Diffusion Bonding Beryllium to Reduced Activation Ferritic Martensitic Steel: Development of Processes and Techniques

Only a few materials are suitable to act as armor layers against the thermal and particle loads produced by magnetically confined fusion. These candidates include beryllium, tungsten, and carbon fiber composites. The armor layers must be joined to the plasma facing components with high strength bonds that can withstand the thermal stresses resulting from differential thermal expansion. While specific joints have been developed for use in ITER (an experimental reactor in France), including beryllium to CuCrZr as well as tungsten to stainless steel interfaces, joints specific to commercially relevant fusion reactors are not as well established. Commercial first wall components will likely be constructed front Reduced Activation Ferritic Martensitic (RAFM) steel, which will need to be coating with one of the three candidate materials. Of the candidates, beryllium is particularly difficult to bond, because it reacts during bonding with most elements to form brittle intermetallic compounds. This brittleness is unacceptable, as it can lead to interface crack propagation and delamination of the armor layer.;I have attempted to overcome the brittle behavior of beryllium bonds by developing a diffusion bonding process of beryllium to RAFM steel that achieves a higher degree of ductility. This process utilized two bonding aids to achieve a robust bond: a copper interlayer to add ductility to the joint, and a titanium interlayer to prevent beryllium from forming unwanted Be-Cu intermetallics. In addition, I conducted a series of numerical simulations to predict the effect of these bonding aids on the residual stress in the interface. Lastly, I fabricated and characterized beryllium to ferritic steel diffusion bonds using various bonding parameters and bonding aids.;Through the above research, I developed a process to diffusion bond beryllium to ferritic steel with a 150 M Pa tensile strength and 168 M Pa shear strength. This strength was achieved using a Hot Isostatic Pressing (HIP) process (at a temperature between 700 °C and 750 °C for 2 hours at 103 M Pa) with 10 mu m of titanium and 20 mum of copper deposited between substrates. Without the copper and titanium interlayers, the bond formed an intermetallic that lead to fracture from internal residual stresses. Also, slowing the rate of cooling and adding an intermediate hold temperature during cool-down significantly increased bond strength. These beneficial effects were confirmed by the numerical simulations, which showed reduced residual stress resulting from all bonding techniques.;Both metals interlayers, as well as the reduced cooling rate were critical in overcoming the otherwise brittle quality of the beryllium to ferritic steel joint. However, the introduced interlayers are not an ideal solution to the problem. They introduced both Be-Ti and Cu-Ti compounds, which proved to be the eventual failure location in the bond. Further optimization of this joint is necessary, and can potentially be achieved with variation of cooling rates. To make the joint ready for implementation will require larger scale fabrication to verify reliability and to test the joint under operational loads.