Research on Digital Core Characterization and Pore Structure Control Factors of Tight Sandstone Reservoirs in the Fuyu Oil Layer of the Upper Cretaceous in the Bayan Chagan Area of the Northern Songliao Basin

Abstract

The tight sandstone reservoir of the Fuyu Oil Layer in the Quantou Formation of the Cretaceous in the Bayan Chagan area displays intricate microscopic pore structures and pronounced heterogeneity, limiting hydrocarbon exploration and development efficiency. Utilizing core CT scanning digital core technology integrated with field emission scanning electron microscopy (FE–SEM) and whole-rock/clay mineral X-ray diffraction (XRD) analysis, this research performs multi-scale quantitative characterization on 15 representative rock samples from the study area, systematically elucidating reservoir pore structure diversity and its formation mechanisms. The study demonstrates that reservoirs in the study area can be categorized into three types: A, B, and C, exhibiting progressively declining reservoir performance. Type A reservoirs are characterized primarily by dissolution-formed large to medium pores, where macropores (radius > 5 μm) account for more than 92% of storage capacity, average coordination numbers reach 0.27~0.45, and connectivity is optimal. Type B reservoirs are influenced by siliceous cementation, featuring developed residual intergranular pores, macropore volume share declining to 88%, and coordination numbers decreasing to 0.11~0.20. Type C reservoirs experience intense compaction and illite cementation modification, where micropores (radius < 1 μm) constitute 5.6% numerically, yet macropore volume share is merely 76%, coordination numbers drop to 0.02–0.03, and connectivity is minimal. Mineralogical analysis reveals that quartz content exhibits a positive correlation with reservoir properties, as its rigid grain framework effectively resists compaction. Illite content rises with increasing burial depth, and plastic illite occupies pores and segment throats, resulting in Type C reservoir permeability reduction to 0.01~0.25 mD. Dissolution intensity (Type A > Type B > Type C) and cementation types (quartz cementation prevailing in Type B, illite cementation prevailing in Type C) represent crucial factors governing reservoir quality differentiation. This research confirms the reliability of digital core technology for tight reservoir classification and assessment, developing a discrimination model founded on “pore structure-mineral composition-diagenesis”. It provides a geological basis for sweet spot prediction and efficient development in the study area.

1. Introduction

At present, as global energy demand continues to rise, the exploration and development potential of conventional oil and gas is gradually depleting, leading to the exploration and development of oil and gas resources entering the unconventional sector [1,2,3]. Tight oil and gas is a crucial type of unconventional oil and gas resource. China is rich in tight oil and gas resources, with substantial proven reserves and a broad distribution range. The exploration and development of tight oil and gas is gradually replacing conventional oil and gas resources as a key area of exploration and development [4]. At present, substantial tight sandstone oil and gas reserves are found in major basins in China, including the Sichuan Basin, Songliao Basin, and Ordos Basin [5,6,7]. However, tight sandstone reservoirs, characterized by complex pore structures, low porosity and permeability, and strong reservoir heterogeneity, present significant challenges for the exploration and development of tight sandstone oil and gas [8,9,10].

Research by previous scholars indicates that the precise representation of the reservoir’s microscopic pore structure is crucial for the evaluation and effective development of tight sandstone oil and gas resources [11,12]. At present, the commonly used methods for characterizing tight sandstone reservoirs include mercury intrusion, gas adsorption, and nuclear magnetic resonance techniques [13,14,15,16]. While these methods can offer refined characterization of the reservoir’s microscopic pore features, they all have limitations. For instance, mercury intrusion can damage rock samples, gas adsorption cannot measure isolated pores, and nuclear magnetic resonance is influenced by reservoir heterogeneity, which can cause a discrepancy between the characterization results and the actual conditions.

At present, digital core technology is extensively used in the evaluation of different types of unconventional reservoirs, quantitative representation of complex microscopic pore structures, and rock physics experiments [17,18]. Digital core technology is a numerical simulation method for rocks that accurately reflects their real structural features. In contrast to traditional reservoir characterization methods, digital core technology has the benefits of being fast, accurate, non-destructive to samples, and offering high resolution, making it highly effective for evaluating rock heterogeneity [19,20,21].

The Bayanchagan region of the Songliao Basin has abundant tight oil and gas reserves, yet systematic research on the tight sandstone reservoirs in this area is still underdeveloped. Current studies primarily focus on the macroscopic characteristics of the reservoirs, and the microscopic structural features are not well understood [22,23,24]. Earlier research has made initial investigations into the reservoir characteristics in this region and has established a classification of the reservoirs [25]. However, a detailed analysis of the causes of reservoir heterogeneity has not yet been conducted. Building on this, the present study incorporates digital core technology, along with field emission scanning electron microscopy (FE–SEM) and XRD mineral composition analysis, to perform digital core-based classification and characterization of the tight sandstone reservoirs in the Bayanchagan region [20]. Additionally, the study further explores the heterogeneity of different reservoir types from a microscopic genesis mechanism perspective to identify the main factors controlling reservoir formation and evolution. This research will provide a geological foundation for future oil and gas exploration and development in the area, as well as help assess the effectiveness of digital core technology in the classification and evaluation of tight reservoirs.

2. Regional Geological Overview

The Songliao Basin, located in northeastern China, is a large Cenozoic continental sedimentary basin with a graben-horst dual structure. Covering an area of about 2.6 × 10⁵ km², it is the largest onshore basin in China with the richest oil and gas resources [26]. The structural division of the Songliao Basin consists of six first-order structural units: the central depression area, the western slope area, the northern tilted area, the northeastern uplift area, the southeastern uplift area, and the southwestern uplift area [27,28]. The stratigraphy of the Songliao Basin is relatively simple, consisting of Jurassic (J), Cretaceous (K), Neogene (N), and Quaternary (Q) strata from bottom to top. The Cretaceous is divided into the Upper and Lower Cretaceous formations. The study area is located in the western Bayanchagan region of the northern central depression zone in the Songliao Basin, with an exploration area of approximately 5.4 × 10² km². The main part is located in the central section of the Longhupao-Daan terrace, with two fault zones, Halahai and Talaha, that extend in the same direction (Figure 1a,b). The main oil layers developed in this area from top to bottom are the Heidi Temple oil layer, the Sartu oil layer, the Putaohua oil layer and the Gaotai Zi oil layer. Among them, the Fuyu oil layer is located between the third and fourth segments of the Qiantou Formation of the Cretaceous System in the Mesozoic Era. It features river-delta and river facies, and the lithology is mainly composed of sandstone and argillaceous sandstone [29,30] (Figure 1c). The rock permeability in the region is mainly ultra-low permeability. This paper primarily focuses on the Upper Cretaceous Q4 Section Fuyu oil layer in the Bayanchagan area, where 12 wells have produced industrial oil flows, demonstrating the good oil-bearing properties of the Fuyu oil layer in the region.

3. Samples and Methods

The experimental samples were chosen from three wells (Y201, Y26, Y28) in the Bayanchagan region, Songliao Basin. Through systematic observation of these wells, 15 core samples were selected, and a series of experimental analyses, including CT scanning, digital core modeling, field emission scanning electron microscopy (FE–SEM), and XRD diffraction, were performed to refine the characterization of the reservoir in the study area.

The field emission scanning electron microscope instrument adopted in this experiment is the Czech TESCAN MIRA LMS. FE–SEM is a high-precision electron microscope that operates by directing a focused electron beam onto the surface of the core sample, causing interactions between atoms that generate various electrical signals, which are used to display high-resolution images of the sample’s microscopic structure. Prior to the experiment, the core samples are cut into 1.5 × 1.5 cm square pieces, then bombarded with argon ions for 3 h under a vacuum. Afterward, the sample surface is chromium-coated, and the smooth rock surface is placed under the FE–SEM for observation of its microscopic morphological features.

The XRD experiment was conducted using the new high-resolution diffractometer D8AA25 from Bruker AXS GMBH in Karlsruhe, Germany. The working voltage is 15 KV, with a resolution of 2.0 nm, ambient temperature of 20 °C, and 30% humidity. The principle of the XRD experiment is to measure the diffraction of X-rays by a substance, generating a diffraction pattern. This pattern is then analyzed to determine the types and quantities of bulk rock and clay minerals.

The CT scanning experiment adopted the German Phoenix Nanotom S digital core scanner (Figure 2). Initially, core samples were processed into small cylindrical plugs with a diameter of 5 mm using a diamond drill bit. The samples were placed in the scanning chamber and scanning parameters were adjusted (resolution: 1.428 μm, voltage: 130 Kv, current: 120 μA, exposure time: 1800 ms, resulting in 1800 two-dimensional grayscale images). After setup, the X-ray source was activated, allowing X-rays to pass through the sample on the stage and be detected by the detector. The sample was rotated by a set angle, X-rays were emitted repeatedly, and the attenuated X-ray signals were recorded. The sample underwent continuous 360-degree rotation (Figure 3), producing the overall two-dimensional grayscale image of the sample (Figure 4). These images were then used for three-dimensional core reconstruction. As X-rays pass through the core, energy attenuation occurs, and minerals with different densities produce varying grayscale on the two-dimensional images. Pores appear dark black, white represents high-density rock components, and gray represents the rock matrix.

After CT scanning, the acquired data require further processing, and three-dimensional pore structure characterization is performed using Avizo 2020 software. Due to factors such as variations in the sample material and electronic component fluctuations, the slice images obtained from CT scanning often contain noise. To accurately reflect the rock’s internal features, denoising and filtering are necessary. In this experiment, a median filtering method was employed. The principle of median filtering involves sorting the pixel values in the neighboring pixels, then choosing the middle value as a substitute, thereby removing noise while preserving as many details of the original image as possible. After median filtering, the two-dimensional grayscale images must undergo image segmentation. The threshold segmentation method is applied, dividing the grayscale values based on the relationship between the grayscale of the image and the rock density, and referencing the measured porosity using a preset threshold to separate the image into two parts, further distinguishing the rock matrix from the pores (Figure 5). Given the large size of the original two-dimensional grayscale image data, after binarization, a 300 × 300 × 300-pixel region with clear pore features is selected. Using the PNM module in Avizo software, a three-dimensional model is created by stacking the binarized 2D slice images into a 3D data volume. The pore and throat model of the sample is then extracted based on the maximum sphere algorithm, where pores are approximated as spherical and throats as tubular to represent the storage space. This process generates a 3D visual model that reflects the pore distribution and connectivity between pores in three-dimensional space. Connected pores in the 3D pore model are extracted using the watershed algorithm from topology to analyze reservoir connectivity. Finally, by constructing the three-dimensional visual model of the sample, the pore-throat features of the original core are intuitively displayed. Statistical analysis of the 3D model yields parameters such as pore radius, pore count, throat radius, and throat count, which are used to quantitatively characterize the reservoir’s micro-pore structure.

4. Results and Discussion

4.1. Petrological Characteristics of Reservoirs

The XRD diffraction results for whole rock and clay minerals indicate that the mineral composition of the tight sandstone reservoir in the Bayanchagan area of the Songliao Basin is mainly quartz, with quartz content ranging from 27.4% to 56.1%, and an average of 42.4%. Plagioclase follows as the second most abundant mineral, with a content range of 15% to 48.1%, averaging 26.8%. Potassium feldspar content ranges from 2.6% to 12.8%, with an average of 5.0%. Calcite has the lowest content, ranging from 0% to 19.9%, with an average of 5.1% (

Table 1. Whole Rock Mineral Composition from Core XRD Analysis.

Sample Number	Well Number	Whole Rock Quantitative Analysis (%)
		Total Amount of Clay Minerals	Quartz	Potassium Feldspar	Plagioclase	Calcite
1	Y26	11.6	39.3	4.8	28.5	15.8
2	Y26	18.7	37.8	3.3	20.3	19.9
3	Y26	25.7	43.4	7.8	20.2	2.8
4	Y26	26.7	34	7.5	31.8	0
5	Y28	28.7	39.6	4.1	27.7	0
6	Y28	24.9	48.9	3.4	22.7	0
7	Y28	21.3	48.1	3	27.5	0
8	Y28	28.4	47.8	2.6	21.2	0
9	Y28	21.2	39.3	5.4	32.6	1.4
10	Y201	18.6	42.5	3.4	26.7	8.8
11	Y201	12.2	56.1	3.7	15	12.9
12	Y201	10.9	27.4	12.8	48.1	0.9
13	Y201	30.1	36.8	2.6	30.5	0
14	Y201	14.3	43.3	6.1	31.7	4.6
15	Y201	17.3	51.2	4	18.8	8.7

). The clay mineral content ranges from 10.9% to 30.1%, with an average of 20.7%. Illite is the most abundant clay mineral, with a content range of 18% to 66%, averaging 38.1%. Chlorite follows, with a content range of 8% to 50%, averaging 32.1%. The content of illite-smectite mixed-layers ranges from 20% to 46%, with an average of 29.9%, where smectite accounts for 23.6% and illite accounts for 76.4% (

Table 2. XRD. Content of clay minerals in core samples.

Sample No.	Well No.	Total Clay Content (%)			I/S Mixed-Layer (%)
		Illite (I) (%)	Chlorite (C) (%)	I/S Ratio	Smectite (S) (%)	Illite (I) (%)
1	Y26	54	21	25	15	85
2	Y26	42	31	27	20	80
3	Y26	42	34	24	20	80
4	Y26	64	8	28	15	85
5	Y28	66	12	22	25	75
6	Y28	18	36	46	25	75
7	Y28	29	44	27	30	70
8	Y28	28	35	37	30	70
9	Y28	30	33	37	30	70
10	Y201	34	37	29	25	75
11	Y201	30	45	25	25	75
12	Y201	26	44	30	25	75
13	Y201	64	16	20	25	75
14	Y201	23	50	27	20	80
15	Y201	22	35	43	25	75

).

4.2. Reservoir Physical Property Characteristics and Classification

A total of 15 samples were selected from 3 wells to analyze and statistically evaluate the physical properties of the Q4 section reservoir in the Bayanchagan area. The permeability ranges from 0.01 mD to 1.24 mD, with an average of 0.41 mD (

Table 3. Basic Information of Samples.

Sample No	Well No	Depth (m)	Lithology	Porosity (%)	Permeability (mD)
1	Y26	2135.01	Argillaceous Siltstone	7.85	0.10
2	Y26	2140.97	Argillaceous Siltstone	3.59	0.0098
3	Y26	2166.42	Siltstone	7.54	0.24
4	Y26	2179.61	Siltstone	4.40	0.01
5	Y28	1653.01	Siltstone	7.25	0.042
6	Y28	1656.52	Siltstone	5.72	0.023
7	Y28	1667.85	Medium-grained Sandstone	13.98	1.24
8	Y28	1669.05	Medium-grained Sandstone	13.68	1.10
9	Y28	1670.53	Fine-grained Sandstone	10.32	0.87
10	Y201	1809.85	Fine-grained Sandstone	8.99	0.90
11	Y201	1834.90	Fine-grained Sandstone	6.30	0.060
12	Y201	1835.50	Fine-grained Sandstone	8.90	0.42
13	Y201	1860.60	Siltstone	5.94	0.015
14	Y201	1869.83	Fine-grained Sandstone	8.006	0.36
15	Y201	1895.09	Medium-grained Sandstone	8.006	0.76

). The porosity ranges from 3.59% to 13.98%, with an average of 8.03%, indicating poor reservoir properties (Table 3). In this study, 9 significant samples from the Bayanchagan area were selected for fault CT scanning, and a 400 × 400 × 400-pixel region was chosen for three-dimensional digital core quantitative characterization. Based on reservoir physical properties, fault CT scan images, and digital core reservoir parameters, the tight sandstone reservoirs in the Bayanchagan area were classified into three types, with pore connectivity decreasing in sequence. The permeability of type A reservoirs ranges from 1.1014 mD to 0.9046 mD, and the coordination number ranges from 0.45 to 0.27 (

Table 4. Reservoir Classification Table.

Reservoir Type	Sample Number	Porosity%	Penetration Rate mD	Pore Quantity	Average Pore Radius μm	Average Pore Volume μm/cm3	Allocation Number
Type A	8	13.685	1.1014	14,230	2.57	878.696	0.45
	9	10.326	0.8754	14,154	2.27	555.946	0.31
	10	8.992	0.9046	4364	2.78	1189.601	0.27
Type B	15	8.006	0.7636	9062	2.02	563.688	0.20
	12	8.901	0.4231	6878	1.99	168.811	0.14
	11	6.3	0.0601	5867	2.31	203.231	0.11
Type C	1	7.851	0.101	65,535	2.02	61.514	0.02
	2	3.594	0.0098	60,272	2.05	82.112	0.03
	3	7.548	0.2453	65,535	2.15	86.092	0.02

). The permeability of type B reservoirs ranges from 0.7636 mD to 0.0601 mD, and the coordination number ranges from 0.20 to 0.11 (Table 4). The permeability of type C reservoirs ranges from 0.2453 mD to 0.0098 mD, and the coordination number ranges from 0.03 to 0.02 (Table 4). From the two-dimensional grayscale images, type A reservoirs mainly consist of intergranular pores (Figure 4a–c), type B reservoirs mainly consist of remaining intergranular pores, with the pores filled by high-density materials (Figure 4d–f), and type C reservoirs show strong compaction, with only small pores remaining (Figure 4g–i).

4.3. Reservoir Pore Structure Characteristics

To further investigate the pore-structure characteristics of the three reservoir types in the BYCG area, we selected a number of typical samples from each reservoir type and redefined subregions for three-dimensional visualization modeling over domains of 300 × 300 × 300 pixels (Figure 6). In the pore-network model, pores and throats of varying sizes are depicted as spheres and rods in different colors; as the color shifts from blue to red, the sizes of pores and throats progressively decrease, and the three reservoir types display distinct pore–throat characteristics. More specifically, Type A reservoirs are characterized by pores arranged in a cross-cutting, network-like configuration (Figure 6a,d). In Type B reservoirs, pores exhibit a partially connected arrangement, whereby only a portion of the pores are mutually connected (Figure 6b,e). In Type C reservoirs, pores are predominantly isolated, and abundant micron-sized pores densely occupy the reservoir volume, collectively forming the storage space (Figure 6c,f).

For quantitative characterization of reservoir space features, Avizo software was employed to calculate detailed pore and throat characteristic parameters. Results demonstrate distinct differences in pore and throat distribution characteristics across the three reservoir types within the study area. For detailed analysis of pore and throat distribution features in the three reservoir types, pore and throat dimensions in the study area were further categorized. Utilizing inflection points from the pore radius distribution curve (Figure 6), pores with radii exceeding 5 μm were classified as “macropores” using 1 μm and 5 μm as demarcation points. Pores with radii ranging from 1 to 5 μm were categorized as “mesopores”. Pores with radii below 1 μm were classified as “micropores”. Accordingly, throats with radii exceeding 5 μm were designated as “coarse throats”. Throats with radii ranging from 1 to 5 μm were classified as “medium throats”. Throats with radii below 1 μm were categorized as “fine throats”.

In the designated test interval of the samples, Type C reservoirs exhibit the maximum pore count (51,613), succeeded by Type B reservoirs (46,384), while Type A reservoirs show the minimum pore count (32,652), demonstrating a progressive increase in pore numbers from Type A to Type C reservoirs. Regarding the percentage distribution of pores by size, all three reservoir types are predominantly composed of “mesopores” (1–5 μm), comprising more than 80% of the total pore population. “Micropores” (radius < 1 μm) account for about 6% of the total pores. “Macropores” (radius > 5 μm) constitute roughly 2% in Type A and B reservoirs (

Table 5. Detailed pore data of Class A reservoirs in Q4 section of Bayanchagan area.

Reservoir Type	Classification	Micropore	Mesopore	Macropore	Total
Type A	Pore radius (μm)	<1	1~5	>5
	Number	2173	29,298	1181	32,652
	Percentage of pore number (%)	6.66	89.71	3.63	100.00
	Percentage of pore volume (%)	0.04	7.25	92.71	100.00
	Number of connected pores	1345	1859	584	2479
	Percentage of connected pore number (%)	0.46	5.32	1.82	7.60
	Percentage of connected pore volume (%)	0	0.23	81.83	82.06
	Number of isolated pores	2137	27,439	597	30,173
	Percentage of isolated pore number (%)	6.20	84.39	1.81	92.40
	Percentage of isolated pore volume (%)	0.04	7.02	10.88	17.94

and

Table 6. Detailed pore data of Class B reservoirs in Q4 section of Bayanchagan area.

Reservoir Type	Classification	Micropore	Mesopore	Macropore	Total
Type B	Pore radius (μm)	<1	1~5	>5
	Number	3123	42,279	982	46,384
	Percentage of pore number (%)	6.74	91.14	2.12	100.00
	Percentage of pore volume (%)	0.05	11.68	88.27	100.00
	Number of connected pores	52	2834	374	3260
	Percentage of connected pore number (%)	0.13	6.1	0.80	7.03
	Percentage of connected pore volume (%)	0	0.51	79.22	79.73
	Number of isolated pores	3071	39,445	608	43,124
	Percentage of isolated pore number (%)	6.58	85.04	1.32	92.97
	Percentage of isolated pore volume (%)	0.05	11.17	9.05	20.27

) (Figure 7), whereas in Type C reservoirs they comprise approximately 5% (

Table 7. Detailed pore data of Class C reservoirs in Q4 section of Bayanchagan area.

Reservoir Type	Classification	Micropore	Mesopore	Macropore	Total
Type C	Pore radius (μm)	<1	1~5	>5
	Number	2905	45,724	2984	51,613
	Percentage of pore number (%)	5.63	88.59	5.78	100.00
	Percentage of pore volume (%)	0.06	23.85	76.09	100.00
	Number of connected pores	273	618	16	907
	Percentage of connected pore number (%)	0.53	1.2	0.03	1.76
	Percentage of connected pore volume (%)	0.01	0.53	39.37	39.91
	Number of isolated pores	2632	45,106	2968	50,706
	Percentage of isolated pore number (%)	5.10	87.39	5.75	98.24
	Percentage of isolated pore volume (%)	0.05	23.32	36.72	60.09

). In terms of pore volume distribution, the pore volume across all three reservoir types is primarily controlled by the limited number of macropores. In particular, macropores constitute more than 80% of the pore volume in Type A and B reservoirs. Conversely, in Type C reservoirs, macropores account for merely 76.09% of the pore volume. Progressing from Type A to Type C reservoirs, the macropore proportion shows a gradual decline, accompanied by a corresponding increase in the volume shares of mesopores and micropores (Figure 8b). In terms of throat characteristics, substantial variations exist in throat counts across the three reservoir types. In the test interval, Type B reservoirs exhibit the maximum throat number (7100), succeeded by Type C reservoirs (6314), while Type A reservoirs display the minimum throat count (5895) (

Table 8. Data of throat channels of three types of reservoirs in Bayanchagan area.

Reservoir Type	Classification	Fine Throat	Medium Throat	Coarse Throat	Total
	Pore throat radius (μm)	<1	1~5	>5
Type A	Number	1514	3574	807	5895
	Percentage of throat number (%)	25.68	60.63	13.69	100.00
Type B	Number	1999	4225	876	7100
	Percentage of throat number (%)	27.27	59.92	12.34	100.00
Type C	Number	1749	4529	36	6314
	Percentage of throat number (%)	27.71	71.72	0.57	100.00

). Generally, the percentage of coarse throats shows a progressive increase from Type A to Type C reservoirs (Figure 8c).

4.4. Reservoir Connectivity Characteristics

For further characterization of connectivity in the three reservoir types, connected pores and isolated pores were extracted based on three-dimensional digital core models, constructing connected pore and isolated pore network models for the reservoirs, with quantitative characterization of pore parameters. The study reveals that connected pores exhibit continuous and dense distribution patterns within the reservoirs, whereas isolated pores display scattered and disordered distribution (Figure 9). All three reservoir types are predominantly composed of isolated pores (>90%), while connected pores constitute a minimal fraction (7.60%~1.59%). Nevertheless, Type C reservoirs exhibit a precipitous drop in connected pore proportion (merely 1.76%), indicating underdeveloped pore systems (Figure 10a). Progressing from Type A to Type C reservoirs, the connected pore percentage shows a gradual reduction, accompanied by a corresponding increase in isolated pore percentage. Regarding pore volume distribution, limited connected pores in Type A and B reservoirs account for the vast majority of effective volume. Conversely, Type C pore volume is primarily composed of isolated pores, with isolated pores comprising as much as 60.09% of the total volume (Figure 10b,c). Progressing from Type A to Type C reservoirs, the connected pore volume share shows a gradual decline, with a corresponding increase in isolated pore proportion (Figure 10). Within connected pores across the three reservoir types, macropores exhibit the smallest numerical percentage but the largest volume share. Meanwhile, isolated pores are mainly comprised of micropores to mesopores (Figure 10a).

4.5. Analysis of the Formation Mechanism of Reservoirs

Reservoir quality exerts a decisive influence on hydrocarbon resource exploration and development, with pore structure representing a key assessment parameter for reservoir quality [4,31]. Consequently, investigation into the formation mechanisms of reservoir pores holds particular significance in reservoir assessment [32]. As widely recognized, diagenetic processes constitute important factors influencing reservoir quality [33,34,35]. Building upon this premise, this research further examines the impacts of various diagenetic processes on reservoir pore structure, aiming to elucidate the formation mechanisms responsible for the diversity observed in tight sandstone reservoirs of the Q4 member in the BYCG region. Furthermore, it seeks to provide additional verification for the accuracy of digital core experimental methods.

Research findings demonstrate that variations among the three reservoir types mainly occur in macropore and coarse throat proportions and connectivity, exhibiting a declining trend: Type A > Type B > Type C. The reservoirs in the study area experienced compaction, cementation, and dissolution during their formation. Type A reservoirs show relatively weak compaction, characterized by pore types consisting mainly of larger intergranular dissolution pores, intragranular dissolution pores, and primary intergranular pores (Figure 11a). The lithology is dominated by medium sandstone and fine sandstone, displaying well-preserved hair-like illite and bridge-like illite under microscopic examination, with observable pores between clay minerals (Figure 11d). Type B reservoirs undergo relatively intense compaction and limited dissolution, featuring pore types predominantly composed of residual intergranular pores (Figure 11e). The lithology consists mainly of fine sandstone and siltstone, exhibiting strong siliceous cementation (Figure 11f,g). Type C reservoirs undergo the most intense compaction, characterized by pore types consisting mainly of minute intergranular pores with limited dissolution pores (Figure 11i). The lithology is dominated by siltstone and argillaceous siltstone, showing illite cementation that blocks pore spaces (Figure 11j). Progressing from Type A to Type C reservoirs, compaction and cementation effects progressively intensify, whereas dissolution processes gradually diminish. Primary intergranular pores show a gradual reduction, with dominant reservoir pores transitioning from large pores to minute pores (Figure 11a,e,i).

Illite demonstrates considerable plasticity and occupies pore spaces during cementation, consequently diminishing reservoir properties and leading to decreasing trends in porosity and permeability (Figure 12a,d). As reservoir burial depth increases, illite-smectite mixed layers gradually convert to illite, with illite content showing a progressive rise (Figure 13a,c). Type C reservoirs feature pore types primarily composed of clay mineral pores, with some pore throats possibly at nanometer dimensions that micro-CT cannot precisely detect, representing key factors contributing to the inferior physical properties of Type C reservoirs. As quartz and chlorite contents increase, both porosity and permeability exhibit rising trends (Figure 12b,c,e,f). This occurs because although quartz cementation in reservoirs occludes pores and diminishes reservoir properties, the rigid framework of quartz grains offers substantial compaction resistance, allowing preservation of primary pores. Within the study area, chlorite predominantly occurs as film-like chlorite; while chlorite films readily obstruct throats and degrade reservoir properties, they also exhibit some compressive resistance. Additionally, chlorite film formation suppresses quartz overgrowth processes [36,37,38]. The coexistence relationship between chlorite and quartz, coupled with their cementation and compaction interactions, enables relatively good preservation of pores in Type B reservoirs.

The reservoir samples utilized in this research exhibit distinct stratigraphic and depth representativeness. Type A reservoir samples were mainly obtained from reservoirs at around 1600 m depth, characterizing the typical properties of relatively shallow-buried reservoirs. Type B reservoir samples were chosen from critical intervals at about 1800 m depth, representing the transitional features of intermediate-depth reservoirs. Type C reservoir samples were derived from shallow sequences at around 2100 m depth, exemplifying the reservoir characteristics in deeper sections of the study area (Table 3). This systematic sampling approach according to depth and stratigraphic position guarantees that research findings can distinctly reveal the vertical distribution patterns of high-quality reservoirs (Type A). Furthermore, it establishes a reliable geological framework for comparative analysis with analogous reservoirs in neighboring regions or other basins. Thus, the derived understanding provides more definitive guidance for exploration practices.

5. Conclusions

Tight sandstone reservoirs in the Fuyu oil layer of the Bayanchagan area can be clearly classified into three types based on pore connectivity, pore-throat distribution, and mineral composition. Type A reservoirs (depth approximately 1600 m) develop networked large to medium dissolution pores, with macropore volume proportion > 92% and connected porosity reaching 7.6%, representing high-productivity sweet spots. Type B reservoirs (depth approximately 1800 m) display locally connected pores, where siliceous cementation causes throat narrowing, exhibiting moderate reservoir performance. Type C reservoirs (depth approximately 2100 m) undergo strong compaction and illite cementation alteration, featuring isolated pore distribution with connected porosity of only 1.76%, constituting low-productivity intervals. The three reservoir types show significant differences in macropore volume proportion (Type A 92.7% → Type C 76.1%), quantitatively revealing the controlling effect of microstructural heterogeneity on macroscopic flow capacity.

The uneven development of compaction, cementation, and dissolution processes constitutes the fundamental cause of reservoir differential evolution. Dissolution: Type A reservoirs develop abundant intergranular and intragranular dissolution pores due to intense feldspar particle dissolution, effectively enhancing storage space. Cementation: Quartz cementation dominates in Type B reservoirs, which, while reducing porosity, enhances rock resistance to compaction. Illite cementation is particularly developed in Type C reservoirs, with content reaching 38.1% (average value), and its flaky, bridging occurrence severely blocks throats, representing the main cause of connectivity deterioration. Compaction: With increasing burial depth (deeper than 2100 m), mechanical compaction causes primary intergranular pores in Type C reservoirs to nearly disappear, reducing porosity to 3.59%~7.85%.

XRD and FE–SEM analyses confirm that quartz content (27.4%–56.1%) shows positive correlation with porosity and permeability, with its rigid support effect being crucial for pore preservation. Illite content demonstrates significant negative correlation with reservoir physical properties and shows linear increasing trend with burial depth, serving as a sensitive indicator for reservoir quality prediction. Although chlorite films possess compaction resistance, excessive development can partition pore space, with their net effect depending on synergistic relationships with quartz cementation.

The quantitative digital core evaluation process established in this study has achieved three-dimensional visualization and parameter extraction of the pore-throat network in the tight reservoirs of the Bayanchagan area, making up for the deficiencies of conventional methods such as mercury injection and nitrogen adsorption in characterizing connectivity. This classification scheme and genesis model can provide a replicable technical framework for the evaluation of similar tight oil reservoirs in the Songliao Basin.