Nuclear Magnetic Resonance Applications in Shale Reservoir Characterization

Nuclear magnetic resonance (NMR) has become an indispensable technique in the characterisation of shale reservoirs, offering non-destructive insights into pore structure, fluid distribution and producibility. In low-field NMR logging and laboratory measurements, the spin–lattice (T1) and spin–spin (T2) relaxation times are analysed to infer pore size distributions, bound and free fluid fractions and wettability. Two-dimensional correlation experiments (for example T1–T2 maps or diffusion–relaxation spectroscopy) further enhance fluid typing and connectivity analysis, discriminating hydrocarbons from water and kerogen within nanometre-scale pores. Advances in pulse sequence design and inversion algorithms have improved resolution of complex pore networks in organic-rich shales. Integration with machine learning frameworks now permits rapid prediction of saturation and viscosity from T2 distributions and bulk density logs. Applications extend from core-scale laboratory studies to field-scale NMR well logging, enabling sweet-spot identification, evaluation of enhanced oil recovery schemes and dynamic monitoring of imbibition or desorption processes. This global toolkit is reshaping exploration, appraisal and production strategies in unconventional hydrocarbon systems.

Research from Nature Portfolio

Recent studies have harnessed low-field NMR and machine learning to refine in situ water saturation estimates in heavy oil and oil-sand analogue samples. A dual-model framework based on extreme gradient boosting utilises empirical feature engineering from T2-relaxation distributions alongside whole-spectrum mutual information extraction. Trained on ambient and reservoir temperature measurements of 82 core samples with true water content determined by standard extraction, the approach achieved a root-mean-square error below 1 % and R2 of 0.90. This methodology demonstrates the feasibility of combining LF-NMR spin–spin relaxation data and density logs for rapid, accurate evaluation of fluid saturations in complex reservoirs.

Nuclear Magnetic Resonance Applications in Shale Reservoir Characterization publication trend

The graph below shows the total number of articles in nuclear magnetic resonance applications in shale reservoir characterization across all publications each year (not limited to Nature Index journals).

Technical terms

T1 relaxation: Time constant for spin–lattice relaxation, reflecting energy exchange between nuclear spins and their environment, sensitive to pore surface interactions.

T2 relaxation: Time constant for spin–spin relaxation, indicating dephasing among spins, often correlated with pore size and fluid viscosity.

T1–T2 map: Two-dimensional correlation spectrum combining T1 and T2 relaxations to distinguish fluid types and pore environments.

Low-field NMR: NMR measurements conducted at magnetic fields below 1 MHz, commonly used for well logging and core analysis.

Bound fluid: Fluid fraction immobilised by capillary and surface forces within small pores or clay matrices.

Movable fluid: Fluid fraction free to flow under pressure gradients, typically residing in larger pores.

Diffusion–relaxation correlation: NMR pulse sequence linking molecular diffusion coefficients with relaxation times to enhance pore and fluid discrimination.

References

1. Application of XGBoost model for in-situ water saturation determination in Canadian oil-sands by LF-NMR and density data. Scientific Reports (2022).
2. NMR-Based Analysis of Fluid Occurrence Space and Imbibition Oil Recovery in Gulong Shale. Processes (2023).
3. Nuclear magnetic resonance response characteristics and quantitative evaluation method of fluid saturation of lacustrine shale oil. Frontiers in Earth Science (2023).
4. Identifying Different Components of Oil and Gas Shale from Low-Field NMR Two-Dimensional Spectra Based on Deep Learning. Magnetochemistry (2024).