Interfacial drag force and inter-subchannel mixing in vertical two-phase flows for subchannel analysis

Subchannel analysis of two-phase flow in the reactor core is crucial for the design and optimization of advanced nuclear reactors. In subchannel analysis, three-dimensional flow is simplified into one-dimensional axial flow within subchannels, and lateral mixing is assumed between adjacent subchannels. The accuracy of subchannel analysis strongly relies on the accurate modeling of the interfacial drag force in axial flow and inter-subchannel mixing processes. This paper conducted a comprehensive summary of existing interfacial drag force models and interchannel mixing models. The interfacial drag force was closely correlated with interphase relative velocity, which was significantly influenced by subchannel geometry. In rod bundles, structural differences among interior, edge, and corner subchannels led to variations in void fraction and volumetric flux distributions, resulting in distinct distribution parameters and void fraction covariances. A comparative assessment of correlations for both parameters across different subchannel types was conducted, revealing that subchannel geometry exerted a significant influence on distribution parameters at low void fraction conditions and on void fraction covariance at high void fraction conditions. The inter-subchannel mixing of mass and momentum arose from turbulent mixing and void drift. This study also provided a comprehensive summary of turbulent mixing and void drift models. Turbulent mixing models were primarily categorized into separated models and two-phase turbulent velocity-based models; void drift models were classified into equilibrium void distribution-based models and relative velocity-based models. The equal-volume-exchange model of turbulent mixing and void drift, which characterizes turbulent mixing using two-phase turbulent mixing velocity and void drift using an equilibrium void distribution, is currently the most widely adopted framework. However, significant discrepancies exist in the predictions of the correlations of key parameters, such as the turbulent mixing coefficient, two-phase turbulent mixing multiplier, and void drift coefficient. Therefore, several suggestions for model improvement were provided. © 2026 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.