Wissenschaftliche Artikel der BAM
Filtern
Erscheinungsjahr
- 2023 (20) (entfernen)
Dokumenttyp
Schlagworte
- X-ray computed tomography (5)
- Additive manufacturing (3)
- Creep (2)
- Microstructure (2)
- Röntgen-Refraktion (2)
- Simulation (2)
- Synthetic Training Data (2)
- dXCT (2)
- 3D deep convolutional neural network (3D DCNN) (1)
- 3D microstructure (1)
Organisationseinheit der BAM
- 8.5 Röntgenbildgebung (20) (entfernen)
Li dendrites penetration through solid electrolytes (SEs) challenges the development of solid-state Li batteries (SSLBs). To date, significant efforts are devoted to understand the mechanistic dynamics of Li dendrites nucleation, growth, and propagation in SEs, and various strategies that aim to alleviate and even inhibit Li dendrite formation have been proposed. Nevertheless, most of these conventional strategies require either additional material processing steps or new materials/layers that eventually increase battery cost and complexity. In contrast, using external fields, such as mechanical force, temperature physical field, electric field, pulse current, and even magnetic field to regulate Li dendrites penetration through SEs, seems to be one of the most cost-effective strategies. This review focuses on the current research progress of utilizing external physical fields in regulating Li dendrites growth in SSLBs. For this purpose, the mechanical properties of Li and SEs, as well as the experimental results that visually track Li penetration dynamics, are reviewed. Finally, the review ends with remaining open questions in future studies of Li dendrites growth and penetration in SEs. It is hoped this review can shed some light on understanding the complex Li dendrite issues in SSLBs and potentially guide their rational design for further development.
The damage mechanisms of metallic components produced by process laser powder bed fusion differ significantly from those typically observed in conventionally manufactured variants of the same alloy. This is due to the unique microstructures of additively manufactured materials. Herein, the focus is on the study of the evolution of creep damage in stainless steel 316L specimens produced by laser powder bed fusion. X-ray computed tomography is used to unravel the influence of the process-specific microstructure from the influence of the initial void distribution on creep damage mechanisms. The void distribution of two specimens tested at 600 °C and 650 °C is analyzed before a creep test, after an interruption, and after fracture. The results indicate that the formation of damage is not connected to the initial void distribution. Instead, damage accumulation at grain boundaries resulting from intergranular cracking is observed.
It is shown that preconditioning of experimental X‐ray computed tomography (XCT) data is critical to achieve high‐precision segmentation scores. The challenging experimental XCT datasets and deep convolutional neural networks (DCNNs) are used that are trained with low‐resemblance synthetic XCT data. The material used is a 6‐phase Al–Si metal matrix composite‐reinforced with ceramic fibers and particles. To achieve generalization, in our past studies, specific data augmentation techniques were proposed for the synthetic XCT training data. In addition, two toolsets are devised: (1) special 3D DCNN architecture (3D Triple_UNet), slicing the experimental XCT data from multiple views (MultiView Forwarding), the i.S.Sy.Da.T.A. iterative segmentation algorithm, and (2) nonlocal means (NLM) conditioning (filtering) for the experimental XCT data. This results in good segmentation Dice scores across all phases compared to more standard approaches (i.e., standard UNet architecture, single view slicing, standard single training, and NLM conditioning). Herein, the NLM filter is replaced with the deep conditioning framework BAM SynthCOND introduced in a previous publication, which can be trained with synthetic XCT data. This leads to a significant segmentation precision increase for all phases. The proposed methods are potentially applicable to other materials and imaging techniques.
The greatest challenge when using deep convolutional neural networks (DCNNs) for automatic segmentation of microstructural X-ray computed tomography (XCT) data is the acquisition of sufficient and relevant data to train the working network. Traditionally, these have been attained by manually annotating a few slices for 2D DCNNs. However, complex multiphase microstructures would presumably be better segmented with 3D networks. However, manual segmentation labeling for 3D problems is prohibitive. In this work, we introduce a method for generating synthetic XCT data for a challenging six-phase Al–Si alloy composite reinforced with ceramic fibers and particles. Moreover, we propose certain data augmentations (brightness, contrast, noise, and blur), a special in-house designed deep convolutional neural network (Triple UNet), and a multi-view forwarding strategy to promote generalized learning from synthetic data and therefore achieve successful segmentations. We obtain an overall Dice score of 0.77. Lastly, we prove the detrimental effects of artifacts in the XCT data on achieving accurate segmentations when synthetic data are employed for training the DCNNs. The methods presented in this work are applicable to other materials and imaging techniques as well. Successful segmentation coupled with neural networks trained with synthetic data will accelerate scientific output.
Regardless of the experimental care practiced in acquiring X-ray computed tomography (XCT) data, artifacts might still exist, such as noise and blur. This is typical for fast XCT data acquisitions (e.g., in-situ investigations), or low-dose XCT. Such artifacts can complicate subsequent analysis of the data. Digital filters can moderately cure extensive artifacts. The selection of filter type, intensity, and order of application is not always straight forward. To tackle these problems, a complete sequential multilevel, multi-scale framework: BAM SynthCOND, employing newly designed deep convolutional neural networks (DCNNs), was formulated. Although data conditioning with neural networks is not uncommon, the main complication is that completely artifact-free XCT data for training do not exist. Thus, training data were acquired from an in-house developed library (BAM SynthMAT) capable of generating synthetic XCT material microstructures. Some novel DCNN architectures were introduced (2D/3D ACEnet_Denoise, 2D/3D ACEnet_Deblur) along with the concept of Assertive Contrast Enhancement (ACE) training, which boosts the performance of neural networks trained with continuous loss functions. The proposed methodology accomplished very good generalization from low resemblance synthetic training data. Indeed, denoising, sharpening (deblurring), and even ring artifact removal performance were achieved on experimental post-CT scans of challenging multiphase Al-Si Metal Matrix Composite (MMC) microstructures. The conditioning efficiencies were: 92% for combined denoising/sharpening, 99% for standalone denoising, and 95% for standalone sharpening. The results proved to be independent of the artifact intensity. We believe that the novel concepts and methodology developed in this work can be directly applied on the CT projections prior to reconstruction, or easily be extended to other imaging techniques such as: Microscopy, Neutron Tomography, Ultrasonics, etc.
The lack of traceability to meter of X-ray Computed Tomography (CT) measurements still hinders a more extensive acceptance of CT in coordinate metrology and industry. To ensure traceable, reliable, and accurate measurements, the determination of the task-specific measurement uncertainty is necessary. The German guideline VDI/VDE 2630 part 2.1 describes a procedure to determine the measurement uncertainty for CT experimentally by conducting several repeated measurements with a calibrated test specimen. However, this experimental procedure is cost and effort intensive. Therefore, the simulation of dimensional measurement tasks conducted with X-ray computed tomography can close these drawbacks. Additionally, recent developments towards a resource and cost-efficient production (“smart factory”) motivate the need for a corresponding numerical model of a CT system (“digital twin”) as well. As there is no standardized procedure to determine the measurement uncertainty of a CT system by simulation at the moment, the project series CTSimU was initiated, aiming at this gap. Concretely, the goal is the development of a procedure to determine the measurement uncertainty numerically by radiographic simulation. The first project (2019-2022), "Radiographic Computed Tomography Simulation for Measurement Uncertainty Evaluation - CTSimU" developed a framework to qualify a radiographic simulation software concerning the correct simulation of physical laws and functionalities. The most important outcome was a draft for a new guideline VDI/VDE 2630 part 2.2, which is currently under discussion in the VDI/VDE committee. The follow-up project CTSimU2 "Realistic Simulation of real CT systems with a basic-qualified Simulation Software" will deal with building and characterizing a digital replica of a specific real-world CT system. The two main targets of this project will be a toolbox including methods and procedures to configure a realistic CT system simulation and to develop tests to check if this replica is sufficient enough. The result will be a draft for a follow-up VDI/VDE guideline proposing standardized procedures to determine a CT system's corresponding characteristics and test the simulation (copy) of a real-world CT system which we call a "digital twin".
The prediction of porosity is a crucial task for metal based additive manufacturing techniques such as laser powder bed fusion. Short wave infrared thermography as an in-situ monitoring tool enables the measurement of the surface radiosity during the laser exposure. Based on the thermogram data, the thermal history of the component can be reconstructed which is closely related to the resulting mechanical properties and to the formation of porosity in the part. In this study, we present a novel framework for the local prediction of porosity based on extracted features from thermogram data. The framework consists of a data pre-processing workflow and a supervised deep learning classifier architecture. The data pre-processing
workflow generates samples from thermogram feature data by including feature information from multiple subsequent layers.
Thereby, the prediction of the occurrence of complex process phenomena such as keyhole pores is enabled. A custom convolutional neural network model is used for classification. Themodel is trained and tested on a dataset from thermographic in-situ monitoring of the manufacturing of an AISI 316L stainless steel test component. The impact of the pre-processing parameters and the local void distribution on the classification performance is studied in detail. The presented model achieves an accuracy of 0.96 and an f1-Score of 0.86 for predicting keyhole porosity in small sub-volumes with a dimension of (700 × 700 × 50) μm3. Furthermore, we show that pre-processing parameters such as the porosity threshold for sample
labeling and the number of included subsequent layers are influential for the model performance. Moreover, the model prediction is shown to be sensitive to local porosity changes although it is trained on binary labeled data that disregards the actual sample porosity.
The osseointegration in/around additively manufactured (AM) lattice structures of a new titanium alloy, Ti–19Nb–14Zr, was evaluated. Different lattices with increasingly high sidewalls gradually closing them were manufactured and implanted in sheep. After removal, the bone–interface implant (BII) and bone–implant contact (BIC) were studied from 3D X-ray computed tomography images. Measured BII of less than 10 µm and BIC of 95% are evidence of excellent osseointegration.
Since AMnaturally leads to a high-roughness surface finish, the wettability of the implant is increased.
The new alloy possesses an increased affinity to the bone. The lattice provides crevices in which the biological tissue can jump in and cling. The combination of these factors is pushing ossification beyond its natural limits. Therefore, the quality and speed of the ossification and osseointegration in/around these Ti–19Nb–14Zr laterally closed lattice implants open the possibility of bone spline key of prostheses. This enables the stabilization of the implant into the bone while keeping the possibility of punctual hooks allowing the implant to be removed more easily if required. Thus, this new titanium alloy and such laterally closed lattice structures are appropriate candidates to be implemented in a new generation of implants.
Die Brechung von Röntgenstrahlen (Röntgenrefraktion) an Grenzflächen zwischen Materialien unterschiedlicher Dichte ist analog zur Ablenk-ung von sichtbarem Licht an z.B. Glasoberflächen. Es gibt jedoch zwei wesentliche Unterschiede:
a) konvexe Grenzflächen verursachen Divergenz (d.h. der Brechungsindex n ist kleiner als 1), und
b) die Ablenkungswinkel sind sehr klein, und reichen von einigen Bogensekunden bis zu einigen Bogenminuten (d.h. n ist nahe bei 1);
Wie auch bei sichtbarem Licht ist die Ablenkungsrichtung der Röntgenstrahlen abhängig von der Orientierung der durchstrahlten Grenzfläche. Aufgrund dieser Eigenschaften eignen sich Röntgenrefraktionsmethoden hervorragend für:
a) die Erkennung und Quantifizierung von Defekten wie Poren und Mikrorissen und
b) die Bewertung von Porosität und Partikeleigenschaften wie Orientierung, Größe und räumliche Verteilung.
Wir zeigen die Anwendung der Röntgenrefraktionsradiographie (2,5D Technik) und der -tomographie (3D Technik) für die Untersuchung verschiedener Probleme in der Werkstoffwissenschaft und -technologie:
a) Sintern von SiC-Grünkörpern
b) Porositätsanalyse in Dieselpartikelfiltersilikaten
c) Faser-Matrix-Haftung in Metall- und Polymermatrixverbundwerkstoffen
d) Mikrorissbildung in Glaskeramik.
Wir zeigen, dass der Einsatz von Röntgenrefraktionsmethoden quantitative Ergebnisse liefert, die direkt als Parameter in Werkstoffmodellen verwendet werden können.
The stress–strain behavior of ceramics, such as aluminum titanate, has certain features that are unusual for brittle materials—in particular, a substantial nonlinearity under uniaxial tension, and load–unload hysteresis caused by the sharp increase of the incremental stiffness at the beginning of unloading. These features are observed experimentally and are attributed to microcracking. Here we compare different degrees of stress–strain nonlinearity of aluminum titanate materials and quantitatively model them. We use advanced mechanical testing to observe the mechanical response at room and high temperature; electron microscopy, and X-ray refraction radiography to observe the microstructural changes. Experiments show that two types of microcracks can be distinguished: (i) microcracks induced by cooling from the sintering temperature (due to heterogeneity and anisotropy of thermal expansion), with typical sizes of the order of grain size, and (ii) much larger microcracks generated by the mechanical loading. The two microcrack types produce different effects on the stress–strain curves. Such microcracks and the features of the stress–strain behavior depend on the density of the cooling-induced microcracks and on the distribution of grain sizes. They are modeled analytically and numerically.