Rock-on-a-chip: A novel method for designing representative microfluidic platforms based on real rock structure and pore network modelling

1. Data for Figures 5, 6, 10, 14, and 15 (Excel file) This dataset contains the numerical values used to generate Figures 5, 6, 10, 14, and 15 of the associated manuscript. The data are organized in a Microsoft Excel (.xls) file with separate sheets corresponding to each figure. These figures represent key results from permeability measurements and flow simulations comparing experimental and modeled pore networks in quasi-2D microfluidic chips derived from CT-scanned rock samples.

2. MATLAB Script for 2D Mosaic Generation from 3D Pore Networks (MATLAB .m file) This MATLAB script implements the multi-step image processing workflow described in the manuscript for generating realistic 2D pore geometries from 3D CT-based pore network data. Starting from a binarized image of unconnected pores, the script includes the following key functions: connect_mosaic: Connects isolated pores with uniform-width throats and outputs an image of separated throats. distrib_throats: Edits throat widths based on a target throat size distribution and specified number of categories (Nmax), allowing control over heterogeneity in sparse mosaics. smooth_image: Applies smoothing to pore-grain boundaries to enhance fabrication compatibility. grain_fix: Removes small dead-end pores and fixes partially connected throats in large grains, with the minimum grain size (L_min) set by the user. All parameters can be adjusted by the user to match the pore and throat size distributions (PSD and TSD) from the 3D subvolume data. This script was used to produce the microfluidic patterns described in the study, ensuring morphological consistency between 3D and 2D representations.