The accumulation of immobile residual water during CO₂ injection for brine displacement significantly impairs storage efficiency, injectivity, and fluid migration—key factors for scaling up CO₂-based energy technologies. This study investigates the factors governing residual water saturation under different CO₂ phases and effective stress conditions in simulated subsurface environments. The results indicate that under constant effective stress, gaseous CO₂ yields the highest residual water saturation, followed by its supercritical and liquid states. As such, an inverse relationship is observed between residual water saturation and storage efficiency/capacity, underscoring the potential for jointly optimizing energy recovery and CO₂ sequestration. The analysis of the CO₂-brine-rock system confirms that capillary forces control residual water saturation. Increased interfacial tension or contact angle cosine value raises capillary entry pressure, hindering displacement and elevating irreducible water saturation. Moreover, higher effective confining pressure reduces capillary radius and creates "dead pores", thereby increasing capillary pressure and enhancing water trapping in the core. The findings give critical insights into how CO₂ phase behavior and confining pressure govern residual water saturation, displacement efficiency and migration in the reservoir, directly informing strategies for optimal CO₂ storage reservoir selection and enhanced oil recovery operations.

The development of geo-energy resources plays a crucial role in transitioning towards a sustainable energy future and achieving carbon neutrality. Conventional experimental approaches, constrained by macroscopic-scale observations and high costs, often fail to capture critical microscale mechanisms. In contrast, microfluidic technology offers distinct advantages through high-resolution visualization, high-throughput screening, and precise simulation of practical conditions such as temperature, pressure, pore structures, and chemical reactions, effectively addressing key challenges in geo-energy extraction. This review systematically examines innovative applications of microfluidics in shale gas reservoir, carbon capture, utilization and storage, chemical enhanced oil recovery, enhanced geothermal system, and natural gas hydrate. It further investigates prevailing challenges regarding material compatibility, scale translation, and data extrapolation methodologies. The study demonstrates that microfluidic systems provide innovative experimental methodologies, enabling unprecedented precision in elucidating complex geological processes through enhanced mass transfer efficiency and high-throughput screening capabilities, thereby bridging microscale mechanisms with macroscale phenomena. In the future advancements, the microfluidic technology demands synergistic convergence with materials science, chemical reactions, artificial intelligence, and physical explanation to promote the geo-energy research. This interdisciplinary convergence will provide scientific foundation for developing efficient and sustainable energy solutions.

Gas permeability, which is measured mainly through gas permeability experiments, is a critical technical index in many engineering fields. In this study, permeability is firstly calculated based on information from a digital image and an improved permeability prediction model. The calculated results are experimentally verified. Subsequently, a self-developed image-processing program is used to extract feature parameters from a scanning electron microscopy image. Meanwhile, an extreme learning machine algorithm is used to input the image feature parameters obtained using the image-processing program into the extreme learning machine algorithm for machine learning. Additionally, we compare several typically used machine learning algorithms, which confirmed the reliability and accuracy of our algorithm. The best activation function can be obtained by comparing the predicted permeability using an appropriate number of neuron nodes. Experimental results show that the program can accurately identify the features of the microscopy image. Combining the program with an extreme learning machine neural network algorithmgas permeability results to be obtained with high accuracy. This method yields good predictions of permeability in certain cases and has been adapted to other geomaterials.