Abstract Accurate estimation of crude oil viscosity is crucial for formulating optimal hydrocarbon recovery strategies. Nuclear magnetic resonance (NMR) is a non-destructive technique for viscosity characterization with a longitudinal and transverse relaxation times (T₁ and T₂), diffusion coefficient, or apparent hydrogen index derived from ¹H signals. In this review, the theoretical methods for crude oil viscosity characterization with NMR systematically are sorted out, which is crucial for novices and petroleum engineers. Firstly, the basic principles of NMR are introduced, and the theoretical basis of NMR-based crude oil viscosity characterization is described in terms of relaxation times, diffusion coefficient, and apparent hydrogen index. Then, a systematic review is conducted for NMR-based crude oil viscosity characterization method developed over the past three decades. Finally, four kinds of influence factors, instrument parameters, temperature, crude oil type, and gas oil ratio, are analyzed. The results emphasize the importance of selecting a suitable viscosity model based on reservoir type and formation conditions. Two-dimensional NMR T1-T2 techniques demonstrate potential for addressing crude oil viscosity characterization challenges. In general, the key to crude oil viscosity characterization with NMR is selecting the appropriate NMR method rationally based on clear reservoir characteristics. This study aims to assess the existing methods and their limitations critically while providing guidance toward more robust viscosity quantification protocols.
Guo et al. (Sat,) studied this question.