High energy-density beam welding, such as electron beam or laser welding, has found a number of industrial applications for clean, high-integrity welds. The deeply penetrating nature of the joints is enabled by the formation of metal vapour which creates a narrow fusion zone known as a "keyhole". However the formation of the keyhole and the associated keyhole dynamics, when using a moving laser heat source, requires further research as they are not fully understood. Porosity, which is one of a number of process induced phenomena related to the thermal fluid dynamics, can form during beam welding processes. The presence of porosity within a welded structure, inherited from the fusion welding operation, degrades the mechanical properties of components during service such as fatigue life. In this study, a physics-based model for keyhole welding including heat transfer, fluid flow and interfacial interactions has been used to simulate keyhole and porosity formation during laser welding of Ti-6Al-4V titanium alloy. The modelling suggests that keyhole formation and the time taken to achieve keyhole penetration can be predicted, and it is important to consider the thermal fluid flow at the melting front as this dictates the evolution of the fusion zone. Processing 2 induced porosity is significant when the fusion zone is only partially penetrating through the thickness of the material. The modelling results are compared with high speed camera imaging and measurements of porosity from welded samples using X-ray computed tomography, radiography and optical micrographs. These are used to provide a better understanding of the relationship between process parameters, component microstructure and weld integrity.

The effect of high-temperature deformation and the influence of hot working parameters on microstructure evolution during isothermal hot forging of Ti-6Al-4V in the alpha phase field were investigated. A series of hot isothermal axis-symmetric compression tests were carried out at temperatures both low and high in the alpha stability field [(1153 K and 1223 K (880 °C and 950 °C), respectively], using three strain rates (0.01, 0.1 and 1.0/s) relevant to industrial press forging. The microstructures and orientation of the alpha laths were determined using optical microscopy and electron backscatter diffraction techniques. The experimental results show that there is a change in lath morphology of the secondary α phase under the influence of the deformation parameters, and that α lath thickness appears to have little influence on flow behavior.

Inertia friction welding has been used across the aerospace, automotive, and power-generation industries for the fabrication of complex axisymmetric components for over forty years. The process involves one axisymmetric piece being held stationary and another piece being brought into contact set to rotate about its axis of symmetry by a flywheel with the system under an applied load across the joint. Plasticization at the joint interface through the frictional heating sees the two pieces bond together. The titanium alloy Ti-6Al-4V has been widely studied for inertia welding applications. A successful selection of processing parameters (flywheel energy and mass, applied load) allows an inertia welding process which produces a very high-integrity weld, with a minimal heat-affected zone (HAZ) and thermomechanically affected zone (TMAZ), formed as a narrow band at the interface and extending further into the material. The width of this narrow band of heated material is dependent upon the process parameters used. A series of experimental inertia friction welds were performed using Ti-6Al-4V, and a finite element (FE) modeling framework was developed using the FE code Deform in order to predict the widths of the HAZ and TMAZ at the weld interface. The experimentally observed HAZ boundaries were correlated with the thermal fields from the FE model, while TMAZ boundaries were correlated with the Von Mises plastic strain fields.

This document is the author's post-print version, incorporating any revisions agreed during the peer-review process. Some differences between the published version and this version may remain and you are advised to consult the published version if you wish to cite from it.

No abstract