EFFECT OF THERMOMECHANICAL PROCESSING AND ALLOY ADDITIONS ON THE EVOLUTION OF MICROSTRUCTURE IN TWO PHASE TITANIUM ALLOYS

Abstract

Two-phase titanium alloys find extensive applications in aerospace, biomedical, architectural, and chemical sectors due to their outstanding mechanical properties, specific strength, corrosion resistance, and bio-compatibility. Through thermomechanical processing, the microstructure can be engineered to achieve desired mechanical properties by transforming it from lamellar to equiaxed or bimodal morphologies. The addition of alloying elements to titanium modifies phase stability, slip/twinning behavior, diffusion characteristics, and strengthening mechanisms, playing a crucial role in attaining the desired microstructures. These factors significantly influence mechanical properties under both ambient and high temperatures, as well as the extent of thermomechanical processing required for globularization and other desirable features. Therefore, it is imperative to investigate unexplored alloy additions that influence the equilibrium phase fractions of α and β, and their effects on the microstructural and mechanical aspects of these two-phase titanium alloys. Recent research has focused on enhancing ductility in alloys with a martensite phase, often achieved by combining α-phase with martensite or achieving finer aspect ratios, as typically observed in additive manufacturing processes with rapid cooling rates. In such alloys, the fraction and morphology of martensite are essential factors in determining mechanical properties. Since titanium alloys typically undergo hot processing to reduce the large as-cast β grain size resulting from low thermal conductivity, it is important to examine the overall effect of hot processing on the evolution of martensite in these alloys. In the present study, the influence of hot rolling in the sub-transus region of Ti64 on microstructure and texture was investigated, with a particular focus on its correlation with martensite formation. As-cast Ti alloys, including V, Fe, and Ge as alloying elements in Ti64, underwent hot rolling and annealing to analyze the effects of alloying on microstructural aspects and martensite formation, as well as their relationship with mechanical properties. This research provides valuable insights into the microstructural and mechanical aspects influenced i by prior deformation or alloying, offering guidance for designing suitable alloys with combinations of α, β, α′ phases, as well as different morphologies, tailored to specific property requirements. The first chapter of this thesis addresses the existing knowledge gap concerning twophase titanium alloys and outlines the motivation behind the present research. The second chapter presents an extensive literature review that delves into the understanding of thermo-mechanical processing and alloy addition on the phases and texture evolution of titanium alloys. It covers key studies and findings in the field. The third chapter describes the materials and methods employed in the current study, offering a comprehensive overview of the experimental procedures and techniques utilized. In the fourth chapter, the effect of rolling temperature within the two-phase region on microstructure and texture evolution is discussed. Contrary to previous reports suggesting that increasing rolling temperature near the β transus temperature weakens crystallographic texture, this study reveals texture strengthening. This phenomenon can be attributed to the variation in phase fraction with temperature and the simultaneous involvement of slip activity, recovery, and phase transformation during deformation. The hot rolling process leads to texture strengthening in both phases. The α-phase texture evolves towards {1 1 2 0}⟨1 0 1 0⟩ (referred to as T-texture) and {0 1 1 5}⟨3 4 1 1⟩ (near B-texture), while a cube component {0 0 1}⟨1 0 0⟩ and ⟨1 1 1⟩ ∥ ND fiber form in the deformed β-phase. The fraction of secondary α-phase increases with rolling temperature, but the overall texture remains qualitatively similar across samples rolled at different temperatures. The results suggest that the transformed regions in the hotrolled materials grow in orientations similar to the primary α-phase. Grains closer to the T-texture component have easier activation of multiple slip systems, while grains oriented towards the B-texture have limited slip systems. Consequently, the T-texture grains deform more easily and undergo early restoration during hot rolling. With increasing rolling temperature, the T-texture strengthens compared to the B-texture due to the combined influence of deformation-induced phase transformation and strong preferred orientation during reverse transformation. The internal stored energy, contributed mostly by dislocations, decreases at higher deformation temperatures due to restoration mechanisms. This decrease in dislocation density aligns with the intra-grain ii misorientation distribution. The α-phase grain size increases with deformation temperature, and at intermediate rolling temperatures, a bimodal grain size distribution is observed. Furthermore, the presence of orthorhombic martensite (α”-martensite) in the hot rolled materials is confirmed through XRD patterns and TEM microstructure analysis. These fine martensite laths have a width of 20 nm and exhibit an orientation relationship with the neighboring α-phase. The orientation relationship between the two phases is observed as (1 1 2 0)α ∥ (2 0 0)α′′ and [0 0 0 1]α ∥ [0 1 1]α′′. The formation of orthorhombic martensite instead of HCP α’-martensite is technically important for the hot rolling process as it possesses greater ductility. Chapter 5 focuses on the effects of prior deformation on martensite formation in Ti64 alloy, particularly in terms of microstructure and mechanical properties. To observe the influence of prior deformation, martensite was generated in the previously mentioned hot-rolled materials through heat treatment at temperatures ranging from 880 °C to 940 °C, followed by water quenching. The stored energy variation resulting from changes in hot rolling temperature led to significant microstructural variations during quenching from the two-phase region. Microstructural investigations revealed that prior deformation greatly influences the morphology of the martensite phase, while the phase fraction is primarily dependent on the quenching temperature. Deviations in composition from equilibrium conditions, arising from differences in element diffusion at various rolling temperatures, play a key role in martensite lath nucleation and aspect ratio. Higher quenching temperatures lead to an increased Al content in the primary α-phase and an increased V content in the martensite phase. The addition of alloying elements, especially those stabilizing the β-phase (V), reduces the martensitic start temperature (Ms). A lower V content results in higher Ms temperature and a greater driving force for martensite transformation. Consequently, more nuclei form, leading to smaller aspect ratios of martensite laths at lower V content. Higher stored energy in deformed samples, confirmed through dislocation density calculations, promotes the formation of twinned plate martensite. A strong correlation between martensite phase morphology and the character of intervariant boundaries is established. The distribution of intervariant boundaries in martensite exhibits three major angle-axis pairs: 60°/[1 1 2 0]α′ , iii 60.83°/[1.377 1 2.377 0.359]α′ , and 63.26°/[10 5 5 3]α′ , which are associated with the Burgers orientation relationship. The hardness of the martensite phase is influenced by solid solution strengthening, martensite morphology, and microstrain within the sample. Titanium alloys are known to display strong anisotropy in mechanical properties due to the combination of their HCP structure and strong texture. To reduce this anisotropy, the present study shows a microstructure with multiple martensite orientations, with fine equiaxed β-phase regions among them. The combined process of cold rolling and heat treatment accelerates the martensite decomposition reaction. The study demonstrates that the variations in nanoindentation-derived hardness and modulus are much smaller compared to reported literature data, which can be attributed to the fine nature and morphology of the constituent phases. Chapter 6 focuses on the microstructural evolution during thermomechanical processing of two-phase titanium alloys. To investigate the effect of alloy additions, V, Fe, and Ge elements were incorporated into Ti-6Al-4V, and five alloys were prepared through vacuum arc melting: Ti-6Al-4V, Ti-6Al-5V, Ti-6Al-5V-1Fe, Ti-6Al-5V-1Fe-2Ge, and Ti-6Al-4V-2Ge (in wt.%). After hot rolling, an increase in the β-phase fraction was observed in the microstructures, ranging from less than 10% for Ti-6Al-4V to 50% for Ti-6Al-5V-1Fe-2Ge. Alloy additions, as well as deformation-induced strain accumulation, played significant roles in stabilizing the β-phase. V and Fe act as β-stabilizers, while Ge is considered a neutral element. Interestingly, the addition of Ge to Ti-6Al-4V reduced the β-transus temperature, resulting in a higher β-phase fraction during hot rolling. Consequently, the addition of Ge should lead to lower strain partitioning to the α-phase due to increased β-phase at the rolling temperature. Although Ti-6Al-5V-1Fe and Ti-6Al-5V-1Fe-2Ge had similar α-phase fractions during deformation (51-55%), the Ge-added alloy, Ti-6Al-5V-1Fe-2Ge, exhibited higher local strain compared to Ti-6Al- 5V-1Fe. Subsequent annealing of the hot rolled alloys at 750 °C for 60 min resulted in extended static recovery in the microstructures, accompanied by coarsening of secondary α within the β-phase. On the other hand, annealing at a higher temperature, 50 °C below the β-transus temperature of these alloys, led to recrystallization facilitated by phase transformation, resulting in the formation of fine secondary α within the β-phase. The secondary α-phase formed during annealing exhibited a strong Burgers orientation iv relationship (BOR). The number fraction of correlated misorientation significantly increased at 10°, 60°, and 90° misorientation angles, which correspond to typical rotation angles of α-variants with BOR. The fine secondary α formed from the β-phase during high-temperature annealing exhibited a very strong variant selection, resulting in only a few variants within each β-grain. The average grain size of Ti-6Al-4V and Ti-6Al-5V alloys increased considerably after hot rolling and annealing, while the presence of germanide precipitation restricted grain growth in Ge-added alloys. The slip activities in these alloys were investigated during low-strain (5%) room temperature rolling in the α-phase. Basal and prismatic slip activities were typically observed in Ti-6Al-4V, but with Ge addition, intense basal slip activity was noticed. V and Fe additions promoted enhanced pyramidal slip activity. Ge-added alloys exhibited a strong basal texture due to dominant basal slip activity, while V and Fe-added alloys displayed a combination of B and T-texture components due to increased contributions from prismatic slip. The observed slip activity was attributed to a reduction in the difference between critical resolved shear stresses among the basal, prismatic, and pyramidal slip systems. Chapter 7 focuses on understanding the formation of martensite in two-phase titanium alloys with alloy additions. The five hot rolled titanium alloys were solution heat treated and water quenched from 50°C above the β-transus temperature. The morphology and mechanical properties of these alloys were investigated. The evolution of martensite is influenced by the alloy compositions, specifically by the changes in the martensitic start temperature (Ms) and the driving force. The addition of alloying elements results in a decrease in lamellae width of martensite. Notably, there is a strong correlation between lath size and volumetric strain. The Ti-6Al-5V-1Fe-2Ge alloy exhibits fine martensite laths with lamellae widths of 100-200 nm, along with the presence of retained β-phase. Higher β-stabilizing elements suppress the complete conversion of β to α’-martensite and promote the formation of stable β-phase at ambient conditions. The characterization of intervariant boundaries in martensite reveals three major angle-axis pairs associated with the Burgers orientation relationship, which aligns with the findings discussed in Chapter 5. The distribution of grain boundary planes in the formed martensite remains similar, regardless of the alloy additions. The martensite laths exhibit pronounced stacking faults, resulting in local structural changes from HCP v to FCC. Micropillar compression testing of martensitic alloys shows an increase in fracture strain with alloy additions, which is attributed to the finer aspect ratio of martensite. Finer laths with multiple orientations impede the continuous propagation of microcracks along similar crystallographic orientations, leading to a decrease in effective crack length and higher fracture strain. With increasing the temperature from ambient conditions to 400°C, the hardness values decrease by 20-28% for most of the studied alloys, while Ti-6Al-5V-1Fe-2Ge retains nearly constant hardness and it is dependent on solid solution strengthening both at room temperature and high temperature. In conclusion, Chapter 8 provides a comprehensive summary of the research findings and offers valuable recommendations for future studies. It highlights the importance of further research to advance our knowledge of two-phase titanium alloys and to enhance their mechanical properties for a wide range of applications.