Wissenschaftliche Artikel der BAM
Filtern
Dokumenttyp
Sprache
- Englisch (3) (entfernen)
Schlagworte
- Dripping (2)
- PFEM (2)
- UL 94 (2)
- Broadband dielectric spectroscopy (1)
- Deep learning (1)
- Electrochemical impedance spectroscopy (1)
- Equivalent circuit (1)
- Fire behaviour (1)
- Flammability (1)
- High heat resistance (1)
Organisationseinheit der BAM
Machine learning-assisted equivalent circuit identification for dielectric spectroscopy of polymers
(2024)
Polymers have become indispensable across fields of application, and understanding their structure–property relationships and dynamic behaviour is essential for performance optimization. Polymer membranes, particularly ion exchange membranes, play a crucial role in renewable energy conversion technologies, fuel cells, solar energy conversion, and energy storage. In this context, broadband dielectric spectroscopy (BDS) offers a powerful, non-destructive approach to investigate the electrical response and relaxation dynamics of polymers. These properties are investigated by parametrizing the system’s impedance response in terms of a network of circuit elements, i.e. the electrical equivalent circuit (EEC), whose impedance resembles the one of the system under investigation. However, the determination of the EEC from BDS data is challenging due to system complexity, interdependencies of circuit elements, and researcher biases. In this work, we propose a novel approach that incorporates a convolutional neural network (CNN) model to predict the EEC topology. By reducing user bias and enhancing data analysis, this approach aims to make BDS accessible to both experienced users and those with limited expertise. We show that the combination of machine learning and BDS provides valuable insights into the dynamic behaviour of polymer membranes, thus facilitating the design and characterization of tailored polymers for various applications. We also show that our model outperforms state-of-the-art machine learning methods with a top-5 accuracy of around 80% for predicting the circuit topology and a parameter fitting error as low as 0.05%.
The dripping behaviour of polymers is often observed experimentally through the UL94 flammability standard test. In this work, polymeric dripping under fire is investigated numerically using particle finite element method. A parametric analysis was carried out to observe the influence of a single property on overall dripping behaviour via a UL94 vertical test model. Surrogates and property ranges were defined for variation of the following parameters: glass transition temperature (Tg), melting temperature (Tm), decomposition temperature (Td), density (ρ), specific heat capacity (Cp), apparent effective heat of combustion of the volatiles, char yield (μ), thermal conductivity (k), and viscosity (η). Polyamide, poly(ether ether ketone), poly(methyl methacrylate), and polysulfone were used as benchmarks. Simulated results showed that specific heat capacity, thermal conductivity, and char yield allied with viscosity were the properties that most influenced dripping behaviour (starting time and occurrence).
Polyamide 4.6 (PA46) is a high-heat-resistant polymer, but it has no dripping resistance under fire. Three commercial grades of PA46 are investigated under UL 94 vertical fire test conditions. Their performances are discussed based on the materials’ structural, thermal, and rheological properties. PA46 presents flaming drops, whereas dripping is prevented in the flame-retarded PA46.
Friction-modified PA46 has increased flaming dripping. Temperature profiles of the specimens under fire and the temperature of the drops are measured by thermocouples. A UL 94 vertical test configuration consisting of two flame applications is designed to assess the quantitative dripping behavior of the set of materials by the particle finite element method (PFEM). Polymer properties (activation energy and Arrhenius coefficient of decomposition, char yield, density, effective heat of combustion, heat of decomposition, specific heat capacity, and thermal conductivity) in addition to rheological responses in high temperatures are estimated and measured as input parameters for the simulations. The dripping behavior obtained by simulated materials corresponds with the experimental results in terms of time and drop size. A consistent picture of the interplay of the different phenomena controlling dripping under fire appears to deliver a better understanding of the role of different materials’ properties