Large deformation behavior of powerlaw fluid-elastomer composites with application to self healing materials: 3D simulations, analytical framework, and ML model

The life of polymeric products can be improved by embedding the self-healing capability into the polymer. In extrinsic self-healing materials, the channels inside the matrix store the healing fluid which is transported to the damaged region by the capillary action for healing the damage and restoring the material integrity. This capillary action is inadequate to cover a large damage region and achieve multiple damage-healing cycles. To overcome this shortcoming, an extrinsic type of composite is proposed which has cylindrical, uni-disperse, parallel, channels positioned in periodic arrangement inside the matrix. These channels store the healing fluid, which is pressurized using an initial compression of the elastomeric matrix. In the event of an external cut, this fluid is delivered under pressure for healing the damaged region. Hyperelastic elastomer is considered for the matrix phase that is subjected to a large initial compression. Healing fluid is taken as a non-Newtonian, power law fluid. The characterization of the flow of healing fluid in terms of elastomer, fluid properties, initial compression, channel geometry, fluid fraction, etc. is derived in a semi-analytical manner. Numerical simulations are performed to validate the results. Lastly, a machine learning model is trained using results obtained from the semi-analytical model to rapidly estimate fluid flow from relevant inputs. The machine learning (ML) model can be used to design the proposed type of self-healing elastomer composites on the basis of the properties of elastomer and healing fluid, geometry of the composite.