TL;DR
This paper introduces a novel simulation approach for modeling the mold-filling behavior of thermoset composites reinforced with recycled materials. By extending existing viscosity models to account for the increased viscosity introduced by these recycled fillers, the researchers developed a tool that can predict processability and guide the sustainable use of recycled thermoset composites in manufacturing.
What Problem Does This Paper Address?
The manufacturing of thermoset composites, particularly through reactive injection molding (RIM), is crucial for lightweight components in industries like automotive and aerospace. However, recycling cured thermoset waste has been a significant challenge due to its cross-linked structure. This paper tackles the problem of effectively reusing mechanically recycled thermoset composite (RTC) materials by blending them with virgin thermoset composite (VTC). While this recycling approach offers sustainability benefits, it introduces complexity in predicting the material's flow behavior during molding. Existing simulation models often fail to accurately capture the increased viscosity and altered flow characteristics caused by the addition of these recycled fillers, leading to potential issues in processability and final part quality.
Methods
The core of this work lies in developing and implementing an enhanced viscosity model for simulating the mold-filling process. The authors extend the established Cross-Castro-Macosko (CCM) model, which is commonly used for thermoset composites, by incorporating a phenomenological approach called the "stress-strain amplification" method. This extension allows the model to account for the increased viscosity due to the presence of recycled fillers, treating the mixture as a polydisperse suspension.
This new constitutive model was implemented within the open-source computational fluid dynamics (CFD) software, OpenFOAM, specifically by extending its standard volume-of-fluid (VoF)-based multiphase solver, compressibleInterFoam. The simulation framework then uses this enhanced model to predict the flow of the VTC/RTC mixture into a mold cavity.
To validate their approach, the researchers conducted experimental trials using a two-cavity mold equipped with pressure sensors. They injected molding compounds with varying proportions of VTC and RTC at different speeds and compared the measured internal pressure profiles with the simulation results.
Datasets
The study utilized experimental data generated from injection molding trials. These trials involved molding compounds with different weight fractions of recycled thermoset composite (RTC) (0 wt.%, 4 wt.%, and 8 wt.%) and were conducted at two different injection speeds (10 cm³/s and 15 cm³/s). The pressure evolution during the mold-filling phase was measured at three distinct locations within the mold. These experimental pressure profiles served as the benchmark for validating the simulation results.
Key Insights and Ideas
- Recycled materials significantly impact viscosity: The addition of RTC to VTC demonstrably increases the overall viscosity of the molding compound. This is a critical factor that needs to be accounted for in process simulations.
- Existing models fall short: Standard viscosity models, like the CCM model, are insufficient for accurately predicting the flow behavior of VTC reinforced with RTC because they don't account for the filler-induced viscosity increase.
- Stress-strain amplification is key: The stress-strain amplification approach provides a viable way to phenomenologically extend existing viscosity models to capture the effects of reinforcing fillers.
- OpenFOAM as a flexible platform: The successful implementation of the novel viscosity model within OpenFOAM highlights the software's adaptability for complex material modeling and multiphase flow simulations.
- Simulation can guide sustainable manufacturing: By accurately predicting mold-filling behavior with recycled materials, simulations can help optimize processing conditions, reduce waste, and promote the wider adoption of sustainable manufacturing practices.
Key Results
The study demonstrated a qualitative agreement between the simulation results and experimental data for mold-filling pressure profiles.
- Increased pressure with RTC: Both simulations and experiments showed that reinforcing VTC with RTC leads to higher injection pressures required for mold filling compared to using VTC alone. This effect was consistent across different sensor locations (P1, P2, P3) and for varying RTC proportions (4 wt.% and 8 wt.%).
- Model reproduces trends: The implemented viscosity model successfully reproduced the observed trends of increasing pressure with higher RTC content and at different injection speeds.
- Quantitative discrepancies exist: While qualitatively accurate, some quantitative discrepancies were observed between simulated and experimental pressure profiles. The authors attribute these to factors like potential errors in rheological characterization, numerical errors, and the assumption of a no-slip boundary condition at the mold wall.
- Shear heating is amplified: Simulations also showed that higher RTC content leads to increased shear heating of the polymer melt during the filling process, resulting in higher melt temperatures.
Limitations and Future Work
The authors acknowledge several limitations and suggest avenues for future research:
- Wall-slip boundary condition: The use of a no-slip boundary condition at the mold wall led to overpredicted pressures, especially at later stages of mold filling. Implementing a non-ideal wall-slip model is suggested to improve accuracy.
- Temperature validation: Direct validation of simulated melt temperatures was not possible due to the lack of integrated temperature sensors in the experimental mold. Future work should include actual temperature measurements.
- Anisotropic filler orientation: The current model primarily considers spherical fillers. While the framework is adaptable to anisotropic fillers, the instantaneous orientation of such fillers and their impact on viscosity during mold filling were not explicitly modeled.
- Experimental data fitting: Potential errors in fitting rheological data and numerical simulation errors could contribute to discrepancies.
- Further model refinement: Future studies could explore incorporating more sophisticated models for wall-slip and mold-melt heat transfer within OpenFOAM to further enhance simulation accuracy.
Summary updated