Open Access

The digital transformation of materials science is enabling a paradigm shift from traditional trial-and-error approaches toward autonomous, data-driven materials discovery and optimization. This contribution presents the development of a digital and autonomous materials pipeline that integrates automated synthesis, high-throughput experimentation, advanced characterization, computational modeling, materials databases, and artificial intelligence into closed-loop research workflows. Examples include self-driving laboratories for nanomaterials, electrochemical coatings, oxide glasses, and bulk metallic materials, where machine learning algorithms autonomously guide experimental decisions and optimize targeted material properties. By combining feedback-controlled experimentation with materials informatics and simulation, the platform enables reproducible, scalable, and accelerated exploration of complex compositional and processing spaces. The presented framework demonstrates how autonomous laboratories can bridge digital and experimental materials science, providing a foundation for next-generation materials design, discovery, and qualification.

As-cast metallic glasses relax continuously as a function of time and temperature, which may lead to significant property changes. In search for a better structural understanding of aging of metallic glasses, we exploit here the ability to track atomic-scale dynamics via x-ray photon correlation spectroscopy (XPCS). We focus on isothermal annealing of a model glass in the microsecond regime, from which the XPCS experiment is simulated. Examination of the one-time intensity cross-correlations reveals a multi-stage structural decorrelation sequence, where different atomic-scale transport mechanisms can be linked to the stepwise decorrelation. Specifically, these range from thermal vibrations, glassy network evolution, and topological and chemical ordering towards crystallization. These insights promote a picture, in which the metallic model glass has dynamically wellseparated microstructural evolution stages during isothermal relaxation. We discuss these findings in the context of our experimental XPCS observations.

Glassy solids are ubiquitous in our daily lives, including silicate glasses, amorphous polymers, or glassy oxides. A more modern addition to this class of materials are metallic glasses, representing among the toughest and strongest metals we know. Structurally, metallic glasses distinguish themselves from their crystalline counterparts by a broad distribution of nearest neighbor distances and a lack of long-range order. This has fundamental consequences in terms of both barrier energies for structural transitions and related relaxation times. In this talk, we will focus on the stress-strain domain prior to macroscopic yielding of metallic glasses and probe transport at the atomic-scale using coherent scattering methods. Specifically, we will first track stress-driven structural dynamics during loading in terms of decorrelation times that give detailed insights into the time scales of microplastic flow (Nature Communications 10 (2019) 5006). These show strong deviations away from the expected time-dependent relaxation dynamics, exhibiting abrupt fluctuations. Finding that such time-scale fluctuations are also present in unstressed metallic glasses (Nature Communications 15 (2024) 6595), we leveragemolecular dynamics simulations in the microsecond time domain to understand this phenomenon. The simulations reveal how extend cluster rearrangements can cause abrupt structural dynamics (Acta Materialia 267 (2024) 119730), which manifests itself as intermittency in the time domain. As a final step, we consider the lowest experimentally accessible stress domain to interrogate the true microplastic limit of metallic glasses. It is found that, in contrast to crystalline metals, flow occurs at even the smallest applied stress (Materials Today 82 (2025) 92). This stands in stark contrast to the several GPa-large yield stresses with Weibull moduli as high as for technical Al- or Fe-based alloys. We discuss these findings in the context of an extended glassy microstructure with strong structural partitioning.

In search for a structural understanding of relaxation and rejuvenation of metallic glasses, we exploit here the ability to track atomic-scale dynamics via x-ray photon correlation spectroscopy (XPCS). Conducted across temperatures and under the application of stress, the results reveal nonmonotonically evolving and fluctuating relaxation times throughout isothermal condions, demonstrating heterogeneous dynamics at the atomic scale (Nature Communications 10 (2019) 5006). In concert with atomistic simulations, we identify an aging-time dependent transiton from classical stretched exponential decorrelation at short waiting times to an emerging power-law signature at long waiting times (Nature Communications 15 (2024) 6595), which suggests sub-diffusive transport. Throughout intermitient aging, relaxation times increase on average, but temporal fluctuations emerge due to collective transport and cluster dynamics, providing a structural explanation for intermitient aging revealed with XPCS experiments (Acta Materialia 267 (2024) 119730). Applying stresses within the elastic domain, with a focus on the low stress regime, reveals considerably accelerated atomicscale transport. Within the resolution of the experiments, this remains true at stresses as small as 0.5% of the yield strength. With increasing stress, the distribution of relaxation times changes from compressed exponential to simple exponential, reminding of a stress-temperature equivalence in the time-scale domain (Materials Today 82 (2025) 92). We discuss our findings in the context of an emerging microstructure in metallic glasses.

Metallic glasses exhibit exceptional mechanical properties, yet their long-term stability is governed by complex atomic-scale relaxation and transport processes. In this work, the structural dynamics of amorphous metals are investigated using X-ray Photon Correlation Spectroscopy (XPCS) in combination with atomistic modeling. The experiments reveal heterogeneous and non-monotonic aging behavior, including intermittent, fluctuating, and stationary relaxation regimes during extended annealing near the glass transition temperature. Molecular dynamics simulations identify correlated string-like excitations and cluster dynamics as key mechanisms underlying thermally activated transport and structural relaxation. Furthermore, the observed waiting-time distributions and decorrelation signatures indicate anomalous sub-diffusive transport that deviates from classical Kohlrausch–Williams–Watts behavior and follows power-law and Mittag–Leffler-type dynamics. These findings support a picture of metallic glasses as dynamically heterogeneous systems with collective atomic transport processes that dominate structural evolution across experimentally relevant timescales.