Figure 1a shows the EBM setup from ARCAM AB, Mölndal, Sweden that was used to produce parts with the dimensions of 30×30×10 mm, as shown in Fig. 1b. The used Ti6Al4V powder has a mean particle size of 71 µm. The chemical composition of the powder was (wt.%); aluminum = 6.04, vanadium = 4.05, carbon = 0.013, iron = 0.0107, oxygen = 0.13 and balance titanium. The surface roughness (Sa) of the as-fabricated EBM part on the top face and side faces is 6 µm and 21 µm, respectively. Although the EBM parts are produced according to optimized parameters in previous works [26, 27], as shown in Table 1, On both side and top surfaces, the average surface roughness values are still too low for many applications. For example, the requirement of surface roughness in aerospace parts range from 0.2–0.25 µm [28], and for femoral medical implants is Ra < 0.2 µm [29]. Therefore, a secondary operation for the EBM parts is necessary in order to provide a good surface finish. The secondary operation in this study is carried out via the milling process. Figure 1c shows the three-axis CNC vertical milling machine, Mori DMG (DMC 635 V Ecoline), Germany, capable of the feed rate of 24 m/min, spindle speed of 8000 rpm, and a positioning resolution of 1 µm. This machine was used for investigating and comparing the machining performance of the as-fabricated and heat-treated EBM Ti6Al4V parts with respect to three possible part/layers’ orientations, as shown in Fig. 1d. Figure 2 shows the detail of the three possible part/layers’ orientations that can be encountered during the milling process to enhance the surface finish of the produced EBM part. In the first case, the tool was fed across the EBM layers (Face A in Fig. 1d), in the second case, the tool was fed parallel to the EBM layers (Face B in Fig. 1d), and in the third case, the tool was fed within the plane of EBM layers (Face C in Fig. 1d). A solid carbide end mill cutter was used in the milling experiments. Table 2 summarizes the cutting parameters for milling tests, which are used to determine the effect on the milling quality of EBM component orientation and heat treatment. Table 2 listed process parameters belong to the range of previous studies for machining Ti6Al4V as provided by [13, 30, 31].
Table 1
Default EBM parameters from ARCAM for producing Ti6Al4V parts [26, 27].
EBM Parameters | Beam current | Acceleration voltage | Focus offset | Line offset | Scan speed | Powder layer thickness | Preheat temperature |
Values | 15 mA | 60 kV | 3 mA | 0.1 mm | 4530 mm/s | 0.05 mm | 750°C |
Table 2
Process parameters used for milling the EBM parts
Parameters | Notation | Values | Units |
Cutting speed | V | 50, 80 | m/min |
Feed rate | f | 30 | mm/min |
Axial depth of cut | dA | 0.4, 0.6 | mm |
Radial depth of cut | dR | 4.8 | mm |
Part/layers orientation | - | Face A, Face B, Face C | - |
Two types of EBM Ti6Al4V parts were used in the milling experiments; (i) as-fabricated EBM parts, and (ii) the heat-treated EBM parts. The purpose of the heat treatment was to check its effectiveness in suppressing the effect of the layers' orientations during machining or in other words to suppress the directional properties of the EBM part. It should be noted that already several studies have been reported on the heat treatment of the as-fabricated EBM parts to acquire the desired microstructures [21, 22]. However, the heat treatment was therefore used as a tool in this study to counteract the negative effects of the orientations of the EBM layers during machining. Furthermore, to ensure a fair distinction between the machining of the as-fabricated and heat-treated parts, the heat treatment was chosen in such a way that it would produce approximately the same microstructure and hardness as that of the as-fabricated EBM parts. Nevertheless, no prior knowledge exists in the literature on such a heat treatment that would result in the same microstructure and hardness as-fabricated EBM parts. Therefore, several heat treatment recipes were employed, as listed in Table 3, to obtain the heat-treated EBM parts with the same microstructure and hardness. This was done to exclusively study the effect of the heat treatment on the machining behavior, instead of the microstructural or hardness changes. As it is well known that, the microstructure and hardness significantly affect the machining of the material [32]. Whereas, the purpose of this work is to study the effect of heat treatment in terms of extra consolidation time given to the samples to diminish the layers’ orientation effect.
For microstructure observation of the as-fabricated and heat-treated EBM parts, the parts were ground with grade P220, P400, P600, P800, P1000, P1500, and P2500 silicon carbide papers, fine polished with alumina suspension, and etched using Kroll’s reagent. Microstructural images were taken using a Metkon IMM 901 metallurgical microscope.
Heat treatment of the as-fabricated EBM parts was carried out in a furnace (Nabertherm), Germany, which can heat parts up to 3000°C, as shown in Fig. 3a. To evaluate the milling performance, four responses were measured including surface roughness, microstructure surface morphology, and micro-hardness. The illustration of the milling setup is shown in Fig. 3b. Figure 3c shows the optical microscope used to obtain the images of the porosity. After milling for each of the three orientations (Face A, Face B, and Face C), scanning of an area of 2.2 mm × 1.7 mm obtained an average surface roughness (Sa) on machined surfaces. For every orientation, the average surface roughness of five areas along the cutting direction was taken. A 3D (Contour GT-K) optical profilometer from Bruker, Germany, was used to scan the surface roughness (Sa) of the machined as-fabricated and heat-treated EBM parts, as shown in Fig. 3d. Jeol JCM 6000 Plus scanning electron microscope (SEM) from Japan was used to examine the surface morphology of the machined as-fabricated and heat-treated EBM parts, as shown in Fig. 3e. Microhardness indentation measurements were taken to see the effect of the heat treatments and to compare the hardness of the heat-treated EBM samples with the as-fabricated EBM parts. A Struers A / S, Ballerup, Austria, Durascan 10 Vickers hardness (HV) device was used to test microhardness, with a load of 100 gf (0.1 N) applied for a dwell time of 15 s, as shown in Fig. 3f. Micro-hardness readings were also recorded on Faces A, B, and C (see Fig. 1c) for both the as-fabricated and heat-treated EBM parts after the milling experiments. The average of five micro-hardness readings was taken from each face.
Table 3
Different heat treatment conditions used for the EBM parts.
| Temperature (°C) | Time (h) | Cooling environment |
Heat treatment-1 [21, 33] | 950 | 2 | In air |
Heat treatment-2 [34] | 850 | 2 | In air |
Heat treatment-3 [4, 34] | 600 | 3 | In air |