Tool Life and Surface Integrity Characteristics of SLM and C&W Inconel 718 in Dry and MQL Condition

In this paper, a tool life test was performed using PVD coated insert on Inconel 718 (IN718) samples. The tool life was tested on IN718 workpiece fabricated through conventional (Cast & Wrought) and additive (Selective Laser Melting) process routes. The as-built SLM samples were subjected to two heat treatments such as Hot Isostatic Pressing (HIP) and Aeronautic Heat Treatment (AHT). This study aimed to evaluate the machining behavior of C&W and SLM IN718 from the machining point of view such as tool wear, chip appearance, surface roughness, and residual stresses. The tool life tests were performed under a dry and near dry environment like using Minimum Quantity Lubrication (MQL). Meanwhile, the tool wear propagation under different cutting conditions is also explored. Under both the cutting conditions ank wear and fracture of the cutting edge are the most predominant failure modes minimizing the tool life. The microhardness, surface roughness, and residual stress measurements were analyzed. The result indicates that the microstructural difference between the C&W and SLM has more inuence on the tool life compared to the machining environment. On machining, the SLM sample has 80% and 43% more tool life than the C&W in dry and MQL machining. Comparing the dry and MQL machining of SLM, on using MQL the tool life is 30% less compared to the dry machining.


Introduction
Additive manufacturing techniques have transformed the nature of manufacturing these complex contoured components layer by layer from a 3D model [1][2][3][4][5]. Additive manufacturing has evolved as an economically viable alternative to conventional manufacturing as it does not require any special tool or xture [6]. Selective Laser Melting (SLM) is one among the Laser powder bed fusion (L-PBF) manufacturing processes [7]. Generally, in the Laser powder bed fusion (L-PBF) technique, optimization of various process parameters such as laser power, laser scan speed, hatch spacing, etc. is essential to realize denser parts [8][9][10][11]. The as-built parts are less dense due to porosity which are introduced from existing pores of the powder, lack of fusion and entrapment of gas. The as-built samples are inferior in mechanical properties due to defects and due to the heterogeneous microstructure concerning C&W parts [12]. Post -processing techniques like standard heat treatments like Hot isostatic processing and age hardening can be employed to minimize these defects inorder get the desired mechanical properties [13,14]. Inconel 718 (IN718) is a precipitation strengthened nickel-based superalloy, has been developed extensively over the past four decades [15,16]. The remarkable weldability property of IN718 is due to low content Ti and Al which makes this material best suited for SLM [17][18][19][20]. IN718 is the preferred alloy for fabricating aerospace components such as turbines disks, compressors disks, blades, casing used at high-temperature applications [21].
IN718 parts maintain their mechanical properties up to 650°C due to the presence of strengthening phases [22]. As a consequence of its mechanical properties and low thermal conductivity, it is classi ed as a di cult material to machine [23,24]. While machining IN718, the low thermal conductivity and rapid work hardening rate result in elevated temperature leads to excessive tool wear [25]. During dry machining the temperature is more than 1000°C, a strong tendency to welding to the cutting tool alters the tool geometry by Build Up Edge (BUE) formation. The abrasive wear occurs due to the presence of hard carbides such as TiC and NbC trapped at the tool-chip interface generates groove marks on the chip and starches on the tool [26].
The severe ank and notch wear are the dominant failure mode that occurs while machining IN718. The ank wear usually develops due to the adhesion of workpiece material on the tool surface [27]. In the milling of IN718, the intermitted thermomechanical load results in the notch wear and microchipping at the depth of cut. An increase in temperature at the cutting tool/workpiece interface directly relates to the increase in tool wear progression [28,29]. To increase the tool life, the type of coating deposited on the tool surface, its hot hardness property should be considered to minimize the plastic deformation at the cutting edge.
The use of cutting uids has a major impact in increasing the tool life. Therefore, to enhance the tool life, cutting uids-assisted machining is the only option to reduce the temperature at the cutting zone.
Concerning environmental awareness Minimum Quantity Lubrication (MQL) was used as an alternative to conventional cutting uids [30]. The use of lubrication minimizes the thermal softening [31], but the cutting tool is prone to plastic deformation which induces compressive residual stresses on the workpiece [32]. On subsequent machining during tool life, progressive deformation of the cutting edge in uences the surface integrity (esp. surface roughness) to a greater extent [33].
The majority of machinability study discusses the machining in uence on surface integrity and tool wear.
According to the best of the authors' knowledge, no previous work has been reported on the in uence of microstructure realized by two different process routes such as conventional (C&W) and additive (SLM) on tool wear.
Therefore, the main objectives of this paper are: The Fabrication of samples and their post-processing methods like heat treatments.
The mechanical characterization and microstructural analysis.
Their in uence on tool life and tool wear mechanisms.
To identify different types of tool wear under dry and MQL machining.
To study the micromorphology of chip pitch, chip segmentation frequency, and equivalent chip thickness in detail.
The methodology shows the in uence of the microstructure and machining environment on the wear mechanisms. Finally, the conclusion summarizes the relevant results and the main parameters controlling tool life.  Table 1. The samples were fabricated using an SLM 125 HL machine, equipped with a Yb-ber laser of 400W capacity. The powder particle size was found to be in the range of 40-90 µm measured using Horiba Camsizer X2. The SEM image of the powder particles is shown in Fig. 1 (a), the dendrites are visible on the magni ed single powder particle in Fig. 1 (b). All the samples were CAD modeled to verify the sample position and the supports on the build platform using Magics software version 21. The process parameters used were laser power -275 W, scanning speed -760 mm/s, the hatch distance -120 µm, and the layer thickness -50 µm with a minimum scanning time of 12 seconds. The laser scan pattern used was of strips type, the angle between each strip was 67°. The build platform was preheated to 200°C, the oxygen content <0.03% was maintained. The vertical (90°) orientation sample shown in Fig. 1 (b) was fabricated in a controlled nitrogen gas atmosphere.

Materials
The samples were machined out from the build platform without any post-processing to study the microstructure and mechanical properties of the as-built samples. The heat treatment like Hot Isostatic Pressing (HIP) was performed at the target temperature of 1160°C and pressure of 102 MPa was maintained for 3hrs. Subsequently, Aeronautic Heat Treatment (AHT) otherwise called as aging heat treatments was performed at 720°C for 8 hours, furnace cooled for 1.5 hours until the temperature reduces to 650°C and maintained for 8 hours followed by air-cooled as shown in TTT diagram of Fig. 2.
Even if this TTT diagram has originally been developed for a cast material and may not be suitable totally for SLMed Inconel 718 given the possible level of dendritic segregation, it already provides a good idea of the precipitation domains of γ′ and γ′′ strengthening phases and it is known to provide almost best static properties for an LPBF-processed Inconel 718 [34].

Machining conditions
The machining tests were performed on the 5 axis Hermle C40U CNC machine. An external generator (Lubrilean Digital Super-developed by SKF) was connected to the CNC machine to supply the oil mist. A computer equipped with Vogel software is connected to the generator to maintain a constant ow rate of 73 ml/hr at a pressure of 2 bar. The Vascomill MMS SE1, a Blaser oil based on synthetic esters, was used as the MQL oil. The samples were machined using a single insert with a tool diameter of 20 mm as in Fig. 3 (a) and the top view of the cutting process using a single insert is shown in Fig. 3 (b). A sample length of 60 mm was machined at a cutting speed (V c ) of 30 m/min for a feed rate (f z ) of 0.1 mm/tooth. The axial depth of cut (a p ) is 0.5 mm and the radial depth of cut (a e ) is 9 mm. The cutting conditions were followed from our previous studies, where the machinabilty study and surface integrity analysis were performed under different cutting conditions [36].
The tool used in this study is of reference R390-020A20-07M with the internal cooling channel for lubrication liquid, along with the insert mentioned in Table 2, which was recommended and provided by Sandvik Coromant. Edge radius, E R 27µm ± 5µm The cutting-edge radius (E R ) of the inserts was measured using the GFM MikroCAD® equipment. At the same time the machined workpiece surface roughness, Ra (µm) was measured three times using 3D pro lometer WYKO NT1100. The residual stress of the machined surface was measured using Bruker XRD D8 equipment. The parameters used for the residual stress measurement were X-ray source Mn, Angle 2 theta 153°, Miller indices 311. Each test was repeated three times to obtain a range of standard deviation. The carbide insert was very small to x the thermocouple to perform the temperature study.  (Fig. 4d) and the undiffuse powder particle after fabrication is shown in Fig. 4e. A homogenized microstructure with a average grain size of 8.9 µm is evolved after HIP heat treatment by the complete dissolution of the melt pool boundaries and closure of the pores as in Fig. 4 (c). At 1100°C during the HIP heat treatment there is no δ precipitation since all the Nb is in solution state. On cooling the orthorhombic δ phase (Ni 3 Nb) is precipitated and distributed at the grain boundary along with the carbides shown in Fig. 4 (f). The precipitated needle-like δ phase at the grain boundary (Fig. 4g) restricts the grain growth [37] and reduces the ductility of the material [38].

Results
Subsequently AHT heat treatment was perfomed where the strengthening phase γ′-Ni 3 (Al, Ti) and γ′′-Ni 3 Nb precipitates using the remaining Nb available in the γ matrix. The average grain size after AHT heat treatment is 14.1 µm which is 37% larger than grain size of the HIP sample.

Evolution of tool wear
According to the ISO 3688, the tool performance was evaluated by measuring the total time in cut (tc) in minutes until the average ank wear (Vb) reaches 0.3 mm. This section presents the results of the tool life test performed on additive SLM sample and conventional C&W sample under dry and MQL machining environment. The PVD tool of (Ti, Al) N 2 coating with reinforced edge, a lead angle of 90° was used for all the machining conditions. The machining tests were interrupted after each time in cut (tc) of 0.294 mins to estimate the tool wear. On machining SLM under the dry condition, the initial phase was from tc 0 to 0.588 mins during this phase the tool wear rapidly progressed to a Vb of 0.08 mm. In the linear phase tc from 0.588 to 1.471 mins, the wear was rather in uniform rate attaining a Vb of 0.103 mm, and in the catastrophic phase tc from 1.471 to 2.941 mins, the ank wears Vb was 0.3mm. On machining SLM using MQL the tool wear progressed to a Vb of 0.08 mm during the initial phase from tc 0 to 0.588 mins. In the linear phase tc from 0.588 to 1.471 mins the wear was uniform until Vb of 0.17 mm. The catastrophic phase was from tc 1.471 to 2.059 mins at this phase the measured Vb was 0.35mm. The tool wear pattern on the machining SLM sample was justi able regarding the C&W.

Effect on surface roughness
The surface roughness, Ra (µm) was measured on samples after each machining pass in both dry and MQL conditions shown in Fig. 6. The surface roughness measured on the C&W sample has an increasing pattern as the tool wear progresses. Whereas the surface roughness decreases with an increase in tool wear on machining SLM samples in both dry and MQL conditions. Especially under dry machining of the SLM sample the lowest surface roughness of 0.15 µm was recorded. Whereas, the highest the roughness value of 0.36 µm was recored while machining C&W sample under dry condition.

Evolution of residual stress in the different machining environment
The residual stresses measured at the middle of the sample were compressive in nature irrespective of the workpiece and machining environment. In dry machining conditions, there is 36% reduction in the compressive residual stress as the ank wear (Vb) increases from 0.06 mm to 0.3 mm concerning the SLM sample. While MQL machining there is a 66% reduction in the compressive residual stress for the same ank wear period shown in Fig. 7.
The thermal load produced during machining has a direct impact on tool wear and the surface quality of the machined workpiece. During machining, the cutting heat is mainly taken away by the chips in dry conditions and by cutting uids in MQL conditions. Apart from that large quantity of the thermal stress is transferred to the tool and only a small quantity is transferred into the workpiece and into the surroundings. The quantity of heat dissipates into the workpiece (Qw) accumulates as residual stress [39].
More heat dissipates into the machined surface results in compressive residual stress during dry and MQL machining. Using MQL resulted in less thermal load (less heat dissipation to the machined surface) and high mechanical load, which maintained the compressive residual stress. The magnitude of the compressive residual stress decreases as the tool wear progresses. This trend is due to machining with a sharp tool tip results in maximum mechanical load and minimum thermal load. Whereas, as the tool wear progresses, the cutting edge becomes blunted which results in maximum thermal load and minimum mechanical load.
The residual stress value also remained more or less in equilibrium irrespective of the different microstructure possessed by the SLM sample compared to the C&W sample.

In uence of heat treatment on mechanical properties
The hardness of the samples was determined by the average of 25 measurements after applying 20 kgf (or 200N) load for 15 seconds to each sample. The hardness value of the as-built, HIP and HIP-AHT samples about the cast and wrought IN718 is shown in Fig. 8.
The pattern of the hardness value is consistent with the work of Zong et al. [13]. In this work, they fabricated Inconel 718 using the Direct Metal Laser Sintering (DMLS) technique, where the hardness value increased following the aging heat treatment. The hardness value has increased linearly after each heat treatment; of about 30% after HIP. On subsequent AHT heat treatment, the hardness increased up to 42% concerning as-built hardness.

Tool wear analysis under dry environment
The tool wear progression was measured after every single pass, using the optical microscope (OM). The tool wear was investigated under Scanning electron microscope (SEM). Build Up Edge (BUE) formation (Fig. 9a) was observed on the insert at tc 0.588 mins while machining C&W sample under dry condition. Excessive heat generated during dry machining, leads to micro-welding of the workpiece at the tool tip eventually leads to BUE formation. For the same tc of 0.588 mins on machining SLM sample under the dry condition the tool wear begins with the abrasion of the coating material in the form of severe grooves.
In addition the diffusion of workpiece material on to the rake face and insert fracture at the cutting edge were observed as shown in Fig. 9 (b).
At tc of 1.1471 mins, the notch wear appears at the depth of cut (Fig. 10a) occurs due to machining of the deformed hardened surface resulting from the previous machining [32]. On futher machining, extensive abrasive wear along with adhesive wear resulted in ank wear (Fig. 10b) at the clearance face.

Tool wear analysis under MQL environment
The tool wear were analysed under MQL assisted machining of SLM and C&W workpiece. The mechanical load is higher during the machining of C&W which results in the fracture of the cutting edge.
The wear mechanisms observed are adhesion of the workpiece material, coating delamination and diffusion of the chip material. Similar observation were made by Ibrahim et al [40] who also reported that, rst the coating delamination occurs followed by the adhesion of workpiece material as in Fig. 11 (a). On further machining the sliding of the adhered material on the rake face resulted in abrasion wear and crack (Fig. 11b) which led to the tool failure at tc 1.177 mins reaching ank wear, Vb of 0.3 mm. In spite of using MQL, the high carbides content in C&W sample along with the sliding of adhered material (sandwiched between the chip and the coating) results in the abrasive wear on the rake face [41]. Fig. 11 (c) shows the rake face of the tool on machining SLM sample using MQL, as a consequence the coating delamination and adhered material zone is minimum. At tc of 1.764 mins crack initiation at the rake face occurs along with the crater wear resulting in a ank wear, Vb of 0.17 mm. On further machining the crack propagates leading to tool fracture as the load withstanding capacity of the tool is reduced at tc of 2.058 mins reaching the ank wear Vb of 0.3 mm. The instability of the cutting tool on machining SLM under MQL condition results in chipping, crack and breakage (Fig. 11d) of the cutting edge leading to a poor tool life with reference to the machining of the SLM sample under dry condition.
Despite using MQL, the tool life is short concerning the dry machining of the SLM sample. The cuttingedge was unpredictable after a tc of 2.058 minutes. This could be due to the uneven thermal load over mechanical load leading to tool fracture.

Chip morphology
Cyclic or serrated types of chips were obtained on machining C&W and SLM samples under dry as well as MQL conditions. This serrated chip formation is due to the occurrence of alternating high shear strain followed by low shear strain [42]. Observation on the shining side and serrations at the free surface of both the SLM and C&W chip is quite interesting. The presence of intensive groove marks (Fig. 12a) at the shining side of the C&W chips under dry conditions is due to the friction produced during the sliding of adhered material. Use of MQL have reduced the friction at the tool chip interface resulted in less intense groove marks (Fig. 12b) unlike dry machining of C&W workpiece. The SLM chips under dry condition have a very smooth shining side, which shows that the machinability of SLM is good concerning C&W. Certain black spots (Fig. 12c) were observed on the chips produced by dry machining of SLM sample, which could be due to the high temperature generated at the tool chip contact. Whereas, on machining under MQL the tool wear was not predictable which is evident from the undulation marks and adhesion of the coating present on the shining side of the chip (Fig. 12d). This could be due to the improper cut and thermal variation at the tool chip contact zone. The presence of groove marks on the shining side of the C&W chips and the smooth shining side surface of the SLM chips show the machinability of SLM IN718 is better than C&W IN718. Figure 13 shows the schematic diagram that represents the characteristic features of the segmented chip. The maximum thickness of the segmented chip (t 1 ), the maximum thickness of the segmented chip (t 2 ), and pitch is the distance between two segmented peaks.
The serrated teeth formation occurs when the chip segmentation frequency is similar to the uctuation frequency of the cutting force. The sheer instability results in a 50% increase in chip segmentation frequency while machining the SLM sample using MQL (Fig. 14a). Whereas, for the C&W sample under dry and MQL condition and SLM sample under the dry condition the chip segmentation frequency is decreased to below 20 kHz. This is due to tool wear resulting in thermal softening and increased frictional forces.
Chip segmentation frequency [43] is calculated from equation 1 In equation 2, t e is the equivalent chip thickness (µm), t 1 is the maximum chip thickness (µm) and t 2 is the minimum chip thickness (µm). While machining the C&W sample under dry conditions, the tool fails by BUE. The equivalent chip thickness (t e ) is high due to increased frictional force which reduced the chip ow velocity (Fig. 14b) lead to microwelding of the workpiece material at the cutting edge. Frictional force gets considerably reduced on using MQL oil resulted in reduced equivalent chip thickness value due to increased chip ow velocity. Therefore, considering the surface integrity for the fatigue samples, the SLM HIP-AHT sample under dry machining is good a choice. To improve the tool life, some other metalworking uids like ood cooling, cryogenic, or hybrid cryogenic -MQL assistance should be performed.

Conclusions
In this study, tool wear of PVD coated tool (Ti,Al)N was investigated using Cast and Wrought (as reference) and SLM processed samples. The SLM samples further had received two different heat treatments. The two-heat treatments were HIP and HIP-AHT leading to differences in hardness, the AHT providing greater hardness thanks to the precipitation of γ′′ + γ′ phases. To check the reliability of the tool, a tool lifetime test was conducted according to the ISO 3688 standard. The tool life test was performed under dry and MQL condition using a PVD-coated reinforced edged insert. The surface roughness and residual stresses of the samples were measured at different tool cut in time, tc and following conclusion were obtained: The additive SLM sample has 80% and 43% more tool life than the C&W sample in both dry and MQL conditions.
The tool life on machining SLM sample under MQL condition was 30% less compared to the dry machining.
It is due to an increase in shear instability which is evident from the 50% increase in chip segmentation frequency resulted in chipping, crack and breakage of the cutting edge leading to a poor tool life.
The decrease in the compressive residual stress as tool wear progresses irrespective of the cutting condition reveals the transition have taken place from maximum thermal load to minimum mechanical load.        Evolution of the residual stress under dry and MQL conditions on machining C&W and SLM sample.