Development of coupled THM models for reservoir stimulation and geo-energy production with supercritical CO2 as working fluid

Abstract

Due to enhanced public awareness, the eco-friendly techniques in geo-energy exploitation, e.g. supercritical CO2 fracturing, have received extensive attention in the last two decades. In this dissertation, two specific numerical models have been developed to address the issues associated with utilization of supercritical CO2, like fracture creation, proppant placement and fracture closure in unconventional gas reservoirs, reservoir stimulation, heat production and CO2 sequestration in deep geothermal reservoirs, respectively. (a) In unconventional gas reservoir, the model consisting of classic fracture model, proppant transport model as well as temperature-sensitive fracturing fluids (CO2, thickened CO2 and guar gum) has been integrated into the popular THM coupled framework (TOUGH2MP-FLAC3D), which has the ability to simulate single fracture propagation driven by different fracturing fluids in non-isothermal condition. (b) To characterize the fracture network propagation and internal multi fluids behavior in deep geothermal reservoirs, an anisotropic permeability model on the foundation of the continuum anisotropic damage model has been developed and integrated into the popular THM coupled framework (TOUGH2MP-FLAC3D) as well. This model has the potential to simulate the reservoir stimulation and heat extraction based on a CO2-EGS concept. Attempting to improve the fracturing ability and proppant-carrying capacity of CO2, the unconventional gas reservoir model is applied in a fictitious model with the parameters of a typical tight gas reservoir. According to results, pure CO2 is inefficient to create a fracture in such tight gas reservoir. Its fracturing ability can be improved significantly by using CO2 thickener. In comparison with traditional guar gum, the leakage of thickened CO2 is higher. But considering the expansion of CO2 in the fracture, accounting for about 37% contribution in this case, the performance of thickened CO2 is comparable with traditional guar gum. Additionally, a linear correlation between the break down pressure and fracturing fluid viscosity has been observed. In non-isothermal conditions, the temporal and spatial leak-off rate and expansion effect distribute unevenly in fracture. Due to high temperature around fracture tip, High leak-off rate and expansion effect is found nearby the tip zone. During fracture initiation, the expansion effect plays a critical role. With continuous injection, the dominating factor in fracture-making is gradually switched to leak-off. Overall, fracturing is more sensitive to leak-off than fluid expansion. For proppant-carrying, thickened CO2 with light proppant can achieve a better proppant placement than heavy proppant, even better than the one transported by guar gum. Due to low specific heat capacity, thickened CO2 has a high gel breaking rate, resulting in better fracture support. In the application of the damage-permeability model at planned Dikili geothermal project, a fracture network with final SRV of 8.7×107 m³ and SRA of 2.2×106 m2 is created after injecting 90,000 kg CO2 in 250 hours. The maximum permeability of 1.3×10-13 m2 has been achieved in x- and z-directions. During heat production, a priority channel with high gas saturation is formed because of the viscosity difference between CO2 and water. The heat is mainly mined in this priority channel. Generally, for both water and CO2 injection, the driven pressure and average thermal capacity shows a positive correlation with injection rate, while inverse relation exists between eventual temperature and injection rate. Under the same mass injection rate, the driven pressure and average thermal capacity of water injection is higher than CO2 injection, whereas the ultimate produced temperature of water injection is lower. Therefore, CO2 as working fluid for heat extraction has the benefits of low driving pressure and is beneficial for realizing relatively stable heat mining. In addition, the injected CO2 is detained in geothermal reservoir due to leak off, which can be regarded as geologically sequestrated CO2. Furthermore, the sequestrated CO2 has a positive relationship with injection rate. After 30 years of geothermal production, up to 950,000 tons CO2 is sequestrated at an injection rate of 100kg/s, demonstrating the potential contribution of CO2-ESG on the sequestration of CO2.