High-efficient low-cost characterization of materials properties using domain-knowledge-guided self-supervised learning

Modern AI-assisted approaches have helped material scientists revolutionize their abilities to better understand the properties of materials. However, current machine learning (ML) models would perform awful for materials with a lengthy production window and a complex testing procedure because only a limited amount of data can be produced to feed the model. Here, we introduce self-supervised learning (SSL) to address the issue of lacking labeled data in material characterization. We propose a generalized SSL-based framework with domain knowledge and demonstrate its robustness to predict the properties of a candidate material with the fewest data. Our numerical results show that the performance of the proposed SSL model can match the commonly-used supervised learning (SL) model with only 5 % of data, and the SSL model is also proven with ease of implementation. Our study paves the way to expand further the usability of ML tools for a broader material science community.

Wireless nanotechnologies light up the next frontier in cell Calcium signalling

Calcium ions impact nearly every aspect of cellular life, playing crucial roles as secondary messengers in regulation of neurotransmission, cell proliferation, migration and differentiation processes, intracellular homeostasis, long-distance signal propagation and stimuli physiological response. Despite its key-role, available techniques to study and selectively regulate Ca2+ signalling largely rely on chemical and electrical approaches, which often cannot ensure the necessary spatial and temporal resolution, specificity, modulation and reversal capability. In this context, Ca2+ modulation based on physical stimuli, such as magnetic, mechanical and optical tools, are emerging ass promising innovative solutions. Here, we focus our attention on a subclass of these approaches, namely wireless-activated techniques, and on functional materials able to act as non-invasive transduction elements. We present an overview of most recent outcomes in the field, and we critically evaluate their advantages and drawbacks. This work is mainly directed to the material science community, but hopefully it will provide a useful perspective also to the broader readership of biotechnologists, physiologists and clinicians.

Ice: the paradigm of wild plasticity

Ice plasticity has been thoroughly studied, owing to its importance in glaciers and ice sheets dynamics. In particular, its anisotropy (easy basal slip) has been suspected for a long time, then fully characterized 40 years ago. More recently emerged the interest of ice as a model material to study some fundamental aspects of crystalline plasticity. An example is the nature of plastic fluctuations and collective dislocation dynamics. Twenty years ago, acoustic emission measurements performed during the deformation of ice single crystals revealed that plastic ‘flow’ proceeds through intermittent dislocation avalanches, power law distributed in size and energy. This means that most of ice plasticity takes place through few, very large avalanches, thus qualifying associated plastic fluctuations as ‘wild’. This launched an intense research activity on plastic intermittency in the Material Science community. The interest of ice in this debate is reviewed, from a comparison with other crystalline materials. In this context, ice appears as an extreme case of plastic intermittency, characterized by scale-free fluctuations, complex space and time correlations as well as avalanche triggering. In other words, ice can be considered as the paradigm of wild plasticity. This article is part of the theme issue ‘The physics and chemistry of ice: scaffolding across scales, from the viability of life to the formation of planets’.

Mavlyanovite, Mn5Si3: a new mineral species from a lamproite diatreme, Chatkal Ridge, Uzbekistan

Mavlyanovite, ideally Mn5Si3, is a new mineral from a lamproite diatreme close to the upper reaches of the Koshmansay river, Chatkal ridge, Uzbekistan. It occurs together with unnamed manganese siliciphosphide and manganese silicicarbide minerals in round to ovoid segregations, up to 10 cm in diameter, in volcanic glass. Segregations of hexagonal prismatic mavlyanovite up to 1–2 mm occur in interstices in the matrix and tiny inclusions (1–2 μm) of alabandite and khamrabaevite occur within mavlyanovite. It is opaque with a metallic lustre, has a dark-grey streak, is brittle with a conchoidal fracture and a near-perfect basal cleavage. VHN100 is 1029–1098 kg/mm2 (Mohs hardness ~7). In plane-polarized reflected light, mavlyanovite is a pale-brownish-grey against the accompanying unnamed manganese silicicarbide (white). Reflectance values and colour data are tabulated. Average results of 19 electronmicroprobe analyses give Mn70.84, Fe 6.12, Si 22.57, Ti 0.15, P 0.18, total 99.86 wt.% leading to an empirical formula of (Mn4.66Fe0.40)5.06(Si2.91Ti0.01P0.02)2.94 based on8 a.p.f.u. The calculated density is 6.06 g/cm3, (on the basis of the empirical formula and unit-cell parameters from the structure determination). Mavlyanovite is hexagonal (P63/mcm) with a 6.8971(7), c 4.8075(4) Å, V 198.05(3) Å3 and Z = 2. The structure has been determined and refined to R1 = 0.017, wR2 = 0.044, GoF = 1.16. Mavlyanovite is the naturally-occurring analogue of synthetic Mn5Si3 which is the parent aristotype structure of the Nowotny intermetallic phases studied extensively by the material-science community. It is also the Mn-dominant analogue of xifengite Fe5Si3. The mineral name honours Academician Gani Arifkhanovich Mavlyanov (1910–1988), for his contributions to the understanding of the geology of Uzbekistan.

High-resolution TEM study of ion beam irradiation induced amorphization in ceramic materials

Radiation damage of nuclear materials (e.g. fast- or fusion-neutron damage in reactor structural components, fission-fragment damage in nuclear fuels and alpha decay damage in nuclear waste forms) has been one of the major challenges faced by the material science community. Ion beam irradiation and implantation experiments have been used extensively in the past few decades not only for simulating these damaging process in materials but also for improving material properties for many technological applications.As an energetic particle traverses a crystalline target, it loses its energy predominantly through electronic (ionization) and nuclear (elastic collision) interactions with the atoms in the lattice. The target atom which receives sufficient energy from the interactions may get displaced from its lattice site and may further displace other target atoms, thus creating a displacement cascade which is usually a few nanometers in scale. Just at the end of the collision phase, which lasts for only a few tenths of a picosecond, a displacement cascade contains a very dense cluster of point defects and the region may be considered amorphous.

Plate and needle-like crystal structure determination by combining electron and synchrotron diffraction

In recent years, significant progress in determining light-element crystal structures using electron diffraction has been made to meet the increasing needs from the material science community. The set of techniques for carrying out electron crystallography, including sample preparation, data collection and recording, diffraction micrograph digitization, and confirmation by direct phasing methods and structure refinements, have being developed. However, only a very limited number of structures have been determinated by electron crystallography because of a number of severe problems and difficulties. Meanwhile, the progress in determining crystal structures using x-ray diffraction has been rapidly increasing, particularly because of the more extensive use of powerful synchrotron diffraction techniques. In this case, the minimum crystal size required for a single-crystal study has decreased from 100 microns to about 10 microns. Even so, many new materials can only be obtained in microcrystalline form with crystallite sizes well below one micron. In this submicron regime, ab initio structure solutions from synchrotron powder-diffraction patterns have proven to be quite powerful.