Slow crack growth resistance and bridging stress determination in alumina-rich magnesium aluminate spinel/tungsten composites

The slow crack growth (SCG) resistance (V–K_I diagrams) of magnesium aluminate spinel and its tungsten composites with different metallic content (7, 10, 14 and 22 vol.%) is reported. It is found that tungsten plays a crucial role in the composite by increasing crack resistance: the higher the W content, the higher the stress intensity factor needed for crack extension at a given rate. The reinforcement is due to the bridging mechanism performed by metal particles, as it strongly affects the compliance of cracked specimens. Its magnitude is estimated by a compliance function Φ(a) from a double torsion test. From the compliance function, R-curve behaviour is predicted for the composite with highest tungsten content. It explains the effect of metal particles on SCG curves. The W–MgAl₂O₄ interface is believed to influence the reinforcement mechanism.     

Atomic Force Microscopy Study and Qualitative Analysis of Martensite Relief in Zirconia

A recent report [S. Deville and J. Chevalier, J. Am. Ceram. Soc ., 86[12], 2225 (2003)] has shown the new possibilities offered by atomic force microscopy (AFM) to investigate martensitic transformation-induced relief in zirconia. In this paper, we studied qualitatively the surface relief resulting from martensitic tetragonal to monoclinic phase transformation in yttria and ceria-doped zirconia by AFM. AFM appears as a very powerful tool to investigate martensite relief with great precision. The phenomenological theory of martensitic crystallography could be successfully applied to explain all the observed features. The formation conditions of martensite are discussed, as well as ways of accommodating locally the transformation strain, i.e., self-accommodating variant pairs and microcracking. Variant growth sequences are observed. These observations bring new insights and explanations on the transformation initiation and propagation sequences.     

Transfer learning from synthetic labels for histopathological images classification

This study introduces a new strategy that combines unsupervised learning (clustering) and transfer learning. Clustering methods are employed to generate synthetic labels for the source dataset (ICAR-2018). The generated dataset is then used for transfer learning to other histopathological datasets (KimiaPath960, CRC, Biomaging??2015, Breakhis, and Lymphoma). The comparative study based on two clustering algorithms (K-means and multi-objective clustering stream) demonstrates the efficiency of MOC-Stream. The generated synthetic histopathological dataset by this clustering algorithm outperformed the original labeled dataset and the imageNet models in transfer learning.

     

Quantitative Analysis of Crack Shielding Degradation During Cyclic Fatigue of Alumina

Grain bridging degradation behind a crack tip is the main cyclic fatigue mechanism in nontransforming ceramics. In this work, a compliance function is used to quantify the shielding capacity of grain bridges during cyclic loading of alumina ceramics with different grain sizes. This allows to identify the different stages occurring during cyclic loading. Significant degradation is observed in the coarse grain material and a marked sensitivity to the loading level is outlined. At moderate loads, bridging degradation occurs prior to fatigue crack growth during an incubation period which can reach several million cycles. At low cyclic loads, the shielding capacity can be entirely degraded, leading to a cyclic fatigue threshold equivalent to that of the fine grain material.