Experimental Investigation on the Milling Performance of a 50 μm D-Shaped Micromilling Cutter with Different Materials

In micromilling, the performance and diameter of the milling cutter directly determine the service life of the milling cutter and the surface quality of the microgroove. Therefore, it is vital for high-precision milling to explore the milling performance of different materials and expand the application scope of micromilling cutter sizes. In this study, the milling performance of three kinds of material cutters—carbide, diamond coating and polycrystalline diamond (PCD)—was evaluated. A series of micromilling tests were carried out to determine the effects of cutter material type on cutter wear, surface quality and burr formation, particularly when a 50 µm micromilling cutter was used in the milling process. A D-shaped milling cutter with a diameter of 50 µm was manufactured on a self-developed high-precision modular machine tool by wire electrode electric discharge grinding (WEDG) technology. From theoretical and experimental perspectives, it is easy to master microgroove quality milled by different material cutters. The results show that the microgrooves processed with PCD cutters have fewer burrs, lower surface roughness values, and a smoother groove bottom morphology.


Introduction
As a kind of micromachining technology, micromilling has attracted increasing attention because of its high processing accuracy, high processing e ciency, strong three-dimensional processing ability, and wide range of workpiece materials [ 1 -2 ]. In recent years, with the continuous progress of science and technology, there has been a higher precision demand for micro-small parts in aerospace, machinery, biomedicine and other industries. Therefore, Zhang et al. [ 3 ] producted that current micromilling technology has been required to become more e cient, economical, and precise.
As the size of the micromilling cutters decreases, the gap between the micromilling process and the traditional milling process has increased [ 4 -5 ]. An et al. [ 6 ] stated the micromilling process usually encounters problems such as poor surface quality, severe surface burrs, rapid cutter wear, and easy fracture. SalmanMumtaz et al. [ 7 ] reported that coated cutters are widely used in the eld of micromilling because of their high wear resistance, high-temperature oxidation resistance, good cutting performance and other advantages. Considering the in uence coating type on cutter performance, Aslantas et al. [ 8 ] researched the effects of nanodiamond (NCD)-coated, TiN-coated, AlCrN-coated and uncoated carbide cutters on cutter wear and rough edge size in Ti6Al4 V alloy micromilling. The result indicated that maximum wear and burr size occurred on uncoated cutter. Their research expanded the application types of cutter materials. Liang et al. [ 9 ] investigated the cutting performance of Ti-6AI-V machined by a micromilling cutter with different coating materials. In contrast to an uncoated micromilling cutter, an AITIN-based coating can result in a reduction in cutting edge chipping and cutter wear length, which gives rise to a longer service life. Moreover, the surface roughness of machined Ti-6Al-4 V is the lowest. In order to verify the failure mechanism of coated carbide cutters in micro-milling experiments, Lu et al. [ 10 ] proposed a method, in which wear and breakage characteristics of cutter are considered. Their results showed that the main failure modes of the cutter were tip fracture, and the coating on the front angle and side fell off. Thepsonthi et al. [ 11 ] noted that polycrystalline diamond (PCD) materials can maintain high hardness at high temperature. Their research indicated that PCD cutters and PCBN cutters have less wear and better cutting performance than carbide cutters at higher milling speeds. Kaitao et al. [ 12 ] developed a Ti(C7N3) ceramic microend milling cutter and discussed its failure mechanism. micro-crack initiation and crack propagation cause cutter failure. Furthermore, they obsered that the tip fracture of Ti(C7N3)-based cermet micro end-mill cutter is lower compaired to WC micro end-mill cutter (176 µm). Ding et al. [ 13 ] asserted that a PCD cutter also showed better application performance than polycrystalline cubic boron nitride cutters due to higher abrasion wear and fracture resistance.
In addition to materials and coatings, micromilling performance also depends on cutter structure design and preparation method. Perveen et al. [ 14 ] carries out experiments for studying difference on the microgrinding performance of four different geometry cutters (circular, D-shaped, triangular, and square), and their experimental result indicated that D-shaped cutter was considered to provide better performance in terms of the achieved surface nish, cutter wear and cutting force, which is conducsive to improve the precision in ultra precision manufacturing. Jan et al. [ 15 ] reported the design and manufacture of singleedge micro end-mills with diameters between 10 and 50 mm and a variable helix angle. Qing et al. [ 16 ] accounting for the effects of micro-texture conducted a comprehensive evaluation to assess cutting performance based on surface roughness, surface hardening, chip formation, and cutter wear. Their experimental result showed that the micro-texture of the cutter surface has an anti-wear effect, which is conducive to obtain superior workpiece surfaces and improve actual processing performance. Peiyuan et al. [ 17 ] proposed new the geometry of micro end mill and carried out experiments on brittle tool steel. The results show that the cutter maximum stress is signi cantly reduced, which is conducive to improve the geometric accuracy and resisted the formation of burrs. Fleischer et al.
[ 18 ] developed a single edge micro milling cutter on the basis of statics. They found that the burr width and surface roughness could be effectively controlled at a milling cutter screw Angle of 30°. Uhlmann et al. [ 19 ] have found that, compared with traditional spiral milling cutters, D-cutters with simple shapes have higher exural rigidity and strength in the micromilling process with decreasing machining size. Lu et al. [ 20 ] observed an increase in cutting temperature due to increase in cutter nose corner radius and main cutting edge radius resulting appearance of residual stress on workpiece and accelerated tool wear. Chern et al. [ 21 ] prepared 31μm diameter carbide micro-milling cutter by using wire electrode discharge grinding and applied it to the aluminum alloy AI6061-T6 milling experiment. Guo et al. [ 22 ] produced a non-coaxial spiral side edge microdrill with the best cross edge using a six-axis CNC grinder. Suzuki et al. [ 23 ] combined a focused ion beam with ultra-precision CNC milling to produce a 25μm diameter carbide micromilling cutter for PMMA process.
At present, Chen et al. [ 24 ] found that research on micromilling technology experiments has mainly focused on spiral milling cutters with diameters of 100 μm and above. To further expand the application range of milling cutter sizes, the present study is the rst to apply a 50 micron micromilling cutter for experiments to compare the milling performance of carbides, diamond coatings and PCD cutters. A carbide cutter and PCD cutter were produced on a self-developed μEM-200CDS2 combined with a highprecision machining machine cutter. On this basis, diamond-coated cutters were prepared by vacuum coating. To master the milling performance of carbide cutters, diamond-coated cutters and PCD cutters, a series of milling experiments on pure copper workpieces were carried out. The cutter wear and bottom morphology of the microgroove for the carbide cutter, diamond-coated cutter and PCD cutter were captured and analysed. Subsequently, the width variation, depth variation, surface roughness and pro le of the microgroove during milling were measured and analysed to obtain a cutter with better milling performance in the eld of micromilling.

Cutter preparation
A D-shaped carbide cutter with a diameter of 50 μm was prepared by wire electrode electrical discharge grinding (WEDG). As shown in Fig. 1a and 1b, a micromilling cutter was fabricated on a machine cutter, and on-machine micromilling processing was performed. During the test, the cutter blank was clamped on a high-speed spindle, and the milling cutter was prepared by WEDG technology. The milling test was then carried out, the z-axis minimum linear feed was 0.1 μm, and the spindle radial run out accuracy was 1 μm. A schematic diagram of the cutter is shown in Fig. 1d; the main features are as follows: the handle D1 is 3 mm in diameter, the bit D is 0.05 mm in diameter, the cutting edge length L1 is 0.15 mm, the cutter length L is 30 mm, the half cone angle of the cutter neck is 15°, the dorsal horn of the bottom cutting edge is 20°, the cutting edge inclination is 20°, and the thickness of the coating is 1μm. The diameters of the carbide cutter and the diamond-coated cutter are 50 μm and 56 μm, respectively. The diameter of the PCD cutter is 45 μm 2.2 Milling scheme Ultra ne grain carbide (grain size of 0.4 μm) and polycrystalline diamond are used as the materials for the micromilling cutter in the test because of their high hardness, wear resistance and bending stiffness. Table 1 shows the main parameters of the carbide, diamond coating and PCD materials.
A schematic diagram of the micromilling process is shown in Fig. 1c. Three groups of different cutters were tested with the same milling parameters, and ve straight grooves were milled in each group. The length of each microgroove was 3 mm, and a milling length of 15 mm was nally achieved in each group of experiments. Table 2 shows the milling parameters in the experiment. The milling depth of each feed was 0.002 mm, and the feed was 5 times. The total milling depth of each groove was 0.01 mm. The PCD material cutter was fed 10 times, and the total milling depth was 0.02 mm because the hardness of PCD was higher than that of the carbide cutter and diamond-coated cutter. The machined surface was cleaned by an ultrasonic machine. The surface roughness is measured along the feed direction by using a white light interferometer. In addition, the surface morphology of the microgroove bottom was observed by scanning electron microscopy.

Five grooves
The surface burr of the microgroove milled by a diamond-coated cutter is shown in Fig. 2. The experiments show that the groove bottom varies from smooth to gradually arise from the burr and arc with increasing milling distance. As shown in Fig. 2a, the rst groove bottom is clean and tidy. This is because the milling cutter is sharp enough at the initial stage of milling. However, there are two discernible boundaries in the second groove shown in Fig. 2b. The reason behind this phenomenon is as follows. With the partial coating falling off, the milling cutter tip begins to wear as a result of a sharp point being produced. In contrast to the second groove, many small burrs exist in the middle of the third groove bottom, while the ank side of the groove is relatively at, as shown in Fig. 2c. The reason is the wear loss of the cutter, as the material cannot be cut effectively, nally leading to chips attached to the sidewall. In addition, it can be seen clearly in Fig. 2(d-e) that the fourth and fth groove bottoms exist in the shape of a circular arc without too much of a burr when the cutter head is basically ground at and assumes a circular arc, as seen in Fig. 4b.   Fig. 3 illustrates the difference in morphology of the fth groove bottom milled by a carbide cutter, diamond-coated cutter and PCD cutter. As shown in Fig. 3a, more burrs were generated at the sidewall of the microgroove milled by the carbide cutter when the uncut material broke at the milling terminal.
Constant cutter wear is a signi cant contributory factor to the formation of burrs. Fig. 3b shows the microgrooves milled by a diamond-coated cutter. The burr is relatively small on the whole. There is no rotation path at the groove bottom. Fig. 3c shows the microgroove milled by the PCD cutter. The burr of the microgroove sidewall is low, and the morphology of the microgroove bottom is tidier. Fig. 4 shows the wear degree for the carbide cutter, diamond-coated cutter and PCD cutter. As shown in Fig. 4a, the tip of the carbide cutter was seriously worn after machining. The top of the contact surface between the cutter and the workpiece is worn out of a wedge shape and spreads around. Although the hardness value of the diamond-coated material is higher than that of the carbide material, the wear area of the diamond-coated cutter, as seen in Fig. 4b, is larger than that of the carbide cutters. The possible explanation for this phenomenon is as follows: rst, the matrix material is exposed to the outside when the coating falls off. The increase in wear rate is mainly attributed to scratching between the matrix material and coating falling off at the groove bottom. Second, the chemical elements (Co, Fe, WC, etc.) of the diamond-coated cutter react with the chemical elements (Cr, Ni, etc.) in the workpiece. In addition, these elements (Co, Cr, Ni, etc.) diffuse into the matrix structure of the diamond-coated cutter under the condition of constant friction between the diamond-coated cutter tip and workpiece. In ltration of new elements reduces the matrix material strength and aggravates diamond-coated cutter wear during the milling process (Huang et al. 25 ). As shown in Fig. 4c, wear hardly occurs at the cutter tip, and fragments of pure copper still adhere to the surface of the cutting edges. The cutter tip shape can be well maintained within a milling distance of 15 mm. Experiments have indicated that PCD cutters have better wear resistance than carbide cutters and diamond-coated cutter.  Fig. 5(a-c). There is an obvious cutter rotation trace pattern at the microgroove bottom, and the residual appears intermittently along the circular arc feed direction.

Microgroove surface morphology 3.2.1 Three-dimensional morphology
Cutter path lines become increasingly obvious with increasing milling distance. This phenomenon is due to the increase in tip radius caused by cutter wear. The morphology of the groove bottom milled by a diamond-coated cutter is presented in Fig. 5(d-f). The rst microgroove has a relatively clear bottom surface, as shown in Fig. 5(d). With the loss of coating and the gradual wear of the cutter, local uncut chips remain on the sidewall of the microgroove, resulting in two traces at the bottom of the groove, as shown in Fig. 5(e). More chips exist in the groove bottom along the cutter feed direction, as shown in Fig.  5(f). These residual chips reduce the surface quality of the microgroove. The trace lines, as shown in Fig.  5g-i, are concentrated in the cutting direction of cutter rotation. The trace lines are uniformly distributed and of the same size, without obvious changes with increasing milling distance. The PCD cutter exhibits superior performance in terms of better microgroove surface morphology and relatively straight sidewalls compared with carbide cutters and diamond-coated cutters. Fig. 6a shows the variation in microgroove width depending on milling distance. When the milling distance was 2 mm, the microgroove width of the carbide cutter was 51.2 μm, the microgroove width of the diamond-coated cutter was 56.8 μm and the microgroove width of the PCD cutter was 43.25 μm.

groove depth and width
Within a certain range, the microgroove width of the carbide cutter and diamond-coated cutter has a positive relationship with the milling distance, and the maximum value of the carbide cutter is 58.8 μm, while the other is 53.7 μm. However, the width of the microgroove milled by the PCD cutter is stably maintained at 44.08 μm. The variation in the width of the microgroove milled by the PCD cutter is more stable than that of the carbide cutter and diamond-coated cutter. Fig. 6b reveals the correlation between the microgroove depth and milling distance. There is a steady decline in the depth of microgrooves milled by carbide cutters, diamond cutters and PCD cutters. When the milling distance is 14 mm, the minimum depth values of the microgroove of the three cutters are 5.8 μm, 5.6 μm and 16.6 μm. The reason for the overall decrease in microgroove depth is that wear of the cutter and passivation of sharpness lead to ineffective cutting, which indicates that the PCD cutter has better wear resistance than the carbide cutter and diamond-coated cutter.
The variations in the pro le during the milling process are described in Fig. 7. The grooved bottom milled carbide cutter has a circular arc shape, and the sidewall angle is small. The inclination angle of the sidewall has a positive correlation with the depth of the microgroove, as shown in Fig. 7a. Fig. 7b shows that the groove bottom milled by a diamond-coated cutter has a relatively at surface and that the microgroove sidewall is relatively vertical. In addition, the microgroove bottom gradually becomes rounded, and the sidewalls gradually tilt. This phenomenon is mainly due to the removal of the coating and deterioration of the cutter tip shape. As shown in Fig. 7c, a at groove bottom occurs on the microgroove bottom milled by the PCD cutter during milling. The sidewall inclination angle of the microgroove varies slightly, and the sidewalls are almost perpendicular to the groove bottom. It can be seen from the comparison results that microgrooves milled by the PCD cutter perform better in maintaining the groove pro le. Fig. 8 shows the variation in surface roughness with milling distance. The roughness value of the grooved bottom milled by the carbide cutter is twice that of the diamond-coated cutter at the beginning, and the surface roughness increases with increasing milling distance. Although the roughness value obtained with the diamond-coated cutter is lowest within a milling distance of 12 mm, the roughness value dramatically increases when the milling distance exceeds 12 mm. Fig. 4b and Fig. 8 show that the wear degree of the diamond-coated cutter corresponds to the roughness of the groove bottom. This is due to peeling of the coating and deterioration of the cutter tip shape, which will generate residual chips and cause poor surface quality. However, the roughness of the groove bottom milled by PCD remains almost constant within a 15 mm milling distance. This indicates that a better surface roughness value is obtained with the PCD cutter and that the PCD cutter has a longer cutter life than the carbide cutter and diamond-coated cutter.

Conclusion
This paper presents an experimental study on the cutting performance of carbide cutters, diamondcoated cutters and PCD cutters in the micromilling of pure copper. An evaluation of the cutter wear, surface morphology and surface burr of microgrooves is conducted to determine the milling cutter material with the best milling performance. The following main conclusions are drawn: 1. Microgrooves milled by PCD cutters tend to have superior performance in terms of milling stability compared to carbide cutters and diamond-coated cutters. The roughness value obtained with the PCD cutter remains almost constant. The milling distance and cutter wear greatly in uence the roughness value of the carbide cutter and diamond-coated cutter.
2. The PCD cutter has a minimum wear rate and longer service life than the carbide cutter and diamond-coated cutter, which is attributed to the good wear resistance of the PCD cutter. In addition, microgroove milling by a PCD cutter has better groove retention ability.

3.
A superior microgroove is obtained by the PCD cutter. The maximal burr exists in the microgroove sidewall of the carbide cutter, and the most obvious pattern of the cutter path is generated on the groove bottom milled by the carbide cutter.
4. The preparation of 50 micron milling cutter has important application value for expanding the range of machining dimensions and machining high precision parts with small geometric features in the eld of ultra precision micro milling.

Declarations
Funding: This work was supported by National Natural Science Foundation of China (51475310).
Competing Interests: The authors declare that they have no competing nancial interests or personal relationships that could have appeared to in uence the work reported in this paper.