Integrating THMCD coupling for CH4 production and CO2 storage in CO2-ECBM: A multiphysics modeling framework

This study develops a Thermo-Hydro-Mechanical-Chemical-Damage (THMCD) framework to investigate CH 4 production and CO 2 storage during CO 2 -ECBM recovery. The framework integrates bidirectional couplings across five physical fields (thermal, hydraulic, mechanical, chemical, and damage) based on energy and mass conservation principles. Model validation utilized gas injection replacement and cryogenic liquid nitrogen freezing tests. Numerical simulations show CO 2 injection exerts dual effects: CO 2 -CH 4 competitive adsorption promotes CH₄ desorption, while reservoir damage enhances multiphase fluid seepage and heat transfer. Compared to the THMC model, the THMCD framework boosts cumulative CH 4 production by 8.54% and CO 2 storage by 24.6%. Parametric analyses reveal that higher injection pressure and reservoir permeability enhance gas flow rates; high injection temperature strengthens well-reservoir temperature gradient and induces coal reservoir fracturing damage (widening CO 2 seepage and CH 4 desorption-migration channels); high initial water saturation delays early peak rates but accelerates them later via damage-induced permeability. Under the adopted proxy economic criterion (CO 2 /CH 4 rate ratio threshold) and the studied conditions, an indicative injection delay of ∼2500 days is recommended to balance sustained CH 4 production and reasonable CO₂ storage. This work advances CO₂-ECBM design by emphasizing damage-coupled multiphysics interactions, providing actionable strategies for synergistic energy extraction and carbon management. • A fully coupled THMCD multiphysics framework for CO₂-ECBM is developed, incorporating competitive adsorption and injection-induced damage feedback. • The model is validated with laboratory data, showing the coupled effects of adsorption and damage on multiphase seepage and heat transfer. • Key reservoir and operational parameters are analyzed, and injection-timing recommendations are given based on a proxy criterion.