Microplasticity in metallic glasses

Metallic Glasses (MGs) are amorphous metallic alloys with exceptional mechanical and physical properties. Like all good things, this class of materials comes with its own problems, in this case (i) a lack of ductility, (ii) degradation in properties over time due to aging, and (iii) abrupt and unpredictable failure. In the past five decades since their discovery, the primary focus related to their mechanical properties has been the structural changes due to plasticity accumulating beyond macroscopic yielding. In contrast, the pre-yield regime has essentially been neglected, only now receiving increased attention due to the realization that potential pre-yield microplastic activity can have substantial effects on both structure and properties. This realization stems from indirect property measurements that reveal large changes in, for example, stored excess enthalpy after mechanical stresses of only a percent of the yield stress. The underlying structural activity and processes, that drive such strong structural evolution under a small far-field bias, remain unclear. To this end, the dissertation aims to probe structural processes taking place in the nominally elastic regime, thereby classifying this structural activity as microplasticity. Linked to the identified structural probing techniques, the microplastic regime is divided up into two parts, a stress range far below yielding, and stresses just prior to macroscopic strain localization and therefore shear banding. We use a coherent x-ray scattering technique (x-ray photon correlation spectroscopy or XPCS) to probe the lower stress range, whereas an in-situ acoustic emission (AE) technique is used to track structural activity close to yielding. As a first experiment, XPCS has been used to probe the atomic dynamics underlying microplasticity in an MG loaded under nominally elastic stresses using a four-point bending geometry. The key findings of this work are that i) relaxation times become increasingly longer with increasing applied stress levels and that ii) the far-field stress breaks a long-known universal aging behavior of disordered solids. The experiments reveal how microplasticity in MGs leads to intermittent structural dynamics instead of the monotonous aging response seen under zero-stress conditions. Combining these experiments with molecular dynamics simulations, a picture emerges that suggests atomic-scale microplastic activity, localized in space and time. As a next step, stress is introduced via a thermal protocol. Specifically, non-affine strains are introduced via a cyclic liquid nitrogen cooling sequence. Under these conditions, different structural states are investigated, of which only the nominally less relaxed state experiences an increase in excess enthalpy and therefore structural disordering (rejuvenation). Focusing on the time-scale domain of the underlying structural dynamics, it is found that cryogenic cycling homogenizes the relaxation time distributions, which suggests a reduction in internal structural heterogeneities. Remarkably, a strong correlation between short time-scale dynamics and long-time scale relaxation times is revealed, which in addition to the gained structural insights demonstrates how the used technique can probe structural activity at the sub-experimental time resolution scale. Complementary to the studies involving stress-driven dynamics, XPCS has been used to study the link between structural state and atomic-scale dynamics, revealing a correlation between stored enthalpy and relaxation dynamics. Furthermore, XPCS is used to probe the structural dynamics during isothermal annealing all the way to the onset of crystallization. These complementary studies are substitutes for scheduled in-situ mechanical experiments that could not be realized due to the COVID-19 pandemic. Approaching macroscopic yielding, AE is used to probe crackling noise from local dilatational or bond-breaking structural processes that precede shear banding. The method is found suitable to resolve compositional differences in this microplastic pre-yielding activity. Applied as a function of temperature, the results suggest athermal processes that lead towards shear banding, even though the latter mechanism is known to be thermally activated. Due to its challenging data interpretation and handling, AE is combined with x-ray tomography (XRT) in order to potentially shed a clearer light on the microplastic pre-yielding processes. Whilst XRT is not able to resolve microplasticity on its own, the combination of both methods shows a separation of internal damage accumulation before and after yielding. Furthermore, XRT is used to track internal damage accumulation in the form of shear-band cavities as a function of strain. This allows, for the first time, the determination of the true stress-strain curve of a metallic glass, revealing significant shortcomings of the typically considered engineering stress. A final exploitation of the AE method targets fracture, where an interrupted crack propagation is captured in real-time. Linked to high-speed pyrometry, a signature of glass transition is revealed during fracture, which extends the known heating rate-dependent glass transition domain to 107 K/s. Overall, this dissertation reveals a series of fundamental microplastic processes that are distinctly different to the vastly studied shear-banding phenomenon. The present work pioneers in-situ tracking of structural dynamics with XPCS under thermo-mechanical loads in the microplastic regime at low homologous temperatures. It also sheds light on the strain-dependent internal micro-damage processes that eventually drive the material to failure. The gained insights have, beyond the extension of our fundamental knowledge base, direct implications for pre-yield and post-yield structural control and macroscopic performance. For example, the here obtained results help to rationalize why homogeneous structural disordering via cryogenic cycling is not successful. Indeed, a specific degree of initial structural heterogeneity – here revealed via strong fluctuations of internal relaxation time-scales – is needed to exploit rejuvenation via cryogenic cycling. Whilst not the primary goal of this work, we thus anticipate the value of this dissertation to more application-focused research of MGs.LimitedAuthor requested closed access (OA after 2yrs) in Vireo ETD syste