The Shulu sag is located in the southern part of the Jizhong depression, being a half-graben developed on Paleozoic basement, extending in northeast direction, and faulted in the east and overlapped in the west, with an exploration area of more than 700 km2. At present, five oil-bearing series (O, C–P, Es1, Es2 and Es3) have been discovered, with many types of reservoirs, including sandstones, conglomerates, marls, carbonate reservoirs and other types (Cui Z Q et al., 2015; Zhao et al., 2015). The Paleogene has three reservoir combinations, namely Es1 sandstone, Es2 and Es3 sandstone and Es3 marl in descending order. During the early depositional period of Es3 Member of Shahejie Formation of Paleogene, two paleo-uplifts (Jingqiu and Tierzhuang) and marginal faults divided the Shuluhu Basin into three water bodies (southern, central, and northern), which were not fully communicated, and gradually became salty from north to south (Jiang et al., 2019; Zhao et al., 2015) (Fig. 1a).
Figure 1. Geological structure and thickness distribution of marl in Shulu sag (adapted from Zhao et al., 2015).
The Es3 Member covers directly on the Paleozoic basement, characterized by thick in the east and thin in the west, with the maximum thickness over 1 200 m, of which the maximum thickness of conglomerate is 520 m, and the maximum thickness of marl is 403 m. Carbonate rock is developed on the steep and gentle slopes, and thick marl is developed in the deep low part of the Shulu sag (Fig. 1b). The thick marl is more than 200 km2 in area, with burial deep over 3 000 m. According to 3D seismic data, Es31 sub-member can be divided into five third-order sequences and four oil groups, of which the five third-order sequences are SQ1, SQ2, SQ3, SQ4 and SQ5 in ascending order, with SQ1, SQ2, and SQ3 corresponding to oil groups Ⅳ, Ⅲ, and Ⅱ, respectively, SQ4 and SQ5 corresponding to Oil Group Ⅰ (Han et al., 2015; Zhao et al., 2015) (Fig. 1c).
In 2013, the micro-reservoir evaluation of the marl in Well ST1 was conducted, and the samples from Es31 marl in oil groups Ⅲ and Ⅱ were collected. The sample information is shown in Table 1.
No. Depth (m) Formation Lithology Barrels of cores Oil group Sequence System tract ST-1 3 940.5 Es3 Marl 1 Ⅱ SQ3 LST ST-3 3 960.3 Es3 Marl 2 Ⅱ SQ3 LST ST-8 3 970.3 Es3 Marl 4 Ⅱ SQ3 LST ST-10 3 973.2 Es3 Marl 4 Ⅱ SQ3 LST ST-13 3 974.2 Es3 Marl 5 Ⅱ SQ3 LST ST-14 3 976.5 Es3 Marl 5 Ⅱ SQ3 LST ST-16 3 979.1 Es3 Marl 6 Ⅱ SQ3 LST ST-18 3 984.7 Es3 Marl 7 Ⅱ SQ3 LST ST-20 3 989.9 Es3 Marl 8 Ⅱ SQ3 LST ST-24 3 994.5 Es3 Marl 8 Ⅱ SQ3 LST ST-26 4 033.7 Es3 Marl 9 Ⅱ SQ3 LST ST-27 4 036.0 Es3 Marl 9 Ⅱ SQ3 LST ST-30 4 074.0 Es3 Marl 10 Ⅱ SQ3 HST ST-31 4 074.9 Es3 Marl 10 Ⅱ SQ3 HST ST-33 4 077.4 Es3 Marl 10 Ⅱ SQ3 HST ST-34 4 077.5 Es3 Marl 11 Ⅱ SQ3 HST ST-36 4 079.0 Es3 Marl 11 Ⅱ SQ3 HST ST-38 4 082.1 Es3 Marl 11 Ⅱ SQ3 HST ST-41 4 084.5 Es3 Marl 12 Ⅱ SQ3 HST ST-43 4 087.7 Es3 Marl 12 Ⅱ SQ3 HST ST-45 4 204.8 Es3 Marl 13 Ⅲ SQ2 TST ST-47 4 206.3 Es3 Marl 13 Ⅲ SQ2 TST ST-48 4 208.0 Es3 Marl 13 Ⅲ SQ2 TST ST-50 4 211.5 Es3 Marl 13 Ⅲ SQ2 TST
Table 1. Sample information
The analysis and testing conditions are described in detail as follows.
X-ray diffraction (XRD) analysis: The test instrument is a TTR type diffractometer (Rigaku Electric Co., Ltd.). This analysis was carried out with the powder pressed method (300 mesh). The mass percentage of each mineral was analyzed using a software, with the K value of international standard sample as a reference. To analyze the composition of the clay minerals, the samples were extracted with the suspension method, and the prepared orientation plate was air-dried at room temperature and then tested. The samples were saturated with ethylene glycol for 8 h at 60 ℃, and then were heated at 550 ℃ for 2.5 h. The obtained high-temperature plate was then tested, and the mass percentage of each clay mineral was calculated using a software.
The TOC analysis was conducted with the LCO CS230 carbon and sulfur analyzer. Following the National Standard GB/T 19145-2003, the analysis was conducted in four steps. The first step was to weigh the sample powder of about 10 mg with the electronic balance, and put it into the porous ceramic crucible (which was heated for 2 h in the muffle furnace at 1 000 ℃). The second step was to add a sufficient amount of 12.5% hydrochloride (HCL) and heat it on the electric heating plate at 60 ℃ for two hours until the reaction was complete. The third step was to put the crucible into the filter container, and add distilled water every half-hour period and rinse for three days at one-hour intervals. The fourth step was to dry the crucible in a furnace at 60 ℃, and then to test the TOC content after cooling.
Scanning electron microscope (SEM) analysis: The test instrument is Quanta 650F thermal field emission SEM (FEI, USA). The samples were cut into small pieces, and polished into pieces with the size of 1 cm×1 cm×1 cm using P50 coarse and P500 fine sandpapers. The observed surface must be the natural section (and not the polished). The ready-made rock fragments were glued to the sample pile using a conductive adhesive, and then were kept for 1 day until the adhesive was dried completely. The samples were then coated with gold and observed under SEM.
The mercury intrusion analysis was conducted using a Quantachrome automatic porosity analyzer, with incremented pressure values of 10 MPa (up to 210 MPa). The N2 adsorption experiment was carried out on Malvern ASAP 2020, being search-level ultra-performance full-auto gas adsorption system (which can work for a pore size of 1.7–300 nm). Before testing, the samples were dried at 110 ℃ under vacuum for 24 h to remove moisture, and cooled to room temperature (~23 ℃) in a desiccator with a relative humidity of less than 10%. The PSD (Pore size distribution) was determined by the Washburn equation (Washburn, 1921), with a contact angle of 140º and a surface tension of 485 dyne/cm (Gregg and Sing, 1982).
An Ultra XRM-L200 microscope was used for Nano CT. The testing process involved sample preparation, CT scanning, image reconstruction, material phase definition (including de-noising and segmentation), 3D model reconstruction, and calculation of parameters such as porosity. The most important step was definition of material phase. That is to say, to determine different materials according to the gray value image, and then set parameters to segment. An XM Controller module was used for sample scanning. The exposure time was set to 1 s. Conducted imaging with the continuous imaging mode, tuned finely the X-axis and the Y-axis to move the target to the center of the field of view, then tuned finely the Z-axis, and set the Z-axis down 200 μm. Stopped image acquisition and moved the sample out of the field of view. AXM Reconstruction module was used for data processing to reconstruct the 3D structure.
1.1. Geological Setting
1.2. Samples and Experiments
Generally, rock classification is mainly based on grain size
and rock composition. However, with the exploration of shale oil and gas, the organic matter of fine-grained sedimentary rock has become an important evaluation parameter (Lazar et al., 2015; Jiang et al., 2013). The marl is characterized by being self-sourced and self-reservoired, with high TOC. Based on the previous classification of rock types, a four-element method has been proposed to name the Es31 marl, namely siliceous content (quartz+feldspar), calcium content (calcite+dolomite), clay content (clay) and organic matter content (TOC). The parameters are mainly based on the related references (Chen, 2016; Ma et al., 2016; Zhou, 2016; Lazar et al., 2015; Yan, 2015; Jiang et al., 2013). Among them, the TOC evaluation value was modified according to the evaluation criteria for source rock, that is, 0.5%TOC < 1% is classified as low organic matter, 1.0%TOC < 2% as medium organic matter, 2.0%TOC < 4% as high organic matter, and 4.0% TOC as rich organic matter (Fig. 6).
Figure 6. Classification chart of rock types based on mineral composition. CCCC. Claystone; CCCSL. calcareous siliceous claystone; SCC. siliceous claystone; CS. mixed claystone and sandstone; SSC. clayey sandstone; CSSSL. clayey calcareous sandstone; SSSS. sandstone; CCL. calcareous claystone; CL. mixed claystone and carbonate rock; CSL. three-component mixed mudstone; SL. mixed sandstone and carbonate rock; SSL. calcareous sandstone; CLL. clayey carbonate rock; CSLLL. clay siliceous carbonate rock; SLL. siliceous carbonate rock; LLLL. carbonate rock.
In the Es31 marl of the Shahejie Formation, several types of TOC have been identified, including low organic matter in carbonate rocks, mixed claystone and sandstone; medium organic matter in claystone and clay siliceous carbonate rock; high organic matter in carbonate rocks and clay siliceous carbonate rocks; and rich organic matter in carbonate rocks (Fig. 6). Among them, the medium-high carbonate rock is dominant. Only in ST-1, the organic matter is low in mixed claystones and sandstones. In ST-33, the organic matter is rich in carbonate rocks. This multi-level classification can be realized by combination of petrophysical plates and neural network methods, especially the preprocessing of primary components, to achieve effective prediction of rock types using geophysical methods such as well logging data (Ma et al., 2016). As an important brittle mineral, carbonate rock is easy to be fractured. According to the equation of shale brittleness index (BI), Jarvie et al. (2007) also believed that the higher the quartz and calcium content, the greater the brittleness.
In fact, in the United States, most of oil and gas producing shales are not composed of more than 50% clay, but are composed of traditional shales and fine-grained biosilicon or carbonates with organic matter. These rocks are more accurately classified into argillaceous limestones, or china stones or wacke limestones with kerogen. It can be seen from the distribution of the samples that the medium-high organic carbonate rocks are dominant. The statistical data reveal that the proportion of the medium-high organic carbonate rocks is over 80%. The Oil Group Ⅲ is mainly composite by organic high carbonate rocks, while the rest is medium organic carbonate rocks.
Oils in the marl in Well ST1H mainly occurred as thin films on the surfaces of pores and particles (Figs. 7a and 7b). In the calcite dissolved pores, oils mainly occurred as thin films which are connected to each other (Fig. 7c); in the pyrite inter-granular pores, oils were enriched and mainly occurred as thin films. It should be noted that there are small droplets of crude oil in the intra-pores of marl calcite particles, which can only be determined by morphology analysis because the droplet pores are nanometer scale, and below the detection limit of energy spectrum (Figs. 7d, 7e, 7f).
Figure 7. Oil occurrence in marl in Well ST1H. (a) ST-50, 4 211.5 m; (b) an enlarged view of area A in picture (a) shows the thin film of crude oil in the pores at the edge of the particles; (c) an enlarged view of area B in Picture (a) shows the thin film of crude oil in the pores at the edge of the particles; (d) ST-41, 4 084.5 m; (e) an enlarged view of area C in (d) shows the thin film of crude oil in the pores; (f) pyrite of the Sample ST-50, 4 211.5 m; (g) small droplets of crude oil in Sample ST27, depth 4 036.0 m; (h) small droplets of crude oil in Sample ST10, depth 3 973.2 m; (i) an enlarged view of area D in (f) shows the small droplets of crude oil in the inter-crystalline pores in pyrite of the Sample ST-50, 4 211.5 m.
In this study, based on mercury intrusion data, the combination of the pore volume from mercury intrusion (throats > 0.1 μm) and that from nitrogen adsorption (throats < 0.1 μm) is taken as the actual pore volume of the samples (Table 2) for reservoir evaluation. At present, for evaluation of the pore distribution of all pore sizes of tight reservoirs, the pore fusion method has been widely used in the world. That is, the pore of the first cross-point (POC) is determinated by the nitrogen adsorption curve and the mercury intrusion curve. Pores smaller than the POC are based on nitrogen adsorption data, and pores larger than the POC are based on mercury intrusion data (Cao et al., 2016; Clarkson et al., 2013, 2012; Schmitt et al., 2013). However, this method is not based on strict mathematics (Clarkson et al., 2012). Because nano-scale throats (< 0.1 μm) detected by mercury intrusion experiments were only found in a small number of samples such as ST-16, it is believed that most nano-scale throats (< 0.1 μm) are isolated.
Hg intruded volume
N2 PORE volume
ST-1 0.76 0.007 5 2.653 8 1.993 4 0.016 0 5.39 ST-8 1.19 0.004 0 2.669 3 1.072 5 0.001 0 1.33 ST-10 1.34 0.002 8 2.678 3 0.751 4 0.002 4 1.39 ST-16 0.61 0.007 1 2.647 7 1.889 2 0.002 0 2.41 ST-18 2.43 0.004 1 2.748 9 1.134 6 0.001 1 1.43 ST-20 2.19 0.004 0 2.669 8 1.074 2 0.002 3 1.68 ST-24 0.79 0.006 9 2.649 7 1.816 3 0.002 2 2.41 ST-30 1.65 0.002 7 2.678 8 0.732 4 0.001 3 1.07 ST-33 4.63 0.003 1 2.673 7 0.831 1 0.001 8 1.31 ST-36 1.64 0.001 4 2.600 3 0.363 8 0.001 4 0.73 ST-43 1.50 0.007 0 2.648 5 1.857 1 0.001 3 2.20 ST-47 1.80 0.006 8 2.650 0 1.797 0 0.001 2 2.12
Table 2. Samples porosity calculated by mercury intrusion and nitrogen adsorption
Production is an important indicator of sweet spots, which is influenced by many factors, such as fracture characteristics, completion indicators and reservoir quality. The related oil properties include reservoir physical properties, gas-to-oil ratio (GOR), TOC, maturity, pay thickness, and formation pressure (Cui J W et al., 2015). According to the exploration experience of marine shale oil at home and abroad, the key parameters of sweet spots of marine shale oil are as follows (Cui J W et al., 2015): Ro ranges from 1.1% to 1.3%, with thermal evolution within a condensate oil window, TOC is over 3%, pay zone is over 25 m, micro-fractures are well developed, being overpressure, and porosity is over 4%. Although the criteria for continental shale sweet spots in China are not uniform, the key parameters of marine shale oil can be used for reference for selection of continental shale sweet spots.
According to the above criteria, no sweet spots have been found in Well ST1H. In Oil Group Ⅲ at 4 100–4 200 m, only four criteria have been met: TOC is over 3%, pay zone is over 25 m, micro-fractures are well developed, and being overpressure (Fig. 8). There is no obvious correlation between organic carbon and porosity. This may be the reason for the low maturity of organic matter. Despite this, we still think that rocks with high organic carbon content are favorable horizons as long as the maturity conditions are met (Mendhe et al., 2017).
In the samples from Well ST1H, the porosity is not higher than 4% because the rock is just matured (the average HI=500 mg/g·C). Although good oil /gas shows have been observed, commercial oil production has not been obtained from the marl interval. It is worth noting that based on the OSI Index (oil saturation index) proposed by Jarvie et al. (2007), the OSI Index in the Sample ST-16 is greater than 100, which may be caused by the micro-fractures in the samples. The CT image of the Sample ST-16 indicates that the reservoir space is lamellar, which may be lamellar or micro-fractures (Fig. 9a), and the pore characteristics are clearly different from the dumbbell-shaped pore structure of the Ⅲ Oil Group in ST-47 (Fig. 9b). Numerous studies have proved that Type Ⅰ-Ⅱ kerogen organic pores and carbonate pores can increase with the increase of maturity (Wu et al., 2015; Cui et al., 2013). Therefore, we believe that the Ⅲ Oil Group is at a lake invasion system where high-quality source rock is well developed, and the controlling factor for sweet spot intervals is the rock maturity. According to the superposition of the marl thickness, formation pressure, organic matter abundance and maturity maps, the potential sweet spot interval is located in the deep buried sag zone.