Ralf Busch

The addition of Cr to Ni–Fe–Cu oxygen-evolving anodes was investigated to promote the formation of a protective Cr2O3 scale to improve oxidation resistance for application in CO2- free aluminium electrolysis. A series of Ni60Fe30Cu10-𝑥Cr𝑥 (𝑥 = 0, 2, 4, 6, 8, 10) wt.% and (Ni60Fe30Cu10)90Cr10 wt.% were selected for microstructural, oxidation and electronic structure characterisation via Thermo-Calc, XRD, SEM–EDX, XPS and NEXAFS. A single-phase 𝛾-fcc microstructure was preserved across all annealed compositions despite Cu–Cr immiscibility. An abrupt transition in oxide morphology occurs at Cr = 10 wt.% where non-hermetic Cr2O3 dominates the scale and effectively shields Ni, Fe and Cu from oxidation, preventing the formation of a (Ni,Fe)3O4 spinel structure. At Cr ≥ 8 wt.% NiO formation was suppressed in regions where chromia had not passivated the oxidation reaction, which correlates with a Ni L3-edge energy shift measured via NEXAFS at Cr = 8 wt.%. This shift is attributed to Cr-induced redistribution of the Ni 3d DOS, suggestive of a reduction in Ni chemical activity. The prevention of NiO formation - known to increase the risk of passivation in Ni–Fe–Cu anodes - may lead to a widening of the viable Ni-rich composition space compared to the base ternary, and subsequent phase stability advantages.

Fe2−xCoxP1−ySiy has attracted significant attention for magnet applications due to its promising intrinsic magnetic properties and the absence of strategic elements, such as rare earths (RE). Most samples reported in the literature are polycrystalline, obtained via mechanical alloying, which makes direct measurement of magnetocrystalline anisotropy challenging. Only a limited number of single-crystal compositions have been characterized. In this work, we have produced Fe2−xCoxP1−ySiy polycrystalline samples by combined induction and arc melting techniques. The obtained samples have a better degree of purity with respect to mechanical alloyed samples reported in literature, with secondary phase appearing only in specimens containing large Si quantities. In addition, this work builds upon prior single crystal studies by providing a comprehensive characterization of a broader range of compositions 0.1 ≤ x ≤ 0.3 and 0.1 ≤ y ≤ 0.4, including the first direct and systematic measurement of the anisotropy field in Fe2−xCoxP1−ySiy polycrystalline samples. Results show that the maximum anisotropy field µ0HA = 2.84 T is obtained in Fe1.90Co0.10P0.80Si0.20, with saturation magnetization MS = 111 Am2/kg and Curie temperature TC = 495 K. The theo retical maximum energy product approaches 200 kJ/m3, which makes this material one of the most promising for RE-free permanent magnets with intermediate performances between fer rite and RE-magnets. Experimental results were complemented by DFT calculations in order to estimate magnetic moment and magnetocrystalline anisotropy energy (MAE) maps for a wider range of compositions. Computational results show that band filling plays a role in the optimization of MAE in the Fe2−xCoxP1−ySiy system.

The electrochemical corrosion behaviour of uncoated, Ni, Ni-Cr20 and Cr-coated Ni42Fe38Cu20 wt.% anodes was investigated with respect to corrosion current density, wear rate and scale morphology in a NaF-KF-AlF3-Al2O3 melt at 800 °C, for up to 1 h of electrolysis. Ni-Cr20 and Cr-coated anodes were found to be the most resistant to oxidation in air due to Cr2O3 formation and reduced interdiffusion with substrate elements, whereas Ni-coated anodes demonstrated significant interdiffusion and oxide penetration. Electrochemical measurements of exchange current densities on the clean and oxidised anode surface were used to calculate the wear rate (WR), where the Ni-Cr20 coating was deemed to be detrimental to anode performance due to the catastrophic tendency towards electrode passivation via NiF2 formation. For the Ni and Cr-coated anodes, whilst a slight increase in the electrochemical parameters and wear rates was observed compared to their uncoated counterparts, SEM-EDX analysis of used-anode cross-sections revealed a significant reduction in electrolyte penetration for these two coatings as well as the formation of a more uniform oxide scale with less extensive blistering. Overall, the increase in corrosion parameters for the Ni and Cr-coated anodes appears to derive from the sacrificial dissolution of the coating material, whilst a noticeable degree of protection to the bulk is provided. The longer-term corrosion resistance implications with respect to changes in corrosion scale and desirable NiFe2O4 formation are discussed, and a robust research methodology for future studies on suitable coatings for inert anodes is displayed.

The mechanical response of metallic glasses depends on their structural state, commonly represented by the fictive temperature Tf. Here, we combine low‑load nanoindentation with statistical analysis of pop‑in events to disentangle the structure‑sensitive onset of plasticity from the structure‑insensitive post-yield response in a Zr‑based bulk metallic glass (Vit105). Specimens were annealed to targeted Tf states and indented at constant strain rates. The first critical pop‑in load decreased with increasing strain rate and showed a non-linear and saturating dependence on Tf. Empirical cumulative distribution functions (ECDFs) revealed that rate‑controlled activation of the onset is consistent with thermally assisted nucleation of shear‑transformation zones (STZs). The activation volume increased with strain rate, and the corresponding STZ size was estimated at (6–20 atoms). We further observed a decrease of the effective activation volume Vact.STZ with increasing indentation depth/number of pop‑ins, suggesting a transition from larger cooperative rearrangements at incipient yielding to deformation along pre‑formed shear paths. The results demonstrate a decoupling of onset and flow metrics, link Tf to rate‑dependent incipient plasticity, and motivate alternative protocols to more robustly capture the incubation regime at very shallow depths.

Preceding investigations on the atomic dynamics in melts of Ti75Ni25 revealed faster atomic dynamics as compared to that found in melts of Zr64Ni36 and Ni66.7B33.3, despite the fact that all these melts are characterized by similar packing fractions. In order to find a structural explanation for the different dynamic behavior, we have studied the short-range structure of stable and undercooled Ti75Ni25 melts. The melts were containerlessly processed under high-purity conditions by application of the electrostatic levitation technique. Partial structure factors of the liquid alloys have been determined by a combination of neutron diffraction (with isotopic substitution) and synchrotron X-ray diffraction. The studies reveal that Ti75Ni25 melts are characterized by a chemical short-range order, where heterogeneous Ti-Ni nearest neighbors are preferentially formed. This chemical short-range order, however, is less pronounced than the chemical short-range order reported for melts of Zr64Ni36 and Ni66.7B33.3. The less pronounced chemical short-range order may account for the faster atomic dynamics reported for liquid Ti75Ni25. For a further analysis of the structure-dynamics relationship the experimentally determined partial structure factors were used as an input for calculations in the framework of the mode coupling theory of the glass transition. These reveal that the Ni self-diffusion is only slightly faster than the Ti self-diffusion with a ratio of the self-diffusion coefficients DNi/DTi ≈ 1.2. Moreover, the Onsager coefficients calculated by MCT are in good agreement with those estimated by use of Darken's equation indicating that cross-correlation effects are negligible when describing the interdiffusion in liquid Ti75Ni25.