melted. The same process continues when adding a new layer, with the surface of the previous layer acting as the substrate surface. This process forms a molten and recrystallized zone in the surface zone and an interlayer zone where the material is deposited, resulting in changes in microstructure and grain size.

Fig. 8
figure 8

Measured points of grain size in different areas in the thin-walled samples fabricated using LMD

Full size image
Fig. 9
figure 9

Optical microscopic images of samples from a group D and b group F

Full size image

3.2.1 Impact of Pulsed Laser Parameters on the Microstructure of LMD-Fabricated Samples

The microstructure morphology and grain size of the LMD-fabricated samples affect their final mechanical properties. The microstructure depends on the thermal history of the molten pool. The lower the cooling rate, the coarser the grain size and the lower the tensile strength. The temperature gradient and solidification rate affect the grain size and growth direction of the samples. Pulsed laser has the characteristics of low heat input, which makes the cooling rate of the molten pool faster and the grain finer. In addition, the solidification of the pulsed laser processing molten pool is a periodic change process.

As shown in Fig. 10, all samples from the four sections (D1, D2, F1, and F2) have a similar trend in grain size in the three regions (beginning, center, and end). However, the grain size of the samples from the same section has different values in the three regions. The grain size of the samples from section D1 is in the order of d11 < d12 < d13, as shown in Fig. 10a. The same pattern is observed in the samples from section D2, i.e., d21 < d22 < d23.

Fig. 10
figure 10

Grain size change trend diagram in LMD thin-walled samples from a group D and b group F

Full size image

The difference in grain size is that the inputted energy in d11 and d21 is lower than that in d12 and d22, which is lower than the inputted energy in d13 and d23 because of the decreasing duty cycle. The low inputted energy contributes to the refined structure and makes the grain generated by the higher cooling rate of the molten pool smaller. Thus, the grain size increases with the increasing duty cycle from 66.7% to 80%.

Figure 10b presents the grain size of the samples from group F. These samples differ in grain size in each zone, although the inputted power is the same in all samples. The temperature histories of the molten pools are different because of the different durations of a single pulse [27], resulting in different energy distributions and grain sizes between the samples in section F1 (f11, f12, and f13) and between the samples in section F2 (f21, f22, and f23). The grain size is in the order of f11 >> f12 > f13. The same pattern can be observed in the samples from section F2, i.e., f21 > f22 > f23, as illustrated in Fig. 10b. Although the duty cycle is the same in the samples from group F (50%), approximately 13 cycles per second can be found in the f11 and f21 samples, which have a 37.5 ms pulse period followed by a 37.5 ms pulse off for solidification; 20 cycles per second in the f12 and f22 samples, which have a 25 ms pulse period followed by a 25 ms pulse off for solidification; and 50 cycles per second in the f13 and f23 samples, which have a 10 ms pulse period followed by a 10 ms pulse off for solidification. The higher the single pulse energy and the fewer the pulses, the larger the grain size, as observed in the samples from f11 and f21 with a frequency of 13.33 Hz. The lower the single pulse energy and the more pulses, as shown in the samples from f13 and f23 with a frequency of 50 Hz, the smaller the grain size.

3.2.2 Impact of Scanning Pattern on the Microstructure of LMD-Fabricated Samples

Figure 10a shows the grain size of the samples from sections D1 and D2. Figure 10b shows the grain size of the samples from sections F1 and F2. The samples from sections D1 and F1 were manufactured with unidirectional scanning patterns, whereas those from D2 and F2 were manufactured with bidirectional scanning patterns. Although the two groups have different pulsed laser parameters, the influence of the scanning pattern is the same in both groups. The grain size values in the three regions show fewer differences in the D1 and F1 samples than in the D2 and F2 samples. This result can be ascribed to the fact that the cooling is more uniform throughout the samples fabricated with the unidirectional scanning pattern (D1 and F1 samples) compared with the samples fabricated with the bidirectional scanning pattern (D2 and F2 samples), and they have a higher cooling rate. The cooling duration is also longer in the D1 and F1 samples than in the D2 and F2 samples. When the cooling period is prolonged, the temperature decreases during the fabrication cycle, reducing the grain size in all layers. Consequently, the D1 and F1 samples have a smaller grain size than the D2 and F2 samples.

The graphs in Fig. 10 show that the grain size of the LMD-fabricated samples has been dropping from the beginning zone until the end zone. When the substrate is irradiated by the laser beam, it stores the heat and works as a heat source, which leads to the grain size of the samples from the beginning zone being much larger than those of the samples from the other zones [28]. The effect of the heat stored in the substrate decreases with the deposition of more layers as the thermal conductivity resulting from the substrate decreases, causing the grain size at the center area of the sample to decrease. The graphs show that the grain size at the top layers continues to decrease. This decrease is because the top layers are conducted with ambient air from the other side, which increases the cooling rate and solidification, causing the grain size on the top layers to decrease.

3.3 Microhardness of LMD-Fabricated Samples

Figure 11 shows a schematic of the microhardness test for the LMD-fabricated samples where the ten sampling points are carried out along the central axis of the deposited layers. Figure 12 shows the distribution curve of Vickers hardness with different pulse parameters and different scanning strategies along the central axis of the deposited layers. The graphs show that the microhardness trend is almost the same in all thin-walled samples fabricated using LMD. The figures show a higher microhardness value on the top of the deposition, where higher cooling rates are achieved than in the center and beginning area of the deposition. As discussed in Sect. 3.2, the remelting and recrystallization of the preceding layer during the addition of a new layer in LMD alter the grain size and the microstructure of the interlayer. Moreover, the microhardness varies as a result of these changes. Furthermore, the LMD, especially in pulsed-wave mode, has a fast cooling rate of fine grain structure and produces a fine grain-strengthening effect, which can significantly improve the hardness of the deposited layers [27].

Fig. 11
figure 11

Microhardness measured points in the LMD thin-walled samples

Full size image
Fig. 12
figure 12

Microhardness change curve of the cross section of LMD thin-walled samples from a group D and b group F

Full size image

3.3.1 Impact of Pulsed Laser Parameters on the Microhardness of LMD-Fabricated Samples

The microhardness curve and the average microhardness of the samples from group D are illustrated in Figs. 12a and 13a, respectively. They represent the significant impact of the changing duty cycle on microhardness during LMD. This variation in microhardness values is because the thermal histories of the LMD components manufactured using a pulsed laser are significantly different because of the varied energy distribution and cooling rate caused by the variation in pulsed laser parameters, which makes the grain size distinct and then affects the microhardness of LMD components. A comparison of the average microhardness of the deposition layers of different samples shows that the average microhardness gradually increases with decreasing duty cycle in both scanning strategies. In the unidirectional scanning pattern, the average microhardness values are 217.4, 229.9, and 237.1 at duty cycles of 80%, 75%, and 66.7%, respectively. In the bidirectional scanning pattern, the average microhardness values are 199, 208.7, and 216.9 at duty cycles of 80%, 75%, and 66.7%, respectively. This variation in microhardness is due to the fact that the heat input decreases with decreasing duty cycle, thereby enhancing the cooling rate, fine microstructure, and microhardness.

Fig. 13
figure 13

Effects of pulse laser parameters and scanning pattern on the microhardness of LMD-fabricated samples from a group D and b group F

Full size image

The effect of changing pulsed laser frequencies on the microhardness of LMD thin-walled samples is illustrated in Figs. 12b and 13b. The average microhardness gradually increases with increasing frequency during unidirectional and bidirectional scanning strategies, as shown in Fig. 13b. The pulse frequency mainly affects the cooling time. The smaller the frequency, the more sufficient the cooling of the deposition layers. The faster the cooling rate, the finer and smaller the microstructure, and the greater the microhardness.

During pulsed laser LMD, the microhardness increases as the duty cycle decreases. As shown in Sect. 3.1, when the duty cycle decreases, the inputted energy decreases, which causes the created molten pool to decrease. The smaller the molten pool, the faster it cools, which leads to smaller grains and higher microhardness [29]. This result is confirmed by the d11 and d21 samples having the highest microhardness among the samples. Although the samples from group F have the same average power (700 W), the single pulse energy is lower at 50 Hz than at 20 and 13.33 Hz, leading to a relatively higher microhardness because a faster cooling rate is maintained in the f13 and f23 samples.

3.3.2 Impact of the Scanning Pattern on the Microhardness of LMD-Fabricated Samples

The microhardness curves of the samples from groups D and F are shown in Fig. 12a, b, respectively. Despite being performed with different pulsed laser parameters (duty cycle and frequency), the impact of the scanning pattern is the same in each group. A comparison of the microhardness behavior of section D1 samples (d11, d12, and d13) that were manufactured with a unidirectional scanning strategy and section D2 samples (d21, d22, and d23) that were manufactured with a bidirectional scanning strategy shows that the samples with the same pulsed parameter and different scanning patterns [(d11 and d21), (d12 and d22), and (d13 and d23)] have similar microhardness variation trends with different values, as evidenced from the average microhardness graphs shown in Fig. 13a. The same effect of the scanning pattern is also observed in the samples from group F, as shown in Figs. 12b and 13b. These differences in microhardness values of the samples are due to the difference in microstructure grain size, as analyzed in Sect. 3.2. Compared with the D2 and F2 samples produced using a bidirectional scanning strategy, the D1 and F1 samples manufactured using a unidirectional scanning strategy not only have a lower temperature during manufacture because of a more uniform cooling rate and longer cooling time but also possess an ultimately smaller grain size [30]. The smaller the grain size in a region, the greater its strength and hardness [31].

4 Conclusions

In this study, a thin-walled AISI316L stainless steel was fabricated through LMD using a low-power pulsed laser and two scanning patterns (unidirectional and bidirectional). This study investigated the geometric properties, microstructure, and microhardness of the samples. The major conclusions can be summarized as follows:

  1. 1.

    In groups D and F, the unidirectional scanning pattern shows more stability than the bidirectional scanning pattern. Higher geometric stability, smaller grain size, and higher microhardness can be acquired in the unidirectional scanning pattern than in the bidirectional scanning pattern.

  2. 2.

    The influence of pulsed laser parameters (duty cycle and frequency) is the same in the unidirectional and bidirectional patterns.

  3. 3.

    The geometric stability increases with increasing duty cycle. However, the geometric stability does not vary considerably using different frequencies.

  4. 4.

    The microstructure grain size is influenced by the duty cycle of the pulsed laser. The grain size increases with increasing duty cycle. The microstructure shows a smaller grain size at 66.7% in d11 and d21, and the larger grain size appears at 80% in d13 and d23. The frequency also affects the microstructure and grain size of the samples. The grain size increases with decreasing frequency; large grains appear at 13.33 Hz in the f11 and f21 samples, whereas small grains appear at 50 Hz in the f13 and f23 samples.

  5. 5.

    The microhardness values of the deposited layers vary considerably with the pulsed laser parameters in each zone. The average value of microhardness increases with decreasing duty cycle of pulse laser but increases with increasing frequency. The maximum microhardness values of the samples from groups D and F are presented in d11 and f13, respectively.

  6. 6.

    The influence of the duty cycle on geometric properties is stronger than that of frequency. However, the effect of frequency on microhardness is stronger than that of the duty cycle.

These results suggest that the pulsed laser parameters and scanning patterns considerably influence the properties of LMD-fabricated components. These parameters and patterns must be regulated to improve the quality and efficiency of the product and the reliability of LMD.

Further studies should investigate the influence of the low-power pulsed laser parameters and scanning patterns on tensile stress, residual stress, and thermal distortion of the substrates during LMD.