Three-dimensional model reconstruction of intact shale gas reservoir based on the composition, structure, and content of different components

Shale has strong compositional and structural heterogeneity， and its molecular structure model accuracy is a bottleneck which limited the molecular simulation results. To improve reproducibility of three-dimensional （3D） intact shale model and enhance the reliability of molecular simulation results. The chemical composition of the carbon skeleton， functional groups， and chemical bonds in organic matter （OM） were quantitatively characterized through a combination of methods including ¹³C nuclear magnetic resonance （NMR）， Fourier transform infrared spectroscopy （FTIR）， and X-ray photoelectron spectroscopy （XPS）. The two-dimensional （2D） OM structure was reconstructed according to above experimental results in software ACD/C+H NMR， until the iterative validation of ¹³C NMR profiles was successfully obtained. Then， the 3D OM structure was reconstructed in software Material Studio （MS） based on the reconstructed 2D structure after geometric optimization and anneal. It is a 3.23 nm×3.23 nm×3.23 nm （α = β = γ = 90°） cell with a molecular formula of C₁₉₀₀H₁₃₄₀O₂₄₀N₁₀₀ and relative molecular mass of 29380. Density and porosity of experimental and calculated results were compared to validate its precision， the results exhibit low relative errors of 2.0% and 1.6%， respectively. Finally， the 3D intact shale structure was reconstructed based on the total organic matter content （TOC） and mineralogical composition data， with the fractions of 936 quartz， 73 illite， 19 kaolinite， five chlorite， and two 2D OM structures， respectively. Its cell size is 6.21 nm×6.21 nm×6.21 nm （α = β = γ = 90°）. The 3D intact shale structure was also validated by experimental and calculated results on porosity and density， with relative errors of 2.8% and 1.3%， respectively. The density deviation is mainly caused by the density differences in quartz and other brittle minerals， whereas the porosity deviation is influenced by both mineralogical composition and different calculation models of the experiments and probe detection. The excellent compatibility of the real shale sample and the reconstructed structure confirms the reproducibility and reliability of this 3D intact shale model， and the effectiveness and feasibility of this construction method. These findings fill the knowledge gap of a systemic 3D construction method for intact shale and provide a highly reproducible 3D structure model for the Longmaxi shale gas reservoir.