Investigation of burr formation and surface integrity in micro-milling of aluminum alloy LF21 slot

In this article, aluminum alloy LF21 is used as the simulation object. The chemical composition of aluminum alloy LF21 is shown in Table 1 and the material parameters are shown in Table 2.

Abaqus, a 3D FE simulation software, was used to simulate the micro-milling of aluminum alloy LF21 slot top burrs and exit burrs. Since this investigation does not include tool wear, the tool is set as a rigid body in the simulation. The accuracy of burr formation in FE simulation is closely related to the material’s constitutive model, and the J-C [1] material model was chosen for the simulation using expression (1). (1) σ = ( A + B ε n ) 1 + C ln ε ̇ ε 0 1 − T − T room T melt 1 − T room m , \sigma =(A+B{{\varepsilon }}^{n})\left(1+C\hspace{.25em}\mathrm{ln}\hspace{.25em}\frac{\dot{{\varepsilon }}}{{{\varepsilon }}_{0}}\right)\left[1-{\left(\frac{T-{T}_{\text{room}}}{{T}_{\text{melt}1}-{T}_{\text{room}}}\right)}^{m}\right], where σ \sigma is the flow stress of the workpiece material; ε ̇ \dot{\varepsilon } is the equivalent plastic strain rate; ε \varepsilon is the equivalent plastic strain; ε 0 {\varepsilon }_{0} is the strain rate reference value, generally taken as 0.01 s⁻¹; n is the strain hardening index; m is the heating softening index; and T _melt is the material melting temperature; T _room is the room temperature; T is the workpiece temperature; A is the stress coefficient; B is the strain intensification coefficient; and C is the strain coefficient. The parameters of the J-C eigenmodel are shown in Table 3.

In this article, a 3D model is used for the simulation of the micro-milling process, where chip formation is more complex than 2D orthogonal cutting, thus making it difficult to accurately represent the bonded area on the front face of the cutter in the model. Referring to the related literature, the contact between the chip and the front tool face as well as the workpiece surface and the rear tool face in the model was set as sliding friction, while the contact between the tool and the workpiece was defined as Coulomb friction with a friction coefficient of 0.2 [1].

Using Solid Works software, the models of the workpiece and a double-edged milling cutter with a diameter of 0.8 mm were created. To save time and improve efficiency, and to ensure that the tool and workpiece met the accuracy requirements, relevant simplifications were made. Then, they were imported into Abaqus software, as shown in Figure 1.

Figure 1

Simulation model.

Micro-milling experiments were carried out on a Beijing Jingdiao Carver 400GA vertical machining center with a spindle speed of up to 30,000 rpm. A monolithic carbide tool with a diameter of 800 μm, a helix angle of 30°, and an edge radius of 1.04 μm was used to machine the aluminum alloy LF21 workpiece. The workpiece was cut into small pieces with dimensions of 30 mm × 15 mm × 15 mm using wire electrical discharge machining and then mounted on the machine table as shown in Figure 2. Before the test, the surface of the workpiece was pre-machined with an end mill with a diameter of 10 mm to ensure flatness, and then the straight groove was milled. The micro-milling parameters are shown in Table 4, and the spindle speed n varies from 6,000 to 18,000 rpm. The milling depth a _p varied between 30 and 70 μm, and the feed rate v _f varied between 0.005 and 0.045 m/min, converted to feed per tooth, which included cases greater than and less than the radius of the cutting edge of the tool. The conversion formula for feed rate v _f and feed per tooth f _z in Table 2 is as follows: (2) v f = n ⋅ f z ⋅ z 60 , {v}_{\text{f}}=\frac{n\cdot {f}_{z}\cdot z}{60}, where n is the spindle speed (rpm), f _z is the feed per tooth (μm/z), and z is the number of tool teeth.

Figure 2

(a) Micro-milling experimental platform diagram, (b) micro-milling cutter (bottom view), and (c) schematic diagram of the micro-milling slot.

The surface morphology and contours of the workpieces were examined using a scanning electron microscope (SEM, HITACHISU-70) and a 3D laser scanning confocal microscope (LSCM, VK-X200, KEYENCE). The 3D LSCM provides detailed measurements of the groove bottom surface roughness. A representative measurement of the slot bottom surface is displayed in Figure 3(a). Five random areas at the bottom of the microgroove were chosen for surface roughness analysis, and the average value from these areas was taken as the final result.

Figure 3

(a) Illustration of the roughness measurement by 3D LSCM. (b) Schematic of burr size on both up-milling and down-milling sides. (c) Measurement of burr size in SEM image.

The burr width w and burr length l were measured five times in different areas of the top of the microgroove under stable milling condition, and the burr size L was calculated by using formula (3), and finally the average value of the burr size L was taken as the measured value. Determine and consider the burr size for the up-milling side (L _u) and the down-milling side (L _d) as shown in Figure 3(b). The exit burr is also measured for the burr width w and burr length l, and the burr size L is calculated using equation (3). The width and length of the burr on the up-milling side are denoted by (w _u) and (l _u), respectively, and the burr size is L _u. The width and length of the burr on the down-milling side are denoted by (w _d) and (l _d), respectively, and the burr size is L _d, as shown in Figure 3(c). The burr and roughness measurements are shown in Table 5. (3) L = w 2 + l 2 . L=\sqrt{{w}^{2}+{l}^{2}}.

Based on the measured data in Table 5, a grade classification bar chart as shown in Figure 4 was drawn to more intuitively represent the trends of burr size and surface roughness in each experimental group. In the graph, the data are classified into large, moderate, and small grades using different color scales, which helps in analyzing the relationship between parameter combinations and machining quality. The classification criteria are as follows: when the burr size is less than 500 μm or the R _a value is less than 0.6 μm, it is defined as “small”; when the burr size is between 500 and 600 μm or the R _a value is between 0.6 and 0.8 μm, it is defined as “moderate”; when the burr size is greater than 600 μm or the R _a value is greater than 0.8 μm, it is defined as “large.” The different quality levels can be clearly identified from the natural color bar graphs. For example, Groups 6, 7, and 13 have a high level of exit burr and roughness (brick orange), while Groups 5 and 10 have the smallest burr size and R _a value (dark lime green) and show better surface integrity. This significantly enhances the visualization of the trends of the numerical results in Table 5.

Figure 4

Burr size and surface roughness distribution with level classification using natural tones across different experimental groups.

A comparison of the experimental and simulation results for the top burr and exit burr is shown in Figure 5, where it can be seen that the top burr can be divided into two parts, the root is formed by the plastic side stream of the material, and the top is formed by the unseparated chips, with a certain amount of curling and scattering. The top burr shape was uniform on the up-milling side, while it varied significantly on the down-milling side.

Figure 5

Top burr and exit burr formation in FE simulation and experiment observations by SEM.

The exit burr has an incomplete shape along the exit surface, similar to a cantilever beam, and it suffers the greatest bending moment at the root. As the tool gradually cuts from the exit surface in several cutting strokes, the exit burr formed in the previous cutting stroke will be severely squeezed by the bottom surface of the tool in the following cutting strokes. This weakest position is prone to micro-cracking due to severe squeezing during subsequent cutting. The cracks will then gradually expand, leading to localized tearing of the exit burr [23].

In order to effectively compare and analyze the FE simulation results with the experimental observation results, different cutting parameters were set for the simulation of slot top burr and exit burr. The simulation parameters for the slot top burr are set as n = 6,000 rpm, v _f = 0.025 m/min, a _p = 0.05 mm; and the simulation parameters for the exit burr are set as n = 12,000 rpm, v _f = 0.005 m/min, and a _p = 0.05 mm, which are able to comprehensively reflect the formation of the top burr and the exit burr in the process of micro-milling, thus effectively comparing the quantitative and qualitative results with the experimental measurements. Thus, we can effectively compare and analyze the quantitative and qualitative results with the experimental measurements, and verify the accuracy of the simulation model.

During the micro-milling process, there are four different deformation zones, as shown in Figure 6(a). The first deformation zone is the main shear zone (P_I), where the material undergoes plastic deformation and shear slip. The second deformation zone (P_II) is the metal fibrillation zone formed by the extrusion and friction of the chip by the front surface of the tool. The third zone (P_III) is the deformation and rebound of the machined surface induced by the lateral compression and friction of the tool, resulting in fibrillation and work hardening of the surface. The fourth deformation zone (P_IV) is the negative shear zone resulting from the change in the flow path of the material, where exit burrs are usually formed [1,23].

Usually burrs during micro-milling can be categorized into two types, i.e., top burrs and exit burrs. The formation and mechanisms behind these two types of burrs differ significantly. In up-milling, the tool’s cutting direction opposes the workpiece’s feed direction. At the beginning, the uncut thickness is smaller than the material’s minimum cutting thickness. As a result, only plowing, squeezing, and elastic recovery occur between the tool and the material, with no actual cutting effect taking place. This process is illustrated in Figure 6(b1) and (b2). With the rotation of the tool to feed the movement, the instantaneous uncut thickness gradually exceeds the minimum cutting thickness, the material undergoes plastic deformation, and enters the shear phase, where most of the material is torn and small chips are formed under the action of large tensile stress. However, the incomplete cutting resulted in part of the material near the end of the workpiece failing to leave with the chips and instead attaching to the top of the micro-groove as a preliminary top burr [30], as shown in Figure 6(b3). As the tool rotates, these initial burrs are gradually pushed to the outside by the tool edge under the action of plowing and squeezing, which eventually become the top burrs on the up-milling side, as shown in Figure 6(b4).

Next on the down-milling side, the cutting direction of the tool is in the same direction as the feed direction of the workpiece, and the material first experiences compression and protrudes to the down-milling side, as shown in Figure 6(c1). In the next process, the thickness of the uncut chips gradually decreases and the material surface undergoes a large plastic deformation, as shown in Figure 6(c2). The deformed material is significantly pushed outward by the cutting edge until permanent plastic deformation occurs and the chips tear away from the workpiece substrate, as shown in Figure 6(c3). As the thickness of the uncut chip is further reduced, the cutting of the material is no longer evident, and there is a lack of sufficient interaction between the tool and the chip, some of the material remains attached to the workpiece surface, where it continues to be squeezed and rubbed, eventually forming a top burr on the down-milling side, as shown in Figure 6(c4).

When the tool reaches the exit region of the workpiece, the shape characteristics of the region change significantly, as shown in Figure 6(d1) and (d2). At this time, the support stiffness in the cutting direction is significantly reduced, the resistance is reduced, resulting in the plastic flow of the material being no longer effectively controlled, and the surface at the exit of the workpiece becomes a free surface without resistance. At this point, the micro-milling process proceeds to the fourth negative shear deformation zone (P_IV), where the plastic flow path of the material changes, no longer flowing in the traditional upward direction, but rotating downward around the intersection of the cutting region, as shown in Figure 6(d3). This results in part of the material at the exit of the workpiece not being effectively cut and remaining on the exit surface as an exit burr, as shown in Figure 6(d4).

Under the cutting parameters corresponding to this figure, the geometrical size information of the top burr and exit burr was extracted from the simulation results and the burr dimensions were calculated based on equation (3). The results of the comparative analysis between the simulation dimensions and the experimental results are shown in Table 6. For the top burr, the simulation parameters are set to n = 6,000 rpm, v _f = 0.025 m/min, and a _p = 0.05 mm, which corresponds to group 1 in Table 5 of the experimental data. The simulation extracted burr sizes of 458.58 μm on the up-milling side and 517.01 μm on the down-milling side, and the experimental values were 493.53 and 564.59 μm, with errors of 7.1 and 8.4%, respectively, which indicate that the simulation model has good accuracy and trend consistency in the prediction of the top burr. For the exit burr, the simulation parameters are set as n = 12,000 rpm, v _f = 0.005 m/min, a _p = 0.05 mm, and the corresponding experimental data are the sixth group in Table 5. The simulated size of the exit burr is 802.68 μm and the experimental value is 750.97 μm with an error of 6.9%. This result indicates that the model also has a good match in exit burr prediction. The data results show that the simulation model can predict the burr size within 10% under different typical machining parameters, which is highly reliable and can provide an effective numerical basis for the optimization of subsequent machining parameters.

From Figure 7(a), it can be seen that the overall trend of burr size decreases with the increase in spindle speed. Figure 7(b) demonstrates the microscopic images of the top burr of the LF21 microgroove at different spindle speeds. It can be found that the top burr size on the down-milling side is generally larger than that on the up-milling side, which may be related to the difference in the burr formation process between the two sides as mentioned in Section 3.1.1. In up-milling, the material undergoes plowing, extrusion, and shearing, and finally the material tears under high tensile stress to form chips. On the contrary, in the down-milling process, the material is first compressed and raised, and then flows outward under the force of the tool, and part of the material is extruded and rubbed, finally forming extrusion burrs at the top. Therefore, the burr on the down-milling side will be larger [13]. From Figure 7(b1) and (b2), it can be observed that when the spindle speed is 6,000–9,000 rpm, the burr sizes are larger and irregular on both the up-milling side and the down-milling side. At lower rotational speeds, the longer contact time between the tool and the workpiece results in an increase in cutting force. However, this also leads to a decrease in cutting efficiency, causing the material to adhere more strongly to the tool. As a result, the cutting-edge radius expands, intensifying extrusion and friction effects. Moreover, due to the alignment of the tool’s milling direction with the feed direction on the down-milling side, these extrusion and friction forces become even more pronounced. Consequently, the burrs formed on this side tend to be significantly larger compared to those on the up-milling side. As can be seen from Figure 7(b3–b5), with the increase in spindle speed, the burr size decreases significantly, the morphology is more uniform, and the burr on the down-milling side is more obvious. At elevated rotational speeds, the reduced contact time between the tool and the workpiece leads to a decrease in the friction and squeezing forces acting between them [31]. However, when the rotational speed is too high, e.g., 15,000 rpm, the machine tool and the tool may experience vibration phenomenon, which leads to the instability of the cutting process, which causes the burr size to go back up, and the vibration will also lead to the abnormal chip shedding and the formation of irregular burrs on the sides, as shown in Figure 7(b4). Further analysis of the mechanism of this trend shows that the spindle speed plays an important regulatory role in the shear deformation process of the material. When the spindle speed is increased, the cutting speed is also increased, the amount of material removed per unit time is increased, and the plastic flow time of the material in the tool entrance area is shortened, thus suppressing the expansion of the top burr. In addition, as the spindle increases, the cutting temperature increases, which in turn softens the material and improves the cutting stability, thus slowing down the accumulation of burrs [35].

Figure 7

(a) Top burr size measurement results; (b1)–(b5) SEM images of top burr formation of the LF21 slot at different spindle speeds.

From Figure 8(a), it can be seen that the size of the exit burr decreases gradually with the increase in the spindle speed. The increased material removal efficiency at higher rotational speeds leads to a shorter duration for the material’s plastic deformation within the negative shear zone region [23]. Additionally, the cutting force and friction between the tool and workpiece are diminished at these higher speeds, further contributing to the reduction in burr size. Figure 8(b) demonstrates the microscopic images of the LF21 slot exit burr at different spindle speeds. And it can be seen from Figure 8(b5) that when the spindle speed is 18,000 rpm, the exit burr appears to be broken and not completely separated. The reason could be that, although the cutting force diminishes at high speeds, the friction between the tool and workpiece at these elevated speeds still produces a notable amount of cutting heat. This heat softens the surface material locally, causing the chips in this area to fail in fully detaching during the dislodgement process. Instead, they break under tensile stress, leading to incomplete separation. To further explain the behavior of the exit burr with rotational speed, it can also be analyzed from the point of view of the thermo-mechanical response of the material. When the spindle speed is high, the tool-workpiece contact time is shortened but the temperature rises faster, which may cause local material softening, while the vibration in the cutting region may be exacerbated due to centrifugal force, resulting in enhanced material mobility at the exit boundary. At this time, if the tool wear is severe or the cooling is insufficient, it may also inversely contribute to the increase in the exit burr size [1]. Overall, increasing the spindle speed helps to reduce the size of the burr and improve the surface machining quality. Therefore, an appropriate increase in spindle speed can effectively inhibit the generation of burrs in the micro-milling process.

Figure 8

(a) Exit burr size measurement results; (b1)–(b5) SEM images of exit burr formation of LF21 slot at different spindle speeds.

From Figure 9(a), it can be seen that on the down-milling side, the size of the top burr shows a tendency of decreasing and then increasing with the increase in the feed rate. Figure 9(b) demonstrates the microscopic images of the top burr of LF21 slot at different feed rates. When the feed rate is 0.005 m/min (0.208 μm/z feed per tooth), the top burr on the up-milling side is 403.19 μm, and the burr size on the down-milling side is 480.27 μm, with larger burr sizes on both sides. This is because the lower feed rate will make the contact time between the tool and the workpiece longer, the negative front angle effect is more obvious, and the cutting process is dominated by plowing and extrusion [25,32], especially in the down-milling process, which will produce a larger cutting force and friction, so that the resulting burrs will be more obvious and present elongated morphology, and there are obvious extrusion traces on the surface. Moreover, the lower feed rate also leads to poorer plastic flow of the material, resulting in some material being left on the side of the slot bottom, which is more likely to form burrs on the slot bottom of the counter milling side due to the higher tensile force applied to the material on that side [15], as shown in Figure 9(b1). When the feed rate was increased from 0.005 m/min (0.208 μm/z) to 0.015 m/min (0.625 μm/z), the burr size was reduced and sawtooth shaped and wave-shaped burrs were produced on the up-milling side and down-milling side, respectively, due to the increase in the tool’s per-tooth feed, which resulted in the weakening of the negative rake angle effect, the reduction in friction and extrusion effects, and the decrease in the cutting force, as shown in Figure 9(b2). When the feed rate is increased from 0.015 m/min (0.625 μm/z) to 0.025 m/min (1.04 μm/z), the burr size is minimized. It may be due to the fact that the feed of 1.04 μm/z is close to the radius of the tool edge, the shear effect is dominant, and the friction and extrusion effects, as well as the plastic deformation of the material, are weakened, so that the burr size decreases, and the morphology is more homogeneous and regular. This phenomenon is similar to the relationship between feed per tooth and tool edge radius in the literature [16]. And when the feed rate continues to increase to 0.045 m/min (1.875 μm/z), the friction and deformation between the tool and the material are again enhanced due to the increased cutting thickness, resulting in a rebound of the burr size and the morphology shows irregular tearing or attachment phenomenon, as shown in Figure 9(b5). The mechanism behind the phenomenon mainly stems from the influence of feed rate on cutting stress distribution and material accumulation behavior. When the feed rate is moderate, the cutting process tends to be more stable, the material removal is more adequate, and the top burr is reduced due to the shrinkage of the plastic buildup area; however, if the feed rate is too high, the tool force increases abruptly, which in turn makes the burr easy to tear and form a large residual edge [36].

Figure 9

(a) Top burr size measurement results; (b1)–(b5) SEM images of top burr formation of LF21 slot at different feed rates.

From Figure 10(a), it can be seen that the exit burr size also shows a trend of decreasing and then increasing with the increase in feed rate. At low feed speeds of 0.005 m/min (0.208 μm/z) and 0.015 m/min (0.625 μm/z), due to the small feed per tooth, the effective rake angle is negative, and the cutting process is dominated by plowing and squeezing, which generates large friction and cutting forces, resulting in severe deformation of the burr surface and larger and irregular burrs, as shown in Figure 10(b1) and (b2). When the feed rate is increased to 0.025 m/min (1.04 μm/z), the feed per tooth is closest to the radius of the tool edge, and the cutting process is mainly shear-dominated, with the friction and extrusion effects weakened, as well as the plastic deformation in the negative shear deformation zone (P_IV) weakened, and the generated burr size is minimized and smoothed, as shown in Figure 10(b3). As the feed rate continues to increase, the cutting thickness and friction deformation effect increase, resulting in the burr surface beginning to appear as irregular tearing phenomenon, the burr size being rebounded, and the burr morphology becoming more rough. In addition, the effect of feed rate on the exit burr is also related to the free plastic flow characteristics of the material as it loses support at the exit. Appropriately increasing the feed rate can accelerate the cutting process and reduce the interaction time between the tool and the workpiece, thus limiting the thermal deformation in the exit region; however, when the rate is too high, the material is released at the exit just after it is cut, in which case the material is particularly prone to outward coiling and stacking, which in turn leads to a significant increase in the exit burr [37]. In summary, the optimum feed rate of 0.025 m/min resulted in the smallest size of the top burr and exit burr.

Figure 10

(a) Exit burr size measurement results; (b1)–(b5) SEM images of exit burr formation of LF21 slot at difierent feed rates.

From Figure 11(a), it can be seen that the top burr size increases with the depth of cut for both the up-milling and down-milling sides. The burr size increases from 369.26 to 530.07 μm for the up-milling side and from 377.42 to 620.13 μm for the down-milling side. Figure 11(b) demonstrates the top burr morphology of the micro-milled grooves at different depths of cut. It can be seen that when the depth of cut is small, the burr size is small and uniform. As the depth of cut increases, the burr size also increases gradually, and the surface morphology is gradually irregular, complex, and even stacking phenomenon. As the depth of cut increases, so does the contact area between the tool and the workpiece [15]. This leads to a rise in cutting force, which in turn causes greater plastic deformation of the material. Consequently, the burr size increases. Thus, a smaller depth of cut proves beneficial in mitigating the formation of top burrs. To explain the trend, one can also look at the composition of the cutting force. When the depth of cut is shallow, due to the smaller unit cutting force, the surface material of the workpiece is more likely to form a clean cutting surface, which contributes to the control of the top burrs; however, if the depth of cut is too small, the friction behavior between the tool and the workpiece is dominant, which in turn is prone to cause the material to be locally extruded upwards, resulting in large fluctuations in the size of the burr [19].

Figure 11

(a) Top burr size measurement results; (b1)–(b5) SEM images of top bur formation of LF21 slot at different depth of cut.

From Figure 12(a), it can be seen that the size of the exit burr tends to increase with the depth of cut. Figure 12(b) shows the exit burr morphology of the micro-milled groove at different depths of cut. When the depth of cut is 0.03 mm, the burr size is relatively small, 557.43 μm. At this time, the burr morphology is relatively neat, and the deformation is less. As the depth of cut increases, the burr becomes larger and irregular, and the edge of the burr also appears to collapse and tear as shown in Figure 12(b4) and (b5). The material removed by a single cutting edge escalates as the depth of cut deepens, leading to a more intense cutting force and stress distribution. This intensification enhances the material’s plastic flow within the negative shear deformation zone (P_IV). Furthermore, a deeper cut extends the contact time between the tool and the workpiece, amplifying both friction and squeezing effects. However, when the depth of cut reaches 0.06 mm, the rate of burr size growth begins to slow down. This could be because the material’s plastic deformation reaches a point of saturation, causing the negative shear deformation zone to weaken and thus diminishing the burr size increase. The phenomenon of increased exit burr with increasing depth of cut is also attributed to increased plastic deformation due to increased cutting load. As the depth of cut increases and the tool cuts into the workpiece, the material at the exit is more likely to undergo extensive plastic outward deformation during the release process and form larger burrs, especially if the exit is not supported by effective stiffness [1,11]. Therefore, in order to control the burr formation, a smaller depth of cut should be selected as much as possible.

Figure 12

(a) Exit burr size measurement results; (b1)–(b5) SEM images of exit bur formation of LF21 slot at different depth of cut.

As illustrated in Figure 13(a), the roughness (Ra) at the groove’s bottom decreases progressively with an increase in spindle speed. This can be attributed to the fact that higher speeds reduce the tool-workpiece contact time, which in turn lowers friction, dampens tool vibrations, and minimizes bending. As a result, the cutting process becomes more stable, significantly enhancing the quality of the machined surface. As can be seen from Figure 13(b1), when the spindle speed is 6,000 rpm, obvious corrugation can be seen, burrs are large, the surface is not smooth enough, and the roughness value is 0.674 μm. As illustrated in Figure 13(b2), when the spindle speed reaches 9,000 rpm, a noticeable reduction in the groove bottom surface roughness occurs. The surface corrugation begins to smooth out, gradually becoming more uniform and regular, with the roughness value dropping to 0.596 μm. As the spindle speed steadily increases to 18,000 rpm, the roughness continues to diminish, and the surface becomes increasingly smoother, showing only a few scattered pits and burrs. Especially at 18,000 rpm, the surface becomes flatter and smoother, and the roughness value is reduced to 0.465 μm. The formation mechanism of the trend can also be further explained from the perspective of the material removal mode and cutting force changes in the micro-milling process. With the increase in spindle speed, the number of cutting times of the cutting edge per unit time increases, which is conducive to reducing the pulsating impact and mechanical vibration between the tool and the workpiece, thus making the cutting process smoother. At the same time, as the spindle speed increases the cutting temperature will also increase, resulting in the aluminum alloy material in the local area being more likely to produce plastic deformation and flow, reducing the tearing and bonding phenomenon, which helps to obtain a smoother groove bottom surface [19]. Overall, the higher spindle speed has an obvious promotion effect on reducing the roughness of the bottom of the groove and improving the surface quality.

Figure 13

(a) Surface roughness R _a measurement results; (b1)–(b5) Surface micromorphology and 3D characteristics of LE21 slot at different spindle speeds.

It can also be seen from the observation of the 2D morphology after machining in Figure 14(a) that the finish of the machined surface improved significantly with the increase in the spindle speed, the grooves between the grains became shallower and narrower, the raised ridges also narrowed from wide to narrow, and the traces of the tool traveler became more detailed. And it can be seen from the EDS mapping (Figure 14(b)) and combined with the elemental content (Table 7) that the content of element C is higher at the cutting bond on the machined surface, which is caused by the fact that the cutting temperature and friction between the tool and the workpiece are increased at higher spindle speeds, resulting in the chips easily adhering to the tip of the tool to form the accumulating chippings. As the cutting process proceeds, the size and presence of the accumulated chippings show instability, sometimes increasing in size, sometimes disappearing, and the dislodged pieces of the accumulated chippings may adhere to the machined surface.

Figure 14

(a) SEM images of LF21 slot bottom under spindle speeds of (a1) 6,000 rpm and (a2) 18,000 rpm; and (b) SEM-EDS mapping of LF21 slot bottom under spindle speed of 15,000 rpm.

From Figure 15(a), it can be seen that the roughness at the bottom of the groove shows a tendency to decrease and then increase with the increase in feed rate. As mentioned in Section 3.1.3, this is because when the feed per tooth is small, the cutting process is mainly plowing and squeezing, resulting in larger friction and hence larger surface roughness. With the increase in feed rate, when the feed per tooth is close to the tool edge radius of 1.04 μm, the cutting process is dominated by shearing, and the surface quality is optimal at this time. When the feed rate is further increased, the friction between the tool and the workpiece as well as the plastic deformation of the material are enhanced, resulting in the surface roughness of the bottom of the groove picking up again [25]. From the 3D surface morphology of the LF21 groove at different feed rates in Figure 15(b), it can be seen that significant plow marks and ravine appear on the machined surface at lower feed rates. As the feed rate increased, the surface topography improved, the plow marks were reduced and the number of ravines decreased. When the feed rate was 0.025 m/min (1.04 μm/z), the texture of the machined surface was clear and uniform, the number of deep grooves and valleys was less, and the roughness value was at its lowest. When the feed rate continued to increase, the peaks of the machined surface valleys increased, as well as more irregular corrugations appeared, and the roughness value picked up again. Therefore, moderate feed rate and feed per tooth can effectively optimize the surface roughness.

Figure 15

(a) Surface roughness R _a measurement results; (b1)–(b5) Surface micromorphology and 3D characteristics of LE21 slot at different feed rates.

Through the observation of the processing 2D morphology surface in Figure 16, it can be seen that when the feed rate is 0.005 m/min, the processing surface is relatively rough, there are obvious scratches, and the width of the furrow is not uniform; with the increase in the feed rate to 0.025 m/min, the surface texture becomes clear and uniform; and when the feed rate is further increased to 0.045 m/min, the texture of the surface is also increased, and the scratch marks also become more obvious.

Figure 16

SEM images of LF21 slot bottom under different feed rates: (a) 0.005 m/min, (b) 0.025 m/min, and (c) 0.045 m/min.

From Figure 17(a), it can be seen that the roughness value of the bottom of the groove increases gradually as the depth of cut increases. This is because when the depth of cut is large, the contact area between the tool and the workpiece increases during milling, and the cutting force increases accordingly as well as triggering vibration and bending of the tool, resulting in an uneven surface and higher roughness. From the 3D surface morphology of LF21 grooves at different depths of cut in Figure 17(b), it can be seen that the machined surface is smoother when the depth of cut is 0.03 mm; when the depth of cut is 0.04 and 0.05 mm, respectively, the machined surface has significant undulations, forming more peaks and projection; when the depth of cut is 0.06 and 0.07 mm, respectively, the machined surface formed more significant peaks and ravines. This corresponds to an increase in the roughness values, especially in Figure 17(b5) where the surface undulations are more dramatic, indicating that at larger depths of cut, the effect on the surface topography of the workpiece is more significant. Therefore, by selecting a smaller depth of cut, the surface topography can be improved to some extent.

Figure 17

(a) Surface roughness R _a measurement results; (b1)–(b5) Surface micromorphology and 3D characteristics of LE21 slot at different depths of cut.

It can also be seen from the 2D surface topography of Figure 18(a1) and (a2) that the machined surface finishes gradually decrease as the depth of cut increases, the walkover marks become more obvious, and the grooves between the textures deepen more and more. From the EDS energy spectrum Figure 18(b) and combined with the elemental content Table 8, it can be seen that the surface chip sticking contains not only the main alloying factors such as Mn, Fe, Si, Cu, etc., but also a high level of O element in it, which indicates that a oxidized layer was formed on the surface of LF21 during the micro-cutting machining process. In conclusion, a better surface finish can be obtained by a smaller depth of cut.

Figure 18

(a) SEM images of LF21 slot bottom under depth of cut of (a1) 0.03 and (a2) 0.07 mm, and (b) SEM-EDS mapping of LF21 slot bottom under depth of cut of 0.07 mm.