Abstract
Selective laser melting (SLM) of titanium–aluminium (TiAl) alloy components has gained significant attention in the modern industrial world. The flexibility of the SLM process in producing complex shapes with minimum utilization of material and energy makes it dominant over other manufacturing techniques. As aerospace and biomedical industries demand complex-shaped TiAl alloy components, part fabrication using SLM becomes the ultimate solution. However, the unacceptable level of surface integrity and anisotropic behavior of SLM components demand post processing operations such as laser polishing, chemical polishing, and conventional polishing methods. In this study, a recently developed polishing method called wire electrical discharge polishing (WEDP) is performed on TiAl alloys for obtaining a smooth and defect-free surface. This study aims to investigate the micro-layer modification occurring to the WEDP-processed surface in detail. The experimental results establish the effectiveness of WEDP method in terms of improved surface integrity. The surface finish (Sa) got enhanced by ~88% after WEDP processing. In addition, the thickness of recast layer formed by WEDP was found to be minimum. Moreover, post-processing of TiAl alloy resulted in better surface morphology specifically at lower settings of peak current. It is noteworthy that the migration of wire material was minimum with zinc-coated brass electrode compared to the normal brass electrode. Hence, coated wire electrodes are recommended for WEDP process. In short, an excellent surface integrity can be achieved using WEDP process through favorable surface modification aided by lower peak current and coated wire electrodes. Furthermore, less electrode wear observed in WEDP process enables the deployment of lower feed rates leading to minimal electrode consumption.
Introduction
Metal additive manufacturing (AM) is a rapidly evolving technology in the modern manufacturing world owing to its inherent advantages such as mass customization, design freedom, and minimal material wastage. 1 Selective laser melting (SLM) is one of the widely used metal AM methods, which can produce high-density parts (~99%) with fine resolution and tolerances. Among the wide range of materials fabricated using SLM, titanium–aluminium (TiAl) alloy establishes dominance in terms of high strength-to-weight ratio, excellent corrosion resistance, and good biocompatibility. 2 In the present scenario, fabrication of complex-shaped TiAl alloy components using SLM is gaining momentum due to its increasing demand in the field of aerospace and biomedical industries. However, the components fabricated using SLM exhibit poor surface integrity 3 leading to high surface roughness values. Thus, finishing/polishing of the fabricated components is highly recommended to eliminate surface irregularities/defects for improving the surface finish. Laser polishing, chemical polishing, and conventional polishing methods are the available post-processing variants employed for enhancing the surface integrity of metal AM components. 4 However, the aforementioned processes are associated with many inherent limitations which are detailed in the below section.
Review on finishing/polishing of additively manufactured TiAl alloys
Several finishing/polishing processes have been attempted on additively fabricated titanium alloy components by researchers till date. In one of the studies, laser polishing was found to improve the surface finish of SLMed Ti6Al4V specimen from ~10.2 µm to ~2.4 µm. 5 However, an increase in the laser energy input led to increased velocity of the melt pool resulting in the formation of periodic striation patterns over the polished surface. In another study, an ablative process performed prior to laser polishing was seen to be effective in removing large-scale structures over the as-built Ti6Al4V surface. 6 Followed by this, nanosecond fiber laser polishing was performed on Ti alloys (TC4 and TC11) by researchers that caused enhancement in surface microhardness owing to the formation of α’ martensitic phase. An excellent surface finish of less than 1 µm was also achieved using laser polishing. 7 Besides these studies, micromachining was employed using femtosecond laser for surface modification of Ti6Al4V parts. The as-built roughness was reduced by 5 times and a final surface finish of ~0.8 µm was attained using a single pass of laser. Anyhow, oxide layer formation was evident over the ablated surface on account of post-processing in atmospheric conditions without inert gases. 8 A recent study reported that polishing of additively fabricated TiAl alloy using a pulsed laser results in crack formation. The reason can be attributed to the low remelting depth achieved on behalf of low energy density of pulsed laser. 9 Although laser polishing can ensure improvement in surface finish, the method experience difficulty in polishing complex internal surfaces and holes where accessibility of laser head is limited. 10 In addition, the requirement of high processing time made researchers to explore the capability of conventional methods for post-processing. Surface mechanical attrition treatment (SMAT) processing of SLM-fabricated Ti6Al4V components received significant attention as it could induce compressive residual stresses on the material. However, the surface finish offered by SMAT was inferior when compared with other methods. 11 On the other hand, milling process on as-built Ti6Al4V surface contributed an excellent surface finish of ~0.3 µm by effectively removing partially melted particles. But, milling process left slight milling stripes over the processed surface. At the same time, roughness valleys were still evident after post-processing in case of vibratory grinding and micromachining. 12 In another study, SLMed Ti6Al4V components subjected to toroidal milling in another study resulted in larger flank wear on major cutting tool edge as compared to conventional alloy material. 13 In addition, a later study emphasized that SLMed Ti6Al4V samples exert a larger axial force of more than 22% in comparison with conventional alloy samples, which can ultimately deteriorate tool life. 14 Taking into account the fact that conventional methods are not suitable for post-processing complex and intricate shapes, chemical polishing was introduced by researchers. A two-step chemical polishing method was introduced later on an additively fabricated Ti6Al4V alloy, where ~71% improvement in surface finish was achieved. Stream tracks visible after first polishing step (erosion) got eliminated after second step of polishing (leveling). The process also featured the formation of a thin passivation layer over the material surface that enhances corrosion resistance. 15 Even so, the problem of excess mass/material loss was reported in due course of time while conducting chemical polishing on Ti6Al4V alloy fabricated using laser powder bed fusion. 16 Thus, loss of dimensional accuracy is one of the big challenges in chemical polishing of metallic AM components. Furthermore, the applicability to limited materials, risk of toxicity and selective phase dissolution also raises concern regarding chemical polishing.17, 18
To overcome the problems associated with existing post-processing methods, researchers started exploring the potential of electrical discharge machining (EDM) technology in post-processing metal AM components. Most of the metal AM alloys used in industries are of difficult-to-cut in nature due to high hardness. Unlike existing finishing methods, the spark-assisted material removal in EDM offers the feasibility to process such materials irrespective of their hardness. 19 Hence, the technology is widely adopted in automotive, aerospace, and biomedical industries for various requirements. 20 In recent times, a low-energy finishing process called wire electrical discharge polishing (WEDP) was introduced by researchers for post-processing metallic AM components. 21 WEDP is a low-discharge energy polishing method derived from EDM technology. The excellent surface finish in submicron range as well as improved surface-related mechanical properties offered by WEDP makes it an appropriate post-processing method for metallic AM components. 22 Moreover, WEDP efficiently performs in processing metal AM alloys independent of the physical properties 23 and non-planar geometrical design. 24 It is noteworthy that the industries rely on wire cut EDM method for removing fabricated metal AM components from the base plate. In this context, WEDP offers the feasibility to polish materials in the same part removal platform using specific energy settings thereby minimizing the workflow and production time. However, the modification induced by WEDP on rough micro-layer at SLM-fabricated TiAl alloy surface using WEDP is not investigated till date. Thus, this study elucidates the surface modification caused by the spark-induced thermal energy in the WEDP process.
Mechanism and methodology
Working mechanism of WEDP
WEDP is a derived variant of EDM that utilizes very small discharge energy for improving surface integrity. The process involves removing a thin rough layer of metallic material in microns via melting and vaporization phenomenon. The thermal energy of continuous electric sparks produced at the interelectrode gap (IEG) between tool electrode and workpiece is utilized for polishing. The peaks and valleys over the rough surface get eliminated and nearly uniform melting and solidification of microlayer subsurface are ensured. The pulse on the duration for WEDP is chosen to be minimum for avoiding deep crater formation over the surface. The schematic of finishing/polishing process using WEDP is depicted in Figure 1.
Schematic of WEDP process (a) in XY plane and (b) three-dimensional view.
SLM fabrication
TiAl alloy specimen chosen for WEDP experiments was fabricated using direct metal laser sintering (DMLS) equipment (Model: EOS 290 DMLS, GmbH Germany) with recommended parameter settings. The equipment utilizes 400 W Ytterbium fiber laser (Wavelength: 1060− 1100 nm). The powder material chosen for SLM was nearly spherical shaped with diameter varying from 15 to 45 µm. Using the powder material, specimens with dimensions of 30 mm × 10 mm × 10 mm were built over the base plate. The chances of oxidation are minimized by conducting SLM fabrication in argon atmosphere.
Experimental details
The fabricated specimen is placed in wire EDM equipment (Model: Electronica Ecocut) with the aid of suitable fixtures. The experimental setup for WEDP is shown in Figure 2. Using ‘edge finding’ option available in the machine, straightness of sample was ensured and zero reference was set. Subsequently, the table is traversed to provide a depth of cut equivalent to the mean value of maximum peak-to-valley height (Sz).
Experimental setup for WEDP process.
Both pulse on time and peak current setting was kept minimum to maintain minimum discharge energy during the process. The pulse on-time values were chosen to be less than 5 µs to ensure machining in the finishing regime. Finally, a very thin microlayer having thickness equivalent to Sz is removed from the rough metallic surface (sidewalls parallel to building direction) in order to eliminate the irregularties/defects present on the surface. The experimental settings fixed for the study are given in Table 1.
Experimental conditions.
Variation in roughness parameters after WEDP post-processing of SLMed TiAl alloy.
Characterization/analysis
The surface morphology of as-built SLMed and WEDP processed samples were captured using a field emission scanning electron microscope (FESEM) equipment (Model: Carl Zeiss Gemini SEM300). The required EDS spectrum was also obtained using FESEM equipment. The associated roughness measurements over the samples were conducted using a non-contact optical profilometer (Model: AEP Nanomap 1000 WLI). The current and voltage waveforms were acquired with the assistance of a digital oscilloscope (Model: Tektronix 3 series MDO).
Results and discussion
Surface morphology
The surface morphology of the as-built SLM sample and WEDP processed sample were evaluated using FESEM images. The presence of pits and voids was evident over the as-built SLM surface. The reason can be attributed to the lack of material fusion or interlayer bonding during the SLM process arising from lower input laser energy. In addition, the presence of unmelted powder particles can be observed within the pits/void spaces. This further confirms that the energy received at the same zones was insufficient for complete melting. However, all such defects and surface irregularities got eliminated after WEDP post-processing of TiAl alloy. Consequently, the non-uniform and rough surface morphology of as-built SLM sample was modified to form a uniform and smooth surface as shown in Figure 3a,b.
Surface morphology of TiAl alloy in (a) as-built SLM and (b) WEDP condition.
The non-uniformity in material fusion is visible from the SEM images at higher magnification (×1.00 K) as shown in Figure 4a. The surface morphology after WEDP is significantly different in contrast with the as-built SLM specimen. The electrical discharges in the form of sparks occurring between the wire electrode and TiAl specimen generate craters over the processed surface (Figure 4b). Moreover, spherical-shaped debris particles can be noticed over the WEDP-processed surface. Although dielectric flushing is ensured during WEDP process, the debris particles exhibit higher tendency to get resolidified on the processed surface due to rapid solidification induced by deionized water quenching. Further, tiny micron-sized pores are prevalent over WEDP surface as shown in Figure 4b. The underlying reason can be due to the vaporization of alloying elements having lower melting point during WEDP process.
Surface morphology of TiAl alloy at higher magnification in (a) as-built SLM and (b) WEDP condition.
Surface roughness analysis
The roughness parameters such as Sa, Sz, and Svk were considered to assess the effectiveness of WEDP. Sa, Sz, and Svk represent the 3D areal roughness, maximum peak-to-valley height and depth of valleys, respectively. The mean values of each parameter based on five readings were chosen to ensure repeatability. Table 2 shows the percentage improvement in surface finish achieved using WEDP in terms of various roughness parameters.
As-built surface roughness (Sa) of ~9.32 µm was improved to ~1.04 µm using WEDP method. The maximum peak-to-valley height (Sz) also got improved by ~50.17% which indicates that WEDP can contribute to excellent surface finish. Moreover, it is significant to note that the Svk value obtained after WEDP finishing is much lesser compared to the as-built sample as shown in Figure 5. Svk value was reduced by ~87.5 % which indicates that the depth of valleys got minimized over the surface. The Svk value signifies the fluid retention capacity of the surface. Thus, the effective volume of surface valleys is minimized, which appreciably prevents the corrosive liquid chemicals from filling the valley gaps thereby hindering the chances of corrosion. Moreover, the tendency of surface valleys to act as microscopic stress raisers is lowered at smaller Svk values. Thus, the reduction in Svk value can limit or minimize local stress concentration over the surface thereby delaying the chances of crack initiation and propagation in practical applications.25, 26
Surface roughness for as-built and WEDP processed TiAl at TON = 2.2 µs.
The effect of pulse on duration (TON) in finishing regime on surface finish during WEDP processing of TiAl alloy was explored. The percentage improvement in surface finish was found to be dependent on the TON value as shown in Figure 6. At higher settings of TON (4.5 µs), the mean roughness of the as-built SLMed sample was reduced to ~1.38 µm contributing to ~85% reduction in surface roughness. However, ~92% enhancement in the surface finish was achieved (Sa = 0.84 µm) at lower settings of TON (2 µs) owing to the reduction in crater dimensions resulting from shortened sparking duration. Roughness value of less than 1 µm represents a smooth surface nearly free from microdefects. 27 Thus, sub-micron surface finish can be achieved with lower TON settings.
Effect of pulse on time in improving surface finish.
Effect of peak current
The post-processed surface morphology is strongly dependent on the nature of the electric spark. The intensity of the spark in turn is strongly dependent on the peak current setting employed in WEDP. The surface morphology obtained is better at lower settings of peak current (~4 A) compared to higher settings of peak current (~11 A) as shown in Figure 7.
Surface morphology at peak current (Ip) of (a) ~11 A and (b) ~4 A.
With higher peak current, the discharge energy will be higher. Thus, bigger craters will be produced by the sparks leading to the deterioration of smooth surface morphology. Moreover, undesirable surface irregularities are generated in the form of micron-sized debris particle re-deposition resulting in degradation of surface quality (Figure 7a). Thus, improvement in the surface finish is hindered at higher peak currents. On the other hand, lower peak current ensures small crater formation and minimal redeposition of melted particles (Figure 7b) leading to smooth surface morphology. Thus, smaller peak current setting is favourable for WEDP process.
The current and voltage signals captured using a digital oscilloscope for higher and lower peak current settings are shown in Figure 8. The actual peak current value is ~4.29 A at machine notch value settings of Ip = 10 Mu, whereas ~11.03 A at Ip = 12 Mu (Mu refers to machine units). As evident from Figure 8, the discharge duration will be in the range of nanoseconds for both current settings. However, the discharge duration remains small (~380 ns at ~4 A) in WEDP finishing regime as compared to ~640 ns at ~11 A. Thus, WEDP involves minimal discharge duration which is responsible for lowering the effective discharge energy during the process leading to smooth surface finish.
Typical current and voltage waveforms in WEDP process at (a) ~4 A and (b) ~11 A.
Recast layer analysis
As WEDP process involves melting and solidification of the material being polished, recast layer can be observed over the surface after post-processing. The recast layer thickness (RLT) was measured at five different regions at the top surface having recast layer. The average RLT (RLTavg) obtained was ~4.23 µm, which is very small compared to normal RLT which can go up to ~30 µm. The underlying reason is that the minimum discharge energy in WEDP method restricts the thickness of recast layer formed on the surface to be minimum. The small recast layer formed over the WEDP-processed TiAl alloy surface is shown in Figure 9.
Recast layer formed on WEDP surface.
The smaller RLT indicates that the impact of thermal energy on the surface microlayer in WEDP process is only minimal. Moreover, minimum thickness of recast layer reduces the chances of sudden fatigue failure due to reduced density of microcracks. Thus, the tendency of recast layer to act as a stress raiser is also minimized in WEDP-processed samples. However, the thermal energy produced during WEDP is responsible for inducing short microcracks over the recast layer as shown in Figure 10. The lower thermal conductivity of TiAl alloy can generate thermal residual stresses on the surface leading to microcrack formation over the surface.
Surface crack density (SCD) evaluation over WEDP surface at Ip = ~4 A.
The average surface crack density (SCD) on the WEDP surface was evaluated at lower settings of Ip (~4 A) to explore the impact of spark-induced thermal energy. Lower settings of Ip were chosen to investigate the surface modification at lower discharge energy levels. SCD represents the total length of micocracks per unit area of the SEM image. Length of notable microcracks is considered for the evaluation of SCD as marked in Figure 10. The area of micrograph is 8778 µm2 and the total length of microcracks is nearly 162.66 µm. Thus, the average SCD on WEDP processed TiAl surface appears to be ~0.018 µm/µm2. The density of microcracks can be lowered by maintaining higher servo voltage settings during WEDP process. 23
Effect of wire electrode material on TiAl surface
Both workpiece and wire electrode are eroded in WEDP process due to thermal energy induced by the spark in IEG. Thus, material constituents can migrate from wire material towards the workpiece and vice versa. This can affect the mechanical properties of the processed component depending on the properties of foreign material coming from the wire electrode. WEDP of SLM TiAl alloy was conducted using two different wire electrode materials. Normal brass wire and zinc-coated brass wire were considered for the study to evaluate the wire material migration to the workpiece. Energy dispersive X-Ray spectroscopy (EDS) analysis was carried out for both cases to determine the elemental composition on WEDP surface. The typical EDS spectrum obtained for uncoated electrode case is shown in Figure 11a.
(a) Typical EDS spectrum for WEDP processed TiAl surface and (b) material migration from wire electrodes to workpiece surface after WEDP.
For both cases, wire material migration to the processed surface was evident as shown in Figure 11b. The elemental weight percentage of copper was ~6.52% while that of zinc was ~1.6% for the surface processed by WEDP using normal brass wire. However, zinc-coated brass wire exhibited minimum migration of wire material towards the processed surface compared to normal brass wire. For zinc-coated brass wire, the weight percent got drastically reduced to ~0.42% and ~0.36% for copper and brass, respectively. The zinc coating prevents the melting of core wire material (brass) and restricts the excess migration of copper towards sample surface. Thus, coated wires are recommended for WEDP process to minimize material migration from wire electrode.
Wire electrode analysis and TiAl migration
The wire electrode utilized during WEDP at lower and higher discharge energy settings was analysed using SEM images. At higher discharge energy, significant wear was observed and the thermal impact of spark energy was spread over a larger area of the wire (Figure 12a). In addition, the higher discharge energy aggravated degradation of the wire in terms of severe crack formation. Such instances of wire surface degradation can promote chances of wire breakage.28, 29 In contrast, the wire electrode was subjected to less wear at lower discharge energy settings as shown in Figure 12b. Moreover, the cracks generated on the wire surface at lower discharge energy were relatively lower. Therefore, lower feed rates can be employed in WEDP that can ultimately minimize wire electrode consumption thereby enhancing process efficiency.
SEM images of wire electrode after WEDP under (a) higher and (b) lower discharge energy settings.
The loss of elements from SLMed TiAl surface in terms of migration/diffusion towards wire electrode was also investigated using areawise EDS analysis. Evidence of material migration towards from work piece towards tool electrode has been reported in some studies.30, 31 Figure 13 shows the EDS spectrum obtained over wire electrode surface after WEDP processing. It can be observed that wire electrode exhibits the presence of Ti, Al, and V elements. The weight percentage compositions of Ti, Al, and V are 0.45%, 0.1%, and 0.07%, respectively. Thus, the elemental weight percent composition of each elements over wire electrode is minimal indicating negligible diffusion/migration of atoms from SLMed alloy surface.
EDS spectrum obtained from wire electrode surface after WEDP.
Conclusions
The present study explored the changes occurring to the SLMed TiAl alloy surface after WEDP post-processing. The surface irregularities as well as defects such as pits, voids, and so on, present over the as-built surface got eliminated through WEDP post-processing. More than 90% reduction in roughness can be achieved at lower settings of pulse on time which yields sub-micron surface finish. The lower Svk values are beneficial in limiting the risk of stress concentration and chances of corrosion in practical applications. The thickness of recast layer formed in WEDP is minimal in comparison with normal rough cut operation owing to the lower discharge energy associated with the process. Moreover, smooth surface morphology with minimal deposition of debris particles is achieved at lower settings of peak current. EDS analysis revealed that the zinc-coated wire electrode exhibited minimal amount of electrode material migration towards WEDP processed surface as compared to the normal brass electrode. Thus, coated wires are recommended for WEDP process. The lower wear of wire electrode during WEDP opens up the possibility to employ lower feed rates which can ultimately minimize the electrode consumption during the process. Furthermore, migration/diffusion of atoms from SLMed alloy surface occured during WEDP post-processing is negligible. In short, a favourable surface modification can be accomplished on TiAl alloy using WEDP which establishes the potential of the method in enhancing the industrial acceptability of metal AM components.
Footnotes
Acknowledgment
The authors are thankful to the staff in Central Facility for Materials and Manufacturing (CFMM) and Central Instrumentation Facility (CIF) at IIT Palakkad for their support during fabrication and material characterization. The authors are also grateful to Dr. D. Chakradhar for the support received in using Wire EDM facility at IIT Palakkad.
Data availability
All the data associated with the study is presented in the manuscript itself.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship and/or publication of this article.
