Automatic dispersion, defect, curing, and thermal characteristics determination of polymer composites using...

Area of Science:

• Materials Science
• Polymer Science
• Non-Destructive Testing

Background:

• Infrared thermography is a non-destructive technique for polymer composite analysis.
• Manual processing of thermal images is prone to errors due to artifacts.
• Identifying composite components, defects, and curing states relies on thermal property variations.

Purpose of the Study:

• To develop and apply an automatic machine learning model for analyzing thermal images of polymer composites.
• To quantify filler dispersion and assess curing states in graphite/graphene-based polymer composites.
• To compare the effectiveness of hand, planetary, and batch mixing techniques on composite properties.

Main Methods:

• Analysis of thermal images from polymer composites using an automatic machine learning model.
• Quantification of filler dispersion with a resolution of approximately 20 µm.
• Comparison of thermal diffusivity and curing times for composites fabricated via different mixing methods.

Main Results:

• The machine learning model accurately identified filler characteristics and quantified dispersion despite image artifacts.
• Batch mixing demonstrated superior filler dispersion (DI=0.07) compared to planetary (0.0865) and hand mixing (0.163).
• Curing times varied, with PDMS taking 500 s and PDMS-Graphene/PDMS Graphite Powder taking 800 s.

Conclusions:

• The developed machine learning approach provides a quantitative and qualitative analysis of polymer composites.
• This automated method enhances the reliability of composite characterization, overcoming limitations of manual processing.
• The study highlights the impact of mixing techniques on filler dispersion and the utility of thermal analysis for quality assessment.