Sustainable Enhanced Oil Recovery(EOR): From Laboratory Coreflooding to AI-Driven Field Deployment

Enhanced oil recovery (EOR) plays a pivotal role in maximizing hydrocarbon recovery from mature reservoirs and offers opportunities to improve energy efficiency and reduce environmental impacts in the upstream sector. However, translating laboratory-scale insights into reliable field-scale performance remains challenging due to reservoir heterogeneity, chemical retention, and modeling uncertainties. This review synthesizes recent advances in thermal, chemical, gas-based, and hybrid EOR strategies, with a focus on coreflooding experiments that underpin energy-efficient recovery mechanisms. Particular attention is given to the use of high-pressure and high-temperature (HPHT) systems, microfluidic platforms, and advanced imaging techniques aimed at improving the representativity of laboratory studies. Key upscaling challenges, including wettability alteration, reactive transport, and timescale mismatches, are critically evaluated from an energy-efficiency perspective. The integration of digital rock physics (DRP), machine learning (ML), and AI-based modeling is discussed as an emerging approach to enhance process understanding, optimize operational parameters, and support more energy-efficient EOR deployment within the broader energy transition.