Five-state framework reveals where movable shale oil accumulates in clay-laminated reservoirs

Oil can be present in shale, but not all of it is stored in the same way or has the same ability to move. Unlike conventional reservoirs, where oil may occupy relatively large connected pores, shale oil is held in micro- to nanoscale pore systems. At these scales, oil molecules interact strongly with kerogen, organic pore walls, mineral surfaces and pore centers.

These interactions mean that conventional indicators such as total organic carbon, porosity, pyrolysis parameters or total oil content cannot fully explain where shale oil is stored or how mobile it may be. Some oil enters the molecular network of kerogen, some attaches to pore walls, and some remains closer to the center of pores. Distinguishing these occurrence states is therefore central to understanding shale-oil behavior.

A research team led by Bo Liu at Northeast Petroleum University developed a quantitative framework for characterizing shale-oil occurrence states. The researchers used clay-laminated shale oil from the Cretaceous Qingshankou Formation in the Songliao Basin, northeast China, as a representative lacustrine shale-oil case. This shale system is marked by source-reservoir coupling, in-situ oil retention, fine lamination, clay-rich composition and nano- to microscale pore networks, making it suitable for testing a multiscale method.

The workflow integrates several levels of analysis. It begins with basic geochemical characterization and hydrocarbon-generation kinetics, then uses kerogen swelling experiments and whole-rock and kerogen chloroform extraction to separate organic- and inorganic-hosted oil. SEM/EDS imaging identifies organic and inorganic pores, while AFM analysis characterizes pore-surface roughness. Molecular dynamics simulations are then used to examine how oil molecules enter kerogen, attach to pore walls or occupy pore centers.

From a physical perspective, the study distinguishes three broad oil states: swelling-bound oil, adsorbed oil and free oil. From a storage-medium perspective, it separates kerogen, organic pores and inorganic pores. Combining these two perspectives yields five quantifiable occurrence types: kerogen-swelling oil, adsorbed oil in organic pores, free oil in organic pores, adsorbed oil in inorganic pores and free oil in inorganic pores.

Kerogen-swelling oil reflects the penetration of oil molecules into the kerogen macromolecular network. This is not simply oil sitting in a pore. It is a coupled process involving molecular diffusion, network flexibility and organic-matter expansion. Adsorbed oil in organic pores forms films or multilayer structures on organic pore walls and is controlled by kerogen type, pore-wall properties, specific surface area, oil-component polarity and nanoscale confinement. Free oil in organic pores occurs closer to pore centers and is more weakly bound to the walls, making it important for mobility evaluation.

Inorganic pores require separate treatment. Adsorbed oil in inorganic pores occurs on clay-mineral, siliceous or other mineral surfaces and is influenced by mineral polarity, wettability, surface roughness and adsorption capacity. Free oil in inorganic pores occurs near pore centers and is affected by pore-size distribution, pore geometry, connectivity and confinement effects. Treating organic and inorganic pores as a single pore system would obscure these differences.

The study also shows why realistic kerogen models are important. Graphene and other ideal carbon surfaces can approximate surface adsorption, but they cannot reproduce kerogen swelling. Natural kerogen contains a three-dimensional macromolecular network, flexible segments, functional groups and internal spaces into which oil molecules can enter. A kerogen-based model therefore provides a more realistic basis for separating swelling-bound, adsorbed and free oil.

The work further emphasizes that shale-oil liquid hydrocarbons cannot be treated in the same way as shale gas. Shale gas is dominated by small gas molecules such as methane. Shale oil, by contrast, is a multicomponent liquid hydrocarbon system containing saturates, aromatics, resins and asphaltenes. These components differ in polarity, molecular geometry and interfacial behavior. Heavier or more aromatic components tend to associate more strongly with pore walls, while lighter and less polar components are more likely to occupy pore centers. Even free oil is therefore affected by compositional fractionation and nanoscale confinement.

Occurrence states also evolve with thermal maturity. At low maturity, kerogen-swelling oil is more important. During the mature stage, swelling-bound oil decreases and free oil in organic and inorganic pores increases. At higher maturity, compaction and pore-structure adjustment can reduce the available space for free oil. This maturity-dependent evolution links molecular-scale oil storage to reservoir-scale interpretation.

The main value of the work is not a basin-specific numerical result, but an occurrence-state characterization framework with strong applicability across study areas. For other continental shale-oil systems, researchers can obtain basic geochemical, mineralogical, maturity and pore-structure data, then calibrate the framework with representative swelling-extraction experiments, SEM/EDS pore identification, AFM roughness measurements and molecular dynamics simulations. The same logic can then be used to identify the local combination of five oil occurrence states, compare organic and inorganic pore contributions, and interpret mobility and sweet-spot formation mechanisms.

By moving beyond total oil content, the framework provides a way to connect oil-rock interfacial processes, nanoscale pore structure and reservoir characterization. It offers a reusable approach for studying how shale oil is stored, trapped and mobilized in continental shale systems.

Liu B, Tian S, Dong Z, et al. 2026. Shale oil occurrence states and quantitative characterization: Theories and application to clay-laminated shale oil evaluation. Science China Earth Sciences, 69(5): 1839-1855. https://doi.org/10.1007/s11430-025-1829-7