Hydrogel has a complex network structure with inhomogeneous and random distribution of polymer chains. Much effort has been paid to fully understand the relationship between mesoscopic network structure and macroscopic mechanical properties of hydrogels. In this paper, we develop a deep learning approach to predict the mechanical properties of hydrogels from polymer network structures. First, network structural models of hydrogels are constructed from mesoscopic scale using self-avoiding walk method. The constructed model is similar to the real hydrogel network. Then, two deep learning models are proposed to capture the nonlinear mapping from mesoscopic hydrogel network structural model to its macroscale mechanical property. A deep neural network and a 3D convolutional neural network containing the physical information of the network structural model are implemented to predict the nominal stress–stretch curves of hydrogels under uniaxial tension. Our results show that the end-to-end deep learning framework can effectively predict the nominal stress–stretch curves of hydrogel within a wide range of mesoscopic network structures, which demonstrates that the deep learning models are able to capture the internal relationship between complex network structures and mechanical properties. We hope this approach can provide guidance to structural design and material property design of different soft materials.

Existing theories cannot predict the mechanical property changes of polyacrylamide hydrogels with different water content because of the absence of side chains. In this study, polyacrylamide hydrogels are prepared and tested to investigate the side chain effect on their mechanical properties. First, the comparison between the effective chain density and total chain density provides proof of the large amount of side chains in the polymer network of PAAm hydrogel. We propose a practical chain density fraction to measure the side chain fraction. Then, the abnormal Young’s moduli-polymer volume fraction relationship reveals that side chains affect the mechanical properties of hydrogel through the insufficient lubrication of water. Water confined in narrow space within a molecular-level size can bear shear force to provide extra deformation resistance. A constitutive mode considering the effect of the insufficient lubrication of water is proposed. Combining this constitutive model with experimental results, we find that this insufficient lubrication of water exists even in equilibrium PAAm hydrogel. Molecular dynamics simulations reveal that this insufficient lubrication of water comes from the constraint of polymer chains. It also demonstrates that when there is insufficient lubrication, the rearrangement of water molecules leads to the persistent energy dissipation in the Mullins effect of PAAm hydrogel.

Co-continuous interlocking lattice structures usually present superior compressive properties and energy absorption characteristics. In this study, co-continuous interlocking polydimethylsiloxane/polylactic acid (PDMS/PLA) lattice composites were designed with different strut diameters, and successfully manufactured by combining the fused deposition modeling (FDM) technique and the infiltration method. This fabrication method can realize the change and control of structure parameters. The effects of the strut diameter on the compressive properties and energy absorption behavior of PDMS/PLA lattice composites were investigated by using quasi-static compression tests. The compressive properties of the co-continuous interlocking PDMS/PLA lattice composites can be adjusted in a narrow density range by a linear correlation. The energy absorption density of the co-continuous interlocking PDMS/PLA lattice composites increases with the increase in the PLA strut diameter and presents a higher efficiency peak and wider plateau region. The PLA lattice acts as a skeleton and plays an important role in bearing the compressive load and in energy absorption. The indexes of the compressive properties/energy absorption characteristics and PLA volume fraction of co-continuous interlocking PDMS/PLA lattice composites show linear relationships in logarithmic coordinates. The effect of the PLA volume fraction increasing on the plateau stress is more sensitive than the compressive strength and energy absorption density.

Traditional single-network hydrogels are limited by trade-offs among key mechanical properties, such as elastic modulus and toughness, and are further compromised by structural defects introduced during synthesis. These constraints significantly hinder their performance in demanding applications. Here, we propose a network-refinement strategy using repeated crosslinking to achieve comprehensive mechanical reinforcement. Through network refinement, structural defects within polymer networks are progressively filled, and both the network homogeneity and effective chain density are improved. The reinforced hydrogels exhibit up to a 6-fold increase in elastic modulus, a 10-fold enhancement in fracture toughness, a 20-fold increase in tensile strength, and a 42-fold improvement in work of fracture, while maintaining high stretchability and negligible hysteresis under moderate deformation. This universal strategy provides an effective route to comprehensively enhance the mechanical properties of hydrogel-like materials, paving the way for robust soft materials in applications such as cardiac healing patches, load-bearing biomedical implants, and wearable electronics.