Ultrafast Laser Asynchronous Processing for Laser Metal Deposition Manufacturing of Ti6Al4V

doi:10.21203/rs.3.rs-4692740/v1

Download PDF

Research Article

Ultrafast Laser Asynchronous Processing for Laser Metal Deposition Manufacturing of Ti6Al4V

https://doi.org/10.21203/rs.3.rs-4692740/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Additive manufacturing technology has rapidly developed to enable the manufacturing of various complex feature parts. However, parts manufactured through Laser Metal Deposition (LMD) often exhibit poor surface quality with the waviness of significant peak-to-valley distances, making it hard to improve surface quality through laser polishing. This study proposes a surface treatment method using ultrafast laser asynchronous processing to solve the issue of large waviness in Ti6Al4V parts manufactured by LMD, in which rough processing is carried out using picosecond laser with a galvanometer and precision processing is conducted using femtosecond laser with an objective lens. The influence of laser energy, defocus distance, and repetition times on material surface quality is investigated. Using the proper processing parameters, the initial surface with a step of 250 µm can be eliminated, which demonstrates the feasibility of ultrafast laser asynchronous processing in reducing peak-to-valley distance in materials.

Ultrafast laser

Galvanometer

Objective lens

Ti6Al4V

Compared to traditional manufacturing processes, additive manufacturing technology is widely used in industries such as aerospace, medical and automotive due to its characteristics of low manufacturing costs, short processing times and high material utilization rates [1]. Selective laser melting (SLM) powder bed technology and laser metal deposition (LMD) are two widely used forming methods in metal additive manufacturing technology. Due to their different forming mechanisms, SLM technology achieves a more uniform and smoother surface compared to LMD [2, 3]. In addition, SLM technology allows for better control of melt pool formation and solidification, resulting in a more stable and uniform melt pool, which helps reduce surface defects and waviness [4]. Therefore, the surface quality and waviness of parts manufactured by SLM technology are superior to those produced by LMD technology.

Many scholars conducted research on addressing the issue of poor surface quality in parts manufactured by SLM technology. Commonly used post-processing methods include milling, sandblasting, machining, magnetic/abrasive/ultrasonic polishing [5–8]. Considering the processing cost, efficiency and environmental pollution, laser polishing is proposed to improve the surface quality manufactured by SLM [9–12], in which the high thermal energy of the laser acts on the material, causing the material surface to melt and rapidly solidify to enhance surface quality. Some scholars used ultrafast lasers to treat the surfaces of SLM parts [13–15], ultimately improving the surface quality of the materials. However, the above surface treatment methods are not applicable to LMD parts because the surface peak-to-valley distance is much larger than that of SLM parts.

Currently, there is limited research on surface treatment for LMD parts. Some scholars used continuous wave lasers for surface treatment of LMD parts [16, 17], but the improvement in surface quality is not significant. Rose et al.[18] found that increasing the number of laser polishing passes can reduce the surface roughness of LMD parts. In addition, Jon et al.[19] used nanosecond laser for LMD parts with laser beam machining (LBM) + laser polishing treatment. Shen et al.[20] used picosecond laser machining + laser polishing treatment for LMD parts. However, the heat-affected zone during the laser remelting process is relatively large and may affect the physical properties of LMD parts.

To address the problem of poor surface quality in LMD parts, this study proposes a strategy of ultrafast laser asynchronous processing for the LMD manufactured Ti6Al4V parts. Preliminary processing is carried out using picosecond laser combined with a scanning galvanometer, in which the rapid scanning and high removal efficiency are utilized. Subsequently, femtosecond laser combined with an objective lens is used for precision processing, in which the femtosecond laser can further improve surface quality.

As shown in Fig. 1, the processing strategy for the LMD manufactured Ti6Al4V includes picosecond laser with scanning galvanometer and femtosecond laser with an objective lens. Figure 1(a) demonstrates an obvious peak-to-valley distance of the surface generated by LMD. There is a significant difference in light intensity between the peak and valley of the laser when processing the material, leading to different removal. The material removal at the peaks is more pronounced, helping to reduce the material peak-to-valley distance. Therefore, by using the long Rayleigh length of the galvanometer for the first step of processing, the large-scale unevenness on the material surface can be quickly treated, minimizing the peak-to-valley distance of the material as much as possible, providing a relatively smooth surface for subsequent processing as shown in Fig. 1 (b). Then, the fine processing is carried out using the short Rayleigh length of the objective lens to further reduce the peak-to-valley distance until the material surface is flat, as shown in Fig. 1 (c).

During the picosecond laser processing, due to the large Rayleigh length of the galvanometer, the difference in removing material peaks and valleys is not obvious. Therefore, to avoid excessive removal of material at the valleys, the energy of the laser reaching the valleys is close to the Ti6Al4V ablation threshold by adjusting the defocus distance, as shown in Fig. 2. When the ultrafast laser is focused at the peak, as depicted in Fig. 2 (a), the removal depth of the ultrafast laser is L_at. At this point, material removal occurs not only at the peaks but also at the valleys. When the defocusing distance is adjusted to L_d, as shown in Fig. 2 (b), the laser energy reaching the valley approaches the ablation threshold of Ti6Al4V. As a result, less material is removed at the valleys, and the material removal is primarily concentrated at the peaks, thereby achieving the goal of reducing the peak-to-valley distance.

The experimental setup used in this study is shown in Fig. 3. The picosecond laser is produced by the German company TRUMPF with the model TruMicro5050, equipped with a scanning galvanometer. It has a maximum pulse energy of 125 µJ, a wavelength of 1030 nm, a repetition frequency of 400 kHz, and a pulse width of 8 ps. The femtosecond laser is equipped with a 10x objective lens, with a pulse width of 290 fs, a wavelength of 1030 nm, a maximum pulse energy of 33 µJ, and a repetition frequency of 598 kHz.

According to the peak-to-valley distance of the LMD Ti6Al4V parts varying from 200 µm to 300 µm, the grooves with the depth of 250 µm machined by milling on the Ti6Al4V plate is selected to simulate the peaks and valleys of the LMD surface, as shown in Fig. 4.

During picosecond laser machining, the scanning area is 4 mm × 4 mm, and the scanning path is illustrated in Fig. 3(a). In order to control the laser energy density reaching the valley below the ablation threshold by defocusing distance, the following calculations are conducted.

$$\:\text{ω}\left(\text{z}\right)\text{=}{\text{ω}}_{\text{0}}\sqrt{\text{1+}{\left(\frac{\text{z}}{{\text{z}}_{\text{R}}}\right)}^{\text{2}}}$$

where $\:{\text{ω}}_{\text{0}}$ is the laser spot radius, z is the defocusing distance, and $\:{\text{z}}_{\text{R}}$ is the Rayleigh length, which can be expressed as:

$\:{\text{z}}_{\text{R}}\text{=}\frac{\text{π}{{\text{ω}}_{\text{0}}}^{\text{2}}}{\text{λ}}$

(2)

where $\:\text{λ}$ is the laser wavelength. The relationship between laser energy density F and laser energy E is as follows:

$\:\text{F=}\frac{\text{2E}}{\text{π}{\text{ω}}_{\text{0}}^{\text{2}}}$

(3)

Based on the preliminary experiments, the ablation thresholds for single-pulse ablation of Ti6Al4V using picosecond laser is determined to be 0.61 J/cm². When the laser energy is 33 µJ, the defocus distance z is solved from equations (1) and (3) to be 1994 µm. When the defocus distance z is set to 2000 µm, the laser energy density at the material valley is 0.58 J/cm², which is below the ablation threshold of Ti6Al4V. The used parameters in this process are shown in Table 1.

Table 1

Experimental parameters of picosecond laser process
Pulse energy (µJ)	Defocusing distance (µm)	Repetition times (/)	Scanning velocity (mm/s)	Pulse frequency (kHz)	Hatch distance (µm)
30 ~ 34	2000	40	100	400	10
33	1850 ~ 2150	40	100	400	10
33	2000	30 ~ 45	100	400	10

The area with 1 mm × 1 mm is processed by femtosecond laser is based on the picosecond laser processing area. The scanning path is the same as that of the picosecond laser processing, as shown in Fig. 3(b). The used parameters in this processing are shown in Table 2

Table 2

Experimental parameters of femtosecond laser process
Pulse energy (µJ)	Defocusing distance (µm)	Repetition times (/)	Scanning velocity (mm/s)	Pulse frequency (kHz)	Hatch distance (µm)
20	0 ~ 100	3 ~ 15	1	4	10

The three-dimensional optical profiler with a spatial resolution of 0.1 µm and a height resolution of 0.05 µm is used to measure the three-dimensional surface profile of the parts before and after ultrafast laser machining.

4.1 Picosecond laser processing

Figure 5 shows the removal results of peaks and valleys of the material by picosecond laser at different energies. With the laser energy increases, the peak-to-valley distance of the material shows a trend of initially decreasing and then increasing, as shown in Fig. 5 (f). The reason can be explained as follows.

Laser material removal is performed in multi-pulse mode, and the number of pulses N in the processing area can be calculated by the following equation [21]:

$$\:\text{N=}\frac{\text{2}{\text{ω}}_{\text{0}}\text{f}}{\text{v}}$$

Where f is the laser frequency, and v is the laser scanning speed. Substituting the data, the number of pulses in a beam size is determined to be 80. The relationship between the multi-pulse ablation threshold and the single-pulse ablation threshold is described as [21]:

$$\:{\varnothing\:}_{N}={\varnothing\:}_{1}{N}^{S-1}$$

where $\:{\varnothing\:}_{1}$ is the single-pulse ablation threshold with a value of 0.61 J/cm², and S is the incubation factor. Taking S as 0.9 [22], the multi-pulse ablation threshold is determined to be 0.48 J/cm².The calculation results for the laser energy density reaching the material valley at a defocus distance of 2000 µm with laser energies ranging from 30 to 34 µJ is shown in Table 3.

During the laser processing, the depth of material removal at both the material peak and valley exhibits a trend of initially increasing and then leveling off, as illustrated in Fig. 6. This is primarily due to the fact that when the laser energy is constant, initially the laser energy density exceeds the material ablation threshold, resulting in an increase in material removal depth. As the removal depth increases, the energy density reaching the material surface decreases, leading to a weakening of material removal capability. Consequently, the increase in removal depth slows down until it stabilizes when it falls below the ablation threshold.

Table 3

The laser energy density at the material valley for different energies
laser energy(µJ)	30	31	33	34
laser energy density(J/cm²)	0.44	0.45	0.48	0.5

From Table 3, it can be observed that when the laser energy ranges from 30 to 33 µJ, the energy density reaching the material valley is less than or equal to the multi-pulse ablation threshold of Ti6Al4V. Consequently, the depth of material removal at the valley remains consistent. As the laser energy increases, the depth of material removal at the peak increases, resulting in a gradual reduction in the peak-to-valley distance. When the laser energy reaches 34 µJ, laser energy density at the material valley is 0.5 J/cm², which exceeds the multi-pulse ablation threshold for Ti6Al4V, the depth of material removal increases at the material valley. At this point, the increment in removal depth at the material valley far exceeds that at the material peak. Therefore, the peak-to-valley distance gradually increases, as shown in Fig. 6.

The effects of the defocusing distance on the peak-to-valley distance is shown in Fig. 7. With the increase of defocusing distance, the peak-to-valley distance shows a trend of initially decreasing and then increasing. The calculation results for the laser energy density reaching the material valley at a defocus distance ranging from 1850 to 2300 µm with a laser energy of 33 µJ are presented in Table 4.

Table 4

The laser energy density at the material valley for different defocusing distances
defocusing distance (µm)	1850	2000	2150	2300
laser energy density(J/cm²)	0.55	0.48	0.42	0.38

From Table 4, it is evident that when the defocusing distance ranges from 2000 to 2300 µm, the laser energy density at the material valley is less than or equal to the multi-pulse ablation threshold of Ti6Al4V. Consequently, there is minimal variation in the depth of material removal at the valley. As the defocusing distance gradually increases, resulting in a decrease in laser energy at the material peak, the laser ability to remove material weakens, leading to a gradual reduction in ablation depth. Thus, the peak-to-valley distance gradually increases. However, when the defocusing distance is 1850 µm, the laser energy density reaching the material valley is 0.55 J/cm², exceeding the multi-pulse ablation threshold for Ti6Al4V. As a result, the depth of material removal at the material valley increases. As indicated in Fig. 6, at this point, the increment in material removal depth at the material valley surpasses that at the material peak. Therefore, when the defocusing distance is 1850 µm, the peak-to-valley distance of the material is greater than that of the defocusing distance of 2000 µm.

The repetition times has a significant influence on reducing the peak-to-valley distance. Figure 8 shows the surface peak-to-valley distance under different repetition times. As the repetition times increases, the peak-to-valley distance of the material exhibits a trend of initially decreasing and then slowly increasing. The main reason for this phenomenon is that at a laser energy of 33 µJ and a defocus distance of 2000 µm, the energy density of the laser reaching the material valley equals the multi-pulse ablation threshold of the material. Therefore, initially, as the repetition times increases, there is little change in the depth of material removal at the valley, while the depth of material removal at the peak gradually increases, leading to a gradual decrease in the peak-to-valley distance. Subsequently, with further increases in the repetition times, the peak-to-valley distance of the material decreases gradually. The peak and valley gradually converge within the same plane, the depth of the ablation pit increases, and the material splashing during laser processing cannot be effectively removed, leading to reattachment on the material surface. This causes the peak-to-valley distance of the material to gradually increase.

4.2 Femtosecond laser processing

After picosecond laser processing with the galvanometer scanning system, the peak-to-valley distance on the surface of machined Ti6Al4V material decreased from the initial 250 µm to about 35 µm, as shown in Fig. 6 (f), Fig. 7 (f), and Fig. 8 (f). The difference in material removal at the peaks and valleys by picosecond laser with galvanometer assistance is not significant, and it cannot eliminate the peak-to-valley distance of the material. During the femtosecond laser processing, in order to provide a basis for selecting the defocusing amount, pre-experiments are conducted on the surface of Ti6Al4V material with single scans at different defocusing distances. The laser ablation morphology and depth curves under different defocusing distances are shown in Fig. 9.

As the defocusing distance increases, the spot diameter of the laser reaching the material surface enlarges, leading to a decrease in laser energy density, and thus the single ablation depth gradually decreases. As shown in Fig. 9 (f), when the defocusing distance increases to 60–100 µm, the ablation phenomenon within the single laser scanning area gradually weakens, until there is no obvious ablation on the material surface. After picosecond laser processing, there is still a peak-to-valley distance of about 35 µm on the Ti6Al4V material surface. When the defocusing distance at the valley is 60–100 µm, there is no obvious ablation at the valley, the defocusing distance at the peak is equivalent to 20–60 µm, and ablation still occurs with an ablation depth, as shown in Fig. 9 (b) and (c). At the same time, in order to reduce the repetition times during the experimental process, a defocusing distance of 25 µm is selected for this experiment, which not only effectively reduces the peak-to-valley distance but also enhances processing efficiency.

By selecting a defocus distance of 25 µm for multiple scans of femtosecond laser experiments, the surface morphology of Ti6Al4V material under different repetition times is shown in Fig. 10. After 9 times of femtosecond laser ablation, the peak-valley structure on the material surface is gradually removed, and the peak-to-valley distance is significantly reduced, as shown in Fig. 10 (c). With further increase in the repetition times, the surface becomes flat, as shown in Fig. 10 (d)-(e). Therefore, choosing 15 times of femtosecond laser can completely remove the 40 µm peak-to-valley distance formed on the material surface after picosecond laser processing, resulting in a Ti6Al4V surface without obvious steps, as shown in Fig. 10 (e).

4.3 Two-step laser processing

Figure 11 shows the surface morphology and 2D profile of Ti6Al4V material processed by milling, picosecond laser processing, and picosecond + femtosecond laser processing. The step depth of Ti6Al4V processed by milling is 250 µm, used to simulate the peak-to-valley distance of LMD material, as shown in Fig. 11 (a). Firstly, rough processing is performed using picosecond laser with galvanometer, and the processing result is shown in Fig. 11(b). From Fig. 11 (d), the peak-to-valley distance on the material surface decreased from 250 µm to 35 µm. Subsequently, fine processing is conducted using femtosecond laser with objective lens to remove the remaining 35 µm peak-to-valley distance, as shown in Fig. 11 (c). The processing range of the femtosecond laser is 1mm × 1mm. From Fig. 11 (d), it can be seen that after picosecond + femtosecond laser processing, the material surface is almost flat, without significant steps.

This study proposes a strategy of ultrafast laser asynchronous processing for LMD manufactured of Ti6Al4V, aiming to address the issue of poor surface quality in LMD parts. Firstly, rough processing is carried out using picosecond laser with a galvanometer. Subsequently, precision processing is conducted using femtosecond laser with an objective lens. The influence of laser energy, defocusing distance, and repetition times on material surface quality is also investigated. It is found that during picosecond laser rough processing, adjusting the laser energy to 33 µJ, the defocusing distance to 2000 µm, and repetition times to 40 can reduce the peak-to-valley distance on the material surface to 35 µm. In femtosecond laser precision processing, using a laser energy of 20 µJ, a defocusing distance of 25 µm, and 15 repetition times can remove the remaining35 µm peak-to-valley distance, resulting in a smooth surface without significant steps.

Conflict of Interest:

The authors declare that they have no conflict of interest.

Tofail SAM, Koumoulos EP, Bandyopadhyay A, Bose S, O’Donoghue L, Charitidis C (2018) Additive manufacturing: scientific and technological challenges, market uptake and opportunities. Mater Today 21(1):22–37
Leary M, Maconachie T, Sarker A, Faruque O, Brandt M (2019) Mechanical and thermal characterisation of AlSi10Mg SLM block support structures. Mater Design 183:108138
Svetlizky D, Das M, Zheng B, Vyatskikh AL, Bose S, Bandyopadhyay A et al (2021) Directed energy deposition (DED) additive manufacturing: Physical characteristics, defects, challenges and applications. Mater Today 49:271–295
Zhang Y, Huang W (2020) Comparisons of 304 austenitic stainless steel manufactured by laser metal deposition and selective laser melting. J Manuf Process 57:324–333
Edwards P, Ramulu M (2014) Fatigue performance evaluation of selective laser melted Ti–6Al–4V. Mater Sci Engineering: A 598:327–337
Rakurty CS, Bowser KM, Rangasamy B, Kakaraparthi N, Dippolito S (2023) Grinding EB-PBF based additive manufactured Ti6Al4V: A surface integrity study. Adv Industrial Manuf Eng 7:100131
Sahlot P, Suthar M, Wankhede VA (2023) Investigation of surface properties for additively manufactured SS316L parts: Effect of post-processing techniques. Materials Today: Proceedings. ;80:395-9
Ye Y, Zhang C, Gao L, Peng L, Liu G, Zhang Y et al (2023) Effect of electropulsing-assisted ultrasonic nanocrystal surface modification on microstructures and hardness of additive manufactured Inconel 718. Eng Fail Anal 153:107611
Kumar A, Saha S, Kumar CS, Nath AK (2020) Laser surface re-melting of additive manufactured samples with a line focused beam. Materials Today: Proceedings. ;26:1221-5
Lee S, Ahmadi Z, Pegues JW, Mahjouri-Samani M, Shamsaei N (2021) Laser polishing for improving fatigue performance of additive manufactured Ti-6Al-4V parts. Opt Laser Technol 134:106639
Li C, Liu D, Liu G, Liu S, Jin X, Bai Y (2023) Surface characteristics enhancement and morphology evolution of selective-laser-melting (SLM) fabricated stainless steel 316L by laser polishing. Opt Laser Technol 162:109246
Zhihao F, Libin L, Longfei C, Yingchun G (2018) Laser Polishing of Additive Manufactured Superalloy. Procedia CIRP 71:150–154
Canacoo S, Contreras Lopez E, Coronel O, Ahmed F, Li J, Srivastava A (2022) Ultrafast Laser Ablation of Inconel 718 for Surface Improvement. Manuf Lett 33:410–414
Hafiz AMK, Bordatchev EV, Tutunea-Fatan RO (2013) Experimental analysis of applicability of a picosecond laser for micro-polishing of micromilled Inconel 718 superalloy. Int J Adv Manuf Technol 70(9–12):1963–1978
Li N, Fan P, Zhu Q, Cui B, Silvain J-F, Lu YF (2023) Femtosecond laser polishing of additively manufactured parts at grazing incidence. Appl Surf Sci 612:155833
Jardon Z, Ertveldt J, Lecluyse R, Hinderdael M, Pyl L (2022) Directed Energy Deposition roughness mitigation through laser remelting. Procedia CIRP 111:180–184
Kim MJ, Saldana C (2023) Post-processing of additively manufactured IN625 thin-walled structures using laser remelting in directed energy deposition. J Manuf Process 88:59–70
Rosa B, Mognol P, Hascoët J-y (2015) Laser polishing of additive laser manufacturing surfaces. J Laser Appl. ;27(S2)
Arrizubieta J, Cortina M, Ruiz J, Lamikiz A (2018) Combination of Laser Material Deposition and Laser Surface Processes for the Holistic Manufacture of Inconel 718 Components. Materials 11(7):1247
Shen H, Liao CH, Zhou J, Zhao K (2021) Two-step laser based surface treatments of laser metal deposition manufactured Ti6Al4V components. J Manuf Process 64:239–252
McDaniel C, Flanagan A, O’ Connor GM (2014) Evidence for increased incubation parameter in multi-pulse ablation of a Pt:SS alloy using a femtosecond laser at high repetition rates. Appl Surf Sci 295:1–7
Orlandini A, Baraldo S, Porta M, Valente A (2022) Ablation threshold estimation for femtosecond pulsed laser machining of AISI 316L. Procedia CIRP 107:617–622

Download PDF

Reviewers agreed at journal
09 Jul, 2024
Reviewers invited by journal
09 Jul, 2024
Editor assigned by journal
08 Jul, 2024
First submitted to journal
05 Jul, 2024

You are reading this latest preprint version

Ultrafast Laser Asynchronous Processing for Laser Metal Deposition Manufacturing of Ti6Al4V

Status:

Version 1

Abstract

Figures

1. Introduction

2. Processing strategy

3. Experimental set-up

4. Results and discussion

4.1 Picosecond laser processing

4.2 Femtosecond laser processing

4.3 Two-step laser processing

5. Conclusions

Declarations

Conflict of Interest:

References

Status:

Version 1