Pore-fluid characterizations and microscopic mechanisms of sedimentary rocks with three-dimensional NMR: Tight sandstone as an example

https://doi.org/10.1016/j.jngse.2020.103392Get rights and content

Highlights

  • The tri-window pulse sequence measurements are first simulated using the RW method.

  • Pore-fluid in tight sandstone is characterized through 3D NMR.

  • Wetting fluid relaxation information varying from saturation are analyzed and microscopic mechanisms are explored.

  • T1-T2 spectra is more useful for identifying gas reservoir than oil reservoir due to gas high T1/T2.

Abstract

Pore-fluid is directly related to the reservoir quality. Three-dimensional (3D) nuclear magnetic resonance (NMR) can be used for charactering pore-fluid due to its sensitivity to fluid components saturated in sedimentary rocks. In this paper, taken a tight sandstone as an example, investigating pore-fluid characterizations and microscopic mechanisms through 3D NMR. Firstly, a commonly used 3D NMR pulse sequence, tri-window pulse sequence, in actual oilfield was presented and its data processing workflow was exhibited briefly. Then, random-walk method was modified to simulate the pulse sequence measurements at different pore-fluid cases in tight sandstone, such as pure water case, oil-water case, and gas-water case. Finally, pore-fluid 3D NMR responses were exhibited and their microscopic mechanisms were explored. The results show that 3D NMR can identify different fluid components saturated in sedimentary rocks; wetting fluid saturation directly affect wetting fluid relaxation time, but the non-wetting fluid relaxation information is independent of fluid saturation; wetting fluid has an obvious restricted diffusion as a lower saturation; an obvious difference between gas signal and water or oil signal in two-dimensional relaxation spectra due to high longitudinal to transverse relaxation time ratio of pore gas. These simulations provide a theoretical basis for interpreting 3D NMR macroscopic responses, which should be helpful for pore-fluid identifying in tight oil and gas reservoirs.

Introduction

Tight oil and gas are one of the most important unconventional hydrocarbon resources (Behmanesh et al., 2018; McGlade et al., 2013; Ross et al., 2013; Xu et al., 2015; Li et al., 2019), and this kind of reservoirs are mainly existed in tight sandstone (TS) with low porosity and micro-pore development (Lai et al., 2018; Wang et al., 2017; Zou et al., 2012). Nuclear magnetic resonance (NMR) logging technology has a distinctive advantage over conventional logging technology in the application of this kind of reservoirs (Akkurt et al., 1995; Prammer et al., 1995).

In recent years, the new three-dimensional (3D) NMR technology has been developed considerably (Liu et al., 2019; Nicot et al., 2017; Zhang et al., 2019), it extends the observation results from a parameter of transverse relaxation time (T2) to longitudinal relaxation time (T1), T2, and diffusion coefficient (D) three parameters, and it is used to identify reservoir fluids based on different distribution states of oil, gas, and water in (T1, T2, D) space. Compared with one- and two-dimensional (1 and 2D) NMR technology, 3D NMR technology improves the accuracy of fluid identification and can obtain petrophysical parameters more accurately, such as porosity, fluid saturation and permeability. Heaton et al. (2004) employed 3D NMR to identify fluid type in wells drilled with oil-based mud and water-based mud, and calculated formation fluid saturation and oil viscosity. Liu et al. (2013) measured 3D NMR data for heavy oil samples using a unilateral NMR sensor, and then characterized the heavy oil components by NMR relaxometry and diffusometry. Nicot et al. (2017) evaluated the saturation, wettability and pore size distribution of rock by 3D NMR data, measured from inversion recovery-pulse gradient stimulated echo-CPMG (Carr-Purcell-Meiboom-Gill) pulse sequence. The rise of unconventional oil and gas reservoirs, such as tight oil and gas, has led to the development of 3D NMR logging technology, and their NMR response mechanisms and characteristics need to be studied urgently.

The difficulties and the high costs of maintaining the existence state of formation pore-fluid (especially with gas) in TS, making the NMR experiments challenging, at this time it is necessary to use the numerical simulation method to investigate the 3D NMR response mechanisms of TS. The numerical method based on multi-exponential model suffers great challenges for this case due to irregular pores. Fortunately, the numerical method based on digital core model works well in this case, and the method mainly contains the finite difference method (Zientara and Freed, 1980), the finite element method (Hagslätt et al., 2003), the random-walk (RW) method (Leibig, 1993; Talabi and Blunt, 2010), and the artificial neural network method (Farzi et al., 2017). The RW method has been widely used in the past because of its high flexibility and advantages for processing the rocks with complicated pore structures. At present, the researchers mainly focused on the simulation of 1D and 2D NMR responses of rocks using the RW method (Toumelin et al., 2007; Talabi and Blunt, 2010), but only Guo et al. (2016) simulated 3D NMR responses of TS saturated with oil and water, unfortunately, they only used the CPMG pulse sequence with different recovery times and echo spacings to obtain 3D NMR spectra.

In this paper, the random-walk method was modified to simulate the NMR measurements using a tri-window pulse sequence. A tight sandstone was taken as a study example, water and oil or gas in pore were characterized by 3D NMR simulations and their characterizations were analyzed, and the NMR microscopic mechanisms for pore-fluid in TS were explored.

Section snippets

Three-dimensional NMR theory

Three-dimensional NMR is a further expansion of 2D NMR, which can simultaneously measure three parameters T1, T2, and D of pore-fluids. Different 3D NMR pulse sequences correspond to different 3D NMR responses, and the most commonly used 3D NMR pulse sequence in actual oilfield was applied in this paper, as shown in Fig. 1 (Heaton et al., 2004; Zhang et al., 2013). The pulse sequence contains three separate windows, named as tri-window pulse sequence. It can edit the information of T1 by

Simulation method

The RW method is used to simulate NMR characterization of pore-fluid bearing rock through simulating Brownian motions of a large number of particles in the digital core model. The bulk relaxation, surface relaxation, and the diffusion relaxation of pore-fluid occur simultaneously and are independent of each other (Kenyon, 1997; Talabi, 2008), i.e., the normalized magnetization b(t) at time t with a recovery time of RT can be expressed asb(t)=[1b1B(RT)b1S(RT)][b2B(t)b2S(t)b2D(t)]

So their

Pore-fluid characterizations

To study the characterization of pore-fluid in rock by simulation, the TS digital core model was prepared as in Guo and Xie (2017). The visualizations of the tight sandstone digital core with different saturations were shown in Fig. 3. The core's porosity is 6.26%, the voxel size is 6003, and the resolution is 2.77 μm. Different pore-fluid cases were set, such as pure water, oil-water and gas-water, and the fluid's properties at ambient temperature were assumed as shown in Table 1. During the

Discussions

Section 4 has exhibited the pore-fluid characterizations of TS by simulation at pure water, oil-water, and gas-water cases. Here would sum up the phenomena, analyze the reason of characterization, and explore the microscopic response mechanisms of pore-fluid in TS.

From Fig. 4, Fig. 5, Fig. 6, we can see that the positions of the water, oil, and gas signals are different in T1-T2-D spectra, and the signal type is very easy to identify according to fluid diffusion coefficient line. From Table 2,

Conclusion and future work

Tight sandstone reservoir as an important reservoir has been paid an increasingly attention with the exploration and development further thorough. Pore-fluid characterization and microscopic mechanisms exploration are of great significance for the TS research. The paper aims at characterizing pore-fluid and giving an insight into the microscopic mechanisms of TS using the 3D NMR method, and the findings lay a theoretical foundation for applications and further research of 3D NMR in TS

CRediT authorship contribution statement

Jiangfeng Guo: Conceptualization, Methodology, Validation, Visualization, Writing - original draft, Writing - review & editing. Ranhong Xie: Supervision, Project administration, Funding acquisition, Writing - review & editing. Lizhi Xiao: Supervision, Writing - review & editing.

Declaration of competing interest

The authors declare no competing financial interest.

Acknowledgment

This work was funded by the National Natural Science Foundation of China (41674126).

References (31)

Cited by (19)

  • Characterizing pore-level oil mobilization processes in unconventional reservoirs assisted by state-of-the-art nuclear magnetic resonance technique

    2021, Energy
    Citation Excerpt :

    Compared with one- and two-dimensional (1 and 2D) NMR technology, 3D NMR technology increases the accuracy of fluid identification. However, the application of this technology usually needs the help of numerical simulation [58]. As a key parameter in the development of an oil reservoir, wettability greatly affects oil saturations, relative permeabilities, electrical properties, capillary pressures, and oil recovery [59].

View all citing articles on Scopus
View full text