Open Access
Add to BibSonomy
Abstract
The microstructures on a diamond surface have attracted extensive attention in microelectronics, ultra-precision machining tools, and optical elements, etc. In this work, microgrooves were fabricated on a single-crystal diamond surface using ultraviolet nanosecond or infrared picosecond laser pulses. The surface and internal morphologies of the microgrooves were characterized. The chemical composition and phase transition of the diamond after laser irradiation were analyzed. Furthermore, the ablation threshold, ablation rate, and material removal rate of the diamond processed by nanosecond or picosecond lasers were also calculated. In addition, the temperature distributions of the diamond ablated by nanosecond or picosecond lasers were simulated. Finally, the material removal mechanisms of a single-crystal diamond processed by nanosecond or picosecond lasers were revealed. This work is expected helpful to provide a guidance for the laser fabrication of microstructures on diamond.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Diamond is widely used for manufacturing cutting tools in automotive and machining industries [1], heat spreaders in electronic packaging [2], fabrication of diffractive optical elements for high power lasers [3], and semiconductors [4] because of its high mechanical hardness, low friction coefficient, high thermal conductivity, extreme chemical inertness, and large optical bandgap. Recently, there is an increasing interest in the micromachining of diamonds for microelectronics, ultra-precision machining tools, and micro-electro-mechanical systems applications. For example, microstructures fabricated on diamond tool surface are enabled to significantly improve cutting performance by reducing the cutting force, average friction coefficient, and cutting tool wear [5,6]. The use of movable diamond micro-gripper and diamond probe can improve the performance of atomic force microscope by offering lower friction and higher resistance to wear [7]. The microstructures generated on the diamond surface are able to reduce the reflectivity for application in high power laser [8]. However, diamond is difficult to machine by conventional mechanical and/or chemical methods because of its extremely high hardness (Mohr’s hardness of 10) and resistance to most chemicals [9]. Electron beam and ion beam have been used to mill diamond, but both processes are constrained by time-consuming and expensive equipment [10]. Laser beam machining has emerged as an advanced approach to fabricate microstructures on diamonds due to its flexibility, high efficiency, and high precision.
Micromachining of diamond by nanosecond lasers has been widely investigated [11–15]. It can be concluded from these literatures that the morphologies and sizes of the microstructures fabricated on diamonds are closely related to the laser parameters. Diamond has a wide energy bandgap of 5.47 eV and is transparent in a wide wavelength range. For the lasers operating at the center wavelength of 1064 nm (photon energy of 1.17 eV), the photon energy was inadequate to overcome the bandgap of diamond. Therefore, using a short-wavelength laser, such as an ultraviolet laser, can improve energy utilization efficiency [16]. Previous studies reported that the microstructures on diamond surfaces fabricated by ultrashort pulse laser have high surface quality and thin graphite layer (<100 nm) [17–20]. However, femtosecond laser is mainly used in laboratory research due to the complex and expensive equipment. While picosecond lasers can be used for diamond processing in industry because they are more reliable and cost less than femtosecond lasers [21]. Consequently, two types of lasers are expected to be used for the precision processing of diamonds in industrial fields. One is picosecond laser with a short pulse duration, and the other is nanosecond laser with a short wavelength. This study mainly focuses on the microgroove fabrication of single-crystal diamond using ultraviolet nanosecond or infrared picosecond laser pulses. The re-deposited particles around the groove, the nanostructures or nanoripples within the groove, and the cracks and chippings at the groove edges were evaluated and compared. The chemical composition and phase transition of diamond after laser irradiation were analyzed. The ablation threshold, ablation rate, and material removal rate of the diamond after ultraviolet nanosecond or infrared picosecond laser irradiations were calculated. The temperature distributions of diamond processed by single-pulse nanosecond or picosecond lasers were obtained by ANSYS software. The material removal mechanisms of ultraviolet nanosecond or infrared picosecond lasers processing of diamond were investigated.
2. Materials and experimental methods
The single-crystal diamond samples synthesized under high pressure high temperature (HPHT) conditions were provided by Henan Huanghe Whirlwind Co., Ltd (Xuchang, China). The diamond samples were cut to square plates with dimensions of 3.0 × 3.0 × 1.0 mm3. Both surfaces of the diamond plate with orientation <100> were polished to average surface roughness parameter Sa < 5 nm.
Ultraviolet nanosecond and infrared picosecond lasers were used for the microgroove fabrication of single-crystal diamond. The laser parameters of nanosecond and picosecond laser pulses are listed in Table 1. Figure 1 illustrates the schematic of the experimental setup. The laser beam first was expanded by a beam expander, and then guided by mirrors into the galvanometer scanner and finally focused normally onto the diamond surface through an F-θ objective lens. The laser beam moves along X- and Y-directions over the sample surface, which was controlled by the galvanometer scanner. In the laser processing experiments, the scan speed v was fixed at 5 mm/s, and the scan number remained at 1. The incident laser fluence F is calculated by F = 2P/(πω02f), where P is the average output power, f is the repetition rate, and ω0 is the beam radius on the sample surface. The equivalent pulse number N could be calculated by N = 2ω0f/v.
Table 1. Laser parameters of nanosecond and picosecond laser pulses.
The surface morphology of the re-deposited particles around the groove, the internal morphology of the groove, the groove width and depth, and the cross-sectional shape of the groove were evaluated by a scanning electron microscope (SEM). The three-dimensional morphology and ablation volume of the groove were measured by confocal laser scanning microscopy (CLSM). The material composition of the laser-treated diamond was analyzed via energy dispersive X-ray spectroscopy (EDX). The phase transition of diamond after laser irradiation was characterized by a WiTec alpha300R confocal Raman microscope. The excitation source is 532 nm laser with a laser power below 10 mW on the sample to avoid laser-induced heating. The temperature fields of diamond ablated by single-pulse nanosecond or picosecond lasers were simulated using the methods introduced in the literature [22–24]. By the way, the laser processing experiments were carried out in the air and conducted three times to verify the reproducibility.
3. Results and discussion
3.1 Surface morphologies of microgrooves
Figure 2 shows the surface morphology of the microgroove on diamond ablated by nanosecond laser with 16.98 J/cm2 fluence. The microgroove without cracks and chippings was obtained by nanosecond laser processing. A large number of particles were generated during nanosecond laser processing of diamond, and then re-deposited on the diamond surface. According to the size and shape of the re-deposited particles around the groove, the outer surface of the microgroove has been divided into five regions (Zone 1-5), as shown in Fig. 2(a). Figure 2(b)-(f) is an enlarged view of the labelled regions (circled b-f) in Zone 1-5. The histogram analysis of the size of particles in Zone 1-5 is shown in Fig. 2(g) [25]. Dense particles with size mainly ranging from 150 to 250 nm were formed in Zone 1 (Fig. 2(b)). As seen in Fig. 2(c), large particles with size mainly ranging from 200 to 400 nm were loosely distributed in Zone 2. As shown in Fig. 2(d), the particles were densely distributed in Zone 3 and the particle size was decreased to 100-180 nm. As presented in Fig. 2(e), many small particles with size of 60-120 nm are attached together to form micro-clusters in Zone 4. In Zone 5, there are some re-deposited particles with size of about 40-60 nm (Fig. 2(f)). As depicted in Fig. 2(c)-(f), the particle size decreases with the increase of distance from the microgroove center. This phenomenon is due to the fact that smaller particles are ejected at higher velocities and thus travel farther.
Fig. 2. (a) SEM images of the particles around the groove produced by nanosecond laser with 16.98 J/cm2 fluence. (b-f) The enlarged views of the labelled regions (circled b-f) in (a). (g) Histograms analysis of the particle size in (b-f).
In order to clarify the material composition of the laser-irradiated diamond, the EDX spectra of the inner and outer surface of the diamond groove were measured [26]. The EDX spectra in Fig. 3 show that the laser-treated diamond consists of carbon (C), oxygen (O) and gold (Au) elements. Here, the gold element appears because the diamond sample was coated with gold film to create the conductive surface layer necessary for SEM analysis. The weight percentage of oxygen element significantly decreases with the increase of distance from the microgroove center. This may be due to the decrease of laser thermal effect as the distance from the groove center increases. Lee’s [27] study shows that a certain percentage of carbon-oxygen (C-O), carbonyl (C = O) and carboxyl (COOH) bonds were formed in the laser-treated diamond. Therefore, the oxygen element was detected in the EDX spectra.
Fig. 3. (b) Energy dispersive X-ray spectroscopy of the laser-treated diamond in (a), which is ablated by nanosecond laser with fluence of 16.98 J/cm2.
Furthermore, the Raman spectra around the groove (Fig. 4(a)) were measured and the results are presented in Fig. 4(b). The Raman spectrum of point “P1” shows the D-band at about 1350 cm-1 and G-band at about 1580 cm-1. The D-band represents the in-plane A1g (LA) zone-edge mode, while the G-band corresponds to the E2g in-plane vibrational modes [13]. The appearance of D-band and G-band at “P1” position indicates the dense particles in Zone 1 are amorphous carbon and graphite. The increasing of diamond-band intensity from point “P2” to point “P4” implies that the thickness of the re-deposited layer decreases as the distance from the microgroove center increases. The Raman spectrum at point “P5” (black line) shows only a clear and intense diamond-band at 1332 cm-1, which indicates that no particles is deposited on the diamond surface.
Fig. 4. (b) Raman spectra of re-deposited particles around the microgroove in (a), which was ablated by nanosecond laser with fluence of 16.98 J/cm2.
As the laser fluence increases to 45.27 J/cm2, large amounts of re-deposited spherical particles are observed around the microgroove (Fig. 5(a)). As shown in Fig. 5(b), the re-solidified particles can be divided into four zones (Zone 1-4). The particles with diameters from 200 to 484 nm were densely distributed in Zone 1 (Fig. 5(c)). It is worth noting that the size of particles in Zone 1 is close to that of particles at the sidewall of the groove (Fig. 11(f)), and the laser-induced plasma prevents particles from depositing on Zone 1 [15]. Therefore, it can be deduced that the particles in Zone 1 come from the volume expansion of molten graphite material in the groove because the mass density of graphite is lower than that of diamond [28]. The particles in Zone 2 to Zone 4 are formed by re-deposited particles ejected from the microgroove. As shown in Fig. 5(d), the Zone 2 is a transient region with particle diameters in the range of 552-732 nm. As presented in Fig. 5(e) and (f), the particles in Zone 3 and Zone 4 like flower clusters. The diameters of the particles in Zone 3 are domain from 470-1023 nm, while the particle size in Zone 4 is relatively uniform with diameters of about 300 nm (Fig. 5(f)).
Fig. 5. (a) SEM images of the particles around the microgroove produced by nanosecond laser with fluence of 45.27 J/cm2. (b) Enlarged view of the red square box in (a). (c-f) are the enlarged views of circled c, square box, circled e, and f in (b), respectively.
In order to find out whether there is a phase transition on the original diamond surface around the groove after laser irradiation, the Raman spectra of the particles around the microgroove produced by nanosecond laser with fluence of 47.91 J/cm2 were measured, as illustrated in Fig. 6. The Raman spectra of the re-deposited particles show D-band and G-band. This indicates that the re-deposited particles are amorphous carbon and graphite. However, the Raman spectrum taken at point “P4” where the re-deposited layer was removed shows only a diamond-band. This indicates that the surface material of original diamond is not graphitized.
Fig. 6. (b) Raman spectra measured at different points around the microgroove in (a), which was produced by 47.91 J/cm2 nanosecond laser fluence.
Figure 7(a) and (d) illustrates the morphologies of the ablation microgrooves created by picosecond laser with fluence of 1.41 and 2.83 J/cm2, respectively. As presented in Fig. 7(a) and (d), cracks and chippings appear at the edge of the microgrooves produced by picosecond laser. Figure 7(b) and (c) is an enlarged view of the circled b in Fig. 7(a) and circled c in Fig. 7(b), respectively. A small amount of ablation debris with size of 325 nm was generated after 1.41 J/cm2 picosecond laser irradiation. Figure 7(e) and (f) is an enlarged view of the circled e in Fig. 7(d) and circled f in Fig. 7(e), respectively. The microgroove without debris was obtained by picosecond laser processing at 2.83 J/cm2 laser fluence.
Fig. 7. Surface morphologies of the microgrooves produced by picosecond laser with (a) 1.41 and (d) 2.83 J/cm2 laser fluence. The red rectangles in (a) and (d) represent the SEM enlarged views of the diamond groove edge. (b) and (e) are the enlarged views of the circled b in (a) and circled e in (d), respectively. (c) and (f) are the enlarged views of the circled c in (b) and circled f in (e), respectively.
The chemical composition of diamond ablated by picosecond laser was analyzed by EDX and Raman spectra. The EDX spectra of picosecond laser-treated diamond is consistent with that of nanosecond laser-treated diamond. The Raman spectra of picosecond laser-treated diamond is shown in Fig. 8. The Raman spectrum at the center of the microgroove (red line in Fig. 8(b)) shows two features: D-band and G-band. This suggests that the thermally induced graphitization process also occurred at the center of the microgroove after picosecond laser irradiation. The Raman spectrum at the sidewall of the microgroove (“P2”) shows diamond-band and G-band, which indicates that the thickness of graphite layer at the sidewall is thinner than that at the center of the microgroove. The Raman spectrum at “P3” and “P4” shows only a diamond-band. This indicates that there is no re-deposited material around the microgroove.
Fig. 8. (b) Raman spectra measured at different points of the microgroove in (a), which was produced by 1.41 J/cm2 picosecond laser fluence.
The Raman spectra measured at the center of the microgrooves ablated by nanosecond and picosecond lasers with different laser fluences are presented in Fig. 9(a) and (b), respectively. The black curves in Fig. 9(a) and (b) represent the Raman spectra of the untreated diamond. As shown in Fig. 9(a) and (b), the G-band position shifts to low wave numbers as laser fluence increases, leading to a red shift. The red shift of the G-band is due to the fact that C-C bond length increases with increasing the laser fluence [29]. The G-band position is red-shifted by 9 cm-1 (from 1588 to 1579 cm-1) for nanosecond laser irradiation, while it is red-shifted by 15 cm-1 (from 1591 to 1576 cm-1) for picosecond laser processing. This indicates that the red-shift of G-band for picosecond laser is larger than that for nanosecond laser. It can be deduced that the thermal stress in the groove ablated by picosecond laser is larger than that ablated by nanosecond laser.
Fig. 9. Raman spectra at the center of the ablation groove produced by (a) nanosecond or (b) picosecond lasers with different laser fluences. The dashed lines are added as guides to the eye to show the peak position shift.
Figure 10 presents the morphologies of the end of the microgrooves produced by nanosecond (a) and picosecond lasers (b). As presented in Fig. 10(a), the microgroove without cracks and chippings was fabricated by nanosecond laser. However, some obvious cracks and chippings were observed at the end of the microgroove produced by picosecond laser. The formation of the cracks and chippings is described as follows. The peak power density I0 is calculated by I0 = Ep/(τπω02) = F/(2τ), where Ep is the pulse energy. Since the pulse duration of picosecond laser is 10 ps, while that of nanosecond laser is 12 ns. Take the maximum laser fluence as an example, the peak power density of nanosecond laser is I0 = 47.91/(2 × 12 × 10−9) = 2.0 × 109 W/cm2, while the peak power density of picosecond laser is I0 = 3.18/(2 × 10 × 10−12) = 1.6 × 1011 W/cm2. Hence, the peak power density of picosecond laser is about two orders magnitude greater than that of nanosecond laser [30]. Therefore, the temperature of diamond material irradiated by picosecond laser is much higher than that irradiated by nanosecond laser. Because the time of the thermal diffusion is about of the order 10−12 s [31], which is much shorter than the pulse duration of nanosecond laser. Therefore, the heat in the irradiated area of nanosecond laser can diffuse to the unirradiated area, while that of picosecond laser can hardly spread to the unirradiated area. The aforementioned two reasons cause the thermal stress in diamond during nanosecond laser processing is smaller than that of picosecond laser. This can explain why cracks and chippings were produced at the edge of the groove for picosecond laser processing of diamond, but no cracks and no chippings were generated at the edge of the groove for nanosecond laser processing of diamond. Hence, nanosecond laser is a more suitable choice to machine crack-free microgrooves on the diamond.
Fig. 10. SEM images of the end of the microgrooves ablated by (a) nanosecond laser with 16.98 J/cm2 fluence and (b) picosecond laser with 1.41 J/cm2 fluence.
3.2 Internal morphologies of microgrooves
The internal morphology of the microgroove ablated by nanosecond laser with fluence 16.98 J/cm2 was shown in Fig. 11(a). The enlarged views of the circled b and c in Fig. 11(a) are presented in Fig. 11(b) and (c), respectively. As seen in Fig. 11(b) and (c), the dense nanobump-like structures at the sidewall of the groove were generated by re-solidified molten material in the groove. The internal morphology of the microgroove created by nanosecond laser with fluence of 45.27 J/cm2 is displayed in Fig. 11(d). The enlarged views of circled e and f in Fig. 11(d) are show in Fig. 11(e) and (f), respectively. The signs of melting were observed at the sidewall of the microgroove (Fig. 11(e)). As illustrated in Fig. 11(f), particles with diameters about 480 nm were generated at the bottom of the microgroove.
Fig. 11. Internal morphologies of the microgrooves produced by nanosecond laser with (a) 16.98 J/cm2 and (d) 45.27 J/cm2 laser fluence. (b), (c), (e) and (f) are the enlarged views of the circled b, c in (a) and e, f in (d), respectively.
Figure 12 illustrates the laser-induced periodic surface structures (LIPSSs, also termed as ripples) at the sidewall and bottom of the microgroove produced by picosecond laser with 1.41 J/cm2 fluence. The formation mechanism of LIPSSs has been discussed extensively in the literature [10,15,32]. It is currently accepted that the LIPSSs originate from the interference between the incident laser light and the surface scattered wave [33]. The period of the LIPSSs is given by λ/n(1±sin θ), where n is refractive index of diamond, θ is the laser beam incidence angle [34]. The orientation and spatial period of LIPSSs depend on the angle between the incident laser light and the surface scattered waves. As seen in Fig. 12(a) and (b), the LIPSSs are formed on the lower part of the microgroove sidewall. The average spatial period of the LIPSSs at the groove sidewall is about 151 nm, which is slightly less than λ/2n (220 nm). Figure 12(c) shows that two distinct types of LIPSSs were formed at the bottom of the microgroove. The enlarged view of the ripples at the groove bottom (Fig. 12(d)) illustrates that, the low spatial frequency LIPSSs (LSFL) with a period Λ = 1144 nm were split into many high spatial frequency LIPSSs (HSFL) with a spatial period of about 274 nm.
Fig. 12. SEM images of the (a) sidewall and (c) bottom of the microgroove produced by picosecond laser with 1.41 J/cm2 fluence. (b) and (d) are the enlarged view of the circled b in (a) and the circled d in (c), respectively.
The LIPSSs on the sidewall and bottom of the microgroove ablated by 2.83 J/cm2 picosecond laser were illustrated in Fig. 13. The enlarged views of the circled b-f in Fig. 13(a) are show in Fig. 13(b)-(f), respectively. As seen in Fig. 13(b), the ripples at the entrance of the microgroove are broken because the large temperature gradient inside and outside the irradiation area causes diamond material at the edge of the groove to chip easily. Two distinct types of LIPSSs are obtained at the microgroove sidewall (Fig. 13(c)-(e)). As presented in Fig. 13(c)-(e), the average spatial period of LSFL is about 495 nm, which is close to λ/n (440 nm). The average spatial period of HSFL is about 255 nm, whose spatial period is close to λ/2n (220 nm). As presented in Fig. 13(f), the LIPSSs with spatial period of 209 nm were generated at the bottom of the microgroove. Compared with picosecond laser, the LIPSSs were not generated in the microgroove processed by nanosecond laser. This may be due to the fact that molten graphite hinders the formation of LIPSSs in the microgroove.
Fig. 13. (a) SEM images of the microgroove produced by picosecond laser with 2.83 J/cm2 fluence. (b-f) The enlarged views of the circled b-f in (a).
3.3 Ablation threshold, ablation rate, material removal rate
3.3.1 Ablation threshold
The ablation threshold is defined as the minimum fluence required to remove material, and it is related to the equivalent pulse number and the ablation phase. The equivalent pulse number is N = 2ω0f/v = 2 × 15.0 × 30/5 = 180 for nanosecond laser, while the equivalent pulse number of picosecond laser is N = 2ω0f/v = 2 × 60.0 × 200/5 = 4800. The ablation phases of nanosecond and picosecond lasers can be judged from the cross-sectional shapes of the microgrooves. As shown in Fig. 14(b) and (c), the cross-sectional shape of the microgroove produced by nanosecond laser with fluence of 16.98 J/cm2 is V-shaped and it develops into a funnel-shape as the laser fluence increased to 45.27 J/cm2. The V-shaped and funnel-shaped microgrooves produced by nanosecond laser correspond to gentle ablation and strong ablation, respectively. Nevertheless, the cross-sectional shapes of the microgrooves created by picosecond laser under 1.41 and 2.83 J/cm2 energies are both V-shaped (Fig. 14(e) and (f)). This means that there is only one ablation phase for picosecond laser ablation of diamond.
Fig. 14. The top and side views of SEM images of the microgrooves created by (a-c) nanosecond or (d-e) picosecond lasers with different laser fluences.
The ablation threshold was determined using the D2-ln(F) method introduced by Wu et al. [35] and Wen et al. [36]. To calculate the ablation threshold, the ablation widths of the microgrooves produced by nanosecond laser were measured from the top view of the SEM images of the grooves (Fig. 14(a)). Figure 15(a) shows the dependence of groove width on fluence for nanosecond laser ablation. As the laser fluence increased from 11.41 to 47.91 J/cm2, the groove width D increased from 17.5 to 49.5 μm. The measured 1/e2-beam radius ω0 is 15.0 μm, which is deduced from the linear fit of D2 versus ln(F) in Fig. 15(b). The logarithm of the threshold fluence ln(Fth) can be obtained by extrapolation to D2 = 0, and the values of ln(Fth) are 1.823 and 2.903 for gentle and strong ablation, respectively. Therefore, the threshold fluences of diamond processed by nanosecond laser with 180 laser shots are 6.19 J/cm2 for gentle ablation and 18.23 J/cm2 for strong ablation.
Fig. 15. (a) The groove width as a function of the nanosecond laser fluence F. (b) Groove width squared D2 versus the logarithm of the laser fluence (lnF) under nanosecond laser irradiation.
The widths of microgrooves fabricated by picosecond laser with different laser fluence are plotted in Fig. 16(a). The groove width increased from 19.2 to 99.1 μm as the laser fluence increased from 0.80 to 3.18 J/cm2. The linear plot of D2 versus ln(F) is illustrated in Fig. 16(b). The spot size on diamond sample is 60.0 μm, which is extracted from the slope of the line in Fig. 16(b). The logarithm of the threshold fluence ln(Fth) of diamond processed by picosecond laser is -0.248, which is extracted from the intercept of the fitted line in Fig. 16(b). Therefore, the threshold fluence of diamond processed by picosecond laser with 4800 laser shots is 0.78 J/cm2.
Fig. 16. (a) Dependence of the groove width on the applied laser fluence F for picosecond laser ablation. (b) D2 versus ln(F) under irradiation of picosecond laser.
3.3.2 Ablation rate
The ablation rate is defined as the ablation depth per laser shot, which is calculated by dividing the ablation depth by the equivalent pulse number. The groove depths were measured from the side views of the SEM images of the grooves (Fig. 14(c)). Figure 17(a) illustrates the ablation depth of groove created by nanosecond or picosecond lasers as a function of laser fluence. The ablation depths are 48.7 μm at 47.91 J/cm2 for nanosecond laser and 125.6 μm at 3.18 J/cm2 for picosecond laser. Since the equivalent pulse numbers are 180 for nanosecond laser and 4800 for picosecond laser, thus the ablation rates are 270.6 nm/pulse at 47.91 J/cm2 for nanosecond laser and 26.2 nm/pulse at 3.18 J/cm2 for picosecond laser. Figure 17(b) shows the ablation rates of diamond processed by nanosecond and picosecond lasers at different laser fluences. As the fluence of nanosecond laser increased from 11.41 to 47.91 J/cm2, the groove depth increased from 13.6 to 48.7 μm, and the ablation rate increased from 75.8 to 270.6 nm/pulse. The groove depth processed by picosecond laser increased from 17.6 μm for 0.80 J/cm2 fluence to 125.6 μm for 3.18 J/cm2 fluence. The ablation rate of diamond processed by picosecond laser increased from 3.6 to 26.2 nm/pulse as the laser fluence increased from 0.80 to 3.18 J/cm2. It can be seen that the ablation rate of diamond processed by nanosecond laser is more than 10 times higher than that processed by picosecond laser. Therefore, nanosecond laser is more suitable for cutting and fabricating high-aspect-ratio microstructures on the diamond than picosecond laser.
Fig. 17. (a) The groove depth vs. laser fluence under nanosecond or picosecond lasers irradiation. (b) Dependences of ablation rate on the applied laser fluence.
3.3.3 Material removal rate
Furthermore, CLSM was employed to measure the 3D morphology and ablation volume of the microgroove. Figure 18(a) and (b) shows the 3D morphologies of the microgrooves ablated by nanosecond laser with 16.98 J/cm2 fluence and picosecond laser with 1.41 J/cm2 fluence, respectively. The ablation volume of the 63.9 μm-long microgroove produced by nanosecond laser is 7710 μm3. The ablation volume of the 127.8 μm-long microgroove created by picosecond laser is 149086 μm3. The material removal rate MRR of diamond is defined as MRR = (2ω0ρV)/(lN), where ρ is the density of diamond material (3.515 g/cm3), V and l are the measured volume and length of the microgroove, respectively. Therefore, the MRR of the diamond ablated by nanosecond laser with 16.98 J/cm2 fluence is 7.07 × 10−11 g/pulse, while that of the diamond ablated by picosecond laser with 1.41 J/cm2 fluence is 1.03 × 10−10 g/pulse. Laser fluence dependence of material removal rate of diamond ablated by nanosecond or picosecond lasers is depicted in Fig. 19. For nanosecond laser processing, when the laser fluence is less than 28.29 J/cm2, the material removal rate of diamond increases slowly due to the gentle ablation. As the laser fluence exceeds 28.29 J/cm2, the material removal rate of diamond increases rapidly due to the strong ablation, and reaches 3.34 × 10−10 g/pulse finally. For picosecond laser processing, the material removal rate of diamond increases linearly with the increase of laser fluence and finally keeps at 3.36 × 10−10 g/pulse, which is close to that of nanosecond laser.
Fig. 18. 3D morphologies of microgrooves created by (a) nanosecond laser with 16.98 J/cm2 fluence and (b) picosecond laser with 1.41 J/cm2 fluence.
Fig. 19. Dependences of the material removal rate of diamond processed by nanosecond or picosecond lasers on the applied laser fluence F.
In summary, the ablation threshold, ablation rate and material removal rate of diamond ablated by nanosecond or picosecond lasers were compared with femtosecond laser, which is presented in Table 2.
Table 2. Comparison of nanosecond, picosecond and femtosecond lasers processing of diamond
3.4 Simulation of temperature field of a diamond under laser irradiation
Furthermore, finite element software ANSYS was applied to simulate the temperature fields of diamond after single-pulse nanosecond and picosecond laser irradiations. Figure 20 shows contour plots of the temperature fields with different nanosecond laser fluences. As seen in Fig. 20(a), the maximum temperature of diamond after 16.98 J/cm2 nanosecond laser irradiation reaches 2906.95°C. This temperature is far higher than the graphitization temperature of diamond in air (700°C), but lower than the melting point of graphite (3550°C). Therefore, the diamond was converted into graphite under nanosecond laser irradiation with 16.98 J/cm2. When the laser fluence is increased to 28.29 J/cm2, the maximum temperature of diamond ablated by nanosecond is 5463.06°C, which exceeds the boiling point and vaporization temperature of graphite (3700°C). Hence, molten graphite was formed and vaporized in the irradiation region. As the laser fluence increases to 45.27 J/cm2, the maximum temperature of diamond reaches 9915.65°C, which indicates that more diamond material was removed via vaporization under high laser fluence. Therefore, the material removal mechanism of nanosecond laser processing of diamond includes heating, graphitizing, melting, boiling, and vaporization of diamond material. Firstly, the diamond is heated and converted to amorphous carbon and graphite, then the materials within the irradiation region are melted. Finally, the molten materials were ejected from the groove and then re-solidified to form particles on the diamond surface. This could result in the generation of re-deposited material on both sides of the microgroove irradiated by nanosecond laser.
Fig. 20. Temperature distribution on the diamond after single-pulse nanosecond laser irradiation; (a) 16.98 J/cm2; (b) 28.29 J/cm2; (c) 45.27 J/cm2.
The temperature distributions of diamond processed by single-pulse picosecond laser with different fluences are illustrated in Fig. 21. It can be seen that the heat is mainly concentrated on the diamond surface and hardly diffuse to the inside of the diamond. Hence, the temperature gradient of diamond ablated by picosecond laser is far larger than that ablated by nanosecond laser, which is consists with the results of Raman and theoretical analysis. Figure 21(a) shows that the maximum temperature of diamond irradiated by picosecond laser with laser fluence of 1.41 J/cm2 is 8061.69°C, which has far exceeded the vaporization temperature of graphite (3700°C). As the laser fluence increases to 2.83 J/cm2, the maximum temperature of diamond is higher than 18000°C. Since, the temperature of picosecond laser processing of diamond is much higher than that nanosecond laser processing of diamond. This leads to the direct sublimation of diamond material under picosecond laser ablation. Therefore, the material removal mechanism of picosecond laser processing of diamond includes heating, graphitizing, and subliming of diamond material. Hence, the microgrooves ablated by picosecond laser almost have no re-deposited particles around the groove.
Fig. 21. Temperature distribution on the diamond surface after single-pulse picosecond laser irradiation; (a) 1.41 J/cm2; (b) 2.83 J/cm2.
The graphite thickness can be obtained from the temperature distribution of diamond. As shown in Fig. 22(a), the graphite thickness along the Z direction is 2.6-3.2 μm. According to Kononenko’s [20] study, the graphite thickness for nanosecond laser ablation can be calculated by dg = ln(Ts/Tg)(χgτ)1/2, where Ts = 3973 K, Tg = 973 K and χg = 2.8 cm2·s-1 are the graphite vaporization temperature, the diamond graphitization temperature, and the thermal diffusivity of diamond, respectively. Therefore, the theoretical graphite thickness is 2.58 μm, which is close to the graphite thickness obtained by simulation methods. This verifies the correctness of the simulation model of laser processing of diamond. Figure 22(b) presents the temperature distribution along the Z direction for different picosecond laser fluences. The graphite thickness at bottom of the groove ablated by picosecond is 109-124 nm, which is much smaller than that ablated by nanosecond laser. Therefore, the picosecond laser is more suitable for high-precision processing of diamond with high quality due to less thermal damage.
Fig. 22. Temperature distribution along the Z direction for different (a) nanosecond or (b) picosecond laser fluences.
4. Conclusion
In this study, we have experimentally investigated the microgrooves on single-crystal diamond fabricated by ultraviolet nanosecond or infrared picosecond laser pulses. The experimental results have revealed that the microgrooves produced by nanosecond laser have a large number of re-deposited particles on both sides of the microgroove. However, the microgrooves ablated by picosecond laser almost have no re-deposited particles around the groove. Some cracks and chippings appeared at the edge of the microgroove for picosecond laser processing, but no cracks and chippings were formed at the edge of the microgroove for nanosecond laser processing. Periodic nanoripples with periods about 255 and 495 nm were generated in the microgroove ablated by picosecond laser, but no nanoripples were observed in the groove processed by nanosecond laser. The EDX spectra show that the laser-treated diamond consists of the carbon and oxygen elements, and the weight percentage of oxygen element significantly decreases with the increase of distance from the microgroove center. Raman analysis shows that the diamond material was converted to amorphous carbon and graphite after laser irradiation. The thermal stress in the groove fabricated by picosecond laser is larger than that created by nanosecond laser. Besides, the ablation threshold, ablation rate, and material removal rate of diamond processed by nanosecond or picosecond lasers were calculated. The threshold fluences of diamond ablated by nanosecond laser were 6.19 J/cm2 for gentle ablation and 18.23 J/cm2 for strong ablation. While the threshold fluence of diamond ablated by picosecond laser was 0.78 J/cm2. The ablation rate of diamond ablated by nanosecond laser is more than 10 times higher than that of diamond ablated by picosecond laser. The material removal rate of diamond processed by nanosecond or picosecond lasers depends on laser fluence and reaches about 3.3 × 10−10 g/pulse finally. Furthermore, the temperature distribution of diamond ablated by single-pulse nanosecond or picosecond laser were simulated. Experimental and simulation results show that the material removal mechanism of nanosecond or picosecond lasers was different. For nanosecond laser processing of diamond, the material removal mechanism includes heating, graphitizing, melting, boiling, and vaporization of diamond material. For picosecond laser processing of diamond, the material removal mechanism includes heating, graphitizing, and subliming of diamond material. The temperature distributions of diamond along the Z direction show that the graphite thickness is 2.6-3.2 μm for nanosecond laser and 109-124 nm for picosecond laser. This work indicates that nanosecond laser is more suitable for cutting and fabricating high-aspect-ratio microstructures on diamond, while picosecond laser is more suitable for high-precision processing of diamond microstructures with high quality.
Funding
National Natural Science Foundation of China (51805176, 51835004); Subsidized Project for Postgraduates' Innovative Fund in Scientific Research of Huaqiao University (18014080023).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. G. Li, M. Z. Rahim, W. Pan, C. Wen, and S. Ding, “The manufacturing and the application of polycrystalline diamond tools-A comprehensive review,” J. Manufact. Proc. 56, 400–416 (2020). [CrossRef]
2. Y. Okamoto, A. Okada, A. Kajitani, and T. Shinonaga, “High surface quality micro machining of monocrystalline diamond by picosecond pulsed laser,” CIRP Annals 68(1), 197–200 (2019). [CrossRef]
3. T. Wildi, M. Kiss, and N. Quack, “Diffractive optical elements in single crystal diamond,” Opt. Lett. 45(13), 3458 (2020). [CrossRef]
4. H. Umezawa, “Recent advances in diamond power semiconductor devices,” Materials Science in Semiconductor Processing 78, 147–156 (2018). [CrossRef]
5. Y. Su, Z. Li, L. Li, J. Wang, H. Gao, and G. Wang, “Cutting performance of micro-textured polycrystalline diamond tool in dry cutting,” Journal of Manufacturing Processes 27, 1–7 (2017). [CrossRef]
6. X. Hao, W. Cui, L. Li, H. Li, A. M. Khan, and N. He, “Cutting performance of textured polycrystalline diamond tools with composite lyophilic/lyophobic wettabilities,” J. Mater. Process. Technol. 260, 1–8 (2018). [CrossRef]
7. T. Shibata, Y. Kitamoto, K. Unno, and E. Makino, “Micromachining of diamond film for MEMS applications,” J. Microelectromech. Syst. 9(1), 47–51 (2000). [CrossRef]
8. E. Granados, M. Martinez-Calderon, M. Gomez, A. Rodriguez, and S. M. Olaizola, “Photonic structures in diamond based on femtosecond UV laser induced periodic surface structuring (LIPSS),” Opt. Express 25(13), 15330 (2017). [CrossRef]
9. N. Takayama and J. Yan, “Laser irradiation responses of a single-crystal diamond produced by different crystal growth methods,” Appl. Sci. 7(8), 815 (2017). [CrossRef]
10. N. Takayama, J. Ishizuka, and J. Yan, “Microgrooving of a single-crystal diamond tool using a picosecond pulsed laser and some cutting tests,” Precis. Eng. 53, 252–262 (2018). [CrossRef]
11. Y. Xing, L. Liu, X. Hao, Z. Wu, P. Huang, and X. Wang, “Micro-channels machining on polycrystalline diamond by nanosecond laser,” Opt. Laser Technol. 108, 333–345 (2018). [CrossRef]
12. F. Mouhamadali, S. Equis, F. Saeidi, J. P. Best, M. Cantoni, P. Hoffmann, and K. Wasmer, “Nanosecond pulsed laser-processing of CVD diamond,” Optics and Lasers in Engineering 126, 105917 (2020). [CrossRef]
13. H. Ohfuji, T. Okuchi, S. Odake, H. Kagi, H. Sumiya, and T. Irifune, “Micro-/nanostructural investigation of laser-cut surfaces of single- and polycrystalline diamonds,” Diamond and Related Materials 19(7-9), 1040–1051 (2010). [CrossRef]
14. M. Wu, B. Guo, and Q. Zhao, “Laser machining micro-structures on diamond surface with a sub-nanosecond pulsed laser,” Appl. Phys. A. 124(2), 1–9 (2018). [CrossRef]
15. N. Takayama and J. Yan, “Mechanisms of micro-groove formation on single-crystal diamond by a nanosecond pulsed laser,” J. Mater. Process. Technol. 243, 299–311 (2017). [CrossRef]
16. M. Gedvilas, J. Mikšys, J. Berzinš, V. Stankevič, and G. Račiukaitis, “Multi-photon absorption enhancement by dual-wavelength double-pulse laser irradiation for efficient dicing of sapphire wafers,” Sci. Rep. 7(1), 5218 (2017). [CrossRef]
17. G. Eberle, K. Jefimovs, and K. Wegener, “Characterisation of thermal influences after laser processing polycrystalline diamond composites using long to ultrashort pulse durations,” Precis. Eng. 39, 16–24 (2015). [CrossRef]
18. M. D. Shirk, P. A. Molian, and A. P. Malshe, “Ultrashort pulsed laser ablation of diamond,” J. Laser Appl. 10(2), 64–70 (1998). [CrossRef]
19. G. Eberle and K. Wegener, “Ablation study of WC and PCD composites using 10 picosecond and 1 nanosecond pulse durations at green and infrared wavelengths,” Phys. Proc. 56, 951–962 (2014). [CrossRef]
20. V. V. Kononenko, T. V. Kononenko, S. M. Pimenov, M. N. Sinyavskii, V. I. Konov, and F. Dausinger, “Effect of the pulse duration on graphitisation of diamond during laser ablation,” Quantum Electronics. 35(3), 252–256 (2005). [CrossRef]
21. A. Nebel, T. Herrmann, B. Henrich, and R. Knappe, “Generation of tailored picosecond-pulse-trains for micro-machining,” Proc. SPIE 6108, 610812–610818 (2006). [CrossRef]
22. Z. Zhang, Q. Zhang, Q. Wang, H. Su, Y. Fu, and J. Xu, “Investigation on the material removal behavior of single crystal diamond by infrared nanosecond pulsed laser ablation,” Opt. Laser Technol. 126, 106086 (2020). [CrossRef]
23. N. Ren, L. Jiang, D. Liu, L. Lv, and Q. Wang, “Comparison of the simulation and experimental of hole characteristics during nanosecond-pulsed laser drilling of thin titanium sheets,” Int. J. Adv. Manuf. Technol. 76(5-8), 735–743 (2015). [CrossRef]
24. M. Li, B. Han, S. Zhang, L. Song, and Q. He, “Numerical simulation and experimental investigation on fracture mechanism of granite by laser irradiation,” Opt. Laser Technol. 106, 52–60 (2018). [CrossRef]
25. S. Kumar, R. Singh, Q. Yang, S. Cheng, B. Zhang, and B. K. Kaushik, “Highly sensitive, selective and portable sensor probe using germanium-doped photosensitive optical fiber for ascorbic acid detection,” IEEE Sens. J. 21(1), 62–70 (2020). [CrossRef]
26. S. Kumar, R. Singh, G. Zhu, Q. Yang, X. Zhang, S. Cheng, B. Zhang, B. K. Kaushik, and F. Liu, “Development of uric acid biosensor using gold nanoparticles and graphene oxide functionalized micro-ball fiber sensor probe,” IEEE Trans.on Nanobioscience 19(2), 173–182 (2020). [CrossRef]
27. G. C. B. Lee, S. Su, J. Li, K. Sugden, N. Roohpour, H. Yan, and H. Ye, “Analysis of femtosecond laser surface patterning on bulk single-crystalline diamond,” Journal of Experimental Nanoscience 7(6), 662–672 (2012). [CrossRef]
28. R. Windholz and P. A. Molian, “Nanosecond pulsed excimer laser machining of chemical vapour deposited diamond and highly oriented pyrolytic graphite: Part I An experimental investigation,” J. Mater. Sci. 32(16), 4295–4301 (1997). [CrossRef]
29. I. Calizo, S. Ghosh, W. Bao, F. Miao, C. Ning Lau, and A. A. Balandin, “Raman nanometrology of graphene: Temperature and substrate effects,” Solid State Commun. 149(27-28), 1132–1135 (2009). [CrossRef]
30. H. M. Shanshool, H. Naser, N. M. Hadi, H. A. Flaih, F. M. Abbas, M. J. Hussin, and S. S. Hindal, “Parameters affecting the welding of transparent materials using femtosecond laser pulses,” Lasers Manuf. Mater. Process. 7(1), 59–73 (2020). [CrossRef]
31. C. Xiong, C. Ho, and D. Qiao, “Analysis of direct optical ablation and sequent thermal ablation for the ultrashort pulsed laser photo-thermal micromachining,” Coatings 10(12), 1151 (2020). [CrossRef]
32. M. Forster, C. Huber, O. Armbruster, R. Kalish, and W. Kautek, “50-nanometer femtosecond pulse laser induced periodic surface structures on nitrogen-doped diamond,” Diamond and Related Materials 74, 114–118 (2017). [CrossRef]
33. Q. Wu, Y. Ma, R. Fang, Y. Liao, Q. Yu, X. Chen, and K. Wang, “Femtosecond laser-induced periodic surface structure on diamond film,” Appl. Phys. Lett. 82(11), 1703–1705 (2003). [CrossRef]
34. S. Sakabe, M. Hashida, S. Tokita, S. Namba, and K. Okamuro, “Mechanism for self-formation of periodic grating structures on a metal surface by a femtosecond laser pulse,” Phys. Rev. B 79(3), 033409 (2009). [CrossRef]
35. M. Wu, B. Guo, Q. Zhao, R. Fan, Z. Dong, and X. Yu, “The influence of the focus position on laser machining and laser micro-structuring monocrystalline diamond surface,” Optics and Lasers in Engineering 105, 60–67 (2018). [CrossRef]
36. Q. Wen, P. Zhang, G. Cheng, F. Jiang, and X. Lu, “Crystalline orientation effects on material removal of sapphire by femtosecond laser irradiation,” Ceram. Int. 45(17), 23501–23508 (2019). [CrossRef]
37. M. S. Komlenok, V. V. Kononenko, V. G. Ralchenko, S. M. Pimenov, and V. I. Konov, “Laser induced nanoablation of diamond materials,” Physics Procedia 12, 37–45 (2011). [CrossRef]
38. C. Wei, Y. Ma, Y. Han, Y. Zhang, L. Yang, and X. Chen, “Study on femtosecond laser processing characteristics of nano-crystalline CVD diamond coating,” Appl. Sci. 9(20), 4273 (2019). [CrossRef]
