Yield-Stress Composite Fluids With High Thermal Conductivity and Tunable Rheology via Hierarchical Filler...

Yield-stress fluid can reversibly transform between the rigidity of a solid and the fluidity of a liquid, providing mechanical adaptability for its reprocessing and long-term stability. For thermal management applications, however, this adaptability must be coupled with high thermal conductivity-an objective hindered by the conventional trade-off between viscosity, and yield stress. Here, we propose a hierarchical filler architecture strategy by incorporating aluminum nitride particles of multiple sizes into a polydimethylsiloxane (PDMS) matrix to construct a multi-level dynamic network. Small particles fill the interstices between larger ones, creating dense, percolated thermal pathways while enabling reversible network breakdown and reconstruction under shear. This design achieves an unprecedented combination of low viscosity (731.3 Pa s at 10 s^-1), tunable yield stress (from 37.3 to 802.3 Pa), and high thermal conductivity (12.0 W m^-1 K^-1) in a yield-stress composite fluid. Comprehensive rheological characterization reveals excellent thixotropic recovery and mechanical robustness, while physics-informed neural network (PINN) modeling. When yielding stress fluid is used as a thermal interface material, the fluidity of liquid-like form can achieve efficient distribution, and the rigidity of solid-like form can achieve long-term reliability. These findings provide a new approach for designing advanced yield-stress fluids with controllable rheological properties and excellent thermal performance for high-performance thermal management applications.