Five model compounds with representative chemical structures were selected for use in simulation experiments of pyrolytic gas production. The gas production and isotopic fractionation characteristics were observed and analyzed. Then, the factors affecting carbon isotope fractionation during natural gas generation were discussed, and a fractionation model was established and calibrated. We concluded that the final hydrocarbon gas (C_1–5) yield of octadecane, octadecylamine, octadecanoic acid, decahydronaphthalene, and 9‐phenylanthracene decreased in turn with the effective hydrogen content. Compared with linear alkanes or alkyl compounds, cycloalkanes have higher thermal stability and generate gas later. The variation in the carbon isotopic composition of natural gas is primarily controlled by the following three factors: (a) the thermal evolution of organic matter results in a gradually heavier isotopic composition for the main gas production stage. (b) Gas inherits the isotopic composition of its parent material, and this effect is evident when the chemical structure and gas generation mechanism between parent materials are similar. (c) The structure of organic matter determines the reaction mechanism of gas generation, which has a significant influence on the range and trend of carbon isotope fractionation in the process of methane generation. An improved chemical kinetic model can accurately characterize carbon isotope fractionation during gas generation.

To better understand the pore structure and fractal character of lacustrine shales and their influence on liquid hydrocarbon occurrences, in this study, a total of 29 lacustrine oil-bearing shale samples collected from the Shahejie Formation in the Dongying Sag, Bohai Bay Basin, were investigated based on nitrogen adsorption (NGA) analysis combined with TOC, Rock-Eval pyrolysis, X-ray diffraction (XRD), and field emission-scanning electron microscopy (FE-SEM) experiments. The relationships among the compositions (TOC, minerals, and oil content), pore structures, and fractal dimensions of the lacustrine shale samples were discussed. The results showed that the adsorption and fractal characteristics of lacustrine oil-bearing shales differ at relative pressures of 0-0.1 and 0.5-1. Two corresponding fractal dimensions and were determined by the FHH model according to the nitrogen adsorption branches. Specifically, varies from 2.4292 to 2.6109 (mean 2.5245), and varies between 2.4680 and 2.8535 (mean 2.6889). The specific surface area (SSA) ranges from 1.512 m²/g to 34.002 m²/g, with an average of 13.656 m²/g, the total pore volume is between 6.0 × 10^-3 cm³/g and 48.4 × 10^-3 cm³/g (mean 24.5 × 10^-3 cm³/g), and the average pore diameter is in the range of 4.22 nm to 19.57 nm (mean 9.35 nm). Both and increase with increasing SSA and increase with decreasing average pore diameters but have no correlation with pore volume. Moreover, and exhibit positive relationships with clay minerals and negative correlations with carbonate minerals (calcite and dolomite). The relationship between fractal dimensions ( and ) and TOC contents is expressed as a U-shaped curve, characterized by the minimum values at approximately 3% TOC. The shale oil content is controlled by the pore structures and fractal dimensions, and lacustrine shales with lower SSAs and smaller fractal dimensions would have more free oil. Therefore, lacustrine shales in the oil window with TOC contents ranging from 2% to 4% are probably the preferred shale oil exploration target in the Shahejie Formation, Dongying Sag, Bohai Bay Basin. The results indicate that fractal analysis can provide insight into the pore structure characteristics and oil storage capacity of lacustrine shales.

To lay the foundation for the exploration of shale gas, we calculated the total organic carbon (TOC) of Shahezi Formation source rocks in the Xujiaweizi Fault Depression in the Songliao Basin by using the measuring data and the logging curves. Because the source rocks in the study area were formed in a lacustrine basin, they are characterized by strong organic heterogeneity, making it difficult to objectively characterize any changes in the TOC of the underground source rocks based on discrete and limited sampling. In addition, because the source rock is relatively rich in shale and poor in organic matter, the logging response features of high natural gamma, medium-high interval transit time, and medium-low resistivity, making it easily identified. However, because the logging parameters to predict the TOC of source rocks are not universal, it is impossible to establish a prediction equation that is universally applicable. To solve the above problems, we used the variable-coefficient ΔLgR technique to predict the TOC of the source rocks in the study area. We defined the two key parameters that affect the TOC prediction in the classic ΔLgR technique as the undetermined coefficients, and the coefficients were determined according to the logging and geologic data of the individual wells. The application results indicate that the variable-coefficient ΔLgR technique has an average relative error of 17.5% in the prediction of the TOC, which is 16.1% lower than that of the classic ΔLgR technique. Thus, the prediction results can effectively reflect the vertical variation in the TOC of source rocks. Based on the logging evaluation results for the TOC of source rocks in 35 wells throughout the study area, the thickness of the high-quality source rocks in the fourth member of the Shahezi Formation was mapped. The thickness of the high-quality source rocks in the fourth member of the Shahezi Formation is generally >30 m. There are two centers of greater thickness in the plane, with the maximum thickness of more than 70 m.

Velocity analysis using the semblance spectrum can provide an effective velocity model for advanced seismic imaging technology, in which the picking accuracy of velocity analysis is significantly affected by the resolution of the semblance spectrum. However, the peak broadening of the conventional semblance spectrum leads to picking uncertainty, and it cannot deal with the amplitude-variation-with-offset (AVO) phenomenon. The well-known AB semblance can process the AVO anomalies, but it has a lower resolution compared with conventional semblance. To improve the resolution of the AB semblance spectrum, we propose a new weighted AB semblance based on principal component analysis (PCA). The principal components or eigenvalues of seismic events are highly sensitive to the components with spatial coherence. Thus, we utilized the principal components of the normal moveout (NMO)-corrected seismic events with different scanning velocities to construct a weighting function. The new function not only has a high resolution for velocity scanning, but it is also a friendly method for the AVO phenomenon. Numerical experiments with the synthetic and field seismic data sets proved that the new method significantly improves resolution and can provide more accurate picked velocities compared with conventional methods.