Cements and concretes materials characterisation using machine-learning-based reconstruction and 3D quantitative mineralogy via X-ray microscopy.

3D imaging via X-ray microscopy (XRM), a form of tomography, is revolutionising materials characterisation. Nondestructive imaging to classify grains, particles, interfaces and pores at various scales is imperative for our understanding of the composition, structure, and failure of building materials. Various workflows now exist to maximise data collection and to push the boundaries of what has been achieved before, either from singular instruments, software or combinations through multimodal correlative microscopy. An evolving area on interest is the XRM data acquisition and data processing workflow; of particular importance is the improvement of the data acquisition process of samples that are challenging to image, usually because of their size, density (atomic number) and/or the resolution they need to be imaged at. Modern advances include deep/machine learning and AI resolutions for this problem, which address artefact detection during data reconstruction, provide advanced denoising, improved quantification of features, upscaling of data/images, and increased throughput, with the goal to enhance segmentation and visualisation during postprocessing leading to better characterisation of samples. Here, we apply three AI and machine-learning-based reconstruction approaches to cements and concretes to assist with image improvement, faster throughput of samples, upscaling of data, and quantitative phase identification in 3D. We show that by applying advanced machine learning reconstruction approaches, it is possible to (i) vastly improve the scan quality and increase throughput of 'thick' cores of cements/concretes through enhanced contrast and denoising using DeepRecon Pro, (ii) upscale data to larger fields of view using DeepScout and (iii) use quantitative automated mineralogy to spatially characterise and quantify the mineralogical/phase components in 3D using Mineralogic 3D. These approaches significantly improve the quality of collected XRM data, resolve features not previously accessible, and streamline scanning and reconstruction processes for greater throughput.

Terrestrial surface stabilisation by modern analogues of the earliest land plants: A multi-dimensional imaging study.

The evolution of the first plant-based terrestrial ecosystems in the early Palaeozoic had a profound effect on the development of soils, the architecture of sedimentary systems, and shifts in global biogeochemical cycles. In part, this was due to the evolution of complex below-ground (root-like) anchorage systems in plants, which expanded and promoted plant-mineral interactions, weathering, and resulting surface sediment stabilisation. However, little is understood about how these micro-scale processes occurred, because of a lack of in situ plant fossils in sedimentary rocks/palaeosols that exhibit these interactions. Some modern plants (e.g., liverworts, mosses, lycophytes) share key features with the earliest land plants; these include uni- or multicellular rhizoid-like anchorage systems or simple roots, and the ability to develop below-ground networks through prostrate axes, and intimate associations with fungi, making them suitable analogues. Here, we investigated cryptogamic ground covers in Iceland and New Zealand to better understand these interactions, and how they initiate the sediment stabilisation process. We employed multi-dimensional and multi-scale imaging, including scanning electron microscopy (SEM) and X-ray Computed Tomography (µCT) of non-vascular liverworts (Haplomitriopsida and complex thalloids) and mosses, with additional imaging of vascular lycopods. We find that plants interact with their substrate in multiple ways, including: (1) through the development of extensive surface coverings as mats; (2) entrapment of sediment grains within and between networks of rhizoids; (3) grain entwining and adherence by rhizoids, through mucilage secretions, biofilm-like envelopment of thalli on surface grains; and (4) through grain entrapment within upright 'leafy' structures. Significantly, µCT imaging allows us to ascertain that rhizoids are the main method for entrapment and stabilisation of soil grains in the thalloid liverworts. This information provides us with details of how the earliest land plants may have significantly influenced early Palaeozoic sedimentary system architectures, promoted in situ weathering and proto-soil development, and how these interactions diversified over time with the evolution of new plant organ systems. Further, this study highlights the importance of cryptogamic organisms in the early stages of sediment stabilisation and soil formation today.

Cryptogamic ground covers as analogues for early terrestrial biospheres: Initiation and evolution of biologically mediated proto-soils.

Modern cryptogamic ground covers (CGCs), comprising assemblages of bryophytes (hornworts, liverworts, mosses), fungi, bacteria, lichens and algae, are thought to resemble early divergent terrestrial communities. However, limited in situ plant and other fossils in the rock record, and a lack of CGC-like soils reported in the pre-Silurian sedimentological record, have hindered understanding of the structure, composition and interactions within the earliest CGCs. A key question is how the earliest CGC-like organisms drove weathering on primordial terrestrial surfaces (regolith), leading to the early stages of soil development as proto-soils, and subsequently contributing to large-scale biogeochemical shifts in the Earth System. Here, we employed a novel qualitative, quantitative and multi-dimensional imaging approach through X-ray micro-computed tomography, scanning electron, and optical microscopy to investigate whether different combinations of modern CGC organisms from primordial-like settings in Iceland develop organism-specific soil forming features at the macro- and micro-scales. Additionally, we analysed CGCs growing on hard rocky substrates to investigate the initiation of weathering processes non-destructively in 3D. We show that thalloid CGC organisms (liverworts, hornworts) develop thin organic layers at the surface (<1 cm) with limited subsurface structural development, whereas leafy mosses and communities of mixed organisms form profiles that are thicker (up to ~ 7 cm), structurally more complex, and more organic-rich. We term these thin layers and profiles proto-soils. Component analyses from X-ray micro-computed tomography data show that thickness and structure of these proto-soils are determined by the type of colonising organism(s), suggesting that the evolution of more complex soils through the Palaeozoic may have been driven by a shift in body plan of CGC-like organisms from flattened and appressed to upright and leafy. Our results provide a framework for identifying CGC-like proto-soils in the rock record and a new proxy for understanding organism-soil interactions in ancient terrestrial biospheres and their contribution to the early stages of soil formation.

Correlating Local Volumetric Tissue Strains with Global Lung Mechanics Measurements.

The mechanics of breathing is a fascinating and vital process. The lung has complexities and subtle heterogeneities in structure across length scales that influence mechanics and function. This study establishes an experimental pipeline for capturing alveolar deformations during a respiratory cycle using synchrotron radiation micro-computed tomography (SR-micro-CT). Rodent lungs were mechanically ventilated and imaged at various time points during the respiratory cycle. Pressure-Volume (P-V) characteristics were recorded to capture any changes in overall lung mechanical behaviour during the experiment. A sequence of tomograms was collected from the lungs within the intact thoracic cavity. Digital volume correlation (DVC) was used to compute the three-dimensional strain field at the alveolar level from the time sequence of reconstructed tomograms. Regional differences in ventilation were highlighted during the respiratory cycle, relating the local strains within the lung tissue to the global ventilation measurements. Strains locally reached approximately 150% compared to the averaged regional deformations of approximately 80-100%. Redistribution of air within the lungs was observed during cycling. Regions which were relatively poorly ventilated (low deformations compared to its neighbouring region) were deforming more uniformly at later stages of the experiment (consistent with its neighbouring region). Such heterogenous phenomena are common in everyday breathing. In pathological lungs, some of these non-uniformities in deformation behaviour can become exaggerated, leading to poor function or further damage. The technique presented can help characterize the multiscale biomechanical nature of a given pathology to improve patient management strategies, considering both the local and global lung mechanics.