Deep shale gas reservoirs buried underground with depth being more than 3500 m are characterized by high in-situ stress, large horizontal stress difference, complex distribution of bedding and natural cracks, and strong rock plasticity. Thus, during hydraulic fracturing, these reservoirs often reveal difficult fracture extension, low fracture complexity, low stimulated reservoir volume (SRV), low conductivity and fast decline, which hinder greatly the economic and effective development of deep shale gas. In this paper, a specific and feasible technique of volume fracturing of deep shale gas horizontal wells is presented.
Jiang Tingxue, Bian Xiaobing, Wang Haitao, Li Shuangming, Jia Changgui, Liu Honglei, Sun Haicheng
State Key Laboratory of Shale Oil and Gas Enrichment Mechanisms and Effective Development, Beijing 100101, China. Sinopec Research Institute of Petroleum Engineering, Beijing 100101, China
Received 3 November 2016; accepted 25 January 2017
In addition to planar perforation, multi-scale fracturing, full-scale fracture filling, and control over extension of high-angle natural fractures, some supporting techniques are proposed, including multi-stage alternate injection (of acid fluid, slick water and gel) and the mixed- and small-grained proppant to be injected with variable viscosity and displacement. These techniques help to increase the effective stimulated reservoir volume (ESRV) for deep gas production.
Some of the techniques have been successfully used in the fracturing of deep shale gas horizontal wells in Yongchuan, Weiyuan and southern Jiaoshiba blocks in the Sichuan Basin. As a result, Wells YY1HF and WY1HF yielded initially 14.1 × 104 m3/d and 17.5 × 104 m3/d after fracturing. The volume fracturing of deep shale gas horizontal well is meaningful in achieving the productivity of 50 × 108 m3 gas from the interval of 3500–4000 m in Phase II development of Fuling and also in commercial production of huge shale gas resources at a vertical depth of less than 6000 m.
Along with the breakthroughs in shale gas exploration in Fuling, Changning and Weiyuan blocks as well as the deepening of commercial development, shale gas development is moving towards deep formations, which usually refers to reservoirs deeper than 3500 m. It is estimated that deep shale gas resources are huge up to 4612 × 108 m3 in the areas such as Jiaoshiba, Dingshan and Nanchuan, exhibiting a bright exploration and development future.
However, the geology of shale gas in deeper formations and its effect on fracturing also change greatly . Firstly, the increase of wellbore friction leads to higher wellhead pressure and restricted injection displacement, which results in small induced fracture width, low sand/fluid ratio, and decreased fracture conductivity. Secondly, stress rises in three axises, and horizontal stress difference rises in two axises; the magnitude order of stress in three axises may also change. The overlying stress is usually moderate in middle–shallow formations and reaches the maximum in deep formations.
Accordingly, the fracture reorientation and lateral extension of bedding fractures become difficult. Thirdly, rock plasticity enhances, but rock brittleness gradually weakens due to the increase of temperature and confining pressure, resulting in difficulties in the fracture initiation and extension. Fourth, fracture conductivity declines quickly. The closing pressure rise causes higher probability of proppant embedding and crushing.
Quick decline of conductivity results in corresponding decrease of fracture length and stimulated reservoir volume. Fifth, since various tectonic movements occurred frequently at the structure margins, early faults interacted with later faults, or even bedding fractures coexisted with natural fractures in No. 1–5 layers at the bottom of Silurian, Longmaxi Fm, implying that the fracture initiation and extension patterns (especially the extension of fracture height) changed greatly.
Therefore, the previous horizontal well volume fracturing mode and technical parameters for middle-shallow shale gas reservoirs are no longer suitable for deep shale gas reservoirs; instead, it is necessary to carry out block-specific reservoir research on the basis of investigation into deep shale gas fracturing technologies around the world, in order to realize the economic and efficient development of deep shale gas in China.
Comparison of fracturing technologies for deep shale gas reservoirs around the world
Fracturing technologies in the United States
As to deep shale gas (with depth >3500 m) in the United States, commercial development has been realized after many years’ efforts in Eagle Ford, Haynesville and Cana Woodford , , , but not in Hilliard–Baxter–Mancos and Mancos. Table 1 shows the reservoir parameters, post-frac production and costs of these plays. It can be seen that the post-frac production of shale gas in formations deeper than 4100 m is relatively low, and the drilling and fracturing costs per well exceed CNY100 million, thus failing to meet the goal of economic and efficient development.
Table 1. Deep shale gas fracturing and development in the USA.
In Cana Woodford, the core technologies used include: (1) multi-cluster and large-diameter perforation (3–6 clusters in single section, perforation diameter of 14.5 mm); (2) combination of high-viscosity fracturing fluids, i.e. pretreatment acid + linear gel + slick water + gel”; (3) low sand ratio for continuous sand fracturing, with an average sand/fluid ratio of 3–6%; and (4) large fracturing scale in single section, with 1800–2800 m3 fluid and 80–110 m3 sand required. Main changes for these technologies involve increasing the clusters and diameter of perforation in single section (corresponding to the high content of brittle minerals in deep shale gas in the US), and improving the sand volume and composite sand/fluid ratio in single section. The medium viscosity linear gel is injected ahead in order to increase the initial fracture height.
Fracturing technologies in China
Chinese researchers have worked a lot on fracturing for deep shale gas within the country , , . Table 2 provides a comparison of major parameters between the deep shale gas reservoirs and the middle–shallow reservoirs , ,  in Well JY1HF in China and Cana Woodford in the US. It shows that deep shale gas in China is relatively poor in porosity, TOC, gas abundance, brittle mineral content, and horizontal stress difference.
Table 2. Comparison of deep shale gas reservoir characteristics between representative shale blocks in the Sichuan Basin and Cana Woodford in the USA.
Deep fracturing technologies mainly comprise: ① conventional perforation clusters and diameter – 2 clusters in single section and diameter of 10.5 mm; ② combination of fracturing fluids, i.e. “pretreatment acid + gel + slick water + gel”; ③ plug sanding with low sand/fluid ratio, with composite sand/fluid ratio of 1.1–4.2% (2.4% on average); and ④ great fluid volumes (2460–3091 m3) and small sand volumes (averagely 26–50 m3).
Compared to the overseas fracturing technologies, the above-mentioned technologies are characterized by less perforation clusters and small diameter in single section, and lower sanding volume and composite sand/fluid ratio in single section, which reflects greater challenges in deep shale gas fracturing in China. In view of post-fracturing results, the production is low (stable production less than 5 × 104 m3/d) and declines quickly (decline rate above 50% in half a year), thus it is uncommercial and restricts the progress of deep shale gas fracturing.
Volumetric fracturing technology of deep shale-gas horizontal wells
Planar perforating technology
The conventional screw perforation is not ideal for fracture propagation during deep shale gas fracturing, since it adopts the mode of single-perforation fracture initiation with the displacement of less than 0.3 m3/min. If multiple fractures are initiated simultaneously in multiple shots, the fractures with small spacing (fracture spacing is 0.06 m when the perforation density is 16 shots/m) will interfere with each other; within the perforation cluster, the induced stress stacks to lead to higher overall stress, restricting the propagation of multiple fractures as a whole.
However, shale is greatly heterogeneous, and there are only 1 or 2–3 fractures initiated in the cluster. In such a case, both the single-perforation flow and friction increase dramatically, which is disadvantageous for the sufficient fracture propagation. Hence this paper proposes a planar perforation mode (Fig. 1), similar to the directional perforation in vertical wells, exceptionally that the latter requires precise fracture azimuth.
Fig. 1. Comparison between screw perforation and planar perforation.
According to the calculation of casing failure strength (Table 3), the casing failure strength under planar preformation mode with the number of perforation less than 10 shots is equivalent to that (i.e. 16 shots/m) under screw perforation. Given the displacement of 12 m3/min, the dynamic fracture propagation simulation shows that the planar perforation (6 shots/circle) can realize an increase in fracture height, fracture width, fracture length and SRV respectively by 6.3%, 4.6%, 7.2% and 19.8% compared with the screw perforation (3 clusters/section, perforation density of 20 shots/m). Moreover, if the displacement under screw perforation keeps constant, the perforation clusters can improve from 2 clusters to 3–5 clusters; if the perforation clusters keep constant, the number of fracturing device can be reduced by about 50%. No matter which mode of planar perforation is adopted, cost reduction or efficiency improvement can be achieved.
Table 3. Mises stress comparison of casing failures between planar perforation and screw perforation.
Multi-scale fracture creation technology
Alternate acid injection or small-scale acid fracturing
Due to the great difference of horizontal stress, the critical pressure for opening natural fractures cannot be achieved simply by adjusting the parameters of fracturing, and thus the satisfactory fracture complexity cannot be achieved. The commonly used intra-fracture diverting agent is also disadvantageous in several aspects. For instance, narrow pressure windows for deep fracturing make it difficult to seal the fractures that extend dynamically, and the fracture height may not be controlled if both high-angle natural fractures and bedding fractures exist.
Therefore, a method of alternate acid injection or small-scale acid fracturing was put forward. The chemical reaction between acid and rock makes the communication of carbonate minerals possible within different natural fractures along the principal fracture length direction. In order to facilitate field operation, it was suggested that the acid volume injected in each section should not exceed the wellbore volume. Furthermore, in order to inject different acids into different positions for acid–rock reaction, it was suggested that the acid displacement operation can be carried out with optimized volume after acid injection, with low displacement near wells and high displacement far from wells.
Since acid injection is impossibly infinite, there is always a certain area not swept by acid where the fracture complexity is difficult to be improved. In this case, the small-scale dilute acid fracturing can be adopted to make the fracture length reach 70% of final gross fracture length. A lot of simulations show that the geometric size of fractures has already reached above 70% during the initial 1/5 of fracturing time (Fig. 2). Therefore, small-scale acid fracturing is relatively limited, and won’t cause a sharp rise of operation cost.
Fig. 2. Extension of fracture sizes in different operation stages.
Multi-stage alternate injection of viscosity-variable fracturing fluid
Through multi-factor simulation analysis for fracture width, fracture length, fracture height and SRV, it is believed that the fracturing fluid viscosity is the most important factor. Fig. 3 shows the examples of fracture width and SRV.
Fig. 3. Multi-factor significance analysis for fracture width and SRV.
Usually low-viscosity fracturing fluid is efficient in penetrating and communicating small and micro fractures, while medium–high viscosity fracturing fluid is difficult to enter small and micro fractures because of high viscous resistivity but has to extend along the principal fracture direction. Therefore, with consideration to the advantages of fracturing fluids with different viscosities, multi-stage alternate injection of fracturing fluids with variable viscosity and variable displacement should be adopted. In this way, the fracturing fluids can fully extend along principal fractures and also effectively connect the complicated fractures along the principal fracture length, so that fracture complicity and SRV can be maximized.
Obviously, single-stage injection of low-viscosity slick water and medium–high viscosity gel can merely form a fracture pattern of near-well complicated fractures and far-well single principal fractures. The injection of simple slick water can merely form near-well complicated fractures but no principal fractures reach far-well zones. The multi-stage alternate injection of slick water and gel depends on the “viscous fingering” effect (with viscosity ratio above 6 times). Specifically, slick water injected in the later stage advances to the front of fractures created by gel injected in the preceding stage, and continues to communicate and extend along small and micro fractures. Then, the medium–high viscosity gel is injected to push the principal fractures forward. This multi-stage circular injection can realize a universal coverage of principal fractures under a complicated fracture system.
For deep formations, since the closing pressure rise results in a relative decline of original bedding and fracture width, the slick water system with lower viscosity is required for effective communication and extension. According to the features of common emulsion slick water, the viscosity can take 2–3 mPa·s. In order to enhance the sand-carrying performance, the powder type slick water with a viscosity of 9–12 mPa·s that is commonly used in medium–shallow formations is selected.
Fig. 4 shows the optimized upper limit of gel viscosity, indicating a recommended value of 100 mPa·s. The gel viscosity of 30–40 mPa·s is also optional, which is usually used in medium–shallow formations, in order to communicate more branch fractures before main sanding with viscosity of 100 mPa·s. In the same way, according to the sensitivity simulation analysis of fracture parameter and SRV, the occupation percentage of two kinds of gel should be 30–40%.
Fig. 4. Optimization of fracturing fluid viscosity limit.
The simulated proportion of fractures with different widths at different well depths is shown in Table 4.
Table 4. Proportion of average fracture width corresponding to different gel percentages at different well depths.
Full-scale fracture filling
After fracture space in different sizes was created, it is essential to realize the full filling of full-scale fractures in order to improve ESRV. According to the current principle that fracture width is 6 times that of average proppant grain size, and the proportion of fractures with different widths obtained from Table 4, the optimization chart for different grain sizes and proportions of proppant was completed (Fig. 5). For the deep shale gas fracturing, the proportion of small-size proppant should be increased greatly, especially in deeper formations. It is also necessary to select the small size proppant with 140–230 meshes.
Fig. 5. Different proppant grain sizes at different well depths and their proportion optimization.
For a complicated fracture system with principal fractures (primary fractures), branch fractures (secondary fractures) and lower-level fractures (tertiary and quaternary fractures), the study demonstrates  that the proppant with 70–140 meshes can enter the secondary, tertiary and quaternary fractures, while the proppant with 20–40 meshes can hardly enter the secondary fracture or fractures of lower level.
Multi-scale small-sized proppant can enter different branch fractures for sealing, filtrate reduction and support. Moreover, when the proppant grain size reduces by one level, the subsidence velocity can decline 1/3–1/2, which is favorable for improving far-well vertical support efficiency of small and micro fractures. In addition, with the increase of closing pressure, the difference of conductivity between small particle diameter proppant and large particle diameter proppant tends to decrease. In field operation, the small particle diameter proppant may have increased sanding concentration, so it may realize higher conductivity than large particle diameter proppant.
For the above purpose, the small grain proppant should be injected for a relatively long time; otherwise, a part of proppant may retain in the principal fractures to choke the conductivity. Therefore, the design of sanding procedure is extremely essential. The small grain proppant should be added mainly by carrying with low-viscosity slick water. If the viscosity of carrying fluid is high, the small grain proppant holds large dragging force and thus can’t easily enter the small-scale fracture system. The high-viscosity gel can be injected along with continuous sanding to improve the composite sand/fluid ratio.
Finally, considering the precise proportion of fracture in different scales calculated in Table 4, the mixed grain sanding technique can be applied in some cases, namely, the small grain proppant mixes with medium or large grain proppant in a certain proportion, and is carried by low- and medium-viscosity slick water. In this way, the grain size range of proppant is improved, and the proppant with different grain sizes can be transmitted into the corresponding multi-scale fractures under natural action.