Original Development of a Multi-Scale Pore Network Model for Wettability Heterogeneity Characterisation and Displacement Efficiency Prediction in Nigerian Heavy Crude Carbonate Reservoirs

Wettability heterogeneity at the pore scale fundamentally controls waterflood and surfactant flood recovery efficiency in carbonate reservoirs containing heavy Nigerian crudes, and its quantitative characterisation requires a multi-scale pore network modelling approach that integrates molecular-scale surface chemistry with pore-throat-scale fluid topology and reservoir-scale displacement patterns. This study develops an original multi-scale pore network model that bridges these scales for Nigerian heavy crude carbonate reservoir systems. At the molecular scale, molecular dynamics simulations are performed to compute contact angles as a function of crude oil asphaltene surface coverage, brine salinity, and mineral surface composition (calcite versus dolomite) under reservoir temperature conditions. These molecular-scale contact angle distributions are used as input to a reconstructed pore network model derived from micro-CT imaging of Nkalagu Formation carbonate core samples from Ebonyi State. Quasi-static pore-scale invasion percolation simulations generate capillary pressure and relative permeability curves for the heterogeneous wettability pore network. An upscaling procedure using representative elementary volume analysis links pore network outputs to Darcy-scale inputs for a Darcy flow reservoir model. The multi-scale framework predicts a 14.2 percent reduction in waterflooding efficiency compared to uniformly water-wet reference simulations, which is validated against core flooding experimental results on preserved wettability core samples. Sensitivity analysis identifies the asphaltene-to-resin ratio of the crude oil as the dominant molecular-scale parameter controlling reservoir-scale wettability effects. Keywords: multi-scale pore network model, wettability heterogeneity, carbonate reservoir, Nigerian crude, displacement efficiency.