Oxford Centre for High Energy Density Science (OxCHEDS)

Abstract:

In their landmark study1, Fecht and Johnson unveiled a phenomenon that they termed the ‘entropy catastrophe’, a critical point where the entropy of superheated crystals equates to that of their liquid counterparts. This point marks the uppermost stability boundary for solids at temperatures typically around three times their melting point. Despite the theoretical prediction of this ultimate stability threshold, its practical exploration has been prevented by numerous intermediate destabilizing events, colloquially known as a hierarchy of catastrophes2, 3, 4–5, which occur at far lower temperatures. Here we experimentally test this limit under ultrafast heating conditions, directly tracking the lattice temperature by using high-resolution inelastic X-ray scattering. Our gold samples are heated to temperatures over 14 times their melting point while retaining their crystalline structure, far surpassing the predicted threshold and suggesting a substantially higher or potentially no limit for superheating. We point to the inability of our samples to expand on these very short timescales as an important difference from previous estimates. These observations provide insights into the dynamics of melting under extreme conditions.

Abstract:

High-quality total scattering data, a key tool for understanding atomic-scale structure in disordered materials, require stable instrumentation and access to high momentum transfers. This is now routine at dedicated synchrotron instrumentation using high-energy X-ray beams, but it is very challenging to measure a total scattering dataset in less than a few microseconds. This limits their effectiveness for capturing structural changes that occur at the much faster timescales of atomic motion. Current X-ray free-electron lasers (XFELs) provide femtosecond-pulsed X-ray beams with maximum energies of ~24 keV, giving the potential to measure total scattering and the attendant pair distribution functions (PDFs) on femtosecond timescales. Here, we show that this potential has been realised using the HED scientific instrument at the European XFEL and present normalised total scattering data for 0.35Å−1 < Q < 16.6Å−1 and their PDFs from a broad spectrum of materials, including crystalline, nanocrystalline and amorphous solids, liquids, and clusters in solution. We analyse the data using a variety of methods, including Rietveld refinement, small-box PDF refinement, joint reciprocal–real space refinement, cluster refinement, and Debye scattering analysis. The resolution function of the setup is also characterised. We conclusively show that high-quality data can be obtained from a single ~30 fs XFEL pulse for multiple different sample types. Our efforts not only significantly increase the existing maximum reported Q-range for an S(Q) measured at an XFEL but also mean that XFELs are now a viable X-ray source for the broad community of people using reciprocal space total scattering and PDF methods in their research.