Augmented semantic segmentation for the digitization of grinding tools based on deep learning

In order to analyze various process characteristics, grinding simulations can be used, which need accurate models of the tool and the individual grains. For this purpose, grinding tools can be digitized. To identify characteristic grains from a large number of measurements, each individual grain has to be analyzed and separated from the bond manually. Therefore, a deep learning-based methodology was developed to achieve a high segmentation accuracy of the grain boundaries efficiently. Additionally, a data augmentation approach was investigated to limit the data necessary for learning. The model transferability was quantified by analyzing different states of tool wear.     

Role of salt on adhesion of an epoxy/aluminum (oxide) interface in aqueous environments

Joints held by polymeric adhesives are commonplace in many engineered products, but normal service can require exposure to environmental conditions that present a significant challenge for maintaining the structural integrity of the interface. In particular, aqueous environments can wreak havoc on the joint strength. Here, a mechanistic approach is used to understand the difference in the debonding behavior of an epoxy/aluminum (oxide) interface when exposed to deionized (DI) water and aqueous sodium chloride by correlating macroscopic failure with the sorption of salt and water into the adhesive and its nanoscale distribution. For the epoxy‐aluminum system examined here, the presence of sodium chloride increases the resistance to crack growth in comparison to DI water. The debonding appears to be controlled by water near the buried interface. Salt water decreases the solubility of water in the epoxy and decreases the concentration of water near the buried interface, but the concentration of salt that enters the epoxy is below the detection limit. Thus, even if ions cannot penetrate or sorb into the adhesive, the presence of salt can significantly alter the water distribution within the adhesive and ultimately the strength of the joint. POLYM. ENG. SCI., 56:18–26, 2016. © 2015 Society of Plastics Engineers     

Reduction of the elongation at break of thermoplastic polyolefins through melt blending with polylactide and the influence of the amount of compatibilizers and the viscosity ratios of the blend components on phase morphology and mechanics

The objective of this work is the synthesis of a polypropylene/ethylene‐propylene‐rubber (TPO)/polylactide (PLA)/compatibilizer (PVM) blend to reduce the elongation at break of TPO by blending TPO with brittle PLA. Three TPO types with different viscosities were melt blended with PLA and an ethylene/n‐butylacrylate/glycidyl methacrylate terpolymer (PVM) as reactive compatibilizer. All blends had a constant PLA amount of 30 wt%. Two parameters were varied in the experiments, viscosity of the TPO types, and amount of PVM used in the blends. Both parameters played important roles in reducing the nominal elongation at break compared to pure TPO foils and influencing the phase morphology of extruded blend foils. The nominal elongation at break could be reduced by 100‐150% through blending TPO with PLA and PVM. Characterization regarding the blend morphology, especially the size and shape of the dispersed PLA phase in the TPO matrix was done by Environmental Scanning Electron Microscopy (ESEM) images. Investigations of the morphology showed that size and shape of dispersed PLA phases are dependent on the viscosity ratios of the blend components and on the amount of compatibilizer in the blend. AFM images of the polymer blends reveal soft rubbery layers around the dispersed PLA phases. POLYM. ENG. SCI., 56:905–913, 2016. © 2016 Society of Plastics Engineers