As a function of the complex thermomechanical processing history that preceded it, the microstructure and microtexure in the final application of an α+β titanium alloy governs the mechanical properties it exhibits. Crucial during the early stages of processing α + β titanium alloys are the α + β pre-strain and β-recrystallisation processes which are carried out to homogenise both chemistry and morphology. The β grain structure generated during the β-recrystallisation process transforms to a lamellar α + β microstructure which is subsequently broken down during secondary processing. Large effective structural units of similarly aligned laths generated during the cooling from the β-recrystallisation process potentially persist through this secondary processing to the final application, reducing the lower mechanical property bounds of the alloy. To investigate the effect of strain, strain rate, pre-strain temperature, β-recrystallisation temperature and time at β-recrystallisation temperature on final β grain size several large scale Ti-6Al-4V and Timetal 575 double cone samples were forged, sectioned and β-heat treated at the Advanced Forming Research Centre, Glasgow. Using industrially relevant parameters and intermediate stage billet material provided by TIMET enabled the generation of a comprehensive data set of final β grain sizes as a function of the five input parameters. Micro and macrostructural investigation of the as received, as-forged and final β-recrystallised state comprehensively characterised the microstructural evolution throughout the process. To determine the minimum effective strain required to initiate dynamic recrystallisation of the β phase in α + β phase field deformation backscatter electron imaging, electron backscatter diffraction and β reconstruction techniques were employed. Very low effective strains (0.1 6 ε) were required to initiate and complete the β recrystallisation of early intermediate, β-workout Ti-6Al-4V material with unsaturated grain growth continuing at extended times at temperature. Despite the coarse, heterogeneous starting material very low α + β effective strain and short times above the β-transus temperature were enough to produce a homogeneous, equiaxed microstructure devoid of the gradient of effective strain produced by the double cone sample geometry. Due to using a more refined Ti-575 starting microstructure and because of slower phase transformation and grain growth kinetics this gradient of effective strain was more apparent in the final β-recrystallised state of Ti-575 with reduced grain size coinciding with an increase in α + β pre-strain. Ultimately, the fine colony basket-weave microstructure generated during the water quench from the industrial β recrystallisation process produced in Ti-575 an effective structural unit size an order of magnitude lower than in Ti-6Al-4V.
|Date of Award||17 Oct 2016|
- University Of Strathclyde
|Sponsors||Timet UK Ltd|
|Supervisor||Paul Blackwell (Supervisor) & William Ion (Supervisor)|