Structure and Proper es of Layered Ti-6Al-4V-Based Materials Fabricated Using Blended Elemental Powder Metallurgy

Due to the high specific strength of Ti, materials on its base are indispensable when high-strength and low-weight requests are a chief demand from the industry. Reinforcement of Ti-alloys with hard and light particles of TiC and TiB is a credible pathway to make metal matrix composites (MMC) with enhanced elastic moduli without compromising the material’s low-weight. However, reinforcement of the alloy with hard particles inevitably lowers the value of toughness and plasticity of material. Yet, in many applications simultaneous high hardness and high plasticity are not required through the entire structure. For instance, parts that need enhanced wear resistance or resistance upon ballistic impact demand high hardness and strength at the surface, whereas their core necessitates rather high toughness and ductility. Such combination of mechanical properties can be achieved on layered structures joining two and more layers of different materials with different chemical composition and/or microstructure within each individual layer. Multi-layered structures of Ti-6Al-4V alloy and its metal-matrix composites (MMC) with 5 and10% (vol.) of TiC and TiB were fabricated in this study using blended elemental powder metallurgy (BEPM) of hydrogenated Ti. Postsintering hot deformation and annealing were sometimes also employed to improve the microstructure and properties. Structure of materials were characterized using light optical microscopy, scanning electron microscopy, electron backscattered diffraction, x-ray microscopy, tensile and 3-point flexural tests. The effect of various fabrication parameters was investigated to achieve desirable microstructure and properties of layered materials. Using optimized processing parameters, relatively large multilayered plates were made via BEPM and demonstrate superior antiballistic performance compared to the equally sized uniform Ti-6Al-4V plates fabricated by traditional ingot and wrought technology.


Introduction
The anti-ballistic protection of land systems, mobility and protection of the fighting vehicles and military personnel is vital in success of defense and anti-terrorist operations. Traditional material for armor is rolled homogeneous or high strength steel [1,2]. Use of steel armor, however, can increase the overall weight of the fighting vehicle on 15-20%, which causes change in vehicle mobility, maneuverability, fuel efficiency and requires stronger breaks and more powerful engines [3]. The Army is in search of lightweight substitute for steel armor. Due to the high specific strength of titanium, materials on its base are contemplated as a viable alternative in low-weight armor production [3,4]. However, when the armor parts are fabricated using traditional ingot and wrought technology the feasibility of implementation is questionable due to the high cost of the armor parts. In this regard the use of blended elemental powder metallurgy (BEPM) of Ti offers an effective cost reduction due to its ability to produce near-net shape parts, while the waste is considerably reduced [5]. Besides, BEPM can facilitate reinforcement of Ti-alloys with hard and light particles of TiC and TiB to make metal matrix composites (MMC) with enhanced elastic moduli without compromising the material's low-weight so desirable for armor. Reinforcement of the alloy with hard particles, however, most likely lowers the value of toughness and plasticity of material. Yet, in many applications, including armor, simultaneous high hardness and high plasticity are not essential through the entire structure: armor parts require high hardness and strength at the surface, whereas their core rather necessitates high toughness and ductility. Such combination of mechanical properties can be achieved on layered structures joining two and more layers of different materials with different chemical composition and/or microstructure within each individual layer. The objective of this study was development of cost-efficient technology for fabrication of low-weight and superior antiballistic properties multilayered structures (ML) made using BEPM of Ti-6Al-4V (wt.%) alloy (Ti-64) and composites on its base with TiB and TiC.

Materials and Experiments
Combinations of high strength and enhanced ductility for anti-ballistic applications can be attained in the same part by creation of layered structures, which combine at least two materials with different properties. In manufacturing of armor, it is common to use a layered material consisting of a front-facing layer, whose purpose is to blunt and abrade the incoming projectile and a back-facing supporting layer absorbing energy by deformation. In our armor design we choose to combine layers of Ti-64, which provide sufficient ductility, along with metal matrix composites (MMC) on a Ti-64 base reinforced with TiC or TiB particles to provide high hardness. Layers with higher concentrations of reinforcement particles have greater hardness. These high hardness layers build the front of the armor plate, to enhance wear resistance and to deform and stop projectiles upon ballistic impact. The Ti-64 alloy layer with higher ductility and toughness forms the backside of the plate to prevent crack propagation and plate destruction. For ML structures with more than two layers, we performed different tests with having the hardest layer in the middle of the plate to prevent extensive spallation of this most brittle layer and in the front of the plate to maximize the abrasion of the projectile.
ML structures of Ti-64 alloy and its MMC with 5 and 10% (vol.) of TiC and TiB were fabricated using BEPM of hydrogenated Ti. Four different powder sets of titanium hydride TiH 2 (3.5 %H, wt.) with different particles sizes (<40, <100, 80-100, 100-125 µm) was used to determine the effect of the base powder size on shrinkage behavior and sintering characteristics. TiH 2 powder was blended with corresponding amounts of a master alloy powder, 60%Al-40%V (wt.), with particles size smaller than 63 μm to form the required Ti-64 composition. To form the MMC layers the required amounts of reinforcement particles powders were added and mixed before the pressing stage. The size of TiC powder particles used was 1-30 μm. In order to obtain TiB inclusions as a part of the composite, we used TiB 2 powder with 1-20 μm size particles expected to chemically transform during the sintering via the following reaction: TiB 2 +Ti=2TiB. For the ML structures, the blends for each layer were prepared separately and added to the die before pressing. The die pressing (DP) and sintering technique were used to fabricate the samples. Plates, with the size 90×90×18 mm were pressed at 150 MPa; bars 65×10×10 mm and cylinders 10 mm diameters, 10-12 mm height were pressed at 150 and 640 MPa. Plates were suitable for ballistic test and bars and cylinders for mechanical tests and structure characterization. Single layer materials were fabricated to test the properties of individual layers. Sintering of all preform samples was conducted in a vacuum furnace (1250 °C, 4h) followed by slow furnace cooling. Postsintering hot deformation and annealing were also tested.
Structures of materials were characterized using light optical microscopy, scanning electron microscopy, electron backscattered diffraction and x-ray microscopy. Mechanical properties of materials have been evaluated using hardness, tensile and 3-point flexural tests. The deformation energy was measured based on engineering stress-strain curves obtained on tensile test. Ballistic test was performed in the NATO certified laboratory at Ivan Chernyakhovsky National University of Defense (Ukraine). The ballistic test conditions: B-32 bullet (caliber 7.62×54 mm, mass 10.4 g), kinetic energy 3500-3800 J. More details on samples fabrication and characterization protocols used in this study are discussed elsewhere [6,7].

Results and Discussion
Light op cal microscopy (Fig 1) and the SEM images obtained on smaller size samples pressed at 640 MPa show the grain size of the alloys is generally below 100 µm. The grain size for the composite layers is commonly smaller (Fig  2). We see nice compac on and good integra on between the layers. EBSD results of TiB composite show presence of three different phases: HCP α-Ti, BCC β-Ti, and orthorombic TiB shown in Fig 2 (c) in red, blue and yellow correspondingly.
All layered samples of large dimension suitable for the ballis c test (plates) demonstrate significant linear shrinkage a�er the sintering. That was result of dehydrogena on of Ti and par cles' sintering. Plates with TiC composite, however, were successfully made using die-pressing protocol. That was due to very close values of shrinkage of the alloy and TiC composites layers. Shrinkage values measured on plates were 15.7, 15.5 and 15.1% for Ti-64, Ti-64+5%TiC and Ti-64+10%TiC layers correspondingly. Insignificant shrinkage mismatch of individual layers didn't cause cracking or distor on of the plates due to the minor interface stress which most likely was compensated by varia on in porosity of adjacent layers. The difference in shrinkage behavior of layered plates with TiC and TiB can be explained by the different ways those two types of par cles evolve within the structures. TiC has very high mel ng point (3140 °C) and doesn't undergo changes during the sintering. However, TiB is a result of in-situ reac on. Dilatometry data (Fig 3) show shrinkage, of TiH 2 powder (9.5-10%) with the temperature raise up to 1200 °C (1). Shrinkage of TiC composite is at close value (3).
For TiB composites (4 and 5) we observe slight elonga on instead of shrinkage above 800 °C where in-situ reac on on Ti monoboride forma ons starts. The shrinkage of TiB composite is smaller at higher content of TiB par cles. We observe two different morphologies of boride par cles: needles and irregular resemble "sea urchin". Images show pores at the base of needles (Fig 4). There is an extensive porosity observed around par cles with the "sea urchin" morphology. Due to the large difference in mutual diffusion mobility of boron and tanium it creates the Kirkendall's porosity effect. The increased porosity of TiB composites causes swelling and reduces shrinkage.
To minimize shrinkage mismatch of individual layers during sintering of ML structures with TiB we tested several experimental parameters. The shrinkage of Ti-based compacts can be strongly affected by the hydrogen presence in powders [8]. We run the test on bi-layered bars to evaluate this effect. It was established that acceptable level of shrinkage mismatch can be reached when TiH 2 powder is used for the composite layer and conven onal CP-Ti powder for the alloy. However, it requires very high compac on pressure, 640 Mpa, to be used, which was not available in our laboratory set up for large-size plates pressing. The powder size is another well-recognized parameter to control the densifica on of sintered products. Our recent study on combined effect of the size and the type of the powder used shows that acceptable shrinkage mismatch can be achieved at 150 MPa with proper selec on of the hydrogenated tanium powder size used [7]. The ML plates with the size suitable for the ballis c test were fabricated using op mized parameters on the powder size and hydrogen content.
Some proper es of studied materials are shown in the Table 1 [10]. It was noted [i.q.] that at higher strain rate the deforma on energy of bi-layered structure with 10% TiC is higher than that of Ti-64 alloy. Also, stress/strain characteris cs and deforma on energy is be�er-quality for ML structures with TiC composites than with TiB ones (Fig. 5). This result implies bi-layered structure of Ti-64//Ti-64+10% TiC can have be�er an -ballis c proper es compared to the uniform single-layer Ti-64 and ML structures with TiB prepared with used powder approach since it capable absorbing higher energy during deforma on.
Results on an ballis c test corroborates with the data on 3-point flexure test (Fig 6). We can see substan ally be�er an -ballis c performance of the layered structures compared to the cast Ti-64 alloy. Plates of about the same thickness, 25-27 mm, were tested in similar condi ons. Bullets of 7.62 caliber shoot the cast plate all the way through and stuck inside of the layered structures without showing damage on the back side of the tested plate. The cast sample on this slide was annealed at 1100 °C for 1 h and cooled with the furnace. That produced a lamellar microstructure [11].
We believe that residual porosity in as-sintered layered structures can have harmful effect on performance of materials ( Table 1). The data show loss of Vickers hardness with increased porosity, even in the presence of highmoduli par cles of TiC and TiB that supposed to contribute to the hardness gain. That is why we a�empted hot plas c deforma on of sintered ML structures for their more complete densifica on aiming in hardness improvement. Ideally such treatment should provide the porosity reduc on without imposing uncontrollable grain growth and layers' delamina on. It is well recognized that the hot deforma on above α+β®β transus temperature (990 °C for Ti-64) significantly so�ens the alloy and eases its deforma on. However, it increases the risk of uncontrollable grain growth in part due to porosity reduc on and gives higher affinity for the alloy's oxida on. For that reason, post-sintering hot plas c deforma on of different structures was conducted at temperatures not far from the beta-transus of Ti-64 alloy. The applied temperature was ranged from 800-900 °C when porosity of the samples was rela vely low and up to 1100 °C for the samples with the highest porosity followed by annealing at 850 o C for 2 hours. We found that the deforma on of layered structures using hot rolling was unsuccessful due to significant disparity in metal flow and degree of accommodated plas c deforma on in Ti-64 alloy and adjacent composite layers. As a result, significant distor on of the samples and delamina on of the layers were observed [6]. We established the hot pressing of the plates were more successful. Such geometry of the load applica on in combina on with the op mized temperature and deforma on parameters provided shape preserva on and integrity of the plates. ML samples with TiC par cles were successfully hot pressed at 1100 °C to a total deforma on degree of 45%, without breaking the integrity between the layers. At such condi ons the deforma on degree of Ti-64 alloy layer in the ML structure reached 65% that resulted in reduc on of total porosity of this layer to less than 1% and decrease in the pore size. The deforma on behavior of the ML plates used MMCs with TiB was fairly different. The hot pressing was performed at the same temperature of 1100 °C and was stopped at 10% of the total deforma on due to the cracks forma on started within the MMC layers. At such degree of total deforma on of the layered plate the layer of Ti-64 alloy was enable of 20 % deforma on and porosity reduc on from 3.6 to 1-1.5%. At such porosity level the grain growth was not cri cal leading to the average grain size of about 150 µm. MMC layers with TiB par cles were barely deformed (no more than 5-10%) and demonstrated minor decrease in porosity.
Thermomechanical processing of the ML structures is s ll ongoing study, but our current results show the poten als on layered structures hardness increase a�er the op mized hot plas c deforma on applied. We expect further improvement of an ballis c performance of ML structures especially with TiC composites. We also believe that an alterna ve approach on applica on of thermomechanical treatment can be used to improve the performance of the layered structures. In such approach op mized thermomechanical treatment could be used separately on each layer followed by layers bonding, i.g., by using diffusion bonding, HIP or fric on welding etc.

Conclusions
Rela vely big size layered structures of Ti-64 alloy and MMC on its base with TiC and TiB were successfully fabricated using BEPM die-pressing and sintering protocol. Sintered materials were characterized with uniform structure and composi on within each layer and complete integra on between the layers. During BEPM processing the shrinkage levels of the base alloy and MMC with TiC are similar, enabling the successful fabrica on of mullayered structures without need for op miza on of the sintering processing parameters for rela vely large plates.
The shrinkage level of MMC reinforced with TiB is significantly different from that of the alloy layer. Effec ve control of shrinkage mismatch between individual layers can be achieved through the careful selec on of base powder size and the use of hydrogenated Ti powders.
Mul -layered as-sintered plates fabricated in the course of this study using BEPM were successfully tested for anballis c applica on. Good an ballis c proper es of ML plates were explained by experimentally measured increase of the flexure stress and flexure strain of the layered structures compared to uniform single layer structures.
We found that hot plas c deforma on of layered structures has poten als on improvement of their mechanical proper es and an ballis c performance. However, the deforma on of the layered structures using hot rolling was unsuccessful due to disparity in the plas c flow of different layers. Hot pressing deforma on at 1100 °C followed by the annealing was successfully applied to reduce the porosity of as-sintered layered structures with TiC MMCs and improve their hardness. The effect of hot pressing and annealing on the proper es of the layered plates with TiB MMCs was less apparent.