Filtern
Dokumenttyp
- Posterpräsentation (4) (entfernen)
Sprache
- Englisch (4)
Referierte Publikation
- nein (4)
Schlagworte
- AFM (3)
- Boehmite (3)
- Nanomechanics (2)
- AFM-IR (1)
- Artificial intelligence (1)
- Data privacy (1)
- Epox (1)
- Epoxy (1)
- Epoxy nanocomposite (1)
- Federated learning (1)
Organisationseinheit der BAM
Boehmite nanoparticles (AlOOH) were recently found to improve properties such as crack resistance, shrinkage and compressive strength in epoxy composites. Concentration and surface modification of boehmite nanoparticles are key factors for abovementioned enhancements. To understand the underlying mechanisms, more detailed research of micro- and nanoscopic mechanical properties is required. The presented study aims to investigate the influence of concentration and surface modification of boehmite on the stiffness of the bulk epoxy by means of AFM-based approaches: Force-Distance curves (FDC) on the sub- microscale and Intermodulation AFM and amplitude-dependent force spectroscopy (ADFS) on the nanoscale. For this purpose, stiffness-maps of epoxy filled with boehmite, with and without surface modification (HAc-boehmite) were obtained by FDC. These measurements showed a slight increase in overall stiffness of composite with increasing the nanoparticle content. The stiffening effect was observed to be intensified with HAc-boehmite. Since the lateral resolution of FDC is not high enough to distinguish nanoparticles, this effect was assumed to be due to the inevitable inclusion of nanoparticles in the measurement. By using Intermodulation-AFM, yielding ADFS stiffness maps with the resolution of <10 nm, we were able to calculate the average stiffness of bulk epoxy without the interference of nanoparticles. It was expected that the stiffness of regions faraway from particles would be equal to neat epoxy. In contrast, the results showed a drastic increase in stiffness of epoxy with increasing boehmite concentration (0, 1, 2.5, and 15%) especially in case of introducing 15wt% HAc-boehmite. Another important observation was formation of a spatial structure with non-homogenous stiffness distribution in bulk epoxy with HAc-boehmite. The underlying mechanisms of described observations are not fully understood yet. One hypothesis is the local increase in crosslinking density which we aim to investigate in our further studies by combining Dynamic Mechanical Thermal Analysis (DMTA) and Intermodulation-AFM.
The application of inorganic nanoparticles as reinforcement agent for polymer composites is constantly growing. Improving the performance of the material with desirable properties requires understanding of the interaction between polymer chains and nanoparticles and the properties of the interphase as well. Boehmite, a mineral of aluminum with basic unit of AlO(OH), is a novel and promising nanofiller which leads to enhanced performance of polymer composites. It has been recently reported that boehmite nanoparticles have reinforcing effect on epoxy matrix in carbon-fiber composites. It was primarily assumed that these improvements are due to very high Young’s modulus of boehmite particles. However, in our latest study we presented new values for the Young’s modulus of boehmite much lower than those reported earlier. This brings up the importance of interphase properties, e.g. the crosslink density, which can have the dominant role in the overall material property. Nevertheless, due to resolution limitations of conventional nanoprobing approaches, characterization of the interphase between individual particles and matrix is a challenge.
In this study, the main goal is to investigate the interphase of the epoxy/boehmite nanocomposites using AFM-based methods. We simplify the three-dimensional nanocomposite system to a two-dimensional horizontally layered sample with a large and easy to access interphase area. For this purpose, 1µm coatings of hydrothermally synthesized boehmite are prepared as the substrate on which the epoxy is later molded and cured. AFM surface potential and force maps were obtained on the cross-sectional cut of epoxy/ boehmite sample. The results show unexpectedly a large interphase area (approx. 1 µm) with different electrical and mechanical properties comparing to bulk epoxy. The average force-distance curves from this region showed more elastic behavior compared to bulk. The underlying mechanism of this influence is not fully understood yet. Thus, further investigation on the interphase region using the novel Nano-IR approach provides more information about the chemical characteristics. Numerical simulation will give complementary information to understand the effect of nanoparticles on the crosslinking density of the interphase. This can be different to bulk epoxy due to either different local temperature gradients or due to a surface selectivity of boehmite towards the monomer molecules.
Inorganic nanoparticles are used to improve the performance of epoxy as the matrix phase in fiber-reinforced composites used for aerospace applications. The effectiveness of nanofillers on property enhancement of thermosetting polymers depends on many factors including the interaction between the functional groups of nanofillers and the polymer reactants. In the current work, we study the effect of boehmite nanoparticles (BNPs) on properties of anhydride-cured bisphenol-A-diglycidyl ether (DGEBA). Dynamic mechanical thermal analysis (DMTA) and a high-resolution force measurement approach called intermodulation atomic force microscopy (ImAFM) were carried out to investigate the thermomechanical and nanomechanical properties of this material, respectively. It was found that BNPs lead to decrease of glass transition temperature (Tg) and crosslink density of the polymer network meanwhile significantly enhancing the Young’s modulus. Besides formation of a soft interphase near the particles, significant changes in local stiffness of polymer matrix far from the interphase was observed with ImAFM. Thus, boehmite induces long-range chemical alteration on the matrix. This effect has a higher impact on overall composite properties compared to the formation of interphase which is only a short-range effect. The local chemical evaluations on the soft interphase using an infrared-AFM method (NanoIR) revealed the accumulation of anhydride hardener near the boehmite interface. Based on these observations the effect of boehmite on the curing of epoxy is hypothesized to be governed by the strong interaction between boehmite and the anhydride. This interaction causes changes the ratio of reactants in the epoxy mixture and hence alteration of curing pathway and the network architecture. In future studies we examine this hypothesis by measuring the thermomechanical properties of cured epoxies in which the epoxy-hardener ratio is systematically altered and further comparing to those properties of nanocomposites shown in the current study.
The deployment of machine learning (ML) and deep learning (DL) in structural health monitoring (SHM) faces multiple challenges. Foremost among these is the insufficient availability of extensive high-quality data sets essential for robust training. Within SHM, high-quality data is defined by its accuracy, relevance, and fidelity in representing real-world structural scenarios (pristine as well as damaged). Although methods like data augmentation and creating synthetic data can add to datasets, they frequently sacrifice the authenticity and true representation of the data. Sharing real-world data encapsulating true structural and anomalous scenarios offers promise. However, entities are often reluctant to share raw data, given the potential extraction of sensitive information, leading to trust issues among collaborating entities.
Our study introduces a novel methodology leveraging Federated Learning (FL) to navigate these challenges. Within the FL framework, models are trained in a decentralized manner across different entities, preserving data privacy. In our research, we simulated several scenarios and compared them to traditional local training methods. Employing guided wave (GW) datasets, we distributed the data among different parties (clients) using IID (independent, identically distributed or in other words, statistically identical) mini batches of dataset, as well as non-IID configurations. This approach mirrors real-world data distribution among varied entities, such as hydrogen refueling stations.
In our methodology, the initial round involves individualized training for each client using their unique datasets . Subsequently, the model parameters are sent to the FL server, where they are averaged to construct a global model. In the second round, this global model is disseminated back to the clients to aid in predictive tasks. This iterative process continues for several rounds to reach convergence.
Our findings distinctly highlight the advantages of FL over localized training, evidenced by a marked improvement in prediction accuracy . This research underscores the potential of FL in GW-based SHM, offering a remedy to similar challenges tied to data scarcity in other SHM approaches and paving the way for a new era of collaborative, data-centric monitoring systems.