Screening tuned water with a tool to quantify changes in wettability and the impact on enhanced oil recovery in core flooding experiments

This work introduces TScreen, a prototype computational tool designed to calculate wettability indicators in water-oil-rock systems and simulate 1D core flooding experiments, with a focus on tuned water injection strategies for Enhanced Oil Recovery (EOR). By integrating relative permeability and capillary pressure curves that adapt dynamically to wettability changes, TScreen offers a comprehensive framework for analyzing the interplay between brine ion composition and flow dynamics. TScreen allows variation of brine ion composition to evaluate how wettability indicators, such as the Bond Product Sum (BPS) and contact angle, are affected. The solver leverages Surface Complexation Modeling (SCM), a widely used method for analyzing ion adsorption and chemical reactions on water-oil and water-rock surfaces (Brady et al., 2012; Erzuah et al., 2018), in conjunction with DLVO theory (Derjaguin; Landau, 1993; Verwey; Overbeek, 1955). By integrating DLVO-based equations with PHREEQC, the tool performs surface complexation calculations. The calculated BPS values are validated against experimental data, and an enhanced, dimensionless definition of BPS is proposed for improved comparability across scenarios. A standout feature of TScreen is its integration of a 1D two-phase flow simulator, which dynamically couples ion mass transport with contact angle calculations across grid cells. This innovative approach ensures that local brine compositions directly influence relative permeability and capillary pressure curves, making flow dynamics dependent on wettability alteration during tuned water injection. The simulator was validated with a multi-stage core flooding experiment, where low-salinity seawater was injected into a core initially saturated with high-salinity formation water. This confirmed TScreen’s ability to simulate wettability alteration during core flooding experiments to find optimal brines for EOR. TScreen also leverages a pretrained neural network to efficiently generate relative permeability curves for low-salinity brines, using formation water curves as input (Czarnobay et al., 2024). This allows for an agile exploration of different brine compositions, streamlining the workflow for EOR studies. By combining state-of-the-art computational techniques, experimental validation, and machine learning, TScreen represents a transformative advancement in the development and optimization of tuned water injection strategies, unlocking new potential for maximizing hydrocarbon recovery in complex reservoir environments.

Audience Take away:

• How to assess the influence of tuned water injection on Enhanced Oil Recovery (EOR) performance through advanced modeling.
• How TScreen can reduce the need for costly core flooding experiments by simulating the optimal brine compositions for EOR.
• How the ionic composition of brines affects wettability in oil-brine-rock systems, as demonstrated by TScreen using DLVO theory.
• How TScreen integrates dynamic contact angle calculations with core flooding simulations to enhance EOR studies.
• How artificial intelligence aids in generating accurate relative permeability curves to support core flooding simulations.