1. Introduction
Currently, additive manufacturing (AM) techniques are among the most dynamic and rapidly developing methods for the production of detailed components with complex geometries. Several AM techniques can be distinguished based on the batch material (polymers or metal alloys) used [
1,
2].
A quite popular additive manufacturing technique is laser engineering net shaping (LENS). LENS was developed and commercialized in 1997 by Optomec of Sandia National Laboratories [
3,
4]. With this technique, it is possible to reduce time and costs in the production of elements by significantly shortening the entire process, from design to manufacturing. The components manufactured from LENS devices are composed of metallic powder that is supplied directly into a location where a high-power laser beam is focused under a protective atmosphere of argon gas. In a layer-by-layer manner, near-net-shaped components are created with a dimensional accuracy comparably to the CAD model. All of the parameters, such as the amount of fed powder, power of the laser, feed rate of the working table and thickness of the layer, are precisely defined and fully controlled throughout the process [
5]. The influences of these parameters on the mechanical properties and microstructures have been investigated [
5,
6,
7]. These advantages result in a wide variety of applications, from regeneration of damaged parts to small batch manufacturing of personalized components.
There is a very wide range of batch materials available for use in LENS devices, and the most commonly used options are commercial engineering materials, such as steels, nickel-based super alloys and titanium and its alloys. A distinctive result of using the LENS technique to produce these materials is often the formation of nonequilibrium structures or additional phases that are impossible to obtain by conventional methods, which in turn leads to the generation of unique properties. This distinction arises due to the high cooling rate in the molten pool during the manufacturing process. A cooling rate of approximately 7 × 10
4 K/s in the direct energy deposition (DED) process, which encompasses the LENS technique, was reported by Qian et al. [
8].
Very popular materials of interest to researchers working in the field of AM techniques are titanium and its alloys. The alloy tested in this study is a lightweight commercial Ti6Al4V titanium alloy that has a number of very beneficial properties, such as a high specific strength, high melting point and low density. In addition, the very good corrosion resistance and biocompatibility of this alloy allow it to be widely used in the field of medical implants [
9,
10,
11]. All of these advantages have led to its application in the medical and aerospace industry. Pure titanium has two allotropic varieties: low-temperature Ti-α crystals in a hexagonal close-packed (HCP) system and high-temperature Ti-β crystals in a body-centered cubic (BCC) system. The chemical composition of the Ti6Al4V alloy includes alloying elements—aluminum and vanadium—which stabilize the apparent α and β phases. For this reason, the structure of the alloy at room temperature consists of an α + β mixture. The nature of a two-phase structure can be very different from that of an equiaxed structure, ranging from bimodal to lamellar. The sizes, distributions and morphologies of these two phases strongly depend on the cooling rates during manufacturing [
5,
12]. According to a diagram illustrating the dependence of the phase transformations on the cooling rate for Ti6Al4V alloys [
13], if the cooling rate is sufficiently high (>410 °C/s), the β phase transforms through a diffusionless process into a martensitic α’ phase. The formation of the α’ phase within long columnar prior β grains occurs primarily during the AM-based production of components composed of the Ti6Al4V alloy [
14]. This formed α’ phase is the result of the fulfillment of the second condition leading to this result, which states that the temperature during the building of the elements must be higher than the martensitic transformation temperature (Ms). The temperature value depends on the amount and type of impurities present in the chemical composition of the alloy, and it ranges from approximately 575 °C to 800 °C [
15,
16]. The presence of the martensitic phase in the Ti6Al4V alloy plays a major role in the strength parameters. The yield point or tensile strength increases as both the elongation and plasticity decrease [
17,
18]. Galarraga et al. [
17] showed that the presence of α’ needles increases the hardness of the material.
To improve the ductility, postprocessing treatment is necessary for the Ti6Al4V alloy-based components produced by AM techniques. The most popular postprocessing technique is heat treatment [
19]. This treatment mainly results in structural changes that alter the material properties. For the Ti6Al4V alloy, according to a pseudobinary phase diagram, the α→β transition temperature is approximately 980 °C [
20]; therefore, for the decomposition of the α’ phase to occur during heat treatment, the process should be carried out at a similar temperature to that mentioned above. Yuan et al. [
21] investigated the heat treatment effect (annealing at three different temperatures: 750 °C, 850 °C and 950 °C) on the compressive fatigue properties of titanium samples manufactured by the EBM method. By comparing the obtained results, the scholars found that annealing at 950 °C leads to the formation of a two-phase structure and significantly improves the ductility and fatigue strength. Kim et al. [
22] studied Ti6Al4V samples produced by the SLM method and reported that samples subjected to heat treatment at 1040 °C for 1 h exhibit a lower steady-state creep rate than the as-built specimens. Moreover, the impacts of postprocessing heat treatment on the corrosion resistance and biocompatibility of the Ti6Al4V alloy are interesting. Reportedly, annealing the samples produced from the alloy at a temperature of 800 °C for 2 h significantly improves their corrosion resistance [
23]. Conversely, heat treatment at a temperature of 820 °C for 4 h for SLM-based samples improves human bone mesenchymal stem cell (hBMSC) adhesion and proliferation [
24].
Another example of a postprocessing heat treatment is hot isostatic pressing (HIP), where a high annealing temperature and pressure are applied simultaneously. This technique is used mainly for powder materials, and some related research has been conducted for Ti6Al4V alloys [
25,
26,
27,
28]. However, there is another route involving the use of the HIP technique as a postprocessing treatment for bulk samples produced by AM. Qiu et al. [
29] studied the effects of using HIP (920 °C/103 MPa/4 h) on the changes in the porosity and microstructure that may occur during an experiment in SLM-based samples composed of Ti6Al4V powder. HIP is confirmed to completely transform martensite into a two-phase α + β structure and to close almost all of the pores created during sample manufacturing. Both the phenomena described above and their impacts on increasing the plasticity and fatigue endurance ratio parameters of the obtained materials have been described in previous papers [
30,
31,
32]. The structural changes occurring in the Ti6Al4V alloy presented in the literature mainly concern the SLM process. Given the significant sensitivity of this material to technological history, it is necessary to investigate the changes occurring during the additive manufacturing process and subsequent thermal treatment.
In a recent paper, the authors attempted to comprehensively analyze results obtained from the use of two postprocessing heat treatments. The treatments—classic annealing and hot isostatic pressing—were carried out at two different temperatures. The researchers bore in mind that the use of a postprocessing heat treatment is necessary for samples produced with the Ti6Al4V alloy by AM techniques due to the microstructure formed during the manufacturing process. The mechanical properties were compared—microhardness, corrosion resistance and, primarily, the newly created microstructure α + β, which naturally has a large impact on the above properties.
2. Materials and Methods
2.1. Manufacturing of the Samples
The Ti6Al4V samples were manufactured using an Optomec LENS MR-7 device (Albuquerque, NM, USA). The scheme of the laser engineered net shaping system has been described in other papers [
33,
34]. The process of sample production by this method consisted of the layer-by-layer building of the elements previously designed by a CAD program from metallic powders, which were melted by a 500-W IPG YLR-500 fiber laser (IPG Photonics Corporation, Oxford, MS, USA). The working chamber was filled with an ultraclean argon atmosphere, and the oxygen content was controlled inside. For research purposes, cylindrical samples with dimensions of 10 mm × 100 mm (diameter × height) were built on a 7-millimeter-thick Ti6Al4V substrate plate, which was sandblasted and cleaned with acetone. The zig-zag printing strategy for a single layer was used. Each of the built layers was rotated by 30 degrees compared to the previous one without any dwell time between them. The manufactured cylinders were removed from the substrate using a BP-97d electrodischarge machine (ZAP-BP, Kutno, Poland). The parameters of the manufacturing process, such as the laser power and powder flow rate or oxygen content, ensured good metallurgical and dimensional quality of the produced samples. These parameters are presented in
Table 1.
The batch material used for building the cylindrical samples was Ti6Al4V powder supplied by the TLS Technik GmBH & Co Company (Bitterfeld-Wolfen, Germany). The chemical composition of the powder was confirmed by a certificate provided by the above company, and the content of individual elements are shown in
Table 2. This powder was made using an argon atomization method, which determined the spherical shape of the particles, and the sizes of the particles ranged from 45 to 105 µm. These findings were confirmed by scanning electron microscopy (SEM) observations (
Figure 1a), which revealed that the surfaces of the particles were mostly smooth, and that small particles were attached to large particles. The cross-sectional micrograph (
Figure 1b) revealed that some small pores were present inside the particles.
To observe the structural changes occurring in the material, the cylindrical samples produced by the LENS technique were subjected to two types of postprocessing treatments (
Table 3).
Annealing after processing was conducted at 950 °C and at 1050 °C for 30 min for both treatments. The process was conducted in a Nabertherm R80/b750/12-B170 tubular furnace (Nabertherm GmbH, Lilienthal, Germany) at low vacuum (×10
−2 mbar), and before heating, the tube was purged with argon. The samples were cooled in the furnace. The conditions of the postprocessing heat treatment were selected based on our experience and on the available literature [
20,
35,
36,
37].
The second type of postprocessing heat treatment of the samples manufactured by the LENS technique was hot isostatic pressing (HIP). The temperature and duration of the HIP process were the same as those used for the pressure-free heat treatment, i.e., 950 °C for 30 min and 1050 °C for 30 min. The pressure in the working chamber was 300 MPa, and a protective pure argon (99.999%) atmosphere was applied. The specimens were cooled in the HIP chamber.
For comparison, a Ti6Al4V reference sample was taken from a commercially manufactured hip joint endoprosthesis.
2.2. Research Methods
Samples obtained during the LENS process and after two different postprocessing heat treatments were subjected to various material testing methods that are commonly used in materials engineering, in order to characterize the microstructure, mechanical properties and corrosion resistance of the manufactured materials.
The specimens for metallographic observation were taken from the centers of cylindrical samples in the “z” plane parallel to the deposition direction during the LENS process (
Figure 2). Afterward, the samples were mounted in thermosetting resin, ground with 600-mesh, 1200-mesh and 2400-mesh grinding papers, and their surfaces were chemically polished using MD-Chem discs with 0.25% colloidal silica in conjunction with 30% hydrogen peroxide. To reveal their microstructures, the specimens were etched with Kroll’s reagent.
The samples were prepared in this manner and subjected to microscopic observation. A Nikon (Tokyo, Japan) MA-200 optical microscope was used for macrostructural and microstructural observations. Additionally, the microstructures, fracture surfaces, chemical compositions and quantitative phases were investigated using an FEI Quanta 3D (Eindhoven, Netherlands) dual-beam field emission gun scanning electron microscope equipped with energy dispersive spectroscopy (EDS) and electron backscatter diffraction (EBSD).
The total porosity fraction in the full volume of each specimen taken from the center of cylindrical samples was determined using a Nikon MA-200 optical microscope equipped with an NIS-Elements BR 3.8 computer image analyzer. The degree of porosity for each sample was determined as the ratio of the measured pore area to the entire observed area. The analysis was carried out in both directions, taking into account statistically selected regions observed at magnification 100×, which resulted in thirty examined areas. The mean values and standard deviations were calculated for all obtained porosity results of each specimen.
The mechanical properties of the samples manufactured by the LENS technique directly after processing and after pressure-free and HIP heat treatments were identified using an INSTRON (Norwood, MA, USA) 8862 tensile testing machine with extensometer 2630-107 at a crosshead speed of 1 mm/min according to ISO 6892–1:2019 [
38]. The tensile tests were performed on three identical samples under each condition. The dimensions of the tensile samples are shown in
Figure 3. The yield point, tensile strength and elongation were estimated from the original stress-strain curves.
The Vickers microhardness measurements were performed using a Shimadzu microhardness tester under a load of 100 g for 10 s. The average microhardness values were calculated from at least 10 indentations, and the standard deviation was determined.
The phase compositions of the specimens were studied via X-ray diffraction (XRD) using a Seifert 3003 TT diffractometer (Rich. Seifert & CO. GmbH, Ahrensburg, Germany) with Cu Kα radiation (α = 1.5418 Å). The 2θ range was 20–120 degrees with a step of 0.02 deg and an exposure time of 3 s.
All of the electrochemical experiments were performed with a 0531 Atlas-Sollich potentiostat (Atlas-Sollich Z.S.E., Gdańsk, Poland). The experiments were performed in an electrochemical cell containing 100 mL of 0.9% NaCl, where the investigated samples were working electrodes, a platinum plate was the counter electrode and Ag|AgCl|sat. KCl was used as the reference electrode. The surface area of the investigated samples was limited by the rubber O-ring seal to 0.196 cm2. Prior to polarization, the open cell potential (OCP) was recorded for 1 h. Immediately after this measurement, a linear potential scan was performed at a 1 mV/s rate, which started 100 mV below the last OCP value. The polarization curves were recorded to 1.2 V vs. Ag|AgCl. The corrosion potential and corrosion current density were estimated using Tafel plots. Each experiment was performed three times.