Researchers from the National Key Laboratory of Green Hydrogen and Fuel Cell Technology at Xi'an Jiaotong University, led by Professor Baifeng Bai and Professor Chengzhen Sun, have conducted a study on the capillary flow characteristics of water within nano channels ranging from sub-nanometer to 30 nm in size using high-precision large-scale molecular simulations. The study found that reducing the size of the nano-constrained space leads to a capillary flow capacity of water lower than predicted by the classical Lucas-Washburn theory, with deviations significantly increasing as the scale decreases, in line with conventional experimental findings. The research reveals a scale-dependent nature of capillary flow, challenging the common belief that smaller channels result in greater resistance and slower flow. The recent findings have been published in Physical Review Letters.
Surprisingly, when the channel characteristic size decreases to 3 nm, water flow exhibits an unexpected reversal, showing anomalous flow enhancement characteristics, leading to a decrease rather than an increase in theoretical deviations. Theoretical analysis indicates that the structure of nano-constrained water depends on the confined scale, giving rise to two opposing scale-dependent effects: a long-range viscosity enhancement effect that increases flow resistance and a short-range separation pressure effect that enhances flow dynamics. The mismatch of scale effects results in a unique non-monotonic scale dependence of capillary flow in nano-scale water, particularly observed in both hydrophilic and hydrophobic nano channels. By introducing separation pressure, scaling corrections to viscosity, and modeling other nano-scale effects (such as dynamic contact angle and wall slip), a multiscale spatiotemporal unified capillary flow theoretical prediction model has been established from nanoscale to macroscale, nanoseconds to seconds, validated through extensive molecular dynamics simulations and literature experimental results.
For more information, please refer to the related paper: https://doi.org/10.1103/PhysRevLett.132.184001
