Summary Knowing the composition, molecular size, and structure of polar compounds in crude oil that affect sandstone wettability is a prerequisite for a better understanding of oil/rock/brine interactions and for better application of enhanced oil recovery (EOR) techniques to increase recovery factors and improve the economic efficiency of field development. The nitrogen-, sulfur-, and oxygen-containing polar compounds in crude oil are key factors for sandstone wettability changes. In this study, an outcrop sandstone core selected from Jurassic formation in Sichuan Basin, China, was aged by crude oil to restore the wettability after being cleaned by hot Soxhlet extraction with an azeotropic solvent mixture of methanol and dichloromethane (MDC, vol/vol = 7:93). Then, Amott-Harvey experiments that were conducted by combining spontaneous imbibition and forced displacement steps of coreflooding were performed to characterize sandstone wettability after it was cut into four core blocks. The wettability index (IA-H) of four core blocks from the inlet to outlet of oil flooding were −0.523, 0.214, −0.087, and −0.861, respectively, which illustrated different degrees of sandstone wettability. The extracts of three sequential hot Soxhlet extraction steps of four core blocks were analyzed by gas chromatography-mass spectrometry (GC-MS) and high-resolution Fourier transform-ion cyclotron resonance-mass spectrometry (FT-ICR-MS) combined with electrospray ionization (ESI) in negative ion mode. Almost no polar compounds were detected in the n-hexane extracts, and a total of seven classes of different polar molecular compounds—namely, N1, N1O1, N1S1, O1, O2, O3, and O4—were detected in dichloromethane (DCM) and MDC extracts. The relative abundances of the N1S1 and O1 classes in the extracts of DCM and MDC were too low to be ignored. Compared to those of polar compounds in DCM extracts, the relative abundance of neutral nitrogen compounds (N1 and N1O1 classes) in the MDC extracts decreased significantly. In contrast, the acidic compounds (O2, O3, and O4 classes) all showed an obvious increase in the MDC extracts compared to the DCM extracts. It was notable that most of the polar compounds in MDC extracts were O2 and O3 compounds with double bond equivalent (DBE) values = 1 from the perspective of DBE distribution. The proportion of these two compound classes was much higher than that of all other polar compounds. Therefore, we believe that these two compound classes are the decisive factors for changing sandstone wettability combined with previous studies. In addition, based on the number of oxygen atoms and DBE values, we inferred that the O2 (DBE = 1) class was the long-chain saturated fatty acids and that the O3 (DBE = 1) class was the hydroxyl acids containing both one carboxyl and one hydroxyl group. Furthermore, the final determination of the wettability degree of the sandstone surface was the amount of all polar compounds, not only the relative abundance of these two types of acids. These two types of acids in crude oil were equivalent to anchor molecules on the surface of sandstone, and other polar compounds were adsorbed onto their surface to make the sandstone preferentially oil-wet.

Hydraulic fracturing is a key technology in unconventional reservoir production, yet many simulators only consider the single-phase flow of shale gas, ignoring the two-phase flow process caused by the retained fracturing fluid in the early stage of production. In this study, a three-dimensional fluid–gas–solid coupling reservoir model is proposed, and the governing equations which involve the early injection water phenomenon and stress-sensitive characteristics of shale gas reservoirs are established. The finite element–finite difference method was used for discretisation of stress and strain equations and the equations of flow balances. Further, a sensitivity analysis was conducted to analyse fracture deformation changes in the production. Fracture characteristics under different rock mechanics coefficients were simulated, and the influence of rock mechanics parameters on productivity was further characterised. The stimulated reservoir volume zone permeability could determine the retrofitting effect, the permeability increased from 0.02 to 0.1 mD, and cumulative gas production increased from 18.08 to 26.42 million m3, thus showing an increase of 8.34 million m3, or 46%. The effect of Young’s modulus on the yield was smaller than Poisson’s ratio and the width and length of the fractures. Production was most sensitive to the length of the fractures. The length of the fracture increased from 200 to 400 m, and the cumulative gas production increased from 26.44 to 38.34 million m3, showing an increase of 11.9 million m3, or 45%. This study deepens the understanding of the production process of shale gas reservoirs and has significance for the fluid–gas–solid coupling of shale gas reservoirs.

Fracturing is a key factor for shale oil and gas enrichment and high production. An accurate fracture model can effectively guide shale oil and gas exploration and development. The establishment of a natural fracture model must address the challenges of difficult data acquisition and poor representativeness of data points. To solve these problems, we propose a method of shale-reservoir natural fracture modeling based on microseismic monitoring data. This method include three steps, first, we establish an initial natural fracture model based on scale classification, vertical stratification, and genetic classification. Second, the shape and density of hydraulic fractures were interpreted by microseismic monitoring data to calibrate the initial model of shale reservoir natural fractures. Third, we verify the rationality of the model by assessment of the fracture porosity and permeability values. Results show that it is possible to calibrate the natural fracture density model using fracture shape and density as determined by microseismic monitoring. And we predict that an ideal hydraulic fracture network can be formed when the body density of natural fracturing is greater than 0.3 m2/m3. The crude production of wells are negatively correlated with the development of large-scale structural fractures and positively correlated with small-scale structural fractures. The well trajectory should run through small-scale fracture development sections as much as possible, avoiding large-scale high-angle fracture areas. This method provides a new approach to model natural fractures in shale reservoirs that has wide applicability and can be used for modeling shale oil and gas reservoirs.