Effect of Stress Relieving Heat Treatment on the Microstructure and High-Temperature Compressive Deformation Behavior of Ti-6Al-4V Alloy Manufactured by Selective Laser Melting
- 287 Downloads
This study aims to investigate the effect of stress relieving heat treatment on the microstructure and high-temperature compressive deformation behavior of the Ti-6Al-4V alloy, manufactured by selective laser melting. Initial microstructural observation confirmed elongated prior β grains in the building direction of both specimens (as-fabricated and heat-treated specimens). Along with such, the as-fabricated specimen only featured α′-martensite phase, while the heat-treated specimen featured α′-martensite and some α and β phases. Compression tests carried out at room temperature gave yield strengths of 1365 and 1138 MPa for the as-fabricated and heat-treated specimens, respectively. Such values are similar or greater than those of commercial wrought materials. The compressive fracture strain significantly increased after heat treatment. There was a general tendency of reducing yield strength as compressive temperatures increased. At temperatures greater than 700 °C, the as-fabricated and heat-treated specimens achieved similar strength. Microstructural observation after deformation confirmed that the initial microstructure was retained up to temperatures of 500 °C. At 700 °C or greater, both specimens showed drastic microstructural evolution.
Powder-bed additive manufacturing (AM) is an innovative technology that builds three-dimensional parts based on continuous powder layering based on a computer-aided design model.[1,2] Compared to conventional manufacturing processes (e.g., casting, hot or cold working, or machining), the AM model has the advantage of fewer production stages and reduced material wastage. Complex parts are also processed easily with an increased flexibility of design.[3, 4, 5]
AM has more than one metal forming process. Selective laser melting (SLM) is the most widely used process. It is a powder bed fusion method where powdered metal is applied and a high-power laser is used for selective melting to manufacture complex and elaborate free-form parts.[6,7] When compared to other metal AM processes, such as direct energy deposition and e-beam melting, SLM is characterized by a relatively smaller layer thickness and fast cooling rate (105 to 108 K/s).[8,9] Due to such unique characteristics, SLM can manufacture parts with excellent dimensional accuracy and fine microstructure. The dislocation density increases during rapid solidification, resulting in increased hardness and strength.
Among many alloys, Ti-6Al-4V (Ti64) alloy is one of the most commonly used materials in the Ti market and it has a variety of advantages such as outstanding specific strength, corrosion resistance, and bioaffinity.[11, 12, 13] With post–heat treatment of α + β phases, we can easily improve the microstructure, making it highly demanded by aerospace, medicine, and automotive industries and the military.[14, 15, 16] However, Ti64 alloy has high oxygen affinity, which requires particular control during the manufacturing processes and postprocessing. Indeed, SLM can suppress oxidation and has excellent dimensional accuracy, making it ideal for manufacturing complex Ti64 alloy parts.
Due to the aforementioned advantage, various AM processes, including SLM, are used to manufacture Ti64 alloy, and multiple studies, including microstructure and mechanical properties, have been conducted to allow its application in various fields. In particular, microstructure control through process variables, process types, layering direction, and heat treatment[17, 18, 19] and various mechanical properties, including the tensile deformation of each building direction (BD), high-cycle fatigue, and fracture toughness, were reported.[20, 21, 22] Also, the authors recently reported heat treatment for improving high-temperature creep properties. Indeed, it is commonly regarded that SLM Ti64 alloy forms full martensite structures due to its fast cooling rate. Such a martensite structure is highly brittle, which limits its application in many structural applications; therefore, heat treatment is beneficial in improving toughness and ductility. However, despite the previous studies, to the authors’ knowledge, there is a lack of sufficient studies on the high-temperature deformation behavior of SLM parts.
The goal of this study is to investigate the influence of heat treatment on the microstructure and compressive properties of SLM Ti64 alloy. The study also aims at analyzing the deformed microstructure and surface after compression, which varies according to temperature to identify the high-temperature deformation/fracture mechanism of SLM Ti64 alloy.
2 Experimental Method
This study used Ti-6Al-4V (Ti64) manufactured by EOS. The specimen had BD, transverse direction (TD), and scanning direction (SD) sizes measuring 74.5, 12, and 7 mm, respectively. In addition, in the present study, a unidirectional scanning strategy was adopted. The chemical composition (wt pct) analysis of Ti64 alloy manufactured by SLM implemented using an X-ray fluorescence spectrometer (ZSX Primus II) confirmed Ti as well as 5.50Al, 3.87V, and 0.22Fe. Also, to identify the influence of stress-relieving heat treatment on the microstructure and mechanical property of the preceding materials, SLM Ti64 alloy was heat treated (stress relieving) in an Ar atmosphere at 650 °C for 3 hours and then cooled under furnace. The heating rate for the process was set at 10 °C/min.
To evaluate the room- and high-temperature compressive properties of SLM Ti64 alloy, a cylindrical compression sample in the BD direction was produced with a size of Φ 4-mm o.d. × 6-mm height. The compression tests were performed using MTS-810 equipment at room temperature, 500 °C, 700 °C, and 900 °C, and tests were conducted until total strain (εtot) = 0.6 with an initial strain rate 1 × 10−3/s. The compression tests were conducted 3 times for each datum point. Hardness measurement was conducted using a Vickers hardness tester (Mitutoyo HV-100) at constant load 2 Kgf and holding time 10 seconds. A total of 12 measurements were made, and the average was calculated.
To analyze the phase transformations of SLM Ti64 alloy, an X-ray diffractometer (XRD, Ultima IV) was used. Further, to observe the microstructure before and after heat treatment, the specimens were cut at a vertical angle in the TD and then mechanically polished using silicon carbide papers (#100 to #2000) and diamond suspension of 1 μm. The specimen was then etched in 50 mL H2O + 25 mL HNO3 + 5 mL HF solution before being observed using an optical microscope (OLYMPUS BX53M) and scanning electron microscopy (SEM, Tescan VEGA II LMU). Internal grain orientation and disorientation angle distribution were analyzed using electron backscatter diffraction (EBSD, Nordlys nano detector, binning: 4 × 4) analysis. The specimen was polished using silicon carbide paper and diamond suspension and was polished further with 0.01-μm colloidal silica before observation. EBSD observation was made at a step size of 100 nm and 15 kV, and the data obtained were analyzed using an imaging microscopy software (AZTecHKL) program. SEM and field-emission–SEM (Tescan MYRA 3 XMH) equipment were used to observe the compression specimen surface and microstructure after deformation, to identify deformation and fracture mechanisms according to temperature, after high-temperature compression.
3 Results and Discussion
3.1 Microstructures of SLM Ti-6Al-4V Alloys
Meanwhile, when the focus was shifted to the interior microstructure of columnar grains, fine acicular type structures were observed. These structures are suspected to be α′-martensite, and in general, Ti64 alloy, which has a low β-stabilizing element concentration, is known to form α′-martensite with a distorted hexagonal crystal lattice during rapid solidification. In other words, it is suspected to be the result of the insufficient time for β phase into the stable α + β phase. Becker et al. reported that the SLM process has a narrow layer thickness, which causes fast cooling rates, and this, in turn, results in the formation of fully martensite structures.
3.2 Room-Temperature Hardness Properties
3.3 Room and High-Temperature Compressive Properties
The room-temperature characteristics of the two specimens used in this study were compared to commercial cast and wrought materials and the other SLM Ti64 alloys.[21,33,34] The as-fabricated and heat-treated specimens achieved 200 MPa or greater yield strength than cast materials and similar or slightly greater strength than wrought materials. The findings on SLM Ti64 alloy in this study were within the yield strength range previously reported. In particular, the as-fabricated specimen exhibited similar or greater yield strength compared to previous SLM materials. The SLM Ti64 alloy achieved greater strength due to the fast cooling rate that occurred during the melting and solidifying of initial solid powder. In other words, the as-fabricated specimen formed fine martensite structures according to the aforementioned fast cooling rate (approximately 105 to 108 K/s)[8,9] that achieved greater strength.
The yield strength at various compression temperatures is shown in Figure 8(b). The yield strengths of the as-fabricated specimen tested at 500 °C, 700 °C, and 900 °C are 930 ± 9, 242 ± 8, and 45 ± 4 MPa and those of the heat-treated specimen are 600 ± 5, 246 ± 5, and 47 ± 3 MPa, respectively. The as-fabricated specimen achieved greater yield strength at room temperature up to 500 °C. However, in the 700 °C to 900 °C range, a drastic yield strength decrease occurred and the strength difference between the two specimens nearly disappeared. This drastic yield strength decrease at high temperature is assumed to be due to the extreme microstructural change around 700 °C. In general, the β recrystallization temperature of Ti-6Al-4V (manufactured by other conventional processes) is known to be approximately 800 °C to 850 °C, but there are incidents of dynamic recrystallization (DRX) occurring during high-temperature deformation (approximately 650 °C to 750 °C) reported. Details of such shall be discussed along with the microstructural observation after deformation in Section III–D.
3.4 Fracture and Compressive Deformation Behaviors
Initial microstructural observation confirmed that prior β grain elongated in the direction where heat is discharged and acicular type α′-martensite structures were formed within the β grain. Also, XRD and misorientation angle distribution analysis confirmed that phase decomposition of some α′ and dislocation density decreased after stress relieving heat treatment
Room-temperature hardness anisotropy appeared in the as-fabricated specimen; on the other hand, the average hardness decreased by 10 pct, and at the same time, the hardness anisotropy disappeared in the heat-treated specimen. Room- and high-temperature compression tests confirmed that the room-temperature yield strength was approximately 1365 MPa for the as-fabricated specimen and 1138 MPa for the heat-treated specimen, which is similar to or greater than those of wrought materials. The as-fabricated specimen had greater yield strength than the heat-treated specimen up to the 500 °C condition, but at temperatures of 700 °C or higher, both specimens showed a drastic strength decrease and differences in mechanical properties disappeared.
Surface observation after the room-temperature compression test confirmed that the as-fabricated specimen failed to accommodate most of the deformation, resulting in fractures, while the heat-treated specimen accommodated most of the deformation. At temperatures of 500 °C or higher, both specimens showed ductile characteristics where fractures or shear cracks were not formed. Meanwhile, for room-temperature to 500 °C compression specimens, which did have differences in mechanical properties, the microstructure after the compression test was similar to that of the initial microstructure. The martensite structure was completely decomposed in the case of the specimens compressed at temperatures of 700 °C or higher with similar mechanical properties
This research was supported by the Korean Institute of Materials Science, Korea.
- 10.M. Simonelli, Y.Y. Tse, and C. Tuck: Annual International Solid Freeform Fabrication Symposium, University of Texas, Austin, TX, 2012.Google Scholar
- 11.R.A. Wood: Titanium Alloy Handbook, Metals and Ceramics Information Center, Battelle, Columbus, OH, Dec 1972, publication no. MCIC-HB-02.Google Scholar
- 13.M.K. Mcquillan: J. Metall. Rev., 1963, vol. 8, pp. 41–104.Google Scholar
- 19.G.M.T. Haar and T.H. Becker: Materials, 2018, vol. 11(1), p. 146.Google Scholar
- 26.T. Becker, M.V. Rooyen, and D. Dimitrov: S. Afr. J. Ind. Eng., 2015, vol. 26, pp. 93–103.Google Scholar
- 35.M. Vanderhasten, L. Rabet, and B. Verlinden: Metalurgija, 2005, vol. 11, pp. 195–200.Google Scholar