Generation of textured diamond abrasive tools by continuous-wave CO2 laser: Laser parameter effects and optimisation

Journal of Materials Processing Tech. 275 (2020) 116279

Contents lists available at ScienceDirect

Journal of Materials Processing Tech. journal homepage: www.elsevier.com/locate/jmatprotec

Generation of textured diamond abrasive tools by continuous-wave CO2 laser: Laser parameter effects and optimisation Hao Nan Lia, a b

⁎,1

T

, Ke Ge Xieb,1, Bo Wub, Wei Qiang Zhub

School of Aerospace, University of Nottingham Ningbo China, Ningbo 315100, China Department of Mechanical, Materials & Manufacturing Engineering, University of Nottingham Ningbo China, Ningbo 315100, China

A R T I C LE I N FO

A B S T R A C T

Associate Editor: R. Mishra

Textured Diamond Abrasive Tools (TDATs) have been attracted multitudes of attentions from both academic scholars and industrial engineers due to the reported superior grinding performances. However, the absence of appropriate fabrication technologies can still be considered as the bottleneck of wide applications of TDATs. To fill this gap, this paper utilised the continuous-wave CO2 laser to produce passive-grinding structures on the diamond abrasive tool surfaces. Not only the theoretical analysis of the non-metallic multi-material ablation threshold for diamond abrasive tools, but also the experimental investigation of the effects of both the laser power and the beam feed rate on both the topographies and morphologies of the produced structures were studied. The different structure generation mechanisms under different laser parameters were analyzed as well. Based on this, the strategy for selecting proper ablation parameters was provided. Various examples of the produced textured wheels were given in the end to prove the study significance and feasibility. The presented work was expected helpful to provide a new method for industries to produce textured grinding wheels in an efficient and low-cost way.

Keywords: Diamond abrasive tools Textured surface Laser ablation CO2 laser

1. Introduction The need of high-performance abrasive tools has been highly emphasized due to the increasingly strenuous working conditions in the modern ultra-high-precision processes featured by intensive load and high temperature (Hao Nan Li et al., 2019; Lin Li et al. 2019). The stateof-the-art research on this was intensively performed, from the fundamental single grain fracture modes (Wang et al., 2018), via the progressive grain wear (Dai et al., 2019), to the high-performance abrasive tool applications in the new materials like metal matrix composites (Zhou et al., 2019). Even so, the most widely-reported issues however have been closely related with various kinds of thermal damage, including residual stress (Nguyen and Zhang, 2009), phase transformation (Li and Axinte, 2018), grinding hardening layer (Silva et al., 2013), and heat-induced surface (Yu et al., 2016) or subsurface (Zhao et al., 2018) microcracks. Based on this, Textured Diamond Abrasive Tools (TDATs), have been proposed and already attracted multitudes of attentions from both academic scholars and industrial engineers (Li and Axinte, 2016). The superior performances of TDATs have been extensively found in the experimental trials including good coolant transportation ability (Mihić et al., 2013), effectiveness to lower

grinding temperature (Butler-Smith et al., 2009), and achievable ultrasmooth surfaces (Denkena et al., 2015a, 2015b). However, fabrication technologies of TDATs can be still considered as an open question due to the fact that the ultra-high hardness of diamonds embedded in porous bond agents would be quite difficult to remove in a controllable way (Li and Axinte, 2016; Ding et al., 2017). To this end, many efforts have been paid so far to try to properly produce macro/microscale structures on diamond abrasive tool surfaces. Based on the basic material removal principles, most of the previous TDAT fabrication methods can be mainly divided into three groups: (i) mechanical, (ii) chemical, and (iii) thermal methods. By introducing geometrical interference between standard diamond tool surfaces and texturing tools (e.g. milling cutters, rollers), mechanical methods in general removed both diamond abrasives and bond agents by consuming the mechanical energy. Mohamed et al. (2013), (2014) not only generated the spiral grooved grinding wheels by the single-diamond dresser so that the achievable wheel cut depth was improved by 120% while the power was reduced by 64%, but also proposed the synchronous system which can refresh the worn groove geometry. The size, the dimension accuracy, and the pattern of the achievable textures were scaled down, improved, and diversified by

⁎

Corresponding author. E-mail address: [email protected] (H.N. Li). 1 These authors equally contributed to this work. https://doi.org/10.1016/j.jmatprotec.2019.116279 Received 13 December 2018; Received in revised form 27 May 2019; Accepted 30 June 2019 Available online 02 July 2019 0924-0136/ © 2019 Elsevier B.V. All rights reserved.

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

Table 1 The details of the used diamond abrasive grinding wheels in the trial. shape

dimension (mm)

abrasive type

abrasive size (#)

wheel structure number

abrasive density (vol.%)

bond type

plate

Φ120× Φ32× 20

Diamond

120

N

75%

resin

threshold for TDATs, but also the experimental investigation of the effects of both the laser power and the beam feed rate on both the topographies and morphologies of the produced structures are studied. The different structure generation mechanisms under different laser parameters are analyzed as well, based on which the strategy for selecting proper ablation parameters is provided. Various examples of the produced textured wheels are given in the end to prove the study significance and feasibility. The presented work is anticipated to be not only meaningful to provide the academic reference to the research field of ultra-high precision machining, but also helpful to provide a new method for industries to produce textured grinding wheel in an efficient and low-cost way.

separately equipping the diamond tool with the external excitation (Oliveira et al., 2010), employing the direct drive dresser (Silva et al., 2013), and using the fly-cutting technology (Denkena et al., 2015a, b). Multiple cutting edge tools were employed as well to produce textured grinding wheels (Aurich et al., 2011), based on which the coolant transportation ability was increased by 5 times (Aurich and Kirsch, 2013). However, intensive texturing tool wear and significant chatter can be considered as the key issues (Li and Axinte, 2016). Chemical methods created textures by adding materials in the controllable way and can mainly include (i) electroplating and (ii) brazing. For electroplating, Aurich et al. (2008) applied the masking technology for the preliminary adhesion of the grains on the wheel hub so that the uniformly-distributed grain pattern was achieved. The bioinspired electroplating grinding tools were generated by both Lyu et al. (2017) and Yu et al. (2018) via using the light sensitive PVC together with the UV exposure technology. However, the failures including abrasive dislodgement, grain fracture, and attrition were still widely reported for the electroplated textured wheels (Upadhyaya and Fiecoat, 2007). To overcome this, the alternative texturing method, brazing, was proposed. Both Chattopadhyay et al. (1991) and Chen et al. (2010) employed the Ni-Cr alloy as the filler due to high strength and resistance to heat, wear and corrosion. Ding et al. (2013) and Chen et al. (2017) found the mismatch of the thermal expansion between abrasives, filler alloys, and substrate can result in substantial residual stress. Huang et al. (2018) introduced the ultrasonic vibration into the diamond brazing process to release residual stress after brazing. Even so, the inevitable issues containing low time efficiency and process controllability, huge environment hazards, and complex fabrication procedures, should be properly solved to make this method more applicable and feasible (Li and Axinte, 2016). Thermal methods employed high power laser or plasma/ion to remove or degrade not only the bond agents but also the abrasives by gasification, oxidation, melt, and decomposition. The structures having the feature size in the order of a few micrometers on the abrasive tool surface was successfully produced by using picosecond pulse laser (Walter et al., 2014). Similar micrometer-scaled textures were also generated by Guo et al. (2014a) by using nanosecond pulse laser. However, the very careful selection of process parameters was required otherwise the discontinuous textures might be generated (Guo et al., 2014b). Butler-Smith et al. (2009) employed the pulse laser to ablate thick-film CVD diamond so that the preferential micro grinding arrays having the rectangular, triangular, and hexagonal shapes were generated. More superior grinding element having the well-designed rake and flake angles was generated later on (Butler-Smith et al., 2011). However, high equipment maintenance cost, limited ablation penetration depth, and relatively low efficiency would be the critical bottlenecks of the current laser-based texturing methods (Li and Axinte, 2016). To fill this gap, this paper aims to utilise the continuous-wave CO2 laser to produce passive-grinding structures on the diamond abrasive tool surfaces considering the superior advantages of this laser source including: (i) the high machining efficiency resulted from both the high average laser power and the stablised laser beam quality due to the continuous-wave nature, (ii) the good ablated surface quality thanks to the good match between the long CO2 laser wavelength and the absorbable light spectrum of non-metallic materials containing abrasives and bonds, and (iii) the superior cost efficiency because of the absence of the pulse-laser generator and any modulation magnifier. Not only the theoretical analysis of the non-metallic multi-material ablation

2. Experiment methodology The standard commercial diamond abrasive grinding wheels (D120N75B771/8, 3M Company) were used in all the trials in this study (see more details in Table 1). The laser beam was generated by the continuous-wave carbon dioxide generator (LE900, Hongfan Tech. Company) with the maximum power of 60 W, the wavelength of 10.6 μm, the circular focus with the diameter of 0.3 mm, and the focal length of 12.5 mm. The linear and angular motions and positions of the ablated diamond wheels were accurately controlled by the specially-made positioning system having the translational and the rotational motion accuracies of 1 μm and 0.05 deg based on the G-code-based SIEMENS NC system. As seen in Fig. 1(a), the positioning system consisted of the controller (BR010-11T8 × 2M, BenT CNC Automation Equipment Company), the driver (ZD-M42, Philips Company), the stepper motor (42HS4013A4CE, Sumtor Company), the reducer (SK4248-19.2, Yixing Technology Company) and other auxiliary hardware, enabling the 2axis simultaneous motions (which were the translational and the rotational motions separately along and around the wheel axis). The translational and rotational motion speeds were separately within the ranges of 0–100 mm/s and 0–15 deg/s. Because this paper aims to study the effects of laser parameters on the ablated surface topography and morphology, the wide ranges of both the laser power and feed rate were employed. To make all the results comparable, all the laser ablation texturing trials using seven different levels of laser power and eight levels of laser feed rates were produced on one diamond wheel surface in air atmosphere. As seen in Fig. 1(b), the whole wheel was equally divided into eight segments (see segment A to H) along the wheel circumferential direction and in each segment there were seven ablated tracks along the wheel axis direction (see track X1 to X7 where X refers to A-H). The laser feed rate was incrementally increased from the segment A to H, while the laser power was incrementally increased from the track X1 to X7 (where X refers to A-H). The used parameters were given in Table 2 in detail. After the laser ablation texturing, the diamond wheel was firstly carefully cut into eight segments (A-H) by the wire-EDM machine (MV2400S, Mitsubishi Electric Company), and the two cross sections of each segment were carefully polished by the alumina with the sizes of #600 (6 h), #1200 (6 h), and #2000 (6 h). Each segment then was cleaned by the ultrasonic distilled water bath for 1 h and air-dried at room temperature. Various analyses based on Scanning Electron Microscope (SEM) (174C CZ, ZEISS Company), Energy Dispersion Spectrum (EDS) (174C CZ, ZEISS Company), and Cu-Kα X-Ray Diffraction (XRD) (D8 Advance, Bruker Company) were performed to observe and measure the ablated surface structure topographies and 2

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

Fig. 1. Experiment methodology. (a) experimental setup including the laser and motion control PCs (left), the specially-designed positioning system (middle) and the laser ablation system setup (right), (b) the ablation strategy where the wheel was equally divided into eight segments (A-H) and in each segment there were seven ablated tracks along the wheel axis direction (X1-X7 where X refers to A-H). The laser feed rate was incrementally increased from the segment A to H, while the laser power was incrementally increased from the track X1 to X7, and (c) the quadratic function fitting by the least square method based on the nine keypoints (P1–P9).

one deepest valley point (where the depth was Hmax , see the Point P5), two end points (see the Point P1 and P9), and six intersecting points between the cross section profile and the horizontal lines having the depth of 25%, 50% and 75% of Hmax (see the Point P2-P4 and P6-P8). For the step (ii), the top width and depth can be separately defined as the distance of the two end points (P1 and P9) and the vertical distance of the end point (P1 or P9) to the deepest valley point (P5). The tilt

morphologies respectively. The ablated structure topography was evaluated by (i) fitting the nine keypoints on the observed cross-section profiles based on the least square method to obtain the quadratic function (the red line in Fig. 1c right), and (ii) comparing the four indicators including top and bottom width, depth and tilt angle based on the fitted quadratic function (see the annotations in Fig. 1c, left). For the step (i), the keypoints included

Table 2 The employed laser parameters in the trials. Trial No.

laser power (W)

laser feed rate (mm/s)

Segment No.

Track No. in each segment

Trial No.

laser power (W)

laser feed rate (mm/s)

Segment No.

Track No. in each segment

1–7 8–14 15–21 22–28

60,54, 48,42, 36,30, 24

1.0 2.0 3.0 4.0

A B C D

A1–A7 B1–B7 C1–C7 D1–D7

29–35 36–42 43–49 50–56

60,54, 48,42, 36,30, 24

5.0 6.0 7.0 8.0

E F G H

E1–E7 F1–F7 G1–G7 H1–H7

3

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

angle was obtained by fitting the key points on one side (P6-P9 for the right side, and P1-P4 for the left side). The bottom width was the length of the interception between the two fitted tilt lines and the horizontal line passing the deepest point (P5).

extent that, the classic laser ablation threshold for homogeneous materials might not be applicable to multi-material structures, and obviously, more powerful laser parameters should be employed. Fig. 2(b) shows one example of the ablated structures by using the parameters given in Table 2. It can find that, not only the resin bond was completely burned out but also the diamond was partially graphitised (proved by later XRD analysis), indicating the high feasibility to produce texture structures when using this set of laser parameters. The equivalent laser fluence F corresponding to this set of parameters can be obtained by Eq. (2) based on the schematic in Fig. 2(c),

3. Results & discussion 3.1. Non-metallic multi-material ablation threshold for textured diamond abrasive tools (TDATs) The first consideration when texturing diamond abrasive tools would be related with the ablation threshold which can be determined by both the laser parameters, which were discussed in the following part, and the target material. Therefore an interesting note here would be the multi-material and non-metallic nature of diamond abrasive tools (made of diamond and polymer resin) when doing laser ablation texturing, because nearly all the previous studies focused on the ablation of either homogenous materials or metallic materials. The ablation threshold refers to the laser fluence by which the target material undergoes a permanent damage and at least one layer of material is removed (Jeschke and Garcia, 2002). Based on the classic laser ablation theory (Jeschke and Garcia, 2002), the threshold of a homogenous material can be described as

F=

⎜

(2)

where Pm is the average power, t is the ablated duration, r is the laser spot radius, and v is the laser feed rate. Surprisingly, the equivalent laser fluence F was 5 times of the theoretical threshold of the resin material. This implies that the multimaterial structure ablation mechanism would be much more complex than the single-material homogeneous structure. This might be because of the uncertain chemical reactions between multiple components, the instable absorption of laser power influenced by other components, and the different ablation threshold for different components. 3.2. Effects of laser power on the ablated structure topography

1

λ π ⎞2 Ith = ⎛ (Tc − T0) A ⎝ 4αt0 ⎠

Pm Pm = 2So1 + SABCD πr 2 + 2rvt

Fig. 3 shows the significant effects of laser power on the ablated structure topography: the power increase in general would lead to both the linearly-increasing top width (from 808.31 μm to 1850.30 μm with the slope of 16 μm/W) and the depth (from 297.44 μm to 1012.79 μm with the slope of 25 μm/W) and the nonlinearly-increasing bottom width (from 378.34 μm to 857.66 μm) and side edge tilt angle (from 55.08° to 69.38°). The large difference of the ablated slot top width and depth under different laser powers indicated the continuously effective ablation thanks to (i) the good match between the long CO2 laser wavelength (10.6 μm) and the absorbable light spectrum of both artificial diamond (around 4.3 μm) and resin (around 7.8 μm) atoms in the provided fluence range, and (ii) the well-focalised and stable CO2 laser beam during the whole ablation process even after a layer of the wheel material with the thickness of 1000 μm (the largest valley depth in Fig. 3a) was removed. One noticeable observation here would be the critical laser power region (see Fig. 3c), in which the rising power led to, not the increasing,

⎟

(1)

where λ is the thermal conductive rate of the material, A is the material absorption rate to laser, α is the thermal diffusivity of the material, t0 is the laser dwelling time, Tc is the crucial temperature, and T0 is the room temperature. Please note the key goal when texturing grinding wheels is to remove abrasives within specific area. The ablation threshold therefore can be considered the ablation threshold of the resin bond, i.e. 2.53 × 104 W/cm2 (when the feed rate is 1.0 mm/s), because (i) it is the bond agent that holds the diamond abrasives to remove materials, and (ii) only the successful removal of the bond agent can enable more coolant to be transported in the grinding zone to lower the temperature. However, the ablated surface based on the theoretical value of 2.53 × 104 W/cm2 indicated that the resin materials presented loosened structure but with very limited material removal, and the diamond retained intact as well (see Fig. 2a). This concluded to a large

Fig. 2. The determination of the ablation threshold for non-metallic multi-material resin-bonded diamond grinding wheel. The ablated structure morphologies (a) when the resin threshold value was employed where the loosened structure but with very limited material removal can be observed and the diamond retained intact as well, and (b) when the 5 times of the theoretical threshold value was employed where the resin bond was completely burned out, and (c) the schematic of the laser fluence calculation.

4

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

Fig. 3. The effects of laser power on the ablated structure topography. (a) The fitting profile of the ablated structure cross section profile when the feed rate was 3 mm/s and the laser powers were 54 W (specimen A), 48 W (specimen B), 42 W (specimen C), 36 W (specimen D), 30 W (specimen E), and 24 W (specimen F), (b) the SEM micrographs of the cross sections of the specimens from A to F, and the effects of laser power on the (c) top width (red) and depth (blue), and (d) bottom width (red) and side edge tilt angle (blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

scattering, and (ii) air field ionization (multi-photon ionization, impact ionization and tunneling ionization) leading to the laser beam defocusing effects. It can be seen from Fig. 3(c) and (d) that, the critical laser power region for resin-bonded diamond wheel might be probably within the range from 36 W to 42 W. Please note that the data in Fig. 3(c) and (d) was the average values of the multiple measurements while the error bars referred to the standard deviation. Besides, Fig. 3(b) shows the varying cross section shapes of the ablated structures, from the inverted triangle (see A and B in the Fig. 3b), via the semi-circle (see C, D, and E in the Fig. 3b), to the shallow crater (see F in the Fig. 3b) when the laser power was decreased. There is no surprise for the semi-circle and shallow crater shapes because they fitted well with the ideal Gauss distribution of the

but the decreasing top width (see Fig. 3c) and the side edge tilt angle (Fig. 3d). This interesting phenomenon might be explained by the following competing active and negative ablation processes: in the active ablation, the increasing laser power not only resulted in intensive radiation fluence leading to oxidation, graphitisation, melt and even gasification, but also generated the thermal shock to diamond leading to breakage. Therefore, substantial material removal was achieved in this phase. However, the over-increased laser power would lead to the negative ablation phase as well because the powerful laser enables (i) the generation of substantial micro-particles (including micro diamond fragment, micro graphite particles, or other ash-like materials), either covering the ablated spots and hindering the laser absorption due to the high reflectivity, or suspending in the air leading to the laser beam 5

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

Fig. 4. The effects of laser feed rate on the ablated structure topography. (a) The fitting profile of the ablated structure cross section profile when the laser power was 54 W and the feed rates were 3.0 mm/s (specimen A), 4.0 mm/s (specimen B), 5.0 mm/s (specimen C), 6.0 mm/s (specimen D), 7.0 mm/s (specimen E), and 8.0 mm/s (specimen F), (b) the SEM micrographs of the cross sections of the specimens from A to F, and the effects of feed rate on the (c) top width (red) and depth (blue), and (d) bottom width (red) and side edge tilt angle (blue). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

(from 857.66 μm to 382.3 μm), but the increasing side edge tilt angle (from 64.25° to 71.01°). The relationships between the four geometrical parameters and the laser feed rate showed clearly nonlinear feature (see Fig. 4c, d). The critical feed rate region can also be observed in Fig. 4(c) and (d), similar to the critical laser power region. In this region, the increased feed rate resulted in not only the relatively constant depth (see Fig. 4c) but also the decreasing side edge tilt angle (Fig. 4d). The statement based on the competing active and negative ablation in Fig. 3 can also be applied here to explain this observation, and the only difference is in the previous statement the high laser power led to the active ablation while in this case the low feed rate did because of the

power density within the laser beam when normal levels of laser power were used. However, when the laser power was over-increased, intensive air ionization, large ablation depth, and substantial micro-particles would take place, resulting in the non-Guass-shaped cross section profile due to the significant re-focusing effect (Shukla et al., 2017).

3.3. Effects of laser feed rates on the ablated structure topography Fig. 4 shows the dominant influence of laser feed rate on the ablated structure topography. The increased feed rates generally led to the decreasing top width (from 1850.30 μm to 761.90 μm), depth (from 1012.79 μm to 521.63 μm with the slope of 25 μm/W), bottom width 6

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

Fig. 5. The effects of laser power on the ablated structure morphology. (a) The fitted profile of the ablated structure cross section profile when the feed rate was 3.0 mm/s and the laser power was separately 24 W (specimen A), 42 W (specimen B), 54 W (specimen C) and the SEM micrographs of the top views from specimen C, and the SEM micrographs, the XRD patterns, and the EDS analysis results of the specimen A (b, c, d), B (e, f, g) and C (h, i, j).

ablated structures, where the profile shapes kept relatively constant, only changing from the inverted triangle (see A, B, D in the Fig. 2b) to the semi-circle (see E and F in the Fig. 2b) when the feed rate was increasing. The inverted triangle shape might probably be because of the significant re-focusing effect resulted from the high laser power

incubation effect (Zimmer et al., 2014). What is even more interesting here would be the encouragingly good match of the laser fluence of both the critical laser power region (from 36 W to 48 W) and the feed rate region (from 4.0 mm/s to 6.0 mm/s). Similar to Figs. 3(b), 4 (b) illustrates the cross-section profile of the 7

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

Fig. 6. The effects of feed rates on the ablated structure morphology. (a) the fitting profile of the ablated structure cross section profile when the laser power was 54 W and the feed rate was separately 8.0 mm/s (specimen A), 5.0 mm/s (specimen B), 3.0 mm/s (specimen C) and the SEM micrographs of the top views from specimen C; and the SEM micrographs, the XRD patterns, and the EDS analysis results of the specimen A (b, c, d), B (e, f, g) and C (h, i, j).

because the employed laser feed rate in this case (5.0 mm/s) was within the critical feed rate region and therefore the opposite effect took place.

density induced by the low feed rate, while the semi-circle shape was attributed to limited material removal behaviors due to the high feed rate. The only abnormal observation here would be the semi-circleshaped ablation profile in the specimen C in Fig. 4, which might be

8

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

the carbide weight percentage reduced from 70.09% via 62.84% to 58.35%) when the laser power was increased.

3.4. Effects of laser power on the ablated structure morphology Fig. 5 shows the effects of laser power on the ablated structure morphologies. Both the types and the numbers of the observed morphologic features tended to be increased with the increasing laser power. This might be because of both (i) the increasingly complex and aggressive thermal, optical, electric, and mechanical cross-interactions between the laser beam, the abrasives, the resin bond and the air and (ii) the nonlinear and steep temperature gradient from the ablated surface to the subsurface covering an increasing area and leading to various material behaviors. The observed features can be mainly categorised into three groups according to different material behaviors of the diamond abrasives, including (i) exposure/loosening, (ii) deterioration and (iii) breakage. The exposure/loosening was the dominant feature of the ablated abrasive tool morphologies when the low laser power was used (see the specimen A in Fig. 5 when the laser power was 24 W). The abrasive exposure/loosening refers to the reduced abrasive-holding structure of the resin bond generated by the laser ablation, and was mostly observed on the ablated wheel surface. The generation of the loosening structure might be because the low laser power can result in the bond melting, rather than the material removal, and generate the ash-like structure after the resolidification (see the specimen A). It might therefore predict that the loosening features negatively (i) generated the ineffective abrasives which can not remove materials in grinding due to the limited abrasive retainment, (ii) increased the uncertainties of the ultra-highprecision grinding process accuracy due to the loose abrasives on the wheel surface, and (iii) raised the possibility of the undesired abrasive fall-off before the diamond normally worn out. The exposure/loosening was the dominant feature as well, although the laser power was increased from 24 W to 42 W. However, the noticeable observation here would be the absence tendency of the loosening features (see the specimen B) and the raising generation of the abrasive exposure feature. This phenomenon might be explained by the full resin bond material removal by the gasification induced by the medium laser power, rather than just the melting and resolidification in the low laser power. The exposure features are anticipated to (i) actively increase the functional cutting abrasives with favorable grain protrusion heights, leading to more sharp grinding edges and therefore more superior tool grinding ability, and (ii) the intact diamond geometry with the appreciable portion buried beneath the resin bond, leading to better grain holding ability and longer tool durability. Parallel to the exposure/loosening features, the unique morphology featured by abrasive breakage started to appear when the high laser power of 54 W was employed (see the specimen C), where the clear evidences of the diamond fragments can be identified. This breakage was probably resulted from the poor diamond heat resistance, i.e. the diamond material would break easily due to the non-uniformly distributed thermal deformation. A very interesting observation here can be the breakage feature locations, i.e. most of the broken diamonds can be found 300 μm ˜ 400 μm beneath the ablated surface, instead of on the surface. The reason might be: the diamonds on the ablated surface were fully burned out because (i) the employed laser energy density was 3.40 × 105 W/cm2 higher than the diamond ablation threshold and (ii) the air atmosphere facilitated the diamond oxidation. On the contrary, the diamonds beneath the surface suffered the relatively mild thermal shock in the oxygen-poor environment, leading to the incomplete ablation and the breakage, and therefore the broken diamond fragments properly preserved by the resin bond can be clearly observed. Except for the physical features mentioned above (i.e. loosening, exposure, and breakage), the chemical deterioration of both the diamonds and the resin bond can also be found in all the levels of laser power. Not only the diamond was deteriorated into graphite (see c, f, i in Fig. 5 where the diamond weight percentage reduced from 18.12% via 14.27% to 12.62%), but also the organic resin bond was oxidized into the various carbides (e.g. CO, CO2, etc.) (see d, g, j in Fig. 5 where

3.5. Effects of feed rates on the ablated structure morphology Fig. 6 presents the effects of feed rates on the ablated structure morphology. The numbers of the morphologic features of the ablated surface tended to be increased with the reduced laser beam feed rate. This might be because of (i) the increasingly intensive laser fluence, leading to the substantial heat accumulation, material removal behaviors, and air ionization, and (ii) the large thermal affected zone and the complex material behaviors within the zone. The diamond exposure features happened in all the levels of the feed rates from 3.0 mm/s via 5.0 mm/s to 8.0 mm/s when the laser power was kept 54 W. Based on the loosening and breakage features, the effects of the feed rates tended to be unstable: on one hand, the low feed rate of 3.0 mm/s resulted in the presence of the abrasive breakage due to the increasing laser energy accumulation; on the other hand, the fast feed rate of 8.0 mm/s did not lead to the presence of the abrasive loosening (see Fig. 6b), somehow indicating the larger influence of the laser power on the ablated structure morphology in comparison with the feed rates. Regarding the deterioration feature, the low feed rate led to the high possibility of the diamond-to-graphite transformation, which can be evidenced by Fig. 6(c, f, i), where the diamond weight percentage reduced from 16.46% via 13.37% to 12.62%. This transformation was performed because of (i) the laser-induced high temperature and (ii) the relatively low atmospheric pressure in comparison with the critical graphite-to-diamond transformation pressure (De Feudis et al., 2017). 3.6. Selection of laser ablation parameters for texturing diamond abrasive tools Based on the understanding of the laser parameter effects on both topographies and morphologies, the ambition of this section can be considered meaningful, which is the usage of the knowledge above to properly select ablation parameters so that the passive-grinding textures having the limited undesired thermal defects can be produced, because to the best knowledge of the authors most of the previous studies chose the laser ablation parameters by the trial-and-error principle and very limited theoretical guidance was given yet. The parameter selection strategy can be explained based on Fig. 7(a), where the vertical and horizontal axes are separately the energy fluence and the laser feed rate. The whole diagram can be divided into three domains according to both the diamond and resin ablation thresholds calculated by Eq. (1), i.e. (i) the non-ablation domain, where neither the bond resin nor the diamond abrasive can be ablated, (ii) the resin-ablated domain, where only the resin is ablated while the diamond is not, and (iii) the full-ablation domain, where both the resin and the diamond were ablated. The energy fluence corresponding to different employed laser power when different feed rates are used can be plotted in Fig. 7(a) as well based on Eq. (2). Based on above, the reasonable laser parameters can probably be selected. On one hand, the ablation parameters within the full domain can be recommended because only the ablated diamonds can enable the diamond-covered resin to be removed so as to create the passive-grinding textures. On the other hand, the usage of high laser power should also be employed because of the efficient ablation material removal evidenced by the enough ablation depth allowing enough coolants to be transported into the grinding zone to lower the grinding temperature. Considering the employment of the maximum laser power of the employed generator (60 W) might lead to unstable laser beam due to fluctuation, only 90% of the maximum power should be used. This concludes the parameter within the recommended domain in Fig. 7(a) can be reasonable. It can also note from the observed ablated morphology that high feed rate can help to 9

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

Fig. 7. (a) Diagram for the selection of laser ablation parameters for proper wheel texturing, and the ablated diamond abrasive tools with different-shaped textures (including tilt line, parallelogram, hexagonal, triangular, and rectangular structures) by using (b) the parameters selected by the proposed method (the laser power of 54 W and the feed rate of 5.5 mm/s) and (c) the improper parameters (the laser power of 54 W and the feed rate of 2.0 mm/s).

10

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

Acknowledgements

minimize the thermal affected zone where very limited diamond loosening structures can be found and therefore can give the TDATs good machining ability. This observation ends with the conclusion that, the laser parameter at the intersecting point C in Fig. 7(a) can be considered the relatively suitable ablation parameters, i.e. the laser power of 54 W and the feed rate of 5.5 mm/s. Fig. 7(b) presents the ablated diamond grinding wheels having different-shaped textures by using the laser parameters selected by the proposed method. Various textures including tilt line, parallelogram, hexagon, triangle, and rectangle were successfully obtained with the high shape and dimensional accuracy, to a large extent proving the proper selection of the processing parameters based on the proposed strategy. However, few evidences of the poor ablated surface integrities such as microcracks at the edges can still be found, which might be due to the unstable ablation process induced by the motion system vibration and the air atmosphere. Fig. 7(c) presents the comparison with the unsuccessful examples of the wheels ablated by the improper parameters, further indicating the effectiveness of the proposed method. Although there was little variance of the feed rate (changed from previous 5.5 mm/s to the current 2.0 mm/s) while the power kept constant (54 W), the obtained geometries were completely different. Neither clear texture geometries nor the sharp edges can be clearly recognized. The neighboring bond and abrasives appeared to be weak and loosened as well implying the large loss of the grinding ability of the tools.

The authors acknowledge the supports from National Natural Science Foundation of China (No.51805281), Research Project of State Key Laboratory of Mechanical System and Vibration (No. MSV201908), Ningbo 3315 Innovation Team Scheme (No. 2018A-08-C), and University of Nottingham Ningbo China (I01180800099, I01190100001). References Aurich, J.C., Herzenstiel, P., Sudermann, H., Magg, T., 2008. High-performance dry grinding using a grinding wheel with a defined grain pattern. CIRP Annals - Manuf. Technol. 57 (1), 357–362. Aurich, J.C., Kirsch, B., 2013. Improved coolant supply through slotted grinding wheel. CIRP Annals - Manuf. Technol. 62 (1), 363–366. Aurich, J.C., Kirsch, B., Herzenstiel, P., Kugel, P., 2011. Hydraulic design of a grinding wheel with an internal cooling lubricant supply. Prod. Eng. 5 (2), 119–126. Butler-Smith, P.W., Axinte, D.A., Daine, M., 2009. Preferentially oriented diamond microarrays: a laser patterning technique and preliminary evaluation of their cutting forces and wear characteristics. Int. J. Mach. Tool. Manuf. 49 (15), 1175–1184. Butler-Smith, P.W., Axinte, D.A., Daine, M., 2011. Ordered diamond micro-arrays for ultra-precision grinding-An evaluation in Ti-6Al-4V. Int. J. Mach. Tool. Manuf. 51 (1), 54–66. Chattopadhyay, A.K., Chollet, L., Hintermann, H.E., 1991. On performance of brazed bonded monolayer diamond grinding wheel. CIRP Annals - Manuf. Technol. 40 (1), 347–350. Chen, J., Fu, Y., Li, Q., Gao, J., He, Q., 2017. Investigation on induction brazing of revolving heat pipe grinding wheel. Mater. Des. 116, 21–30. Chen, J., Shen, J., Huang, H., Xu, X., 2010. Grinding characteristics in high speed grinding of engineering ceramics with brazed diamond wheels. J. Mater. Process. Technol. 210 (6), 899–906. De Feudis, M., Caricato, A.P., Taurino, A., Ossi, P.M., Castiglioni, C., Brambilla, L., Maruccio, G., Monteduro, A.G., Broitman, E., Chiodini, G., Martino, M., 2017. Diamond graphitization by laser-writing for all-carbon detector applications. Diam. Relat. Mater. 75, 25–33. Denkena, B., Grove, T., Göttsching, T., 2015a. Grinding with patterned grinding wheels. CIRP J. Manuf. Sci. Technol. 8, 12–21. Denkena, B., Grove, T., Göttsching, T., da Silva, E.J., Coelho, R.T., Filleti, R., 2015b. Enhanced grinding performance by means of patterned grinding wheels. Int. J. Adv. Manuf. Technol. 77 (9), 1935–1941. Ding, W.F., Li, H.N., Zhang, L.C., Xu, J.H., Fu, Y.C., Su, H.H., 2017. Diamond wheel dressing: a comprehensive review. J. Manuf. Sci. Eng. - Trans. ASME 139 (12), 121006–121025. Ding, W.F., Xu, J.H., Chen, Z.Z., Miao, Q., Yang, C.Y., 2013. Interface characteristics and fracture behavior of brazed polycrystalline CBN grains using Cu-Sn-Ti alloy. Mater. Sci. Eng.: A 559, 629–634. Dai, C.W., Yu, T.Y., Ding, W.F., Xu, J.H., Yin, Z., Li, H., 2019. Single diamond grain cutting-edges morphology eff ;ect on grinding mechanism of Inconel 718. Precis. Eng. 55, 119–126. Guo, B., Zhao, Q., Fang, X., 2014a. Precision grinding of optical glass with laser microstructured coarse-grained diamond wheels. J. Mater. Process Technol. 214 (5), 1045–1051. Guo, B., Zhao, Q., Yu, X., 2014b. Surface micro-structuring of coarse-grained diamond wheels by nanosecond pulsed laser for improving grinding performance. Int. J. Precis. Eng. Manuf. Technol. 15 (10), 2025–2030. Li, H.N., Yang, Y., Zhao, Y.J., Zhang, Z., Zhu, W., Wang, W., Qi, H., 2019. On the periodicity of fixed-abrasive planetary lapping based on a generic model. J. Manuf. Process. 44, 271–287. Huang, G., Huang, J., Zhang, M., Mu, D., Zhou, G., Xu, X., 2018. Fundamental aspects of ultrasonic assisted induction brazing of diamond onto 1045 steel. J. Mater. Process Technol. 260, 123–136. Jeschke, H.O., Garcia, M.E., 2002. Theoretical description of the ultrafast ablation of diamond and graphite: dependence of thresholds on pulse duration. Appl. Surf. Sci. 197–198, 107–113. Li, H.N., Axinte, D., 2016. Textured grinding wheels: a review. Int. J. Mach. Tool. Manuf. 109, 8–35. Li, H.N., Axinte, D., 2018. On the inverse design of discontinuous abrasive surface to lower friction-induced temperature in grinding: an example of engineered abrasive tools. Int. J. Mach. Tool. Manuf. 132, 50–63. Li, L., Qi, H., Yin, Z., Li, D., Zhu, Z., Tangwarodomnukun, V., Tan, D., 2019. Investigation on the multiphase sink vortex Ekman pumping effects by CFD-DEM coupling method. Powder Technol. https://doi.org/10.1016/j.powtec.2019.06.036. In Press. Lyu, Y., Yu, H., Wang, J., Zhao, G., Liu, Z., 2017. Improved performance of electroplated grinding wheels using a new method of controlled grain size sorting. J. Korean Soc. Manuf. Process. Eng. 30, 336–342. Mihić, S.D., Cioc, S., Marinescu, I.D., Weismiller, M.C., 2013. Detailed study of fluid flow and heat transfer in the abrasive grinding contact using computational fluid dynamics methods. J. Manuf. Sci. Eng. - Trans. ASME 135 (4) 041002-041013. Mohamed, A.M.O., Bauer, R., Warkentin, A., 2013. Application of shallow circumferential grooved wheels to creep-feed grinding. J. Mater. Process Technol. 213 (5), 700–706. Mohamed, A.M.O., Bauer, R., Warkentin, A., 2014. A novel method for grooving and re-

4. Conclusions Unlike most previous studies using pulse laser, this paper utilised the continuous-wave CO2 laser to create passive-grinding structures on the diamond abrasive tool surfaces. The key findings and conclusions of this study might include:

• The employment of the resin ablation threshold did not generate •

• •

satisfying texture structures on the tool surface, implying the laser ablation mechanism of multi-material non-metallic structure was more complex than the ablation of any individual single-material components; The increasing laser power resulted in the increase of all the geometrical parameters of the ablated structures including top/bottom width, depth, and side edge tilt angle. Decreasing feed rate showed similar effects except for the decreasing tilt angle. The interesting critical ranges of both laser power and feed rate were found, where they showed the opposite effects on the ablated topographies due to the competing active and negative ablation processes; There were three basic morphological features on the ablated abrasive tools including (i) diamond exposure/loosening when using low laser power and low feed rate, (ii) diamond breakage when using large laser power and low feed rate, and (iii) the diamond-tographite deterioration no matter what the parameters were; The strategy to select proper laser ablation parameters for texturing diamond abrasive tools was proposed as well based on the observations and knowledge above, and the successful fabrication of the diamond abrasive tools with different-shaped textures proved the method feasibility and reasonability;

With this, the reported findings are anticipated to be not only meaningful to provide the academic reference to the multi-material laser ablation process, but also offer a promising method to produce textured grinding wheel considering the CO2 laser advantages including high machining efficiency, stablised laser beam quality, low cost, and more importantly, good match between the CO2 laser wavelength and the absorbable light spectrum of diamond abrasives and resin bonds.

11

Journal of Materials Processing Tech. 275 (2020) 116279

H.N. Li, et al.

Wang, J.W., Yu, T.Y., Ding, W.F., Fu, Y.C., Bastawros, A.F., 2018. Wear evolution and stress distribution of single CBN superabrasive grain in high-speed grinding. Precis. Eng. 54, 70–80. Yu, H., Lyu, Y., Wang, J., Wang, X., 2018. A biomimetic engineered grinding wheel inspired by phyllotaxis theory. J. Mater. Process Technol. 251, 267–281. Yu, T., Li, H., Wang, W., 2016. Experimental investigation on grinding characteristics of optical glass BK7: with special emphasis on the effects of machining parameters. Int. J. Adv. Manuf. Technol. 82 (5), 1405–1419. Zhao, Y.J., Yan, Y.H., Song, K.C., Li, H.N., 2018. Intelligent assessment of subsurface cracks in optical glass generated in mechanical grinding process. Adv. Eng. Softw. 115, 17–25. Zhou, H., Ding, W.F., Liu, C.J., 2019. Material removal mechanism of PTMCs in highspeed grinding when considering consecutive action of two abrasive grains. Int. J. Adv. Manuf. Technol. 100, 153–165. Zimmer, K., Ehrhardt, M., Lorenz, P., Wang, X., Vass, C., Csizmadia, T., Hopp, B., 2014. Reducing the incubation effects for rear side laser etching of fused silica. Appl. Surf. Sci. 302, 42–45.

grooving aluminum oxide grinding wheels. Int. J. Adv. Manuf. Technol. 73 (5), 715–725. Nguyen, T., Zhang, L.C., 2009. Performance of a new segmented grinding wheel system. Int. J. Mach. Tool. Manuf. 49 (3), 291–296. Oliveira, J.F.G., Bottene, A.C., França, T.V., 2010. A novel dressing technique for texturing of ground surfaces. CIRP Annals - Manuf. Technol. 59 (1), 361–364. Shukla, A.K., Akash, K., Palani, I.A., Manivannan, A., 2017. Laser assisted wet texturing of flexible polyethylene terephthalate substrate using Nd3+:YAG laser for photovoltaics devices. Mater. Sci. Eng.: B 226, 78–85. Silva, E.Jd., Oliveira, J.F.Gd., Salles, B.B., Cardoso, R.S., Reis, V.R.A., 2013. Strategies for production of parts textured by grinding using patterned wheels. CIRP Annals Manuf. Technol. 62 (1), 355–358. Upadhyaya, R.P., Fiecoat, J.H., 2007. Factors affecting grinding performance with electroplated CBN wheels. CIRP Annals - Manuf. Technol. 56 (1), 339–342. Walter, C., Komischke, T., Kuster, F., Wegener, K., 2014. Laser-structured grinding tools – generation of prototype patterns and performance evaluation. J. Mater. Process. Technol. 214 (4), 951–961.

12