Search Results
An AI-based open recommender system for personalized labor market driven education
Attaining those skills that match labor market demand is getting increasingly complicated, not in the last place in engineering education, as prerequisite knowledge, skills, and abilities are evolving dynamically through an uncontrollable and seemingly unpredictable process. Anticipating and addressing such dynamism is a fundamental challenge to twenty-first century education. The burgeoning availability of data, not only on the demand side but also on the supply side (in the form of open educational resources) coupled with smart technologies, may provide a fertile ground for addressing this challenge. In this paper, we propose a novel, Artificial Intelligence (AI) driven approach to the development of an open, personalized, and labor market oriented learning recommender system, called eDoer. We discuss the complete system development cycle starting with a systematic user requirements gathering, and followed by system design, implementation, and validation. Our recommender prototype (1) derives the skill requirements for particular occupations through an analysis of online job vacancy announcements
Resolving mobility anisotropy in quasi-free-standing epitaxial graphene by terahertz optical Hall effect
In this work, we demonstrate the application of terahertz-optical Hall effect (THz-OHE) to determine directionally dependent free charge carrier properties of ambient-doped monolayer and quasi-free-standing-bilayer epitaxial graphene on 4H–SiC(0001). Directionally independent free hole mobility parameters are found for the monolayer graphene. In contrast, anisotropic hole mobility parameters with a lower mobility in direction perpendicular to the SiC surface steps and higher along the steps in quasi-free-standing-bilayer graphene are determined for the first time. A combination of THz-OHE, nanoscale microscopy and optical spectroscopy techniques are used to investigate the origin of the anisotropy. Different defect densities and different number of graphene layers on the step edges and terraces are ruled out as possible causes. Scattering mechanisms related to doping variations at the step edges and terraces as a result of different interaction with the substrate and environment are discussed and also excluded. It is suggested that the step edges introduce intrinsic scattering in quasi-free-standing-bilayer graphene, that is manifested as a result of the higher ratio between mean free path and average terrace width parameters. The suggested scenario allows to reconcile existing differences in the literature regarding the anisotropic electrical transport in epitaxial graphene. © 2020 Elsevier Ltd
A novel approach to fabricate load-bearing Ti6Al4V-Barium titanate piezoelectric bone scaffolds by coupling electron beam melting and field-assisted sintering
A critical-size bone defect in load-bearing areas is a challenging clinical problem in orthopaedic surgery. Titanium alloy (Ti6Al4V) scaffolds have advantages because of their biomechanical stability but lack electrical activity, which hinders their further use. This work is focused on the fabrication of Ti6Al4V-Barium Titanate (BaTiO3) bulk composite scaffolds to combine the biomechanical stability of Ti6Al4V with electrical activity through BaTiO3. For the first time, a hollow cylindrical Ti6Al4V is additively manufactured by electron beam melting and combined with piezoelectric BaTiO3 powder for joint processing in field-assisted sintering. Scanning electron microscope images on the interface of the Ti6Al4V-BaTiO3 composite scaffold showed that after sintering, the Ti6Al4V lattice structure bounded with BaTiO3 matrix without its major deformation. The Ti6Al4V-BaTiO3 scaffold had average piezoelectric constants of (0.63 ± 0.12) pC/N directly after sintering due to partial dipole alignment of the BaTiO3 tetragonal phase, which increased to (4.92 ± 0.75) pC/N after a successful corona poling. Moreover, the nanoindentation values of Ti6Al4V exhibited an average hardness and Young's modulus of (5.9 ± 0.9) GPa and (130 ± 14) GPa, and BaTiO3 showed (4.0 ± 0.6) GPa and (106 ± 10) GPa, respectively. It reveals that the Ti6Al4V is the harder and stiffer part in the Ti6Al4V-BaTiO3 composite scaffold. Such a scaffold has the potential to treat critical-size bone defects in load-bearing areas and guide tissue regeneration by physical stimulation.
Laser powder bed fusion of a superelastic Cu-Al-Mn shape memory alloy
Dense and crack-free specimens of the shape memory alloy Cu71.6Al17Mn11.4 (at.%) were produced via laser powder bed fusion across a wide range of process parameters. The microstructure, viz. grain size, can be directly tailored within the process and with it the transformation temperatures (TTs) shifted to higher values by raising the energy input. The microstructure, and the superelastic behavior of additively manufactured samples were assessed by a detailed comparison with induction melted material. The precipitation of the α phase, which inhibit the martensitic transformation, were not observed in the additively manufactured samples owing to the high intrinsic cooling rates during the fabrication process. Fine columnar grains with a strong [001]-texture along the building direction lead to an enhanced yield strength compared to the coarse-grained cast samples. A maximum recoverable strain of 2.86% was observed after 5% compressive loading. The first results of our approach imply that laser powder bed fusion is a promising technique to directly produce individually designed Cu-Al-Mn shape memory parts with a pronounced superelasticity at room temperature.
Monitoring the thermally induced transition from sp3-hybridized into sp2-hybridized carbons
The preparation of carbons for technical applications is typically based on a treatment of a precursor, which is transformed into the carbon phase with the desired structural properties. During such treatment the material passes through several different structural stages, for example, starting from precursor molecules via an amorphous phase into crystalline-like phases. While the structure of non-graphitic and graphitic carbon has been well studied, the transformation stages from molecular to amorphous and non-graphitic carbon are still not fully understood. Disordered carbon often contains a mixture of sp3-, sp2-and sp1-hybridized bonds, whose analysis is difficult to interpret. We systematically address this issue by studying the transformation of purely sp3-hybridized carbons, that is, nanodiamond and adamantane, into sp2-hybridized non-graphitic and graphitic carbon. The precursor materials are thermally treated at different temperatures and the transformation stages are monitored. We employ Raman spectroscopy, WAXS and TEM to characterize the structural changes. We correlate the intensities and positions of the Raman bands with the lateral crystallite size La estimated by WAXS analysis. The behavior of the D and G Raman bands characteristic for sp2-type material formed by transforming the sp3-hybridized precursors into non-graphitic and graphitic carbon agrees well with that observed using sp2-structured precursors.
From microfluidics to hierarchical hydrogel materials
Over the past two decades, microfluidics has made significant contributions to material and life sciences, particularly via the design of nano-, micro- and mesoscale materials such as nanoparticles, micelles, vesicles, emulsion droplets, and microgels. Unmatched in control over a multitude of material parameters, microfluidics has also shed light on fundamental aspects of material design such as the early stages of nucleation and growth processes as well as structure evolution. Exemplarily, polymer hydrogel particles can be formed via microfluidics with exact control over size, shape, functionalization, compartmentalization, and mechanics that is hardly found in any other processing method. Interestingly, the utilization of microfluidics for material design largely focuses on the fabrication of single entities that act as reaction volume for organic and cell-free biosynthesis, cell mimics, or local environment for cell culturing. In recent years, however, hydrogel design has shifted towards structures that integrate a large variety of functions, e.g., to address the demands for sensing tasks in a complex environment or more closely mimicking architecture and organization of tissue by multiparametric cultures. Hence, this review provides an overview of recent literature that explores microfluidics for fabricating hydrogel materials that go well beyond common length scales as well as the structural and functional complexity of microgels necessary to produce hierarchical hydrogel structures. We focus on examples that utilize microfluidics to design microgel-based assemblies, on microfluidically made polymer microgels for 3D bioprinting, on hydrogels fabricated by microfluidics in a continuous fashion, like fibers, and on hydrogel structures that are shaped by microchannels.
Machine learning for additive manufacturing: Predicting materials characteristics and their uncertainty
Additive manufacturing (AM) is known for versatile fabrication of complex parts, while also allowing the synthesis of materials with desired microstructures and resulting properties. These benefits come at a cost: process control to manufacture parts within given specifications is very challenging due to the relevance of a large number of processing parameters. Efficient predictive machine learning (ML) models trained on small datasets, can minimize this cost. They also allow to assess the quality of the dataset inclusive of uncertainty. This is important in order for additively manufactured parts to meet property specifications not only on average, but also within a given variance or uncertainty. Here, we demonstrate this strategy by developing a heteroscedastic Gaussian process (HGP) model, from a dataset based on laser powder bed fusion of a glass-forming alloy at varying processing parameters. Using amorphicity as the microstructural descriptor, we train the model on our Zr52.5Cu17.9Ni14.6Al10Ti5 (at.%) alloy dataset. The HGP model not only accurately predicts the mean value of amorphicity, but also provides the respective uncertainty. The quantification of the aleatoric and epistemic uncertainty contributions allows to assess intrinsic inaccuracies of the dataset, as well as identify underlying physical phenomena. This HGP model approach enables to systematically improve ML-driven AM processes.
Effect of pore geometry on ultra-densified hydrogen in microporous carbons
Our investigations into molecular hydrogen (H2) confined in microporous carbons with different pore geometries at 77 K have provided detailed information on effects of pore shape on densification of confined H2 at pressures up to 15 MPa. We selected three materials: a disordered, phenolic resin-based activated carbon, a graphitic carbon with slit-shaped pores (titanium carbide-derived carbon), and single-walled carbon nanotubes, all with comparable pore sizes of <1 nm. We show via a combination of in situ inelastic neutron scattering studies, high-pressure H2 adsorption measurements, and molecular modelling that both slit-shaped and cylindrical pores with a diameter of ∼0.7 nm lead to significant H2 densification compared to bulk hydrogen under the same conditions, with only subtle differences in hydrogen packing (and hence density) due to geometric constraints. While pore geometry may play some part in influencing the diffusion kinetics and packing arrangement of hydrogen molecules in pores, pore size remains the critical factor determining hydrogen storage capacities. This confirmation of the effects of pore geometry and pore size on the confinement of molecules is essential in understanding and guiding the development and scale-up of porous adsorbents that are tailored for maximising H2 storage capacities, in particular for sustainable energy applications.
Microbiome-based biotechnology for reducing food loss post harvest
Microbiomes have an immense potential to enhance plant resilience to various biotic and abiotic stresses. However, intrinsic microbial communities respond to changes in their host's physiology and environment during plant's life cycle. The potential of the inherent plant microbiome has been neglected for a long time, especially for the postharvest period. Currently, close to 50% of all produced fruits and vegetables are lost either during production or storage. Biological control of spoilage and storage diseases is still lacking sufficiency. Today, novel multiomics technologies allow us to study the microbiome and its responses on a community level, which will help to advance current classic approaches and develop more effective and robust microbiome-based solutions for fruit and vegetable storability, quality, and safety.
Thermal annealing to influence the vapor sensing behavior of co-continuous poly(lactic acid)/polystyrene/multiwalled carbon nanotube composites
With the main purpose of being used as vapor leakage detector, the volatile organic compound (VOC) vapor sensing properties of conductive polymer blend composites were studied. Poly(lactic acid)/polystyrene/multi-walled carbon nanotube (PLA/PS/MWCNT) based conductive polymer composites (CPCs) in which the polymer components exhibit different interactions with the vapors, were prepared by melt mixing. CPCs with a blend composition of 50/50 wt% resulted in the finest co-continuous structure and selective MWCNT localization in PLA. Therefore, these composites were selected for sensor tests. Thermal annealing was applied aiming to maintain the blend structure but improving the sensing reversibility of CPC sensors towards high vapor concentrations. Different sensing protocols were applied using acetone (good solvent for PS and PLA) and cyclohexane (good solvent for PS but poor solvent for PLA) vapors. Increasing acetone vapor concentration resulted in increased relative resistance change (Rrel) of CPCs. Saturated cyclohexane vapor resulted in lower response than nearly saturated acetone vapor. The thermal annealing at 150 °C did not change the blend morphology but increased the PLA crystallinity, making the CPC sensors more resistant to vapor stimulation, resulting in lower Rrel but better reversibility after vapor exposure.