Failure Prediction of Notched Composites Using Multiscale Approach

Area of Science:

• Materials Science
• Mechanical Engineering
• Composite Materials

Background:

• Polymer composites are susceptible to failure initiated by defects.
• Predicting composite failure, especially with complex defect geometries, remains a challenge.
• Existing models often lack the multiscale approach needed for accurate micro-level analysis.

Purpose of the Study:

• To develop and present multiscale-based failure criteria for polymer composites.
• To enable accurate prediction of composite failure irrespective of defect shape.
• To identify failure locations and predict propagation directions.

Main Methods:

• Utilized a multiscale technique involving upscaling and downscaling processes to link microscale (fiber/matrix) and macroscale (homogenized composite) levels.
• Applied recently proposed unified failure criteria to micro-stresses and micro-strains.
• Incorporated two conditions for failure: stress/strain and stress/strain gradient.

Main Results:

• The developed failure criteria can predict the failure of polymer composites with any defect shape.
• Failure prediction is based on micro-stresses and micro-strains at the fiber and matrix level.
• The criteria successfully determine both the location and direction of failure propagation.

Conclusions:

• The novel multiscale failure criteria offer a robust method for predicting composite failure.
• The dual-condition approach (stress/strain and gradient) enhances accuracy for diverse defect shapes.
• This work advances the understanding and prediction of failure mechanisms in polymer composites.