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

Microfluidics is a key tool for studying pore-scale phenomena in porous media, with applications in oil recovery and carbon storage. However, accurately replicating rock pore structures in quasi-2D microfluidic platforms remains a challenge. Existing design strategies, including regular and irregular networks, fractal geometries, thin-section imaging, and multi-step methods using CT scans and SEM images, often fail to capture real pore space morphologies. To address these issues, we developed a multi-step workflow that preserves pore morphology and size distributions in quasi-2D microchips (rock-on-a-chip) by generating 2D pore throats from 3D network data of CT-scanned rock samples. The method showed strong agreement between 2D and 3D pore and throat size distributions in both designed patterns and fabricated microchips. A critical factor in achieving accurate pore geometry was precise mask alignment, which enabled the fabrication of microchips with narrower throats for relatively tight reservoir patterns. Permeability regulation was achieved by adjusting inlet areas while maintaining pore and throat size distributions similar to the original 3D subvolume. Flow simulations using the Hagen-Poiseuille equation within the OpenPNM framework showed differences between simulated and experimental permeability, especially in low-permeability designs, which were more sensitive to the etching process. Despite these challenges, the proposed approach minimizes common discrepancies between rock pore space morphologies and quasi-2D microchips, significantly improving the reliability of microfluidic studies for applications requiring accurate pore-scale structures.