The mission for fusion and fission
Five years into the Materials for Fusion and Fission Power project, Simon Frost finds out how researchers have progressed in the processing, characterisation and testing of advanced materials for next-generation nuclear power plants.
Dr Xiaoou Yi, postdoctoral researcher at the University of Oxford, outlined the challenges facing materials for nuclear reactors at the Materials for Fusion and Fission Power (MFFP) Showcase at the University of Oxford in December 2014. Subject to neutron exposure and increasingly high temperatures, the materials in nuclear reactors risk displacement damage and transmutations, giving rise to point defects, clusters, dislocations, voids and precipitates. Massive changes in a material’s microstructure often result in the degradation of properties, which manifest in radiation hardening and embrittlement, radiation-induced precipitation, high-temperature hydrogen embrittlement, irradiation creep and volumetric swelling. These defects compromise functionality and, crucially, safety.
Since 2009, researchers on the MFFP project, core-funded by a £5.8m EPSRC grant and led by Steve Roberts, Professor in Materials at Oxford Univeristy, have been working to understand the mechanisms of such defects and develop ways to predict and prevent them when designing radiation-resistant materials – predominantly tungsten alloys and oxide dispersion-strengthened (ODS) steels. The next generation of nuclear reactors is dependent on the development of new materials that can withstand increasingly hostile conditions. Researchers at the University of Oxford Department of Materials presented their work at the showcase.
Martina Meisnar, Nanoanalysis Group, described a multi-scale approach to characterising environmentally assisted cracking in ODS steels. ‘The ultimate goal we set ourselves is to link the microstructure to the local chemistry and mechanical behaviour of the material,’ she said. ‘Usually we start with optical microscopy, in the range of millimetres down to a couple of micrometres, we then apply electron microscopy – mainly scanning and transmission electron microscopy (SEM and TEM, respectively) – from micrometres down to nanometre level and, finally, atom probe tomography, which has atomic resolution and very high chemical sensitivity.’
Measuring the progress of stress corrosion cracking (SCC), Meisnar and the group recorded a surprising result, showing that crack growth rate increases with decreasing temperature. The chief cause of SCC is commonly thought to be oxidation, which depends on temperature-controlled diffusion of oxygen in metals, suggesting that propagation should be faster at higher temperatures. The counterintuitive result led the group to question whether oxidation really is the driving force in SCC or if mechanical influences dominate. A new development in electron backscatter diffraction (EBSD) was a vital tool towards finding a possible answer.
Transmission Kikuchi diffraction (TKD) is a variant of conventional TEM that uses electron transparent samples coupled with conventional EBSD hardware. Its spatial resolution is in the order of 5–10nm – much higher than conventional EBSD – so it can be used effectively to examine highly deformed materials. Using TKD, the researchers assessed the relation between temperature and plastic deformation, observing that plastic deformation was also higher at lower temperatures, suggesting that it may be the governing cause of SCC in ODS steels.
Power in nanoclusters
Patrick Grant, Professor of Materials, described his area of interest as the interface between advanced materials and manufacture. His group, including collaborators at Queen Mary University, London, and University of Liverpool, looked at why adding tiny yttrium nanoclusters to a ferritic steel matrix hugely improves steel’s performance in terms of strength and creep resistance. ‘There are two questions that are interesting here – why do these have such a transformative effect, and how can we manufacture them at scale?’ he asked.
The standard route is to mix oxide powder of around 100–500nm in diameter with steel powder, around 30 microns diameter, through an aggressive mechanical ball milling process that causes the breakup and exchange of materials, then applying hot isostatic pressing (HIP) to consolidate the powders into a single-phase material, before final thermomechanical processing. ‘When you add the oxide powder, the grains measure around 100nm, and very quickly break down to 10nm. After 60 hours of milling, they effectively become invisible to all but the most powerful atomic resolution techniques. It’s not simply a mixing process – there’s something more profound, from a metallurgical point of view, going on. You’re actually dissolving a ceramic into a metal and this will only happen in the highly defective, high-energy microstructures that are contrived by this milling process. So, if you get this stage wrong, it doesn’t matter what you do in the thermomechanical processing – you’ll never get the desirable properties,’ he said.
These desirable yield stress and creep resistance properties are not the only benefit of oxide nanoclusters – they also protect the material from radiation damage. ‘The principal interest is that the material properties are more stable under neutron flux in the presence of nanoclusters. Damage that is induced in the alloy in the form of vacancies segregates to these tiny little particles, and all these interfaces and particles effectively act as sink traps for the defects caused by the neutron bombardment. So we’re not stopping the damage – you can’t make an alloy that will stop it – but we’re making an alloy that manages the damage by having the right microstructure to soak it up.’
An alternative process Grant has been investigating with colleagues at Queen Mary University is spark plasma sintering (SPS). A misleading name, as it involves neither plasma nor a spark. Grant explains, ‘It looks a lot like HIP – you apply a uniaxial force to the powder in a vacuum, and the combination of sintering and the application of force leads to densification. However, in SPS, you pass a very high current through the material at the same time – something like 15,000A, but only at 3V. Some of this huge current flux goes around the outside through the die, producing a heating effect, and some of it goes through the powder. This allows you to take what would be a four-hour HIP cycle and replace it with a five-minute SPS process. If we can consolidate the material quickly in this way, we may be able to retain some of the metastable effects such as ultra-fine grain size into bulk materials.’ Alternative routes to producing a family of materials at scale with the properties achieved through mechanical alloying are something that Grant continues to explore, and he warns that ‘there are still big question marks over the scalability of mechanically alloyed materials economically’.
Tungsten testing upgrade
David Armstrong, Royal Academy of Engineering Research Fellow, said the most important outcome of his group’s work with tungsten-rhenium alloys was that they had made it possible to carry out micromechanical testing of tungsten at reactor-specific temperatures of around 750°C – the highest ever published by a margin of 100°C. The key difficulties for testing tungsten at high temperatures are that it oxidises from around 400°C and the oxide starts to sublime at 700°C – the solution was to design a nanoindentation instrument that is housed within a high vacuum chamber. ‘We can get to a vacuum level of about 1x10-6 mbar, which is necessary for running tests at reactor-specific temperatures. We mount thermocouples all over them to monitor the temperature, and the key thing is that we can match the temperature of the sample with the temperature of the tip – that’s vital for these small-scale, high temperature tests.’
Armstrong and co-researchers tested the performance of a commercially available tungsten-rhenium alloy as supplied and when implanted with helium. The unimplanted material has a hardness of approximately 6.5GPa at room temperature, which drops off rapidly as it is heated up to 300°C, where it stabilises at around 3GPa. The implanted material is extremely hard at room temperature, more than 10GPa, which is consistent up to around 200°C, at which point this rapidly decreases, though remains harder than the unimplanted material. They were able to recover this superior hardness, however, on cooling, showing that the helium remained within the alloy. ‘The helium is still there, trapped in some sort of obstacle’, Armstrong said.
For the latest on the MFFP, visit mffp.materials.ox.ac.uk