Relationship Between Pore-Throat and Pressure in Oil Charging in Tight Conglomerate Reservoirs of the Xiazijie Formation in Manan Area, Junggar Basin: Proved from NMR Experiment

Abstract

The Xiazijie Formation in the Manan Area, Junggar Basin, is an extra-low- to ultra-low-porosity and -permeability deep conglomerate reservoir. The relationship between pore-throat and pressure in oil charging in this tight reservoir is undefined at the initial exploration stage. By combining high-pressure Hg injection with one-dimensional nuclear magnetic resonance (NMR) experiments, a conversion relationship between relaxation time in NMR and pore-throat radius during high-pressure Hg injection was established. Online NMR displacement experiments were conducted at pressures of 2, 5, 15, 25, 35, and 45 MPa. The variation patterns of reservoir oil saturation and the lower limit of the oil-charged pore-throat radius were investigated. The results show that the oil saturation of conglomerate reservoirs ranges from 8% to 32% at a charging pressure of 2 MPa, and from 19% to 52% at a pressure of 45 MPa. The average oil saturation follows a monotonically increasing power function with charging pressure. Oil saturation rapidly increases as charging pressure increases at low pressures, while it gradually increases at high pressures. The lower limit of the oil-charged pore-throat radius ranges from 12 to 82 nm at a charging pressure of 2 MPa, and from 8 to 20 nm at 45 MPa. This lower limit follows a monotonically decreasing power function with charging pressure. The lower limit of the oil-charged pore-throat radius rapidly decreases as charging pressure increases at low values, whereas it gradually decreases with a further increase in charging pressure. Understanding the relationship between oil saturation and the oil-charged pore-throat radius lower limit at different charging pressures could improve the understanding of the reservoir-forming rules and liquid-injection development in the tight conglomerate reservoir of the Xiazijie Formation in the Manan Area.

1. Introduction

The Junggar Basin is an important hydrocarbon-rich area in northwestern China. In the basin, the Mahu Sag is the largest hydrocarbon-rich sag. Two large oil zones, Kewu and Mahu, have been discovered in the peripheral thrust belt and the slope area of the sag. Several areas with reserves of 100 million tons have been established, which confirms the possibility of exploring a whole petroleum system in which conventional and unconventional oil and gas coexist in an orderly manner [1,2,3]. The Lower Permian Fengcheng Formation is the main source rock interval in the Mahu Sag. It is continuously and widely distributed and has a unique hydrocarbon-generation type. The source rock has undergone multiple stages of intense hydrocarbon generation and expulsion, providing an important material basis for large-scale hydrocarbon accumulation [4,5]. Coarse-grained fan-delta deposits developed in the Middle Permian Xiazijie Formation, which is the first reservoir interval overlying the Fengcheng Formation source rocks [6,7]. With large depositional thickness, the Xiazijie Formation was vertically charged by whole hydrocarbon and laterally distributed throughout the whole area. The overlying Lower Wuerhe Formation serves as a regional caprock. This results in a favorable source-reservoir-caprock assemblage in the Xiazijie Formation [8]. At present, hydrocarbon exploration and development of the Xiazijie Formation are in the initial stage and are mainly concentrated in the southern part of the sag, the Manan Area. Multiple oil reservoirs have been proven near the fault zone, with developed producing petroleum reserves of 1.3 × 10⁸ t. A favorable area of 372 km² has been confirmed, with controlled reserves of 0.3 × 10⁸ t and predicted reserves exceeding 1.1 × 10⁸ t. It is one of the major replacement domains for large-scale reserve growth and production increase in the Mahu Sag. With an average daily oil production of 15.7 t, the reservoirs have been tested in low oil saturation in more than 30 wells. However, the burial depth of the Xiazijie Formation generally exceeds 3500 m in the Manan Area. The conglomerate lithology of the Xiazijie Formation is complex; the reservoir physical properties are poor; the heterogeneity is strong; and the differences in burial depth and formation pressure are large. The dynamic mechanism of oil saturation and the lower limit of the oil-charged pore-throat radius in different hydrocarbon charging pressures remain unclear.

Unlike conventional oil and gas reservoirs, hydrocarbon charging in tight reservoirs must overcome greater capillary forces before entering nanoscale pore-throats [9,10,11]. The lower limit of the pore-throat radius for hydrocarbon charged under accumulation pressure is one of the key parameters in the exploration and development of tight oil and gas reservoirs [12,13,14]. By integrating rock mechanics, mud logging, effective pore-throat analysis, and other methods, Deng et al. [15] determined that the lower limits of the charged pore-throat diameter at the source-reservoir interface and in the reservoir of the Chang 8 tight sandstone interval in the Fuxian Area, Ordos Basin, are 15.77 nm and 24 nm, respectively. Li [16] used multiple analytical and testing datasets to establish a relationship model between charging pressure and the lower limit of the charged pore-throat radius and identified the lower limit of the effective charged pore-throat radius as approximately 100 nm in the Chang 7 tight sandstone reservoir of the Yanchang Formation in the Ordos Basin. Wang et al. [17] integrated high-pressure Hg injection, NMR experiments, mud logging, oil testing, and other data to determine that the lower limit of the charged pore-throat radius of the Chang 8 reservoir in the Maling Area of the Ordos Basin is 77 nm. Jiang et al. [18] developed an online three-dimensional microscopic imaging system for core-fluid displacement and established a quantitative evaluation method for the oil charging degree of image-resolvable pores and pore-throats. However, few studies have yet addressed the coupling relationship between the oil-charged pore-throat radius and charging pressure in tight conglomerate reservoirs in the Junggar Basin.

In recent years, increasingly improved online NMR displacement experiments have provided a new method for quantitatively characterizing the occurrence states of oil (gas) and water in tight reservoirs under different charging pressures [19,20,21]. To understand the relationship between pore-throat and pressure during oil charging in the tight conglomerate reservoirs of the Xiazijie Formation in the Manan Area, 16 tight conglomerate samples were acquired from four cored wells. High-pressure Hg injection and one-dimensional NMR experiments were combined to establish the conversion relationship between the relaxation time in NMR and pore-throat radius in high-pressure Hg injection. Online NMR displacement was then used to simulate the variation patterns of reservoir oil saturation and the lower limit of the charged pore-throat radius under oil charging pressures of 2 MPa, 5 MPa, 15 MPa, 25 MPa, 35 MPa, and 45 MPa. The results would support the exploration and development of favorable tight conglomerate reservoirs in the Xiazijie Formation of the Manan Area.

2. Geological Setting

2.1. Stratigraphic Characteristics

Structurally, the Mahu Sag is located on the northwestern margin of the Junggar Basin. It is adjacent to the Xiayan Uplift and Yingxi Sag to the east, the Kebai and Wuxia Fault Zones to the northwest, and the Zhongguai and Dabasong Uplifts to the south (Figure 1a) [22]. From bottom to top, the stratigraphic succession of the Mahu Sag consists of the Carboniferous (C), Permian (P), Triassic (T), Jurassic (J), and Cretaceous (K). In the Permian, the Jiamuhe Formation (P₁j), Fengcheng Formation (P₁f), Xiazijie Formation (P₂x), Lower Wuerhe Formation (P₂w), and Upper Wuerhe Formation (P₃w) were deposited successively. Controlled by paleogeomorphology and provenance, a restricted lacustrine fan-delta depositional system developed in the Xiazijie Formation (Figure 1b). The formation lithology is characterized by interbedded conglomerate, fines, and mudstone (Figure 1c).

2.2. Reservoir Characteristics

The grain size of the conglomerate in the Xiazijie Formation in the Manan Area is dominated by pebbles and granules, with abundant sand as supporting grains (Figure 2a). The conglomerates are mainly derived from volcanic clasts such as andesite, tuff, and dacite. The interstitial materials contain high proportions of sand and mud, and laumontite, calcite, and clay minerals are the main cements [22]. The conglomerate reservoirs of the Xiazijie Formation mainly develop intergranular dissolution pores and intragranular dissolution pores (Figure 2b), with porosity mainly ranging from 3% to 10% (averaging 5.1%) and permeability ranging from 0.01 to 10 mD (averaging 0.1 mD). They are classified as extra-low to ultra-low porosity–permeability reservoirs, with a poor porosity–permeability relationship (Figure 2c) [7]. According to the high-pressure Hg intrusion experiment, the maximum pore-throat radius ranges from 0.5 to 1.3 μm, the average pore-throat radius is ~15 nm, Hg ejection efficiency is low, and the median and drainage pressures are high. The reservoir characteristics of the conglomerate in the Xiazijie Formation indicate a complicated pore-throat structure (Figure 2d).

3. Databases and Methods

3.1. Research Procedure

To understand the relationship between pore-throat characteristics and oil charging pressure in tight conglomerate reservoirs based on NMR experiments, it is crucial to elucidate the primary petrophysical features of the Xiazijie Formation. This study comprises five stages (Figure 3). The first stage involves sample acquisition. Based on the reservoir characteristics of the Xiazijie Formation in the Manan Area, 16 samples with a diameter of 2.5 cm were drilled from full-diameter cores collected from four wells, including one pair of parallel samples. The samples were cut by wire to obtain standard cylindrical samples with a height of 5 cm.

The second stage is experimental preparation. Sample pretreatment and subsequent experimental analyses were performed at the Hubei Key Laboratory of Oil and Gas Drilling and Production Engineering (Yangtze University). The cylindrical samples were successively subjected to oil washing and salt washing by soaking to remove residual oil and salts. After drying and weighing, helium porosity and permeability tests were conducted using the CMS-400 automatic overburden measurement system manufactured by Core Laboratories.

The third stage involves Hg injection and NMR experiments. One sample from the pair of parallel samples was selected for the high-pressure Hg injection experiment using the AutoPore IV-9510 fully automatic Hg porosimeter manufactured by Micromeritics Instrument Corporation. The other sample was saturated with water under a pressure of 30 MPa for 2d and then centrifuged for one-dimensional NMR testing. The remaining 14 samples were used for two-dimensional online NMR displacement experiments. The fourth stage establishes the conversion relationship between relaxation time and pore-throat radius. Finally, variation patterns of oil saturation and pore-throat radius under different charging pressures are discussed in the fifth stage.

3.2. NMR Experiment

At a room temperature of 25 °C, one-dimensional NMR experiments and two-dimensional online NMR displacement experiments were conducted using the MacroMR12-150H-I large-bore NMR analysis and imaging system manufactured by Niumag Analytical Instrument Corporation, Suzhou, Jiangsu Province, China. The experimental procedure followed the Chinese petroleum and natural gas industry standard ‘Specification for measurement of rock NMR parameter in laboratory’ (SY/T6490—2023). The instrument sampling frequency was set to 200 kHz, waiting time to 3000 ms, echo spacing to 0.12 ms, and the number of sampling echoes to 12,000. Aviation kerosene with a density of 0.795 g/cm³ (25 °C) was selected as the charging medium for the two-dimensional online NMR displacement experiment. The procedures of the two-dimensional online NMR displacement experiment were as follows. (a) The two-dimensional NMR spectrum and total signal intensities of oil and water were standardized with volumes of 0.1 mL, 0.5 mL, 1.0 mL, 1.5 mL, and 2.0 mL, respectively, to determine the conversion relationship between the NMR signal and fluid volume. (b) Dry samples were saturated with water under a pressure of 30 MPa for 2d, and then the T₂–T₁ two-dimensional NMR spectra of the water-bearing samples were measured. (c) After oil charging for 5 h under displacement pressures of 2 MPa, 5 MPa, 15 MPa, 25 MPa, 35 MPa, and 45 MPa, respectively, the T₂–T₁ two-dimensional NMR spectra were measured again. (d) Based on the conversion relationship between the NMR signal and the fluid volume, the oil saturation of the samples under different charging pressures was calculated.

3.3. Conversion Relaxation Time and Pore-Throat Radius

High-pressure Hg injection experiments can yield the pore-throat radius distribution of cores. One-dimensional NMR experiments can measure relaxation time and porosity component data. Li et al. [23] established the conversion relationship between relaxation time and pore-throat radius using the cumulative distribution curves of high-pressure Hg injection pore-throat radius and relaxation time. Liu et al. [24] and Li et al. [25] used this conversion relationship to quantitatively characterize the fluid mobility and pore structure characteristics of tight conglomerate reservoirs. Firstly, cumulative distribution curves of high-pressure Hg injection pore-throat radius and relaxation time were plotted simultaneously. Secondly, both cumulative distribution frequencies below the maximum Hg intrusion saturation were selected. Thirdly, taking the scale of the cumulative distribution data of high-pressure Hg injection pore-throat radius as the standard, the cumulative distribution data of relaxation time were interpolated. Finally, the conversion formula between NMR relaxation time and high-pressure Hg injection pore-throat radius was fitted [23,24,25]:

$$r=C{T}_2^{{\displaystyle \frac{1}{n}}}$$

(1)

where r is the pore-throat radius, μm; T₂ is the NMR relaxation time, ms; and C and n are the dimensionless conversion coefficient and exponent, respectively. In this study, a cubic spline function program was developed using Python 3.10 to interpolate the cumulative distribution data of the relaxation time. The parameters C and n were fitted in Microsoft Excel. Interested readers can contact the corresponding author to obtain the related code and declassified experimental datasets free of charge.

4. Results

4.1. Online NMR Charging

After kerosene was displaced into tight conglomerate samples for 5 h under charging pressures of 2 MPa, 5 MPa, 15 MPa, 25 MPa, 35 MPa, and 45 MPa, the NMR signals of oil and water molecules were measured. In the T₂–T₁ two-dimensional NMR spectra, the area of the oil NMR signal gradually increased, indicating that oil molecules progressively entered the rock pore-throats. For example, the conglomerate sample from 4715.84 m in Well JL55 has a porosity of 6.92% and a permeability of 0.131 mD. Its T₂–T₁ two-dimensional NMR spectra under different charging pressures and the variation in oil saturation are shown in Figure 4a–g. When kerosene began to enter the sample, oil signals appeared in the upper part of the NMR spectrum. When the charging pressure was 2 MPa, the oil saturation of the sample rapidly increased to 18.9%; when the charging pressure was 5 MPa, the oil saturation was 27.3%; when the charging pressure exceeded 15 MPa, the oil saturation increased slowly from 31.4% to 38.7% under a charging pressure of 45 MPa.

4.2. NMR Pore-Throat Radius

High-pressure Hg injection experiments provide an effective method for converting NMR relaxation time into pore-throat radius. The T₂ spectrum of a water-saturated core reflects the response characteristics of all pore-throats in the rock. In high-pressure Hg injection experiments, however, Hg cannot be completely injected into small pores. The pore-throat radius distribution obtained from Hg intrusion cannot characterize pore-throat information beyond the maximum Hg intrusion saturation. Therefore, only the cumulative distribution data of the T₂ spectrum corresponding to the Hg intrusion pore-throat radius distribution are selected to establish the conversion relationship between relaxation time and pore-throat radius. The cumulative distribution curves of high-pressure Hg injection pore-throat radius and relaxation time from one pair of parallel samples are shown in Figure 5a. The maximum Hg saturation in the high-pressure Hg injection experiment is 76.64%. The cumulative NMR relaxation time corresponding to Hg intrusion pore-throat radii ranging from 4 nm to 2.5 μm is extracted. The cubic spline interpolation method is used to match the relaxation time and Hg intrusion pore-throat radius at the same cumulative frequency scale. A power-function formula was then fitted by the least squares method, resulting in C = 0.039 and 1/n = 1.3659 for Formula (1) (Figure 5b):

$$r=0.039T_2^{1.3659}$$

(2)

Finally, the relaxation time of the NMR T₂ spectrum was converted into pore-throat radius (Figure 5c). The results show that the NMR pore-throat radius distribution has a wider range, but the curve shapes of the two methods are generally consistent.

5. Discussion

5.1. Variation Pattern of Oil Saturation Under Different Charging Pressures

In the NMR experiments on oil charging in tight conglomerate reservoirs, online NMR displacement experiments were performed on all 14 samples. The oil saturations of the samples ranged from 8% to 32% under a charging pressure of 2 MPa. As the charging pressure increased to 45 MPa, the oil saturations ranged from 19% to 52%. The oil saturation of tight conglomerate reservoirs in the Xiazijie Formation showed a monotonically increasing relationship with charging pressure in each sample. Overall, the greater the sample porosity, the higher the oil saturation under different charging pressures (Figure 6a).

Using the least squares method, the relationship between charging average oil saturation (S_o) of all samples and the charging pressure (P_c) is a power function: S_o = 15.14 × P_c^0.1998. The fitted determination coefficient R² = 0.9867 (Figure 6b). In this power function, the root mean square error (RMSE) between the experimental and calculated charging average oil saturation is just 0.62% at 2 MPa, 5 MPa, 15 MPa, 25 MPa, 35 MPa, and 45 MPa (Figure 6c). When the charging pressure is relatively low, the reservoir oil saturation increases rapidly as the charging pressure increases; whereas at relatively high charging pressures, it increases slowly. Formation charging pressure is the primary driving force for fluid migration within reservoirs. In initially water-bearing formations, the greater the oil charging pressure and the longer the charging duration, the more oil molecules can enter the reservoir [26,27,28].

5.2. Minimum Oil-Bearing Pore-Throat Radius Under Different Charging Pressures

The relationship between the NMR signal and the relaxation time for the conglomerate sample from a depth of 4715.84 m in Well JL55 under different charging pressures is shown in Figure 7a. When the charging pressure exceeded 5 MPa, the shortest relaxation time of the oil NMR signal remained unchanged. Using the conversion relationship between the NMR relaxation time and the pore-throat radius established for the Xiazijie Formation based on Formula (2), the relationship between the NMR signal and the reservoir pore-throat radius under different charging pressures was obtained (Figure 7b). Under water-saturated conditions, the minimum pore-throat radius accessible to water molecules is 7.1 nm. When the charging pressure was 2 MPa, the lower limit of the charged pore-throat radius was 20.5 nm. When the charging pressure exceeded 5 MPa, the minimum pore-throat radius accessible to oil molecules was consistently 17.2 nm. This indicates that even with a continuous increase in charging pressure, oil molecules cannot break through the lower limit of the charged pore-throat radius of 17.2 nm in this rock sample.

The lower limits of the charged pore-throat radius of the 14 samples under different charging pressures are shown in Figure 8a. Under a charging pressure of 2 MPa, the lower limits of the charged pore-throat radius ranged from 12 nm to 82 nm. As the charging pressure increased to 45 MPa, they ranged from 8 nm to 20 nm. The lower limit of the charged pore-throat radius of tight conglomerate reservoirs in the Xiazijie Formation showed a monotonically decreasing relationship with charging pressure in each sample.

Fitted by the least squares method, the relationship between the average lower limit of the charged pore-throat radius (r_min) and the charging pressure is a power-function: r_min = 30.903 × P_c^−0.18. The fitted determination coefficient is R² = 0.8448 (Figure 8b). When the charging pressure is relatively low, the lower limit of the charged pore-throat radius decreases rapidly as the charging pressure increases; whereas at relatively high charging pressures, it decreases slowly. In this power function, the root mean square error (RMSE) between the experimental and calculated average lower limit of the charged pore-throat radius is only 2.02 nm at 2 MPa, 5 MPa, 15 MPa, 25 MPa, 35 MPa, and 45 MPa (Figure 8c). Theoretically, reservoirs are water-bearing at the early stage. In tectonic processes and hydrocarbon accumulation, crude oil begins to charge under different formation pressures. Water in large pores is gradually displaced by crude oil. As the oil charging pressure progressively increases, water in small pore-throats is replaced by crude oil [29,30]. After oil charging pressure stabilizes in the late stage of accumulation, oil molecules can no longer enter smaller pore-throats. Finally, the two-phase fluids of oil and water reach a dynamic equilibrium.

It should be emphasized that practical physical variables controlling oil charging in buried tight reservoirs are highly complex, including crude-oil properties, rock wettability, formation temperature, and geological charging timescales, among others. Addressing all of these variables within a single study is impractical at the initial exploration stage of the Xiazijie Formation in the Manan Area. Therefore, constrained by the practical data and experiment conditions, this study focuses on the coupling relationship between pore-throat and pressure in oil charging in the tight conglomerate reservoirs based on NMR experiments. These results are expected to contribute to the understanding of the reservoir-forming rules and liquid-injection development in future exploration.

6. Conclusions

(1)

The oil saturation of the tight conglomerate reservoirs of the Xiazijie Formation in the Manan Area ranges from 8% to 32% under a charging pressure of 2 MPa. As the charging pressure increases to 45 MPa, the oil saturation rises to 19–52%. The average oil saturation of the reservoir shows a monotonically increasing power-function relationship with charging pressure.

(2)

The lower limit of the charged pore-throat radius of the tight conglomerate reservoirs of the Xiazijie Formation in the Manan Area ranges from 12 nm to 82 nm under a charging pressure of 2 MPa. As the charging pressure increases to 45 MPa, this lower limit ranges from 8 nm to 20 nm. The lower limit of the charged pore-throat radius exhibits a monotonically decreasing power-function relationship with charging pressure.