Impact assessment of interlayers on geological storage of carbon dioxide in Songliao Basin

. Reservoirs in the Songliao Basin are characterized by strong heterogeneity, which increases the dif ﬁ culty of exact reservoir prediction. The clay interlayer developed in the reservoir is an important factor affecting the heterogeneity of the reservoir. Using the reservoir numerical simulation technology, an attempt has been made to investigate the storage ef ﬁ ciency during CO 2 sequestration in Songliao Basin considering different types of interlayer in underground formations. Results indicate that type I interlayer, with a large thickness embedded between the two sand bodies has function of shunting and blocking to alleviate the impacts on cap rock. The type II interlayer has a small thickness and locates inside a single sand body, with poor physical properties and continuity, which has the same blocking effect on CO 2 distribution and moderating in ﬂ uence on the cap rock. The physical properties of type III interlayer are same as the type II interlayer, but it has uneven distribution and poor continuity. In addition, three schemes of perforated zone were designed and their effects on CO 2 storage ef ﬁ ciency and stability were studied. For a single reservoir, the scheme I is to perforate a whole reservoir, which is more conducive to maintain the reservoir ’ s stability. For multiple sets of “ single-reservoir ” , the scheme II can be preferentially selected to perforate the reservoir section below the interlayer when the injection volume is small. However, the scheme III can be used to perforate the interlayer and the reservoir below that when the injection volume is large. The study is bene ﬁ cial to provide guidance and advice for selecting a suitable CO 2 geological storage and reduce the risk of CO 2 leakage.


Introduction
Songliao Basin is a large Meso-Cenozoic sedimentary basin in eastern China and a composite basin with dual faultdepression structure (Fig. 1). The basin is divided into seven first-level units, namely the Western Slope, Northern Dumping, Central Depression, Northeast Uplift, Southeast Uplift, Southwest Uplift and Kailu Depression. The tectonic evolution of the basin generally controls the distribution of source rocks, reservoir distribution and combination characteristics of generation, reservoir and cap (Han et al., 2019;Hu et al., 2019). Songliao Basin has the characteristics of multi-source and multi-sedimentary systems, forming a wide distribution of sandstone reservoirs. The large delta sand body from the northern provenance area extends to the central depression of Changyuan, which provides a good reservoir condition for forming a huge oilfield (Wu et al., 2001). Nearly 100 oil and gas fields have been discovered in the north and south areas of the basin, where is the most important petroliferous basin in China. The Songliao Basin is the most primary petroliferous basin in China. Since 1995, the field test of CO 2 flooding has been implemented in Jilin Oilfield. In 2007, the research and demonstration project of 100 000 tons/year CO 2 flooding and storage was officially put into operation (Zhang et al., 2009). In the recent years, Jilin Oilfield has gradually carried out the research and construction of the project expansion. At present, it has equipped with about supporting injection facilities of about 500 000 tons/year and realized annual CO 2 utilization of about 200 000 tons/year (Ren et al., 2016;Zhang et al., 2014). However, in the process of sedimentation and diagenesis of the reservoir, the interlayer with different genesis and distributions are formed. It will affect the distribution of sand bodies, oil and water in the reservoir, which makes the detailed description of the reservoir more complicated. In the later stage of the reservoir prediction, if the interlayer is not considered, the flooding process of reservoir will be abnormal, and the prediction can be not accurate (Shan et al., 2007;Zhang et al., 2009).
The different definitions of interlayer have been put forward since the last century (Qiu, 1992;Wang and Li, 1996). At present, it is generally accepted that a non-permeable rock in the reservoir that blocks fluid's movement is aquifuge; a relatively non-permeable layer within the sandstone is impermeable interbed, which together constitute the structure of interlayer. Due to the irregular spatial distributions of interlayer, it is significant to study interwell prediction and its effect on post-development prediction (Rao et al., 2013). The previous studies were centered on the identification technology of interlayer and divided it into three classic models, namely the continuous aquifuge, the continuous impermeable interbed and the discontinuous impermeable interbed (Yan and Duan, 2008). Additionally, the interlayer's characteristics of distribution and morphology were also the research focuses. The above series of studies will eventually serve the development of oil and gas reservoirs, for investigating the effects of interlayer on waterflooding, drilling and reservoir distribution. To clarify the characteristics of interlayer, some scholars have adopted the scheme of selective perforation and partial subdivision to carry out the repair production, such as profile control, water plugging and layered water injection (Zhu, 2013). Currently, most oil and gas reservoirs belong to post-development stage and the interlayer is an important factor affecting and even controlling the distribution of remaining oil. Moreover, the strong randomness and heterogeneity of the interlayer increase the difficulty on its research.
Based on previous studies, for oilfield development, the interlayer has a blocking effect and affects the reservoir connectivity and oil production (Fan et al., 2015;Sun, 2013). And the interlayers are classified into three basic types based on the typical characteristics in Songliao Basin. In this paper, the interlayer is introduced into Geological CO 2 Storage (GCS) engineering project to study its role in CO 2 storage process to facilitate the optimization of GCS project storage scheme. In this paper, the effect of interlayers including three types on CO 2 geological storage was studied. Where the reservoir is defined as "singlereservoir collective". The numerical simulation method was used to explore the influence of interlayer on CO 2 storage efficiency, for further analysis and evaluation of the space utilization and stability of CO 2 storage.

Characteristics of interlayer
The previous studies have shown that the north and south oilfields of the Songliao Basin are developed with different interlayers, including argillaceous, sandy and calcareous interlayers in the North Sabei III area (Zhang, 2014). There are three types of interlayer like argillaceous, calcareous and gravelly in a single distributary channel of the II-I group in the Sanzhao Depression. The blocking effect of the interlayer that belongs to Putaohua oil layer in South Daanbei Oilfield controls the distribution of remaining oil (Yu, 2009).
The oilfields of Jilin Oilfield Company are mainly located in the south of the basin. Many sets of gas-water cycles in Denglouku Formation of Wangfu Fault Depression are controlled by the interlayers (Deng et al., 2015). The interlayer of Xinli Oilfield affects the remaining oil in a plane, which deteriorates the physical properties of the reservoir and causes the remaining oil to be enriched (Hu et al., 2013). The strong heterogeneity caused by the non-permeable interlayer in the reservoir is a factor in the formation of ultra-low permeability reservoirs in Haituozi Oilfield (Li et al., 2015). The dawsonite cement in the Cretaceous sandstone of Honggang Oilfield is widely distributed, and the reservoir is characterized by multi-layer cementation separated by mudstone interlayer in a longitudinal direction (Liu et al., 2016). Three sets of interlayers are developed stably in the Heidimiao oil layer in block 208 of Honggang Oilfield, which is mainly composed of interbedded mudstone and calcareous cemented compact siltstone with thicknesses of 10-15 m, 15-20 m and 30 m, respectively (Fig. 2). In addition, three different distribution patterns of interlayer exist in Hei59 Block, where is the CO 2 -EOR pilot test zone of the original reservoir (Yan et al., 2016), including the type of mud rolling in the sand, the stable interbed type of sand mud, and the type of sand rolling in the mud.
Considering the above lettuces, the complex reservoir was simplified using a single-reservoir group and the effect of interlayers during the process of CO 2 storage was analyzed. Finally, the research method was applied to the complex reservoir. In order to facilitate the calculation, the interlayer is simplified according to the real stratigraphic data of the existing oil area and the interlayer was classified into three types in this study: Type I interlayer, thickness greater than 0.4 m, between two sand bodies, with continuity and isolation. Type II interlayer, thickness greater than 0.4 m, located in a single sand body Internal, with a certain continuity and isolation. Type III interlayer, thickness less than 0.4 m, located inside a single sand body, without continuity and isolation, and the distribution is unstable.

Numerical model
Based on previous studies, for oilfield development, the interlayer has a blocking effect and affects the reservoir connectivity and oil production (Fan et al., 2015;Sun, 2013). In this paper, the interlayer is introduced into GCS engineering project to study its role in CO 2 storage process to facilitate the optimization of GCS project storage scheme. In this study, a CO 2 based on three types of interlayer, injection well was set, with the same injection conditions and the injection volume. To study the effect of interlayer on CO 2 storage, we simulate different schemes in this paper by a commercial simulator (ECLIPSE 2012, The Eclipse Foundation).
A model was set up with length, width and height of 2000 m Â 2000 m Â 54 m in three directions, respectively. The grid block in x, y, and z directions is 200 Â 200 Â 18, respectively. In other words, each grid in the x and y directions is 10 m, the z direction is 3 m. The permeability of 1-2 and 17-18 layers along the z direction is set as 1.0 Â 10 À5 mD, the permeability of other grids is set as 8 mD. For interlay, a layer is added between layers 12 and 13 of the model, specific parameter Settings are shown in Table 1. The injection well located in the middle of the model, as shown in Figure 3. In this study, the current reservoir pressure is 16.7 MPa, reservoir temperature is 65°C, water saturation is 0.32, total compressibility is 1.1 Â 10 À4 MPa À1 . The liquid-gas relative permeability curve used in this simulation is shown in Figure 4. In this paper, the injection rate of CO 2 is 3000 m 3 /Day. The injection time was considered for duration of 20 years.

Effect of the interlayer on CO 2 distribution
According to the distribution of residual oil in reservoir engineering, the effect of interlayer is mainly manifested as the blocking, sealing ability and high displacement pressure, which can form the multiple seepage barriers. There is less residual oil in a positive rhythm reservoir area, while the negative rhythm reservoir area belongs to a residual oil rich area (Wang, 2010). By referring to the above ideas and combining the objectives of GCS project, it is critical to investigate the effect of the interlayer on CO 2 storage. Through the reservoir numerical simulation software ECLIPSE, the influence of interlayer on the migration and distribution of CO 2 was studied and the distribution pattern of CO 2 in sand bodies was predicted.
The red line in Figure 5 corresponds to the interlayers in Figure 3. According to the simulation results of gas saturation distribution after CO 2 injection for 20 years (Fig. 5), CO 2 is isolated by the interlayer into two V-shaped plumes and some CO 2 penetrates through the interlayer to form two inverted V-shaped plumes on both sides of the interlayer. For the same type of interlayer, when the injection time is longer, the more CO 2 will penetrate the interlayer, which indicates that the blocking effect of the interlayer is weaker. For the three different types of interlayer, the type I has the best blocking effect. However, the blocking effect of the type III interlayer is the weakest. Generally, the stronger the blocking effect, the maximum utilization of plane space. For example, the plane migration distance (Y direction) of the small V-shaped CO 2 plume at the type I interlayer is the longest. On the contrary, the V-shaped CO 2 plume plane (Y direction) migrates the shortest distance at the type III interlayer. On the other hand, the stronger the blocking effect, the maximum utilization of vertical space. For example, the longitudinal migration distance (Z direction) of inverted V-shaped CO 2 plume is the longest at the type I interlayer, that in the type III interlayer is the shortest. Among the three kinds of interlayer set up in this model, the physical property and continuity of the type I interlayer is the best, the type II interlayer is the second and the type III interlayer is the worst.

Effect of perforation on the interlayer (GSAT)
The interlayer affects the distributions of the reservoir and oil-water, which also makes the selection of the perforation scheme difficult. Xu et al. (2012) pointed out that if the perforation section is in the interlayer, a certain amount of injected water may change its microstructure, permeability and blocking effect, even causes water logging. In this paper, three perforation schemes are designed based on the interlayer in the single-reservoir group (Fig. 6), among which the scheme 1 is to perforate the whole reservoir; the scheme 2 is only to perforate the lower section of the interlayer; the scheme 3 is to perforate the interlayer and the reservoir section. From Figures 7-9, for each type of the interlayer, after perforating the whole reservoir (scheme 1), the CO 2 plume occupies the smallest space and has the lowest spatial utilization rate. The CO 2 plumes takes up the same space in scheme 2 and 3, that is, they have the same space utilization rate. The permeability of packer was set as the same as the caprock. There is still a pressure difference in the packer section and CO 2 leaks out of the packer after 20 years.
Therefore, it is recommended to adopt the scheme 2, which is only to perforate the reservoir section under the interlayer. For different types of the interlayer under the same scheme, the less CO 2 penetrates the stability   interlayer when the blocking effect tends to be stronger. At the same injection volume, the space above the type I interlayer occupied by CO 2 is the smallest, indicating that its blocking effect is the strongest. However, the CO 2 occupies the space above the type III interlayer is the largest, indicating that the blocking effect of the type III interlayer is the weakest.

Effect of the interlayer on stability
The interlayer not only affects the distribution of CO 2 , but also has an impact on the distribution of reservoir pressure. Zou et al. (2010) studied the effects of interlayer's type and thickness on sealing capacity. During the development process, the interlayer with poor sealing   capacity will be broken through due to the rising pressure, causing the oil and gas upwelling and coexisting with water. Based on the single-reservoir group (Fig. 10), the interlayer can mitigate the upper pressure and reduce the risk of cap breakthrough under the consideration of reservoir stability and caprock security.

Sand body
Injection well Perforation Fig. 10. Diagram of the upper pressure.  As shown in Figure 11, the pressure changes at the center of the upper layer in a single-reservoir group. From  Figures 11a-11c, for the three types of single-reservoir group, the ultimate overlying pressure all attains to be maximum under the action of scheme 1. From Figures 11d-11f, at three schemes of perforation, the overlying pressure of singlereservoir group III is the highest. Although in the beginning, the pressure of single-reservoir group II is slightly larger than the type I, but both are higher than that of type III. However, the pressure of type III gradually exceeds the other types and reaches the maximum from the 18th year.
For a single-reservoir group, the scheme 1 is more conducive to reservoir stability, blocking effect on gas and oil, and sequestration of CO 2 . In addition, it can alleviate or even avoid the coexistence of oil, gas and water in the same layer. For multiple sets of single-storage groups, the scheme 2 can be preferred when the injection volume is small, because the interlayer can be as a barrier layer blocking CO 2 migration. However, when the injection amount is large, the scheme 3 is the optimal solution. Because the blocking effect of the interlayer is limited when the injected volume is huge. At the time, a large amount of CO 2 penetrates the interlayer, indicating the reservoir space utilization will increase and more CO 2 will be storage.
In a word, although the pressure change in the singlereservoir group is small, the actual reservoir is composed of thousands of single-reservoir groups. Therefore, such a small change cannot be ignored, which is of research significance.

Conclusion
In this paper, the numerical simulation method was used to study the effects of three different types of interlayer on CO 2 storage efficiency in a single-reservoir group. In addition, the different perforation schemes are discussed for the effects on CO 2 distribution and overburden pressure changes as well. The following conclusions can be obtained from the studies: 1. Among the three kinds of interlayer set up in this model, the physical property and continuity of the type I interlayer is the best, the type II interlayer is the second and the type III interlayer is the worst. 2. Based on the effect of the interlayer on CO 2 distribution, the stability interlayer can play a role in blocking the migration of CO 2 . 3. If the reservoir storage capacity is considered, the unstable interlayer should be selected first in the target layer; if the reservoir stability is considered, the stability interlayer should be preferentially selected as the target layer. 4. For CO 2 storage, it is recommended to select the interlayer and its below reservoir as the perforation layer. 5. Considering the safety and stability, it is necessary to analyze the pressure of the upper pressure in a singlereservoir group. Generally, perforating the whole reservoir is more stable and secure.