We rely on metallic objects every day, from bicycles and bridges to the solder joints in our electronics. In each case, a key step in manufacturing is the solidification of liquid alloy, and it is through controlling solidification that we can control grain structure and defects. Solidification is at the heart of current challenges facing the UK: steel and aluminium production contributes more than 10% to global industrial CO2 emissions, and new solder technologies are required to enable the manufacturing of smaller, more powerful portable electronics. In all these industries, advances will involve controlling the solidification microstructure and controlling solidification defects. Key to the development of grain structure in solder joints and structural castings are the earliest stages of solidification when the number-density of grains is determined by the number density of nucleation events. The project will use new microscopy techniques which combine focussing an ion beam to micro-machine into the centre of crystals and find nucleant particles with electron diffraction to understand how the particles catalyse nucleation. With this information, new ways to control nucleation will be explored. After nucleation, the semi-solid grain structure goes on to significantly affect the formation of defects in castings and solder joints. Part of tackling this challenge is to develop a deeper understanding of how and why casting defects form. It is known that the origin of semi-solid cracking is the stresses and strains that develop during solidification but, to understand the details, we need to observe and measure how numerous solidifying crystals respond to loads during solidification. Metals and alloys are opaque to visible light and their inner structure is therefore hidden from our eyes. By pouring liquid alloy, we can see that they have a low viscosity and that the viscosity increases considerably as alloys solidify, but we cannot see or measure what structural changes are causing these changing flow properties. X-rays can be transmitted through metals, offering the potential to observe the development of microstructure, but it is only in the last decade that X ray sources have become available with sufficient flux and coherence to allow real-time imaging of crystal growth in alloys. This was an enormous step forward as it became possible to test solidification theories developed in 'post-mortem' studies using real metallic samples. This project will extend these synchrotron techniques to observe and measure the solidification of intermetallic grains in solder joints, and to study how deformation of the semi-solid grain structure leads to casting defect formation. We aim to observe and measure for the first time where intermetallics nucleate in solder joints and how they grow during solder reactions. This will give us insights that we can use to engineer solder joint microstructures and tackle the final frontiers in the transition to Pb-free soldering such as a replacement for high-Pb solder for use at T>180C. Similar techniques will be applied to imaging the formation of inter-columnar cracking in experiments analogous to the continuous casting of steel, a process used to produce more than one billion tonnes of steel annually. An exciting aspect of this part of the research is that much about semi-solid alloy deformation is unknown: How is force transmitted from crystal to crystal? What happens when two crystals are pushed into one another? Do they bend? Do they fragment? Do they behave as rigid bodies? Why do strain instabilities develop? Where do cracks begin and how fast do they grow? These questions can only be fully answered with in-situ observations of deformation at the scale of the microstructure. We have begun to address these questions in pilot studies and now we aim to expand this to crack movement in the mush.

The reliability of electronics depends to a large degree on the reliability of the solder joints that interconnect the circuitry. Most solder joints contain tin as the majority phase to enable soldering at a temperature tolerable to the electronic components, but the tin must then operate at up to ~80% of its melting point due to resistance heating in service. As a percentage of melting point, this is as demanding as a turbine blade in an aeroengine and there is a similar ongoing desire to increase the operation temperature. In service, the joints regularly cycle between ~50 and 80% of melting point due to cycles of resistance heating and natural cooling, which causes thermal expansion and contraction of all phases and, therefore, thermal fatigue due to the mismatch in the coefficient of thermal expansion (CTE) at interfaces. Joints can also experience shock impacts, vibration and surges in current density, all of which must be withstood to ensure successful operation. Solder joints contain only up to a few tin grains and are highly heterogeneous with anisotropic properties. Therefore, to understand and predict the performance of solder joints it is necessary (i) to link mechanical measurements to the microstructure and crystallographic orientations in the joint and (ii) to develop crystal-level deformation and damage models that explicitly account for the evolving microstructure and link through to component and PCB-level models of thermal cycling, shock impact etc. Furthermore, to capitalise on the understanding generated by such an approach, it is necessary to develop the capability to reproducibly create the microstructures and orientations during the soldering process that are predicted to give optimum performance in service. To deliver this vision, we bring together expertise in controlling solidification kinetics in solder alloys, in-situ micromechanical measurement of crystal slip and slip transfer across interfaces, defect nucleation and growth, and micromechanical modelling at the crystal and microstructure level and at the component and board-level. With this team, we seek a step change improvement in the understanding, prediction and manufacturing of solder joints that are optimised for high reliability in high value UK industry and in the consumer electronics industry. The work addresses using solidification processing to generate single crystal and structurally representative units (e.g. intermetallic crystals (IMCs) with the desired facets, beta-Sn micro-pillars, or BGA joints with a single known beta-Sn orientation etc.). These are to be studied in carefully instrumented micromechanical tests to extract key material properties, and mechanistic understanding of defect nucleation at the crystal level. The properties and defect nucleation mechanisms are to be implemented in crystal plasticity models and, where necessary, discrete dislocation plasticity models to provide validated quantitative prediction of solder performance under thermo-mechanical and impact loading. The models are then to be exploited to design solder microstructures for optimal performance. The work will then develop methods to manufacture these optimum microstructures within the soldering process, building on recent advances in microstructure control made by the team. These optimised joints will then be tested and modelled such that optimally designed, high reliability joints may ultimately be achieved.