Adhesion Properties and Machine Learning Modeling of Multilayer Thermoplastic Composites
Bibliographic record
Abstract
The advancement of multilayer thermoplastic composites necessitates the development of robust adhesive materials that can withstand high temperatures and diverse mechanical stresses. This thesis presents a comprehensive approach to enhancing thermoplastic adhesives through material innovations and machine learning modeling, aiming to improve the performance and reliability of multilayer composites in demanding applications. First in this study, an immiscible blend adhesive comprising Polyethylene of Raised Temperature, Polyamide 12 was developed. By optimizing the adhesive layer composition, the resulting trilayer composite demonstrated significantly enhanced barrier properties, and mechanical strength in Young’s modulus, creep resistance, and impact absorption, highlighting the blend's suitability for high-temperature, high-pressure applications. Secondly, the incorporation of carbon fibers into adhesive matrix was investigated to address weak adhesive properties at elevated temperatures. Utilizing a novel T-peel test under controlled conditions, CF reinforcement achieved remarkable increases in peel strength. The enhancement mechanisms were elucidated through macro-level improvements such as an expanded peel zone and elimination of crazing, and micro-level factors including stress transfer and energy dispersion into micro peel zones, thereby significantly boosting the adhesive performance under thermal stress. Thirdly, the interface between carbon fibers and thermoplastic matrices was strengthened through nanostructure surface modification by graphene nanoplatelet coating. The coated carbon fibers exhibited an improvement in interfacial shear strength with polyethylene matrices, while a reduction with PA6 due to differing failure mechanisms. Comprehensive morphological, chemical, and wettability analyses, supported by machine learning-based image segmentation, X-ray photoelectron spectroscopy, and contact-angle measurements, provided a detailed understanding of the interfacial enhancements at the micro and nanoscale. Lastly, an Advanced Multilayer Perceptron Regressor model was developed to predict the peel strength of coextruded multilayer thermoplastic composites. This machine learning approach effectively captured the complex relationships between various input parameters and composite properties, despite being trained on a limited dataset. The model demonstrated robust predictive capabilities, validated through benchmark metrics and k-fold cross-validation. Additionally, feature importance analysis and dimensionality reduction facilitated a deeper insight into the key factors influencing adhesive strength, thereby enabling optimized design strategies for multilayer composite manufacturing. This thesis integrates material science innovations with advanced machine learning techniques to develop high-performance thermoplastic adhesives for multilayer composites. The synergistic enhancements in adhesive formulations, fiber interfaces, and predictive modeling contribute to the creation of composites with superior properties. These findings provide a solid foundation for future advancements in the design and optimization of thermoplastic composite materials for various industrial applications.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".