J. Cent. South Univ. (2019) 26: 695-710

DOI: https://doi.org/10.1007/s11771-019-4040-8

Quantitative characterization of sealing integrity by caprock of Paleocene Artashi Formation gypsolyte rock in Kashi Sag of Tarim Basin, NW China

LIAO Xiao(廖晓)¹, WANG Zhen-liang(王震亮)², FAN Chang-yu(范昌育)²,YU Chang-qing(于常青)³, YU Zhu-yu(余朱宇)²

1. College of Science, Chang’an University, Xi’an 710064, China;

2. State Key Laboratory of Continental Dynamics, Department of Geology, Northwest University,Xi’an 710069, China;

3. The Institute of Geology, Chinese Academy of Geological Sciences, Beijing 100037, China

Central South University Press and Springer-Verlag GmbH Germany, part of Springer Nature 2019

Abstract: Maintaining caprock integrity is prerequisite for hydrocarbon accumulation. And gypsolyte caprock integrity is mainly affected by fracturing. Composition, damage behavior and mechanical strength of Paleocene Artashi Formation gypsolyte rock that seals significant petroleum in the Kashi Sag of Tarim Basin had been revealed via X-ray diffraction and triaxial compression test. The results indicate the Artashi Formation can be lithologically divided into the lower and upper lithologic members. The lower member comprises gypsum as the dominant mineral, and the cohesion and friction coefficient are 8 MPa and 0.315, respectively. Similarly, the upper lithologic member consists mainly of anhydrite at the cohesion and coefficient of internal friction values of 18 MPa and 0.296. Given that the failure criterion and brittle-ductile transition factors during burial, the sealing integrity of Artashi Formation can be quantized for seven different stages. The reservoirs at the bottom of Artashi Formation caprock buried from 2285 m to 3301 m are expected to be the most favorable exploration target in the Kashi Sag.

Key words: brittle-ductile transition; failure criterion; sealing integrity; gypsolyte caprock; Artashi Formation; Kashi Sag

Cite this article as: LIAO Xiao, WANG Zhen-liang, FAN Chang-yu, YU Chang-qing, YU Zhu-yu. Quantitative characterization of sealing integrity by caprock of Paleocene Artashi Formation gypsolyte rock in Kashi Sag of Tarim Basin, NW China [J]. Journal of Central South University, 2019, 26(3): 695–710. DOI: https://doi.org/10.1007/s11771- 019-4040-8.

1 Introduction

Argillaceous rock and gypsolyte-saline rock are the major seals in hydrocarbon accumulation around the world. Argillaceous caprock is widely distributed and accounts for more than 80% of petroleum caprock area. However, petroleum reserves sealed by argillaceous rock accounts for only 22% of the total petroleum reserves worldwide [1]. In contrast, caprock consisting of gypsolyte-saline rock occupys just 8% of all caprock area, but seals 55% of petroleum reserves worldwide [1]. The Mesozoic-Cenozoic Kazhdumi- Asmari-Gachsaran play is an important source- reservoir-caprock play of Zagros Basin in the Middle East [2, 3]. Asphaltic mudstone of the Middle Cretaceous Kazhdumi Formation serves as the hydrocarbon source rock, fracture-rich limestone of the Oligocene-Miocene Asmari Formation acts as reservoir, and gypsolyte-saline rock of the Miocene Gachsaran Formation functions as the caprock. The petroleum and gas recoverable reserves of this play are 148.66×10⁸ t and 5.6×10¹² m³ [2]. Compared with argillaceous rock, the smaller pore-throat, lower porosity- permeability, and higher capillary entry pressure make gypsolyte-saline rock more efficient as hydrocarbon caprock.

However, there have been studies indicating that gypsolyte-saline rock can also act as conduit for hydrocarbon migration in recent years. Amu Darya Basin, which covers Turkmenistan, Uzbekistan, Afghanistan, and Iran, is a large basin with abundant hydrocarbon resource in central Asia. The gypsolyte-saline rock of Upper Jurassic is an important regional caprock [4]. The studies by NIE et al [5] and ZHANG et al [6] claimed that the Upper Jurassic gypsolyte-saline rock located in the southern Zarzhu Uplift was thin and had experienced relatively low extrusion stress. Controlled by early faults, there was strike-slip basal faulting that penetrated through the gypsolyte-saline rock. And the integrity of caprock was destroyed. As a result, it acted as the migration conduit for hydrocarbon generated from lower-middle Jurassic coal source rock. LV et al [7] and ZHUO et al [8] documented that hydrocarbon generated by the Triassic and Jurassic source rock in the Kuqa Depression of Tarim Basin migrated upward along fractures and faults cutting through gypsolyte-saline rock of Tertiary.

Gypsolyte-saline rock has such low matrix permeability, and can prevent hydrocarbon from migrating through their matrix. To become conduits for secondary hydrocarbon migration, the permeability must be enhanced. The only way for hydrocarbon leakage through seals and migration is petroleum movement through fractures generated after fracturing [9–17]. The main uncertainties in determining the possibility of hydrocarbon leakage along fractures in caprocks are the timing of fracture formation and the fracture network geometry [11]. The timing of fracture formation is determined by failure criterion. Fractures are formed when the pressure exceeds peak strength of the rock. Geometrical properties of fractures which include persistence, length, orientation and connectivity are mainly affected by brittleness and ductility of rocks subjected to the same loading [11, 18]. Rocks are prone to accommodate deformation in a brittle manner at low pressure, typically at shallow depth in the crust. Brittle deformation is characterized by dilative response. When the lithostatic pressure reaches at peak strength, there is a suddenly localized failure followed by strain softening down to residual strength [19, 20]. Meanwhile, brittle response is accompanied by microcrack opening and coalescence forming distinctly extensional and crossed fractures [11]. Once formed, extensional and crossed fractures in brittle deformation will dilate at low effective normal stresses and have increased permeability with increasing brittle deformation [18]. There is a transition from brittle to ductile behavior for rocks with increasing depth, i.e., with increasing lithostatic pressure. Ductile behavior is characterized by overall contractive response and a homogeneously distributed deformation throughout the rock. Above all, ductile deformation is defined as the ability to undergo large strains without fracturing of the rock [21]. And fractures generated in ductile deformation will contract at high effective stresses and thereby have reduced permeability with increasing ductile deformation [18, 20, 22]. Therefore, the sealing integrity of caprock is determined by fracturing, which includes failure and brittleness or ductility factors.

The surface oil sand, liquid oil seepage, gas seepage, and asphalt imply abundant petroleum resource in the Kashi Sag of Tarim Basin. And the sealing condition is shown not very well. Based on abundant oil and gas sources, Kashi Sag is accompanied by good reservoir conditions, whereas caprock quality becomes the vital factor for hydrocarbon accumulation [23, 24]. Petroleum exploration was conducted in 1952 in the Kashi Sag. And the drilling of Ake1 well resulted in a high yield of industrial natural gas with daily production of 1.19×10⁵ m³ in 2001. Eventually, the Akemomu gas field had been discovered. This event marked the first time that an iconic breakthrough of petroleum exploration was achieved in the Kashi Sag. Meanwhile, a new field for natural gas exploration was opened up in this area and even across the whole Tarim Basin. Akemomu gas field is also the only known natural gas field up to now in the Kashi Sag. Gypsolyte rock of Paleocene Artashi Formation (E₁a) acts as the crucial seal role of Akemomu gas field. Previous studies [23, 25–28] of this regional caprock focused mainly on the tectonic background and sedimentary characteristic aspects. Unfortunately, there are still unclearly scientific problems on the evolutive process of this caprock functioning as the seal role during the formation and even subsequent preservation of Akemomu gas field.

In this study, compositions of the Paleocene Artashi Formation gypsolyte rock in the Kashi Sag of Tarim Basin were determined separately by X-ray diffraction (XRD) test on the lower and upper lithologic samples. And then the failure modes, failure envelopes, and stress-strain characteristics of the lower white gypsum rock and upper offwhite anhydrite rock were investigated using triaxial compression test under various confining pressures according to the lithologic difference. Eventually, a evolutional model of Artashi Formation sealing integrity is proposed with the failure criterion and brittle-ductile transition factors considered comprehensively. And this model was demonstrated reasonably and objectively by analyzing the sealing process of typical Ake1 gas pool. Meanwhile, the favorable exploration target is pointed out for later hydrocarbon exploration in this area.

2 Geological setting

The Tarim Basin, located in northwestern China (Figure 1(a)), is the largest inland basin in the world with an area of approximately 56×10⁴ km². The Kashi Sag is the westernmost structural unit of Tarim Basin (Figure 1(b)). With an area of 2.4×10⁴ km², Kashi Sag is located to the south of the Southern Tianshan, north of the Western Kunlun Mountains, west of the Maigaiti Slope and Yecheng Sag (Figure 1(c)). Kashi Sag is consisted of the lower Precambrian metamorphic crystalline basement and the upper Phanerozoic caprock and had experienced three main structural evolution stages since Sinian [29–33]. During the Nanhuan- Early Paleozoic, it was a passive basin on the continental margin owing to the expansion of the South Tianshan Ocean and the southern of Kudi Ocean. In the late Paleozoic, due to the closed event of the South Tianshan Ocean, Kashi Sag gradually developed into a foreland basin. In the Mesozoic-Cenozoic period, the Kashi Sag entered a stage of inland adjustment with the adjacent Southern Tianshan and Western Kunlun orogenic zones formed.

Due to the stretch of Kepeitetage region in southern central Asia, Kashi Sag experienced eastward transgression from Late Cretaceous [31, 34]. And the corresponding distribution of sedimentary formations gradually expanded [28, 35]. During early Paleocene, Kashi Sag was generally in closed bay and lagoon environment [36] dominated by subtropical-tropical arid climate [37]. An regional distribution gypsolyte, namely, Artashi Formation, was generated on the basis of this. Present outcrops of Artashi Formation are limited and scattered in the piedmont fold belts of the Southern Tianshan and Western Kunlun (Figure 1(c)).

3 Samples and experiments

3.1 Samples preparation

In this study, a representatively geological section in northwestern Wuqia region was investigated (Figure 1(c)). The outcrop of Artashi Formation (E₁a) in this section measures 110 m and shows the distinctly sedimentary difference of the lower and upper lithologic members. The lower lithologic member is 63.7 m, which is characterized by thickly bedded white gypsolyte with thin layers of dolomite (Figures 2(a)–(c)). Meanwhile, it lies conformably on the Upper Cretaceous Tuyiluoke Formation (K₂t), which is composed primarily of sandstone with thin layer pebbly sandstone (Figure 2(a)). Similarly, the upper lithologic member is 46.3 m, which consists of thickly bedded offwhite gypsolyte with thin layers of mudstone (Figures 2(a), (d), (e)). Furthermore, the upper contact with the Paleocene-Eocene Qimugen Formation (E_1-2q) appears to be conformable (Figure 2(a)). And the Qimugen Formation is composed primarily of siltstone, mudstone and sandstone (Figure 2(a)). The representative samples of Artashi Formation were collected on the basis of the lower and upper lithologic members. The detailed sample locations of the measured geological section were shown in Figure 2(a). The samples of the lower white gypsolyte and upper offwhite gypsolyte were designated from w16 to w9 and w8 to w1, respectively. And the X-ray diffraction (XRD) and triaxial compression test were performed on these samples.

Figure 1 Distribution of structural location(a, b) and geological section position(c) in Kashi Sag of Tarim Basin in northwestern China ((a) is from GS(2016)2885)

In this study, sixteen samples based on the lower and upper lithologic difference of Artashi Formation were selected for the XRD experiment. And approximate 50 g of each fresh gypsolyte samples were crushed and sieved to less than 74 μm grain size. In addition, the fresh cylindrical samples with 50 mm in length and 25 mm in diameter were produced using drill and cutting machines. In order to reduce the difference among different samples, cylindrical samples of the same gypsolyte were produced as many as possible, and were designated, for example, from w12-1, w12-2, to w12-3. And then the cylindrical samples were placed in a drying oven for three hours and wrapped by heat shrink tube prior to the triaxial compression test.

3.2 Experimental conditions and procedures

The XRD experiment was performed at the Northwest China Supervision and Inspection Center of Mineral Resources (Center of Experimental Test, Xi’an Geological Survey Center), Ministry of Land and Resources of China using a D/max-2500 X-ray diffractometer with a graphite monochromator filter manufactured by Rigaku of Japan. A copper X-ray tube was used at 40 kV and 200 mA. The experimental conditions were as follows: a constant room temperature of 25 °C, and humidity of 49%. The components of samples were determined via standard powder diffraction data provided by the Joint Committee on Powder Diffraction Standards.

Figure 2 Measured geological section and sampling location distribution of Paleocene Artashi Formation in Wuqia region:

And then the quantitative analysis of mineral composition was conducted on the basis of the basic intensity correlation method.

The triaxial compression test was performed at the State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University of China using RTR-1000 triaxial rock mechanics servo testing system manufactured by GCTS of United States. The conditions provided by this testing system are as follows: the maximum axial pressure of 1000 kN, the maximum confining pressure of 140 MPa, the maximum pore pressure of 140 MPa, a constant dynamic frequency of 10 Hz, the highest temperature of 150 °C, an efficient pressure precision of less than 0.01 MPa, liquid volume within 0.01 cm³, and the deformation within 0.001 mm. The triaxial compression tests were performed at confining pressures ranging from 5 to 50 MPa (5, 10, 20, 30, 40 and 50 MPa), at 20 °C. In this study, we adopted the convention that compressive stresses and compressive axial strains were measured positive. The principal compressive stresses recorded in our tests were accordingly denoted with σ₁>σ₂=σ₃=Pc, where Pc is the confining pressure. The detailed experimental conditions and results are shown in Table 1.

Table 1 Triaxial compression test conditions and results of Artashi Formation gypsolyte rock in Wuqia region

4 Results

4.1 Mineral compositions

The quantitative analysis results of Artashi Formation gypsolyte mineral component in Wuqia region are shown in Table 2. Bulk composition data indicates that the lithology and mineralogy of Artashi Formation gypsolyte are relatively homogeneous. These samples comprise gypsum or anhydrite as the co-dominant mineral species with minor talcum and other minerals differentiated difficultly. However, the detailed mineral compositions of the lower and upper lithologic members are obviously different. The results indicate that the lower white gypsolyte comprise gypsum as the co-dominant mineral species (above 98% each) with minor talcum (ranging from 0 to 0.7%, 0.33% in average), anhydrite and other minerals. In contrast, the upper offwhite gypsolyte consists mostly of anhydrite (about 99% each) with minor gypsum (up to 1.7%, 0.4% in average), talcum (averaged at 0.11%) and other minerals. In summary, the white gypsolyte is mainly composed of gypsum, whereas the absolutely dominant mineral species in the offwhite gypsolyte is anhydrite.

4.2 Failure modes

The failure characteristics of white and offwhite gypsolyte under different confining pressures at room temperature are shown in Figure 3. When the confining pressure was 5 MPa, the white and offwhite gypsolyte displayed split failure on the fracture surfaces with a few strong extensional fractures parallelled to the direction of maximum principal stress (Figures 3(a), (d)). And the number of fractures in offwhite gypsolyte samples was more than white gypsolyte samples. With the confining pressure increasing, both the white and offwhite gypsolyte showed crossed fractures at Pc=20 MPa (Figures 3(b), (e)). The density of crossed fractures in offwhite anhydrite samples was larger than white gypsum samples. Moreover, the crossed fractures of offwhite anhydrite samples were mostly of small X-type with local destruction, whereas white gypsum samples were a larger scale deformation. At 40 MPa confining pressure, the white gypsum and offwhite anhydrite samples were both dominated by shear failure with shear fractures (Figures 3(c), (f)). The connectivity of those shear fractures in white gypsum samples was better than offwhite anhydrite with discontinuous fractures formed.

Table 2 X-ray diffraction results of mineral composition of Artashi Formation gypsolyte in Wuqia region

Figure 3 Sketch of failure modes for white gypsum and offwhite anhydrite samples of Artashi Formation under different confining pressures:

4.3 Failure envelopes

The failure critical conditions can be plotted in the form of Mohr diagrams that can represent the state of stress within the sample at the point of failure. The failure envelopes are determined from the stress conditions (σ₁, σ₃) at the time of failure. The Mohr circles and associated failure envelopes for white gypsum and offwhite anhydrite samples of Artashi Formation under different confining pressures are shown in Figure 4. Figure 4 shows the failure envelopes for white gypsum (1) and offwhite anhydrite (2), defined by the equations:

τ=8+0.315σ                              (1)

τ=18+0.296σ                             (2)

Figure 4 Mohr circles and associated failure envelopes for white gypsum rock (a) and offwhite anhydrite rock (b) samples of Artashi Formation

The failure envelopes appear to be linear with the stress changing, even though the peak strength appeared to be rising with the increasement of confining pressure. The cohesion and friction coefficient of white gypsum sample are 8 MPa and 0.315, whereas the corresponding values of offwhite anhydrite sample are 18 MPa and 0.296. The cohesion of offwhite anhydrite rock is higher than that of white gypsum rock, while the friction coefficient is lower in comparison to white gypsum rock. It indicates that the offwhite anhydrite rock is more steady than white gypsum rock under the same given load.

4.4 Stress-strain characteristics

The differential stress versus axial strain curves obtained of the white gypsum rock and offwhite anhydrite rock samples are shown in Figure 5. When the gypsolyte is initially subjected to an external load, the stress–strain curve exhibits linear relationship, and the gypsolyte is at the stage of elastic deformation. It means that the gypsolyte can rebound to its original shape when the external stress is removed. At low confining pressures(5 MPa, 10 MPa), when the external load reached at peak strength, failure occurred by abrupt brittle fracturing. Hence, the stress decreased rapidly and the differential stress versus axial strain curves were very sharp at the failure points followed by strain softening behavior. At the high confining pressure circumstances (20 MPa, 30 MPa), the stress–strain curves showed gradual strain hardening beyond the peak stress, falling slowly to a steady axial strain condition. When the confining pressure was at 40 MPa and 50 MPa, the stress–strain curves showed a broad yield at failure points and the peak stress stages were followed by strain hardening behavior. With the confining pressure increasing, gypsolyte had transformed from brittle to ductile, strain softening to strain hardening behavior. Furthermore, the peak strengths of offwhite anhydrite rock samples under different confining pressures were higher than white gypsum rock. This may be because the cohesion of offwhite anhydrite rock is higher than white gypsum rock, whereas the friction coefficient is lower. The offwhite anhydrite rock is more stable than white gypsum rock under the same condition. As a result, the gypsum rock is more prone to failure.

Figure 5 Relationship between stress and strain of white gypsum rock (a) and offwhite anhydrite rock (b) samples of Artashi Formation under different confining pressures (differential stress=σ₁–σ₃)

5 Quantitative characterization of sealing integrity

5.1 Separately quantitative characterization of lower and upper Artashi Formation sealing integrity

The gypsolyte of Artashi Formation will fracture during burial, due to the effect of the overlying load pressure and formation pressure. Failure envelopes on the plane of effective normal stress and shear stress for both the lower white gypsum and upper offwhite anhydrite of Artashi Formation can also be plotted in terms of the confining pressure (σ₃) and differential stress (σ₁–σ₃):

σ₁–σ₃=0.86σ₃+21.8                        (3)

σ₁–σ₃=0.79σ₃+48.2                        (4)

where Eq. (3) is for the lower white gypsum rock; Eq. (4) is for the upper offwhite anhydrite rock.

The curve of the overlying load pressure is shown as:

σ₁–σ₃=ρgh                               (5)

where ρ is the average density of overlying stratum (kg/m³), g is the gravitational acceleration (9.8 m/s²), and h is the burial depth (m).

The average density of Cenozoic stratum is 2.32 g/cm³ in the Kashi Sag [38], and the formation pressure gradient is 1.04 MPa/100 m [39]. Based on those parameters, the curve of overlying load pressure is deduced as:

σ₁–σ₃=2.2σ₃             (6)

With the overlying load pressure curve and failure envelopes considered comprehensively, the critical confining pressure for failure of the lower Artashi Formation gypsum rock in this region is 16 MPa, namely, a burial depth of 1553 m (Figure 6(a)); while the confining pressure for failure of the upper Artashi Formation anhydrite rock is 34 MPa, which corresponds to burial depth 3301 m (Figure 6(b)).

Transition from brittle to brittle-ductile is formed when the Mohr-Coulomb failure envelope of rock intersects with the Byerlee friction sliding curve [40]. Specifically, the Mohr-Coulomb failure envelope is the quadratic fitting curve of peak strength in the stress-strain curve, and the condition is met by Byerlee friction sliding curve at the coordinate plane of σ₁ and σ₃ [41, 42]:

σ₁–σ₃≈3.7σ₃ (σ₃<100 MPa)                  (7)

With the confining pressure increasing, gypsolyte transforms gradually brittle-ductile to ductile deformation stage. For this change, transition will occur when the Goetze rule is met by the rock [43, 44]:

σ₁–σ₃≈σ₃                                (8)

Thus, the critical confining pressure and corresponding burial depth of the transition from brittle to brittle-ductile and even ductile deformation for Artashi Formation gypsum and anhydrite can be obtained. The results indicate that the critical confining pressure for the transition from brittle to brittle-ductile deformation of the lower gypsum rock is 11 MPa, which is equivalent to burial depth 1068 m. Moreover, the critical confining pressure for the transition from brittle-ductile to ductile deformation is 94 MPa, namely, a burial depth of 9126 m (Figure 6(a)). Similarly, the critical confining pressure for transition of the upper anhydrite rock from brittle to brittle-ductile deformation is 29 MPa (a burial depth of 2816 m). Besides, 118 MPa which corresponds to burial depth 11456 m is the key confining pressure for the transition from brittle- ductile to ductile deformation of the upper Artashi Formation anhydrite (Figure 6(b)).

Figure 6 Failure and brittle-ductile transition conditions of white gypsum rock (a) and offwhite anhydrite rock (b) of Artashi Formation in the Kashi Sag

5.2 Evolutional model of Artashi Formation sealing integrity

Artashi Formation can be clearly subdivided into the lower and upper lithologic members. The lower member comprises mainly gypsum, while the upper member is composed primarily of anhydrite. The evolution of Artashi Formation sealing integrity is divided into seven different stages with the failure criterion and brittle-ductile transition factors considered comprehensively (Figure 7).

Stage I is divided from deposition to burial depth 1068 m. Both the lower gypsum and upper anhydrite are at the stage of brittle deformation, and local splitting failure is formed with scatteredly extensional microfractures and crossed-fractures. And then the lower gypsum enters the stage of brittle-ductile deformation generating discontinuous shear fractures, namely, stage II (burial depth from 1068 m to 1553 m). Stage III ranges from burial depth of 1553 m to 2816 m when the failure occurs in the lower gypsum. This stage is characterized by brittle-ductile deformation after failure of gypsum, and the shear fractures generated previously connect with each other. As a result, the sealing integrity of the lower gypsum is damaged. With the burial depth increasing, the upper anhydrite gradually enters the stage of brittle-ductile deformation generating discontinuous shear fractures, namely, stage IV (burial depth from 2816 m to 3301 m). Subsequently, when the burial depth ranges from 3301 m to 9126 m (stage V), the upper anhydrite experiences failure with the fractures connected with each other. In this case, both the lower gypsum and upper anhydrite lose their sealing integrity. And the sealing capability of Artashi Formation gypsolyte is the worst. Stage VI spans burial depth from 9126 m to 11456 m when the lower gypsum enters the period of ductile deformation, which is an overall contractile response to the overlying load pressure. The best sealing integrity is shown in the lower gypsum. Eventually, when the lower gypsum and upper anhydrite undergo ductile deformation buried in excess of 11456 m (stage VII), the sealing capability of Artashi Formation gypsolyte is the best of the whole evolution process.

Figure 7 A evolutional model of Artashi Formation sealing integrity in the Kashi Sag

6 Sealing process of Ake1 gas pool

The Lower Carboniferous mudstone is the major source rock of Ake1 gas pool, Cretaceous sandstone acts as the reservoir, and the overlying gypsolyte of Paleocene Artashi Formation serves as the caprock [25, 45, 46]. The present burial depth of the Artashi Formation gypsolyte in Ake1 gas pool ranges from 3120 m to 3180 m (Figure 8).

WANG et al [25] reported that the Ake1 gas pool began to hydrocarbon accumulation at 18 Ma by using inclusion thermometry and K-Ar dating of authigenic illite. The burial depth of Artashi Formation ranged from 1800 m to 2300 m at the crucial period of hydrocarbon accumulation (18 Ma) based on the curve of burial history (Figure 9).And the Artashi Formation gypsolyte was at stage III of the whole sealing evolution according to the established model. The lower gypsum was at the stage of brittle-ductile deformation after failure and had lost sealing integrity. Fortunately, the upper anhydrite was at the stage of brittle deformation without failure and played a key role of sealing during the hydrocarbon accumulation. Thereafter, Artashi Formation entered a stage of rapid burial (Figure 9). The present burial depth of Artashi Formation in Ake1 gas pool ranges from 3120 m to 3180 m, which corresponds to stage IV. During this process, the lower gypsum was still at the stage of brittle-ductile deformation after failure. The connectivity and persistence of fractures generated in this process were strong (Figure 10(a)). Hence, the lower gypsum lost sealing integrity. Meanwhile, the upper anhydrite gradually transformed into the stage of brittle-ductile deformation. And the most important of all, failure had not occurred in the upper anhydrite rock so far. The fractures generated were discontinuous microfractures (Figure 10(b)). Moreover, the connectivity and persistence of those microfractures were weak. Eventually, the sealing integrity of the upper anhydrite was maintained. Therefore, the upper anhydrite of Artashi Formation continually played a critical sealing role in the period of Ake1 gas pool hydrocarbon accumulation and subsequent preservation.

Figure 8 Gas pool section of Ake1 well in Kashi Sag (modified after Ref. [23])

Figure 9 Thermal history of Lower Carboniferous source rock of Ake1 gas pool (modified after Ref. [25])

Figure 10 Typical fracture characteristics of Artashi Formation gypsolyte:

The present burial depth of Artashi Formation gypsolyte ranges from 2285 m to 6650 m in the Kashi Sag [28]. Combined with the evolutional model of Artashi Formation sealing integrity, we can conclude that the effective sealing depth of Artashi Formation is 2285 m to 3301 m. The upper anhydrite plays the key role of sealing all the way, whereas the lower lithologic member of gypsum has lost sealing integrity. Therefore, the reservoir below Artashi Formation gypsolyte buried from 2285 m to 3301 m will be the next step of favorable exploration target in the Kashi Sag.

7 Conclusions

1) The gypsolyte of Paleocene Artashi Formation can be clearly divided into the lower and upper members based on the lithologic difference in the Kashi Sag of Tarim Basin. The lower lithologic member is composed mainly of white gypsum whose cohesion and coefficient of internal friction are 8 MPa and 0.315, respectively. Whereas the upper lithologic member comprises primarily offwhite anhydrite with the cohesion and coefficient of internal friction of 18 MPa and 0.296. In general, the offwhite anhydrite rock is more steady than white gypsum rock under the same load.

2) With the failure criterion and brittle-ductile transition factors considered comprehensively during the burial, the critical confining pressure for failure of the lower gypsum is 16 MPa in this area, and the important confining pressures for the transition from brittle to brittle-ductile and brittle-ductile to ductile deformation are 11 MPa and 94 MPa, respectively. Similarly, the confining pressure for failure of the upper anhydrite is 34 MPa, and the key confining pressures for transition from brittle to brittle-ductile and even brittle-ductile to ductile deformation are 29 MPa and 118 MPa.

3) Based on the differently geomechanical properties of the lower gypsum and upper anhydrite, the evolution of Artashi Formation gypsolyte sealing integrity can be divided into seven stages, which are designated from stage I to stage VII. In particular, both the lower gypsum and upper anhydrite are in ductile deformation at stage VII, when the sealing capability is the best. At stage V, i.e., the period of brittle-ductile deformation after failure, both the lower gypsum and upper anhydrite lose their integrity generating the worst sealing capability.

References

[1] JIN Zhi-jun, LONG Sheng-xiang, ZHOU Yan, WO Yu-jin, XIAO Kai-hua, YANG Zhi-qiang, YIN Jin-yin. A study on the distribution of saline-deposit in southern China [J]. Oil & Gas Geology, 2006, 27(5): 571–583. DOI: 10.3321/j.issn: 0253-9985.2006.05.001. (in Chinese)

[2] BAHROUDI A, KOYI H A. Tectono-sedimentary framework of the Gachsaran Formation in the Zagros foreland basin [J]. Marine and Petroleum Geology, 2004, 21(10): 1295–1310. DOI: 10.1016/j.marpetgeo.2004.09.001.

[3] VAZIRI-MOGHADDAM H, SEYRAFIAN A, TAHERI A, MOTIEI H. Oligocene-Miocene ramp system (Asmari Formation) in the NW of the Zagros basin, Iran: Microfacies, paleoenvironment and depositional sequence [J]. Revista Mexicana de Ciencias Geológicas, 2010, 27(1): 56–71. DOI: 10.1016/j.quascirev.2010.01.005.

[4] BROOKFIELD M E, HASHMAT A. The geology and petroleum potential of the North Afghan platform and adjacent areas (northern Afghanistan, with parts of southern Turkmenistan, Uzbekistan and Tajikistan) [J]. Earth-Science Reviews, 2001, 55(1): 41–71. DOI: 10.1016/S0012- 8252(01)00036-8.

[5] NIE Ming-long, WU Lei, SUN Lin, GAO An-hu. Salt-related fault characteristics and their petroleum geological significance in Zarzhu terrace and its adjacent areas, the Amu Darya Basin [J]. Oil & Gas Geology, 2013, 34(6): 803–808. DOI: 10.11743/ogg20130613. (in Chinese)

[6] ZHANG Chang-bao, LUO Dong-kun, WEI Chun-guang. Controlling factors of natural gas accumulation in the Amu Darya Basin, Central Asia [J]. Oil & Gas Geology, 2015, 36(5): 766–773. DOI: 10.11743/ogg20150507. (in Chinese)

[7] LV Xiu-xiang, JIN Zhi-jun, ZHOU Xin-yuan, PI Xue-jun. Oil and gas accumulation related to evaporite rocks in Kuqa depression of Tarim basin [J]. Petroleum Exploration and Development, 2000, 27(4): 20–21. DOI: 10.3321/j.issn:1000- 0747.2000.04.004. (in Chinese)

[8] ZHUO Qin-gong, MENG Fan-wei, SONG Yan, YANG Hai-jun, LI Yong, NI Pei. Hydrocarbon migration through salt: evidence from Kelasu tectonic zone of Kuqa foreland basin in China [J]. Carbonates and Evaporites, 2014, 29(3): 291–297. DOI: 10.1007/s13146-013-0177-y.

[9] DEWHURST D N, JONES R M, RAVEN M D. Microstructural and petrophysical characterization of Muderong Shale: application to top seal risking [J]. Petroleum Geoscience, 2002, 8(4): 371–383. DOI: 10.1144/petgeo.8.4.371.

[10] DEWHURST D N, HENNIG A L. Geomechanical properties related to top seal leakage in the Carnarvon Basin, Northwest Shelf, Australia [J]. Petroleum Geoscience, 2003, 9(3): 255–263. DOI: 10.1144/1354-079302-557.

[11] NYGRD R, GUTIERREZ M, BRATLI R K, HEG K. Brittle-ductile transition, shear failure and leakage in shales and mudrocks [J]. Marine and Petroleum Geology, 2006, 23(2): 201–212. DOI: 10.1016/j.marpetgeo.2005.10.001.

[12] PETRIE E S, EVANS J P, BAUER S J. Failure of cap-rock seals as determined from mechanical stratigraphy, stress history, and tensile-failure analysis of exhumed analogs [J]. AAPG Bulletin, 2014, 98(11): 2365–2389. DOI: 10.1306/ 06171413126.

[13] HAN Liang, ZHOU Yong-sheng, HE Chang-rong, LI Hai-bing. Sublithostatic pore fluid pressure in the brittle-ductile transition zone of Mesozoic Yingxiu-Beichuan fault and its implication for the 2008 Mw 7.9 Wenchuan earthquake [J]. Journal of Asian Earth Sciences, 2016, 117: 107–118. DOI: 10.1016/j.jseaes.2015.12.009.

[14] YAZDI A K, SMYTH H D. Implementation of design of experiments approach for the micronization of a drug with a high brittle-ductile transition particle diameter [J]. Drug Development Communications, 2016, 43(3): 364–371. DOI: 10.1080/03639045.2016.1253727.

[15] NEVITT J M, WARREN J M, POLLARD D D. Testing constitutive equations for brittle-ductile deformation associated with faulting in granitic rock [J]. Journal of Geophysical Research: Solid Earth, 2017, 122(8): 6269–6293. DOI: 10.1002/2017JB014000.

[16] YUAN Yu-song, JIN Zhi-jun, ZHOU Yan, LIU Jun-xin, LI Shuang-jian, LIU Quan-you. Burial depth interval of the shale brittle-ductile transition zone and its implications in shale gas exploration and production [J]. Petroleum Science, 2017(4): 1–11. DOI: 10.1007/s12182-017-0189-7.

[17] HU Kun, ZHU Qi-zhi, CHEN Liang, SHAO Jian-fu, LIU Jian. A micromechanics-based elastoplastic damage model for rocks with a brittle-ductile transition in mechanical response [J]. Rock Mechanics and Rock Engineering, 2018, 51(6): 1729–1737. DOI: 10.1007/s00603-018-1427-z.

[18] PAPESCHI S, MUSUMECI G, MAZZARINI F. Evolution of shear zones through the brittle-ductile transition: The Calamita Schists (Elba Island, Italy) [J]. Journal of Structural Geology, 2018, 113: 100–114. DOI: 10.1016/j.jsg.2018.05. 023.

[19] WONG T, BAUD P. The brittle-ductile transition in porous rock: A review [J]. Journal of Structural Geology, 2012, 44: 25–53. DOI: 10.1016/j.jsg.2012.07.010.

[20] ZHANG Lei, HE Chang-rong. Frictional properties of phyllosilicate-rich mylonite and conditions for the brittle-ductile transition [J]. Journal of Geophysical Research: Solid Earth, 2016, 121(4): 3017–3047. DOI: 10.1002/ 2015JB012489.

[21] WANG Ming-yang, FAN Peng-xian, QIAN Qi-hu, DENG Hong-jian. Elastoplastic model for discontinuous shear deformation of deep rock mass [J]. Journal of Central South University of Technology, 2011, 18(3): 866–873. DOI: 10.1007/s11771-011-0775-6.

[22] DEHANDSCHUTTER B, VANDYCKE S, SINTUBIN M, VANDENBERGHE N, WOUTERS L. Brittle fractures and ductile shear bands in argillaceous sediments: inferences from Oligocene Boom Clay (Belgium) [J]. Journal of Structural Geology, 2005, 27(6): 1095–1112. DOI: 10.1016/j.jsg.2004.08.014.

[23] HE Deng-fa, LI De-sheng, HE Jin-you, WU Xiao-zhi. Comparison in petroleum geology between Kuqa depression and Southwest depression in Tarim Basin and its exploration significance [J]. Acta Petrolei Sinica, 2013, 34(2): 201–218. DOI: 10.7623/syxb201302001. (in Chinese)

[24] LI Shi-zhen, KANG Zhi-hong, QIU Hai-jun, MENG Miao-miao, FENG Zhi-gang, LI Shi-Chao. Hydrocarbon accumulation modes of the southwest depression in Tarim Basin [J]. Geology in China, 2014, 41(2): 387–398. DOI: 10.3969/j.issn.1000-3657.2014.02.007. (in Chinese)

[25] WANG Zhao-ming, ZHAO Meng-jun, ZHANG Shui-chang, SONG Yan, XIAO Zhong-yao, WANG Qing-hua, QIN Sheng-fei. A preliminary study on formation of Akemo gas field in the Kashi Sag, Tarim Basin [J]. Chinese Journal of Geology, 2005, 40(2): 237–247. DOI: 10.1016/j.molcatb. 2005.02.001. (in Chinese)

[26] DU Jin-hu, WANG Zhao-ming, LEI Gang-lin, HU Jian-feng. A discovery in well Kedong-1 and its exploration significance [J]. China Petroleum Exploration, 2011, 16(2): 1–5. DOI: 10.3969/j.issn.1672-7703.2011.02.001. (in Chinese)

[27] XING Hou-song, LI Jun, SUN Hai-yun, WANG Hai, YANG Qing, SHAO Li-yan, YANG Dong, YANG Shen. Differences of hydrocarbon reservoir forming between southwestern Tarim basin and Kuche mountain front [J]. Natural Gas Geoscience, 2012, 23(1): 36–45. (in Chinese)

[28] ZHANG Liang, HAN Er-bin, ZHU Li-chun, ZENG Chang-min, FAN Qiu-hai, WU Kun, CAO Yang-tong, JIAO Peng-cheng. Characteristics of evaporites sedimentary cycles and its controlling factors of Paleocene Aertashi Formation in the southwestern Tarim depression [J]. Acta Geologica Sinica, 2015, 89(11): 2161–2170. (in Chinese)

[29] LUO Jin-hai, ZHOU Xin-yuan, QIU Bin, YANG Zhi-lin, YIN Hong, SHANG Xin-lu. Structural features of fold-thrust zone in Kashi depression, western Tarim basin [J]. Oil & Gas Geology, 2004, 25(2): 199–203. DOI: 10.11743/ ogg20040214. (in Chinese)

[30] LUO Jin-hai, ZHOU Xin-yuan, QIU Bin, YIN Hong, LI Jian-li. Mesozoic-Cenozoic five tectonic events and their petroleum geologic significances in west Tarim Basin [J]. Petroleum Exploration and Development, 2005, 32(1): 18–22. DOI: 10.3321/j.issn:1000-0747.2005.01.005. (in Chinese)

[31] ZHOU Xin-yuan, LUO Jin-hai, MAI Guang-rong. Structural features and petroleum geology of Kashi Sag and its adjacent area in western Tarim Basin [M]. Beijing: Petroleum Industry Press, 2005. (in Chinese)

[32] ZHAO Wen-zhi, ZHANG Guang-ya. Evolution and hydrocarbon geology of passive continental margin: Taking southwest region of Tarim Basin as an example [M]. Beijing: Petroleum Industry Press, 2007. (in Chinese)

[33] FANG Ai-min, MA Jian-ying, WANG Shi-gang, ZHAO Yue, HU Jian-min. Sedimentary tectonic evolution of the southwestern Tarim Basin and west Kunlun orogen since Late Paleozoic [J]. Acta Petrologica Sinica, 2009, 25(12): 3396–3406. (in Chinese)

[34] LUO Jin-hai, ZHOU Xin-yuan, QIU Bin, YANG Zhi-ling, YIN Hong, LI Yong, LI Jian-li. Petroleum geology and geological evolution of the Tarim-Karakum and adjacent areas [J]. Geological Review, 2005, 51(4): 409–415. DOI: 10.3321/j.issn:0371-5736.2005.04.007. (in Chinese)

[35] SANG Hong, CAO Yang-tong, ZHU Li-chun, ZHANG Hua, ZHANG Liang, YAO Fo-jun, ZENG Chang-min, JIAO Peng-cheng, JIANG Hong, LIANG Hua. Preliminary study of evaporites deposition from the Mesozoic to Cenozoic in southwestern Tarim Depression [J]. Journal of Palaeogeography, 2014, 16(4): 473–482. DOI: 10.7605/ gdlxb.2014.04.039. (in Chinese)

[36] MA Hua-dong, YANG Zi-jiang. Evolution of the Cenozoic in southwestern Tarim Basin [J]. Xinjiang Geology, 2003, 21(1): 92–95. DOI: 10.3969/j.issn.1000-8845.2003.01.015. (in Chinese)

[37] SHAO Long-yi, HE Zhi-ping, GU Jia-yu, LUO Wen-lin, JIA Jin-hua, LIU Yong-fu, ZHANG Li-juan, ZHANG Peng-fei. Lithofacies palaeogeography of the Paleogene in Tarim Basin [J]. Journal of Palaeogeography, 2006, 8(3): 353–364. DOI: 10.1016/S1872-2040(06)60004-2. (in Chinese)

[38] ZHUANG Dao-ze, LAN Xian, LI Yan-qing. Geophysical exploration in Xinjiang [M]. Beijing: Geological Publishing House, 2015. (in Chinese)

[39] LI Xian-qing, XIAO Xian-ming, XIAO Zhong-yao, HU Guo-yi, MI Jing-kui, TANG Yong-chun. Geochemical characteristics and origin of natural gas from Ake-1 gas pool in the Tarim Basin [J]. Natural Gas Geoscience, 2005, 16(1): 48–53. DOI: 10.3969/j.issn.1672-1926.2005.01.011. (in Chinese)

[40] KOHLSTEDT D L, EVANS B, MACKWELL S J. Strength of the lithosphere: Constraints imposed by laboratory experiments [J]. Journal of Geophysical Research Solid Earth, 1995, 100(B9): 17587–17602. DOI: 10.1029/ 95JB01460.

[41] BYERLEE J D. Brittle-ductile transition in rocks [J]. Journal of Geophysics Research, 1968, 73(14): 4741–4750. DOI: 10.1029/JB073i014p04741.

[42] BYERLEE J D. Friction of rocks [J]. Pure and Applied Geophysics, 1978, 116(4, 5): 615–626. DOI: 10.1007/ BF00876528.

[43] GOETZE C. High temperature rheology of westerly granite [J]. Journal of Geophysics Research, 1971, 76(5): 1223–1230. DOI: 10.1029/JB076i005p01223.

[44] FU Xiao-fei, JIA Ru, WANG Hai-xue, WU Tong, MENG Ling-dong, SUN Yong-he. Quantitative evaluation of fault-caprock sealing capacity: A case from Dabei-Kelasu structural belt in Kuqa Depression, Tarim Basin, NW China [J]. Petroleum Exploration and Development, 2015, 42(3): 300–309. DOI: 10.1016/S1876-3804(15)30023-9. (in Chinese)

[45] ZHAO Meng-jun, XIA Xin-yu, QIN Sheng-fei, SONG Yan, LIU Shao-bo, LIU Sheng, QIU Bin, YANG Zhi-lin. Gas source study of Ake well 1 resource in Tarim Basin [J]. Natural Gas Industry, 2003, 23(2): 31–33. DOI: 10.3321/j.issn:1000-0976.2003.02.008. (in Chinese)

[46] LIU Sheng, WANG Dong-liang, WANG Zhao-ming, XIAO Zhong-rao, SU Xue-feng. Geochemical analyses on the natural gas formation of well Ak 1 in the Tarim Basin [J]. Petroleum Geology & Experiment, 2004, 26(3): 273–280. DOI: 10.11781/sysydz200403273.(in Chinese)

(Edited by HE Yun-bin)

中文导读

塔里木盆地喀什凹陷古近系阿尔塔什组膏岩盖层封盖完整性量化表征

摘要：盖层保持自身的封盖完整性是油气聚集成藏中的关键因素，膏岩盖层的完整性主要受破裂作用的影响。本文主要通过X射线衍射和三轴压缩实验研究了塔里木盆地喀什凹陷古近系阿尔塔什组膏岩盖层的矿物成分、变形过程、岩石力学强度、应力–应变等特征。结果表明，阿尔塔什组具有明显的上下两段式沉积特点，下段主要由白色石膏组成，内聚力和内摩擦因数分别为8 MPa和0.315；上段则主要为灰白色硬石膏，内聚力和内摩擦因数分别为18 MPa和0.296。综合考虑埋藏过程中的破裂和脆塑性转换因素，阿尔塔什组膏岩的封盖过程可以量化表征为7个阶段，埋藏深度在2285~3301 m的膏岩之下的储集层是本区最有利的勘探目标。

关键词：脆塑转换；破裂准则；封盖完整性；膏岩盖层；阿尔塔什组；喀什凹陷

Foundation item: Project(41672121) supported by the National Natural Science Foundation of China; Project(D1438) supported by the China Geological Survey

Received date: 2018-04-23; Accepted date: 2018-11-20

Corresponding author: WANG Zhen-liang, PhD, Professor; Tel: +86-29-82334929; E-mail: mianxnwu@163.com; ORCID: 0000-0002- 1971-4893