- Email: [email protected]
A novel stimulation strategy for developing tight fractured gas reservoir
Petroleum xxx (2017) 1e8
Contents lists available at ScienceDirect
Petroleum journal homepage: www.keaipublishing.com/en/journals/petlm
A novel stimulation strategy for developing tight fractured gas reservoir* Zhifeng Luo a, Nanlin Zhang a, Liqiang Zhao a, *, Xuefang Yuan b, Yang Zhang b a b
State Key Laboratory of Oil & Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, 610500, China PetroChina Tarim Oil Company, Korla, 841000, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 May 2017 Received in revised form 30 October 2017 Accepted 8 December 2017
Conventional stimulation methods such as matrix acidizing, acid fracturing, or proppant fracturing have resulted in products that perform poorly and/or fail within months. Other options, such as water fracs with light sand, give better results but are prohibitively expensive. Mineral composition, brittleness index, stress regime, and petrophysical properties, which are favorable for creating complex fracture networks, can be obtained by geochemical and geomechanical analysis. The extended Reshaw and Pollard criterion shows that hydraulic fractures tend to be arrested by pre-existing natural fractures, and complex fracture networks would be created during fracturing. Additionally, the critical stressed faults theory indicates that the pre-existing natural fractures tend to slip with the shear mode as the pore fluid pressure increases. Rotating disk experiments and conductivity tests with artificial sheared plates have shown that flow channels can be etched at the location of scratches on fracture surfaces. Meanwhile, the carbonate cement in natural fractures can be chelated to form wormhole likely flow channels. Complex fracture networks with sufficient acid etched conductivity can be generated by water fracs with acid. A novel and economical volume stimulation strategy known as network acid fracturing has provided Tarim Oil Company the means to develope ultra-deep, ultra-high pressure, high temperature, and ultralow permeability but fractured gas reservoirs. Post-stimulation production performances of numerous wells with network acid fracturing are comparable to those with stimulated reservoir volumes. © 2017 Southwest Petroleum University. Production and hosting by Elsevier B.V. on behalf of KeAi Communications Co., Ltd. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Network acid fracturing Pre-existing natural fracture Carbonate cement Ultralow permeability gas reservoir
1. Introduction The great success of Barnett shale triggered the shale gas revolution in North America. Fisher et al. [1], Maxwell et al. [2], and Fisher et al. [3] were the first researchers to discuss the creation of fracture networks in Barnett shale. Their work gave rise to the basic principles of Stimulated Reservoir Volume (SRV). The generation of
* Project funding: sponsored by National natural science funding "Propagation model simulation studies of volume acid fracturing in forming complex joint networks for tight oil & gas reservoirs" (51404207). * Corresponding author. E-mail address: [email protected] (L. Zhao). Peer review under responsibility of Southwest Petroleum University.
Production and Hosting by Elsevier on behalf of KeAi
complex fracture networks with SRV has been being the key technology for the development of unconventional resources, such as tight gas, tight oil, shale gas and ultralow permeability reservoirs [4]. Pre-existed natural fractures [5], formation rock brittleness [6], stress anisotropy, and interfacial friction [7] are the most important factors in creating network fractures. Gu and Weng [8] quantitatively analyzed the influence of stress anisotropy and intersection angles on crossings in order to interpret the creation of complex fracture networks with the extended Reshaw and Pollard criterion [9]. Olson and Dahi-Taleghani [10] and, later, Dahi-Taleghani and Olson [11] detailed the influence of hydraulic fracture on debonding or shearing of cemented natural fractures. Hydraulic fracture is arrested by and propagated along dilated natural fractures and/or multiple fracture fronts propagations while crossing will form complex fracture networks. Slick-water is most commonly used in SRV to connect hydraulic fractures and pre-existing natural fractures to form complex fracture network, and a small amount of proppant with a low sand ratio is mostly used to prop fracture networks [12]. Hydrochloric acid was utilized to ensure fracture
https://doi.org/10.1016/j.petlm.2017.12.002 2405-6561/© 2017 Southwest Petroleum University. Production and hosting by Elsevier B.V. on behalf of KeAi Communications Co., Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
2
Z. Luo et al. / Petroleum xxx (2017) 1e8
initiation in Eagle Ford shale via dissolving acid-soluble cement and alleviating all perforation and near-wellbore friction [13]. Li and Dai [14] analyzes the feasibility by the comparison of reservoir characteristics of shale gas with tight-gas carbonate, meanwhile, analyzes the validity and limitation of the volume acid fracturing technology and gives the solution for the limitation. However, there has been no research on whether acid can be used in SRV to obtain network fracture conductivity for tight but fractured sandstone reservoirs in which most of the pre-existing natural fractures have been cemented by carbonate. The targeted Cretaceous tight sandstone of the Kuqa gas reservoir in the northwest Tarim Basin is 5300e7000 m buried, and characterized by low porosity (about 7%) and ultralow permeability (about 0.07 103mm2 for matrix porous). However, it is well developed with pre-existing natural fractures, while almost all of these natural fractures have been cemented by carbonate. Conventional stimulation strategies, such as matrix acidizing, acid fracturing and proppant fracturing, had been practiced. However, the results have been disappointing. The introduction of fiber assisted re-orientation SRV with propped fracture networks performed well, but proved very costly [15,16]. Microseismic fracture mapping of SRV pilot tests in the Kuqa gas reservoir has shown that slick-water can stimulate natural fractures adequately to form complex network fractures. Acid can be used to dissolve carbonate cement and etch fracture surfaces selectively. It can also provide adequate network fracture conductivity in such ultralow permeability reservoirs. Therefore, multistage network acid fracturing with diversion between stages and temporary plugging within fractures can be considered as an alternative stimulation strategy for fiber assisted re-orientation SRV. Dozens of wells have been implemented with such technology by PetroChina Tarim Oil Company, and their production performances have been comparable to fiber assisted re-orientation SRV. 2. Generation of complex fracture networks Research into hydraulic fracturing in shale gas has indicated that mineral components, brittleness, natural fractures, stress regime, horizontal stress anisotropy, and intersection angles between hydraulic fracture and pre-existing natural fractures play key roles in creating complex fracture networks. The geochemical and geomechanical properties of Kuqa tight sandstone were investigated to show whether complex fracture networks can be created during hydraulic fracturing. 2.1. Geochemical and geomechanical considerations X-ray diffraction analysis showed that the quartz group (including quartz, feldspars, and pyrites), the carbonate group (including calcite, dolomite, and siderite), and the clay group (including total clays) of rock samples from six wells ranged from 70.33 wt% to 95.45 wt%, 1.64 wt% to 16.87 wt%, and 2.3 wt% to 21.77 wt%, respectively. The BI (Brittleness Index) of Kuqa tight sandstone was about 83.4 from the mineralogy view [17]. Alternatively, it was about 54.8 when calculated with Young's modulus and Poisson's ratio interpreted from well logging datum [18]. Pre-existing natural fractures were analyzed in 13 cores drilled from K2 well blocks. The total length of the cores was 447.07 m, and they covered the whole targeted intervals. Fracture dimensions and mineral fill were recorded for each individual fracture, and fracture pattern characteristics were also noted. Fracture orientation data was collected with a Fullbore Microscan Imager (FMI) for Water base Mud (WBM), and with an Earth Image (EI) for Oil Base Mud (OBM). Fig. 1 shows the FMI logging and core observations of well K2-14. Based on the appearance of natural fractures in the high-
resolution electrical images, the pre-existing natural fractures were interpreted and classified as conductive, resistive, critically stressed, fault, and drilling enhanced. The classification indicates the relative strength of the natural fractures. The resistive fractures were closed and mineralized, while active faults and critically stressed fractures were more open and conductive, even under the original in-situ stress state. These fractures can be stimulated easily during hydraulic fracturing. Almost all of the natural fractures in this core were cemented with carbonate (Fig. 1b). Most of the observed fractures from these 13 cores were sub-vertical, with dip angles range predominantly between 60 and 90 (Fig. 2a). Fracture azimuth could be observed to have two distinct trends, heterotropic to one another. Statistical analysis on the fracture azimuth data revealed a bimodal normal distribution, with modes being approximately 120 and 330 (Fig. 2b). Vertical stress was calculated by integrating formation density, which is obtained from wireline logs. The orientation and magnitude of the horizontal principal stresses was determined from wellbore breakouts [19]. The averaged vertical stress, maximum horizontal stress, minimum horizontal stress, and stress anisotropy (defined as the difference between maximum and minimum horizontal stresses) were 142 MPa, 176 MPa, 164 MPa, and 34 MPa, respectively. The relative magnitude of the three principal stresses indicates the stress regime to be strike slip faulting. 2.2. Probability of generation complex fracture networks A comparison of the geochemical and geomechanical properties of typical shale gas and Kuqa tight sandstone is shown in Table 1. The brittleness index, stress regime, petrophysical properties, and pre-existing natural fractures of Kuqa tight sandstone are favorable for the creation of complex fracture networks during hydraulic fracturing. However, the relatively high stress anisotropy makes formation of complex fracture networks more difficult than in shale gas. The extended Renshaw and Pollard criterion proposed by Gu and Weng [8] was used to investigate the interaction behaviors between pre-existing vertical natural fractures and hydraulic fractures under given in-situ horizontal stresses and interfacial properties. The results indicated that hydraulic fractures tended to be arrested by natural fractures where intersection angles b < 60 , coefficient of friction mf ¼ 0.6, and in-situ horizontal stress ratio 1.14 sH/sh 1.32. The shear (t) and normal (sn), stresses on any three dimensional fracture can be determined by its strike and dip angle. When the Coulomb Failure Function, CFF ¼ t - mfsn, becomes positive with an increase of pore pressure, Pp, during hydraulic fracturing, the originally stable fractures would be activated in shear mode. The critically stressed faults theory proposed by Barton and Zoback [26] was employed to calculate the pore fluid pressure needed to slip natural fractures with arbitrary dip and strike angles [27]. The results showed that the pre-existing natural fractures cannot slip under averaged original stress states (shmin ¼ 142 MPa, sHmax ¼ 176 MPa, sv ¼ 164 MPa), but started to dilate in shear mode when pore fluid pressure, Pp, increased from 120 MPa up to 126 MPa, and most of the fractures dilated for Pp ¼ 135 MPa. The strike-slip faulting stress regime, geochemical properties, and mineral and natural fracture orientation distribution of Kuqa tight sand reservoir demonstrated that complex fracture networks would be created during hydraulic fracturing. This paper integrates the shear slippage criterion with closed natural fractures proposed by Warpinski and Teufel [28], the crossing criterion proposed by Blanton [29], and the re-initiation criterion at the natural fracture tip in order to investigate the complicated intersection behaviors and required conditions. The shear slippage criterion for vertical natural fracture was
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
Z. Luo et al. / Petroleum xxx (2017) 1e8
3
Fig. 1. Pre-existing natural fracture in well K2-14. (a) Examples of natural fractures observed on electrical image, (b) Carbonated cemented natural fracture observed on core sample.
given by Warpinski and Teufel [28], and is expressed in Equation (1).
pðxo Þ > To þ
sHmax þ shmin
jtj t0 þ mf sn pjunc
(1)
Replacing the joint pressure, pjunc, with net pressure, pnet, and expressing the stresses, t and sn, with far-field stresses, shmin and sHmax, the criterion can be expressed as:
pnet
t0 ðsHmax shmin Þ sin 2 b þ mf cos 2 b mf mf
2
sHmax shmin 2
cos 2 b
vðxo Þ sin b a
nðxo Þ ¼
. 2
(5)
"
xo þ l þ a 2 xo l a 2 þ ðxo lÞln xo þ l xo l p 2 # xo þ l þ a þ a ln xo l a 1
ðxo þ lÞln
(6)
(2) p
where t0 is the shear strength of the natural fracture. The crossing criterion for closed vertical natural fractures was given by Blanton. To re-initiate a fracture on the other side of the natural fracture, the pressure, p(xo), would have to overcome the stress, st, acting parallel to the natural fracture plus the tensile strength of matrix rock, To. The stress, st, is composed of the stress, st1, and determined by far-field stresses, and the stress, st2, is determined by the shear stress along the natural fracture [29].
pðxo Þ st þ To ¼ st1 þ st2 þ To
þ
(3)
xo ¼
ð1 þ aÞ2 þ e2mf
!1=2
p
1 þ e2mf
(7)
where △popen is the pressure loss along the length of the opened natural fracture, l. △pshear is the pressure loss along the length of sheared natural fracture, a, and xo is the length between reinitiating point and intersection point. The re-initiation criterion at the dilated natural fracture tip can be depicted as follows.
ptip ¼ pjunc Dpnf sn þ To
(8)
The crossing criterion can be expressed in the form of net pressure, pnet.
where ptip is the fluid pressure at the natural fracture tip, and △pnf is the pressure loss along the natural fracture. Similarly, equation (8) can be given as:
pnet pðxo Þ þ Dpopen þ Dpshear shmin
pnet
Also,
(4)
1 ðs shmin Þð1 cos 2 bÞ þ To þ Dpnf 2 Hmax
(9)
The integrated intersection criteria indicate that hydraulic
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
4
Z. Luo et al. / Petroleum xxx (2017) 1e8
fracture tends to be arrested and re-initiated at the tip of natural fractures with small intersection angles. Fracture also tends to cross natural fractures with large intersection angles in Kuqa tight sand, as Fig. 3 shows. It is easier for fractures with intersection angles above 65 to cross than to slip. The required shear slippage net pressure is negative for fractures with intersection angles between 2 and 55 , which indicates that these fractures are much more likely to be sheared with an increase of pore pressure. All of these criteria indicated that complex fracture networks would be generated during fracturing. Furthermore, the microseismic fracture mapping technology implemented with water fracs in this well block also confirmed the generation of complex fracture networks.
3. Fracture conductivity gained from acid etching
Fig. 2. Statistical analysis: histogram and probability density distribution plot for fracture dip and azimuth, Kuqa fractured tight gas reservoir. (a) fracture dip, (b) fracture azimuth.
In addition to fracture geometry, network fracture conductivity also is a key parameter for hydraulic fracturing. Although sheared fractures have a certain degree of self-support capacity, it is necessary to improve network fracture conductivity in such high closure stress reservoirs. Proppant with low sand ratio has been used in conventional water fracs to prevent fracture from closing and gain propped fracture conductivity. Instead, acid can be used as an alternative stimulation strategy to selectively dissolve carbonate cement and etch fracture faces for most of the pre-existing natural fractures, many of which are cemented with carbonate. Acid etched fracture conductivity is greatly dependent on etching patterns. Referring to the acid etching pattern experiments of carbonate, a rotating disk apparatus and conductivity measurement apparatus with laser scanning were used to observe acid etching patterns of sandstone rock samples [30]. Fig. 4 shows the pre and post etching laser scanning photos of the rock sample from well K2-5 that was etched with a combination of 5 wt% formic, 7 wt % phosphonic, and 2 wt% hydrofluoric with other additives. This combined acid formula has little capacity to etch such sandstone rock samples, which implied that fracturing with this acid formula will form insufficient conductivity. However, the tensile and/or sheared hydraulic fracture face would be rough rather than smooth [31]. Therefore, the acid etching pattern with the rotating disk
Table 1 Geochemical and geomechanical comparison between typical shale gas and Kuqa tight sand. Comparison Items Depth, m Thickness, m Petrophysical Geochemical
Natural Fracture Characteristic
Geomechanical
a b c d e f g h i j
Porosity, % Permeability, 103mm2 Quartz, % Clay, % Carbonate, % Brittleness Fracture Density, 1/m Fracture Aperture, mm Characteristic Young's Modulus, GPa Poisson's Ratio Brittleness
Kuqa Tight Sand 5300e7000 50e300 7 0.07 70.33e95.45 2.3e21.77 1.64e16.87 83.4 0.15e3 <3.42 Sealed with Calcite 21e36 0.23e0.3 54.8
Barnett Shale 1891e2591a 106.7a 5.5 0.00007e0.005 57b 27b 8b 62.0b 2.207c <0.05e0.265c Sealed with Calcitec 33.0c 0.2e0.3c 60e70b
Middle Bakken Shale 2225e3230d 6.1e21.3d 5e 0.04e 26.4f 13f 47.4f 30.4f NAj >0.03e Non-mineralizede 19.9d 0.24d NAj
Eagle Ford Shale 762e4267g 6.1e152.4g 8e18g 1 109-8 104g 15g 20g 55g 16.7g >999/NAh,i,j NAj NAj 6.89e13.79g 0.25e0.27g 40e60g
Data are from Pollastro [20]. Data are from Bowker [6]. Data are from Gale [5]. Data are from Ostadhassan [21]. Data are from Pitman [22]. Data are from Havens [23]. Data are from Stegent [24]. Data are from Offenberger [25]. At least 999 natural fractures intersected by the wellbore of well 1# in Eagle Ford. NA: Not Available published Data.
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
Z. Luo et al. / Petroleum xxx (2017) 1e8
5
Fig. 3. Integrated intersection criteria with averaged geomechanical parameters, SHmax ¼ 176 MPa, Shmin ¼ 142 MPa, To ¼ 5 MPa, t0 ¼ 2 MPa, mf ¼ 0.6 for Kuqa tight sand.
Fig. 4. Laser scanning photos of rock sample from well K2-5 etched 30 min with acid under the condition of 120 centigrade and 7 MPa. (a) Pre-etched, (b) Post-etched.
apparatus is insufficient to draw the conclusion that acid is improper to obtain sufficient conductivity. A pair of sheared rock samples without carbonate cemented fractures from well K2-7 were sheared in order to evaluate the acid etched fracture conductivity with acidizing core flood apparatus at 120 centigrade and 12 MPa. The conductivity experiment indicated that the acid would etch the sheared scratches to generate fracture conductivity, and acid etching could be a viable alternative approach to proppant to gain network fracture conductivity. The acid etched fracture conductivity of Kuqa tight sand was about 1.5 mm2-cm, which is much smaller than the 30e60 mm2-cm for dolomite or limestone etched by hydrochloric acid, as Fig. 5 shows (see Fig. 6). A core penetrated with carbonate cemented fractures from well K5-1 with a diameter of 2.478 cm and a length of 2.711 cm was used to test the effects of acid stimulation at 130 centigrade. There was no fluid at the outlet after 3 h with displacing pressure of 18 MPa for base fluid composed of 4 wt% NH4Cl. Then, chelation acid with a pH of 1.7 was injected, and the displacing pressure decreased sharply from 15.992 MPa. Moreover, the fluid started to flow out after injection acid at a time of 25min, while the flow rate was found to increase obviously and up to 1 ml/min at 31.5min and 80min, respectively. Eventually, the base fluid composed of 4 wt% NH4Cl was injected at a pumping rate of 1 ml/min and a displacing
Fig. 5. Acid etched fracture conductivity of sheared core without carbonate cemented fracture from well K2-7 and well K2-5.
pressure of 4e6 MPa. The pre- and post-acid photos of this carbonate cemented fractured core indicated that some wormholelikely flow channels could be formed with acid. The etching experiment with the rotating disk, as well as the acidizing core flood tests for artificial sheared fractures and
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
6
Z. Luo et al. / Petroleum xxx (2017) 1e8
carbonate cemented fracture, demonstrated that acid could be used to generate fracture conductivity in networks. Because shear slippage and acid etching can create complex fracture networks, it can be concluded that network acid fracturing could be a viable alternative stimulation strategy in certain tight gas reservoirs.
4. Pilot test PetroChina Tarim Oil Company has been introducing fiber assisted re-orientation SRV to its wells. Thus far, this technique has been applied in 7 wells, including 2 wells with microseismic fracture mapping [15] in Kuqa tight gas since 2013. The degradable fiber enhanced the performance of SRV with average post-frac open-flow capacity about 320 104 m3/d, but the technique is very costly. Fiber-ball and fiber-grain assisted multistage network acid fracturing could be an alternative stimulation strategy for fiber assisted water fracs. In order to verify the feasibility of network acid fracturing, 17 wells have been selected to be tested with such a stimulation strategy. The post-frac production performances were shown to be comparable to fiber assisted SRV. Well K2-18 is located in the K2 structure, within the Cretaceous tight sandstone range from 6676 to 6858 m. The well has 12 zones with a total thickness of 13.5 m and porosity between 3 and 5%, and 47 zones with thickness of 107 m and porosity above 5%. The in-situ stress regime is strike-slip faulting with high horizontal stress anisotropy, where the maximum and minimum horizontal stress gradients and vertical stress gradient are about 2.68, 2.18, and 2.45 MPa/100 m, respectively. Imaging logging showed 47 pre-existing natural fractures with apertures larger than 1 mm in the targeted interval, and oriented nearly EW with a density of 0.258 per meter. Core sample observation showed that almost all of these fractures were either partially or completely cemented with carbonate. The azimuth of natural fractures indicated that the intersection angles between natural fractures and hydraulic fracture, which was parallel to SHmax oriented SW-NE, range from 40 to 50 . Most of the intersected pre-existing natural fractures will dilate as bottom hole pressure gradient is increased to 2.15 MPa/100 m, as shown in Fig. 7. The stimulation strategy is intended to link natural fractures with hydraulic fractures to form complex fracture networks and generate fracture conductivity with sequential injection of hydrochloric and hydrofluoric contained acid to dissolve carbonate cement and drilling mud solids. Fig. 8 shows the pressure response of the acid treatment of well K2-18. The tubing pressure decreased sharply at the pumping rate when formation rock came into contact with pad acid, which is composed of 9 wt% hydrochloric and 3 wt% acetic. This suggests that the carbonate cement was dissolved by acid, and fracture conductivity was greatly enhanced. Degradable fiber grains and fiber powder were used as diverters to improve zone acid coverage. The well head tubing pressure increased by 4.6 MPa at a rate of 1.2 m3/min in the first diversion, and 7.6 MPa at a rate of 4.2 m3/min in the second diversion. After the successful execution of network acid fracturing, the well was flowed back for cleanup and rate assessment. The well initially flowed 43.64 104 m3/d with 84.8 MPa of flowing wellhead pressure. This initial result is comparable to the flow-back results of offset wells in the area that were stimulated with fiber assisted SRV. Table 2 displays and compares the net-pay thickness, porosity
Fig. 6. Acidizing core flood experiment for sample with carbonate cemented fractures from well K5-1. (a) Displacing pressure versus time, (b) lateral faces of pre-test core, (c) lateral faces of post-test core from almost the same view of (b), (d) pre- and post-test inlet face, and (e) pre- and post-test outlet face.
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
Z. Luo et al. / Petroleum xxx (2017) 1e8
7
Fig. 7. Coulomb failure function of natural fractures in well K2-18 with increasing pore fluid pressure, white dots represent activated natural fractures: (a) 1.95 SG, (b) 2.05 SG, and (c) 2.15 SG.
Fig. 8. Acid treatment of well K2-18. The dotted circle shows the increasing of well head tubing pressure for injection of fiber grains with powder.
Table 2 Comparison of network acid fracturing and fiber assisted SRV. SS
Well Name
NPT, m
(4H)5, m-%
(4H)3, m-%
NFD, 1/m
q,
AOF, 104 m3/d
Fiber Assisted SRV
K1-1 K1-8 K1-14 K2-8 K2-12 K2-14 K5-1 K1-6 K2-3 K2-18 K8-1 K8-3
114 112 104.5 178.5 140 148.5 157 167.5 129.5 140.5 150.5 117.5
606.9 618.15 604.95 706.65 654.75 739.2 383.25 586.8 323.85 615.2 520.85 665.45
864.1 833.9 733.3 1253.25 933.35 1103.85 871.95 1067.25 795.35 943.9 949.6 863
0.21 0.24 0.67 0.63 0.7 0.291 0.43 0.67 0.1 0.258 0.10 0.31
60 35 26 33 20 30 15 11 70 40 15 15
55 223 208 686 407 447 217 550 60 465 350 570
Network Acid Fracturing
multiplied by thickness with porosity higher than 3% and 5%, natural fracture density, intersection angle between hydraulic fracture and pre-existing natural fracture, and post-stimulation open flow capacity of wells located in K1-K2 and K8 structure. The overall post stimulation open flow capacity of wells with network acid fracturing is comparable to those stimulated with fiber assisted SRV.
Furthermore, the adjacent wells of K2-14 and K2-18 located in the K1-K2 structure with similar reservoir properties obtained almost the same stimulation effect. These pilot wells demonstrated that the network acid fracturing strategy is an effective technology for Kuqa tight gas reservoir. However, either SRV or acid fracturing performed disappointingly for wells with improper natural fracture
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
8
Z. Luo et al. / Petroleum xxx (2017) 1e8
orientation, such as well K1-1 and K2-3, suggesting higher intersection angles between hydraulic fracture and natural fractures in these wells. SS: stimulation strategy; NPT: Net-pay thickness; (4H)n: porosity multiplied by thickness with porosity higher than n%, n ¼ 3, 5; NFD: natural fracture density; q: intersection angle between pre-existing natural fracture and hydraulic fracture; AOF: absolutely open flow capacity. 5. Results and discussion For high horizontal stress anisotropy reservoirs, interface slipping of natural fractures is profitable for stimulations to create complex network fracture, but natural fractures with intersection angles higher than 60 are difficult to dilate and contribute little to stimulation effects. For carbonate-cemented natural fractures, acid can be used as an alternative approach to proppant for obtaining fracture conductivity. A novel and economic stimulation strategy, network acid fracturing, has been proposed and implemented in carbonate-cemented fractured tight gas reservoirs. Poststimulation production performances of wells stimulated with network acid fracturing are comparable to those treated with fiber assisted SRV. The fiber assisted SRV is more suitable for poor formation dominated reservoirs, while network acid fracturing results in higher production for reservoirs with more desirable properties. The dilation of natural fractures with shear mode at high stress anisotropy was concluded only from intersection criteria in this paper, and should be verified with further laboratory experiments. The present designs of volume of acid and slick water for network acid fracturing were determined empirically rather than theoretically. The simulation of network fracture propagation with acidrock reaction for optimized design is in progress and will be published in a coming paper. Acknowledgments The authors thank financial support from the National Natural Science Fund(51404207) and the management of the PetroChina Tarim Oil Company for permission to publish this article. Special thanks to the technical teams in the Oil & Gas Engineering Institute of PetroChina Tarim Oil Company for their assistance in providing and evaluating data, assessing well performance, and implementation of new technology in the field. References [1] Fisher, M.K., C.A. Wright, and B.M. Davidson et al., Integrating Fracture Mapping Technologies to Optimize Stimulations in the Barnett Shale. Paper SPE 77441 Presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, 29 Sept.-2 Oct. 2002. [2] Maxwell, S.C., T.I. Urbancik, and N.P. Steinsberger et al., Microseismic Imaging of Hydraulic Fracture Complexity in the Barnett Shale. Paper SPE77440 Presented at the SPE Annual Technology Conference and Exhibition, San Antonio, Texas, USA, 29 Sept.-2 Oct. 2002. [3] M.K. Fisher, J.R. Heinze, C.D. Davidson, et al., Optimizing horizontal completion techniques in the Barnett shale using microseismic fracture mapping, in: Paper SPE 90051 Presented at the SPE Annual Technical Conference and Exhibition, Houston, Sept. 2004, pp. 26e29. [4] M.J. Mayerhofer, E.P. Lolon, N.R. Warpinski, et al., What is stimulated reservoir volume? SPE Prod. Oper. 25 (1) (2010) 89e98. [5] J.F. Gale, R.M. Reed, J. Holder, Natural fractures in the Barnett shale and their importance for hydraulic fracture treatments, AAPG Bull. 91 (4) (2007) 603e622. [6] K.A. Bowker, Recent development of the Barnett shale play, Fort Worth Basin
West Texas Geol. Soc. Bull. 42 (6) (2003) 1e11. [7] W. Weng, O. Kresse, C. Cohen, et al., Modeling of hydraulic-fracture-network propagation in a naturally fractured formation, SPE Prod. Oper. 26 (04) (2011) 368e380. [8] H. Gu, X. Weng, J. Lund, et al., Hydraulic fracture crossing natural fracture at nonorthogonal angles: a criterion and its validation, SPE Prod. Oper. 27 (01) (2012) 20e26. [9] Gu, H., and X. Weng, Criterion for fractures crossing frictional interfaces at non-orthogonal angles. Paper ARMA 10e198 Presented at the 44th U.S. Rock Mechanics Symposium and 5th U.S. Canada Rock Mechanics Symposium, Salt Lake City, Utah, USA, 27-30 June 2010. [10] Olson, J.E. and A. Dahi-Taleghani. Modeling simultaneous growth of multiple hydraulic fractures and their interaction with natural fractures. Paper SPE 119739 presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas, USA, 19e21 Jan. 2009. [11] A. Dahi-Taleghani, J.E. Olson, Numerical modeling of multistranded-hydraulicfracture propagation: accounting for the interaction between induced and natural fractures, SPE J. 16 (03) (2011) 575e581. [12] G. Schein, The Application and Technology of Slickwater Fracturing, 2005. [13] Stegent, N.A., L.M. Leotaud, and W. Prospere et al. Cement technology improves fracture initiation and leads to successful treatments in the Eagle Ford shale. SPE 137441 Presented at the SPE Tight Gas Completions Conference Held in San Antonio, Texas, USA, Nov.2010. [14] N. Li, J. Dai, C. Liu, et al., Feasibility study on application of volume acid fracturing technology to tight gas carbonate reservoir development, Petroleum 1 (3) (2015) 206e216. [15] F. Gui, K. Rahman, D. Moos, et al., Optimizing hydraulic fracturing treatment integrating geomechanical analysis and reservoir simulation for a fractured tight gas reservoir, Tarim Basin, China, in: ISRM International Conference for Effective and Sustainable Hydraulic Fracturing, International Society for Rock Mechanics, 2013. [16] Xuefang Yuan, Fei Liu, et al., Pilot tests of degradable fiber assisted reorientation stimulated reservoir volume (in Chinese), in: Conference for Unconventional Tight Oil & Gas Exploration Technologies, Xinjiang Petroleum Institute, 2013. [17] Sondergeld, C.H., and K.E. Newsham, et al. Petrophysical considerations in evaluating and producing shale gas resources. SPE-131768-MS presented at the SPE unconventional Gas Conferences held in Pittsburgh, Pennsylvania, USA, Feb.2010. [18] Rickman, R., and M. Mullen, et al. A practical use of shale petrophysics for stimulation design optimization: all shale plays are Not clones of the Barnett shale. SPE-115258-MS presented at the SPE Annual Technical Conference and Exhibition held in Denver, Colorado, USA, Sept.2008. [19] M.D. Zoback, Reservoir Geomechanics, Cambridge University Press, 2007. [20] R.M. Pollastro, R.J. Hill, D.M. Jarvie, et al., Assessing Undiscovered Resources of the Barnett-paleozoic Total Petroleum System, Bend Arch-Fort Worth Basin Province, Texas, 2003. [21] M. Ostadhassan, Geomechanics and Elastic Anisotropy of the Bakken Formation, Doctor Dissertation, University of North Dakota, Williston Basin, 2013. [22] J.K. Pitman, L.C. Price, et al., Diagenesis and Fracture Development in the Bakken Formation, Williston Basin: Implications for Reservoir Quality in the Middle Member, US Department of the Interior, US Geological Survey, 2001. [23] J. Havens, Mechanical Properties of the Bakken Formation, Master Dissertation, Colorado School of Mines, 2012. [24] Stegent, N.A., and A.L. Wagner et al. Engineering a successful fracturestimulation treatment in the Eagle Ford shale. Paper SPE-136183-MS presented at the SPE Tight Gas Completions Conference held in San Antonio, Texas, USA, Nov.2010. [25] Offenberger, R., and N. Ball et al. Integration of natural and hydraulic fracture network modeling with reservoir simulation for an Eagle Ford well. Paper SPE 168683 /URTeC1563066 Presented at the Unconventional Resources Technology Conference held in Denver, Colorado, USA, Aug. 2013. [26] C.A. Barton, M.D. Zoback, et al., Fluid flow along potentially active faults in crystalline rock, Geology 23 (8) (1995), 683e683. [27] M.W. [26] McClure, R.N. Horne, Investigation of injection-induced seismicity using a coupled fluid flow and rate/state friction model, Geophysics 76 (6) (2011). WC181-WC198. [28] N.R. Warpinski, L.W. Teufel, Influence of geologic discontinuities on hydraulic fracture propagation (includes associated papers 17011 and 17074), J. Petrol. Technol. 39 (2) (1987) 209e220. [29] T.L. Blanton, Propagation of hydraulically and dynamically induced fractures in naturally fractured reservoirs, in: SPE Unconventional Gas Technology Symposium, SPE15261, 1986. [30] C. Malagon, M. Poumik, A.D. Hill, The texture of acidized fracture surfaces: implications for acid fracture conductivity, SPE Prod. Oper. 23 (3) (2008) 343e352. [31] C. Dong, D. Zhu, A.D. Hill, Modeling of the acidizing process in naturally fractured carbonates, SPE J. 7 (04) (2002) 400e408.
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
Contents lists available at ScienceDirect
Petroleum journal homepage: www.keaipublishing.com/en/journals/petlm
A novel stimulation strategy for developing tight fractured gas reservoir* Zhifeng Luo a, Nanlin Zhang a, Liqiang Zhao a, *, Xuefang Yuan b, Yang Zhang b a b
State Key Laboratory of Oil & Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, 610500, China PetroChina Tarim Oil Company, Korla, 841000, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 May 2017 Received in revised form 30 October 2017 Accepted 8 December 2017
Conventional stimulation methods such as matrix acidizing, acid fracturing, or proppant fracturing have resulted in products that perform poorly and/or fail within months. Other options, such as water fracs with light sand, give better results but are prohibitively expensive. Mineral composition, brittleness index, stress regime, and petrophysical properties, which are favorable for creating complex fracture networks, can be obtained by geochemical and geomechanical analysis. The extended Reshaw and Pollard criterion shows that hydraulic fractures tend to be arrested by pre-existing natural fractures, and complex fracture networks would be created during fracturing. Additionally, the critical stressed faults theory indicates that the pre-existing natural fractures tend to slip with the shear mode as the pore fluid pressure increases. Rotating disk experiments and conductivity tests with artificial sheared plates have shown that flow channels can be etched at the location of scratches on fracture surfaces. Meanwhile, the carbonate cement in natural fractures can be chelated to form wormhole likely flow channels. Complex fracture networks with sufficient acid etched conductivity can be generated by water fracs with acid. A novel and economical volume stimulation strategy known as network acid fracturing has provided Tarim Oil Company the means to develope ultra-deep, ultra-high pressure, high temperature, and ultralow permeability but fractured gas reservoirs. Post-stimulation production performances of numerous wells with network acid fracturing are comparable to those with stimulated reservoir volumes. © 2017 Southwest Petroleum University. Production and hosting by Elsevier B.V. on behalf of KeAi Communications Co., Ltd. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Network acid fracturing Pre-existing natural fracture Carbonate cement Ultralow permeability gas reservoir
1. Introduction The great success of Barnett shale triggered the shale gas revolution in North America. Fisher et al. [1], Maxwell et al. [2], and Fisher et al. [3] were the first researchers to discuss the creation of fracture networks in Barnett shale. Their work gave rise to the basic principles of Stimulated Reservoir Volume (SRV). The generation of
* Project funding: sponsored by National natural science funding "Propagation model simulation studies of volume acid fracturing in forming complex joint networks for tight oil & gas reservoirs" (51404207). * Corresponding author. E-mail address: [email protected] (L. Zhao). Peer review under responsibility of Southwest Petroleum University.
Production and Hosting by Elsevier on behalf of KeAi
complex fracture networks with SRV has been being the key technology for the development of unconventional resources, such as tight gas, tight oil, shale gas and ultralow permeability reservoirs [4]. Pre-existed natural fractures [5], formation rock brittleness [6], stress anisotropy, and interfacial friction [7] are the most important factors in creating network fractures. Gu and Weng [8] quantitatively analyzed the influence of stress anisotropy and intersection angles on crossings in order to interpret the creation of complex fracture networks with the extended Reshaw and Pollard criterion [9]. Olson and Dahi-Taleghani [10] and, later, Dahi-Taleghani and Olson [11] detailed the influence of hydraulic fracture on debonding or shearing of cemented natural fractures. Hydraulic fracture is arrested by and propagated along dilated natural fractures and/or multiple fracture fronts propagations while crossing will form complex fracture networks. Slick-water is most commonly used in SRV to connect hydraulic fractures and pre-existing natural fractures to form complex fracture network, and a small amount of proppant with a low sand ratio is mostly used to prop fracture networks [12]. Hydrochloric acid was utilized to ensure fracture
https://doi.org/10.1016/j.petlm.2017.12.002 2405-6561/© 2017 Southwest Petroleum University. Production and hosting by Elsevier B.V. on behalf of KeAi Communications Co., Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
2
Z. Luo et al. / Petroleum xxx (2017) 1e8
initiation in Eagle Ford shale via dissolving acid-soluble cement and alleviating all perforation and near-wellbore friction [13]. Li and Dai [14] analyzes the feasibility by the comparison of reservoir characteristics of shale gas with tight-gas carbonate, meanwhile, analyzes the validity and limitation of the volume acid fracturing technology and gives the solution for the limitation. However, there has been no research on whether acid can be used in SRV to obtain network fracture conductivity for tight but fractured sandstone reservoirs in which most of the pre-existing natural fractures have been cemented by carbonate. The targeted Cretaceous tight sandstone of the Kuqa gas reservoir in the northwest Tarim Basin is 5300e7000 m buried, and characterized by low porosity (about 7%) and ultralow permeability (about 0.07 103mm2 for matrix porous). However, it is well developed with pre-existing natural fractures, while almost all of these natural fractures have been cemented by carbonate. Conventional stimulation strategies, such as matrix acidizing, acid fracturing and proppant fracturing, had been practiced. However, the results have been disappointing. The introduction of fiber assisted re-orientation SRV with propped fracture networks performed well, but proved very costly [15,16]. Microseismic fracture mapping of SRV pilot tests in the Kuqa gas reservoir has shown that slick-water can stimulate natural fractures adequately to form complex network fractures. Acid can be used to dissolve carbonate cement and etch fracture surfaces selectively. It can also provide adequate network fracture conductivity in such ultralow permeability reservoirs. Therefore, multistage network acid fracturing with diversion between stages and temporary plugging within fractures can be considered as an alternative stimulation strategy for fiber assisted re-orientation SRV. Dozens of wells have been implemented with such technology by PetroChina Tarim Oil Company, and their production performances have been comparable to fiber assisted re-orientation SRV. 2. Generation of complex fracture networks Research into hydraulic fracturing in shale gas has indicated that mineral components, brittleness, natural fractures, stress regime, horizontal stress anisotropy, and intersection angles between hydraulic fracture and pre-existing natural fractures play key roles in creating complex fracture networks. The geochemical and geomechanical properties of Kuqa tight sandstone were investigated to show whether complex fracture networks can be created during hydraulic fracturing. 2.1. Geochemical and geomechanical considerations X-ray diffraction analysis showed that the quartz group (including quartz, feldspars, and pyrites), the carbonate group (including calcite, dolomite, and siderite), and the clay group (including total clays) of rock samples from six wells ranged from 70.33 wt% to 95.45 wt%, 1.64 wt% to 16.87 wt%, and 2.3 wt% to 21.77 wt%, respectively. The BI (Brittleness Index) of Kuqa tight sandstone was about 83.4 from the mineralogy view [17]. Alternatively, it was about 54.8 when calculated with Young's modulus and Poisson's ratio interpreted from well logging datum [18]. Pre-existing natural fractures were analyzed in 13 cores drilled from K2 well blocks. The total length of the cores was 447.07 m, and they covered the whole targeted intervals. Fracture dimensions and mineral fill were recorded for each individual fracture, and fracture pattern characteristics were also noted. Fracture orientation data was collected with a Fullbore Microscan Imager (FMI) for Water base Mud (WBM), and with an Earth Image (EI) for Oil Base Mud (OBM). Fig. 1 shows the FMI logging and core observations of well K2-14. Based on the appearance of natural fractures in the high-
resolution electrical images, the pre-existing natural fractures were interpreted and classified as conductive, resistive, critically stressed, fault, and drilling enhanced. The classification indicates the relative strength of the natural fractures. The resistive fractures were closed and mineralized, while active faults and critically stressed fractures were more open and conductive, even under the original in-situ stress state. These fractures can be stimulated easily during hydraulic fracturing. Almost all of the natural fractures in this core were cemented with carbonate (Fig. 1b). Most of the observed fractures from these 13 cores were sub-vertical, with dip angles range predominantly between 60 and 90 (Fig. 2a). Fracture azimuth could be observed to have two distinct trends, heterotropic to one another. Statistical analysis on the fracture azimuth data revealed a bimodal normal distribution, with modes being approximately 120 and 330 (Fig. 2b). Vertical stress was calculated by integrating formation density, which is obtained from wireline logs. The orientation and magnitude of the horizontal principal stresses was determined from wellbore breakouts [19]. The averaged vertical stress, maximum horizontal stress, minimum horizontal stress, and stress anisotropy (defined as the difference between maximum and minimum horizontal stresses) were 142 MPa, 176 MPa, 164 MPa, and 34 MPa, respectively. The relative magnitude of the three principal stresses indicates the stress regime to be strike slip faulting. 2.2. Probability of generation complex fracture networks A comparison of the geochemical and geomechanical properties of typical shale gas and Kuqa tight sandstone is shown in Table 1. The brittleness index, stress regime, petrophysical properties, and pre-existing natural fractures of Kuqa tight sandstone are favorable for the creation of complex fracture networks during hydraulic fracturing. However, the relatively high stress anisotropy makes formation of complex fracture networks more difficult than in shale gas. The extended Renshaw and Pollard criterion proposed by Gu and Weng [8] was used to investigate the interaction behaviors between pre-existing vertical natural fractures and hydraulic fractures under given in-situ horizontal stresses and interfacial properties. The results indicated that hydraulic fractures tended to be arrested by natural fractures where intersection angles b < 60 , coefficient of friction mf ¼ 0.6, and in-situ horizontal stress ratio 1.14 sH/sh 1.32. The shear (t) and normal (sn), stresses on any three dimensional fracture can be determined by its strike and dip angle. When the Coulomb Failure Function, CFF ¼ t - mfsn, becomes positive with an increase of pore pressure, Pp, during hydraulic fracturing, the originally stable fractures would be activated in shear mode. The critically stressed faults theory proposed by Barton and Zoback [26] was employed to calculate the pore fluid pressure needed to slip natural fractures with arbitrary dip and strike angles [27]. The results showed that the pre-existing natural fractures cannot slip under averaged original stress states (shmin ¼ 142 MPa, sHmax ¼ 176 MPa, sv ¼ 164 MPa), but started to dilate in shear mode when pore fluid pressure, Pp, increased from 120 MPa up to 126 MPa, and most of the fractures dilated for Pp ¼ 135 MPa. The strike-slip faulting stress regime, geochemical properties, and mineral and natural fracture orientation distribution of Kuqa tight sand reservoir demonstrated that complex fracture networks would be created during hydraulic fracturing. This paper integrates the shear slippage criterion with closed natural fractures proposed by Warpinski and Teufel [28], the crossing criterion proposed by Blanton [29], and the re-initiation criterion at the natural fracture tip in order to investigate the complicated intersection behaviors and required conditions. The shear slippage criterion for vertical natural fracture was
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
Z. Luo et al. / Petroleum xxx (2017) 1e8
3
Fig. 1. Pre-existing natural fracture in well K2-14. (a) Examples of natural fractures observed on electrical image, (b) Carbonated cemented natural fracture observed on core sample.
given by Warpinski and Teufel [28], and is expressed in Equation (1).
pðxo Þ > To þ
sHmax þ shmin
jtj t0 þ mf sn pjunc
(1)
Replacing the joint pressure, pjunc, with net pressure, pnet, and expressing the stresses, t and sn, with far-field stresses, shmin and sHmax, the criterion can be expressed as:
pnet
t0 ðsHmax shmin Þ sin 2 b þ mf cos 2 b mf mf
2
sHmax shmin 2
cos 2 b
vðxo Þ sin b a
nðxo Þ ¼
. 2
(5)
"
xo þ l þ a 2 xo l a 2 þ ðxo lÞln xo þ l xo l p 2 # xo þ l þ a þ a ln xo l a 1
ðxo þ lÞln
(6)
(2) p
where t0 is the shear strength of the natural fracture. The crossing criterion for closed vertical natural fractures was given by Blanton. To re-initiate a fracture on the other side of the natural fracture, the pressure, p(xo), would have to overcome the stress, st, acting parallel to the natural fracture plus the tensile strength of matrix rock, To. The stress, st, is composed of the stress, st1, and determined by far-field stresses, and the stress, st2, is determined by the shear stress along the natural fracture [29].
pðxo Þ st þ To ¼ st1 þ st2 þ To
þ
(3)
xo ¼
ð1 þ aÞ2 þ e2mf
!1=2
p
1 þ e2mf
(7)
where △popen is the pressure loss along the length of the opened natural fracture, l. △pshear is the pressure loss along the length of sheared natural fracture, a, and xo is the length between reinitiating point and intersection point. The re-initiation criterion at the dilated natural fracture tip can be depicted as follows.
ptip ¼ pjunc Dpnf sn þ To
(8)
The crossing criterion can be expressed in the form of net pressure, pnet.
where ptip is the fluid pressure at the natural fracture tip, and △pnf is the pressure loss along the natural fracture. Similarly, equation (8) can be given as:
pnet pðxo Þ þ Dpopen þ Dpshear shmin
pnet
Also,
(4)
1 ðs shmin Þð1 cos 2 bÞ þ To þ Dpnf 2 Hmax
(9)
The integrated intersection criteria indicate that hydraulic
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
4
Z. Luo et al. / Petroleum xxx (2017) 1e8
fracture tends to be arrested and re-initiated at the tip of natural fractures with small intersection angles. Fracture also tends to cross natural fractures with large intersection angles in Kuqa tight sand, as Fig. 3 shows. It is easier for fractures with intersection angles above 65 to cross than to slip. The required shear slippage net pressure is negative for fractures with intersection angles between 2 and 55 , which indicates that these fractures are much more likely to be sheared with an increase of pore pressure. All of these criteria indicated that complex fracture networks would be generated during fracturing. Furthermore, the microseismic fracture mapping technology implemented with water fracs in this well block also confirmed the generation of complex fracture networks.
3. Fracture conductivity gained from acid etching
Fig. 2. Statistical analysis: histogram and probability density distribution plot for fracture dip and azimuth, Kuqa fractured tight gas reservoir. (a) fracture dip, (b) fracture azimuth.
In addition to fracture geometry, network fracture conductivity also is a key parameter for hydraulic fracturing. Although sheared fractures have a certain degree of self-support capacity, it is necessary to improve network fracture conductivity in such high closure stress reservoirs. Proppant with low sand ratio has been used in conventional water fracs to prevent fracture from closing and gain propped fracture conductivity. Instead, acid can be used as an alternative stimulation strategy to selectively dissolve carbonate cement and etch fracture faces for most of the pre-existing natural fractures, many of which are cemented with carbonate. Acid etched fracture conductivity is greatly dependent on etching patterns. Referring to the acid etching pattern experiments of carbonate, a rotating disk apparatus and conductivity measurement apparatus with laser scanning were used to observe acid etching patterns of sandstone rock samples [30]. Fig. 4 shows the pre and post etching laser scanning photos of the rock sample from well K2-5 that was etched with a combination of 5 wt% formic, 7 wt % phosphonic, and 2 wt% hydrofluoric with other additives. This combined acid formula has little capacity to etch such sandstone rock samples, which implied that fracturing with this acid formula will form insufficient conductivity. However, the tensile and/or sheared hydraulic fracture face would be rough rather than smooth [31]. Therefore, the acid etching pattern with the rotating disk
Table 1 Geochemical and geomechanical comparison between typical shale gas and Kuqa tight sand. Comparison Items Depth, m Thickness, m Petrophysical Geochemical
Natural Fracture Characteristic
Geomechanical
a b c d e f g h i j
Porosity, % Permeability, 103mm2 Quartz, % Clay, % Carbonate, % Brittleness Fracture Density, 1/m Fracture Aperture, mm Characteristic Young's Modulus, GPa Poisson's Ratio Brittleness
Kuqa Tight Sand 5300e7000 50e300 7 0.07 70.33e95.45 2.3e21.77 1.64e16.87 83.4 0.15e3 <3.42 Sealed with Calcite 21e36 0.23e0.3 54.8
Barnett Shale 1891e2591a 106.7a 5.5 0.00007e0.005 57b 27b 8b 62.0b 2.207c <0.05e0.265c Sealed with Calcitec 33.0c 0.2e0.3c 60e70b
Middle Bakken Shale 2225e3230d 6.1e21.3d 5e 0.04e 26.4f 13f 47.4f 30.4f NAj >0.03e Non-mineralizede 19.9d 0.24d NAj
Eagle Ford Shale 762e4267g 6.1e152.4g 8e18g 1 109-8 104g 15g 20g 55g 16.7g >999/NAh,i,j NAj NAj 6.89e13.79g 0.25e0.27g 40e60g
Data are from Pollastro [20]. Data are from Bowker [6]. Data are from Gale [5]. Data are from Ostadhassan [21]. Data are from Pitman [22]. Data are from Havens [23]. Data are from Stegent [24]. Data are from Offenberger [25]. At least 999 natural fractures intersected by the wellbore of well 1# in Eagle Ford. NA: Not Available published Data.
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
Z. Luo et al. / Petroleum xxx (2017) 1e8
5
Fig. 3. Integrated intersection criteria with averaged geomechanical parameters, SHmax ¼ 176 MPa, Shmin ¼ 142 MPa, To ¼ 5 MPa, t0 ¼ 2 MPa, mf ¼ 0.6 for Kuqa tight sand.
Fig. 4. Laser scanning photos of rock sample from well K2-5 etched 30 min with acid under the condition of 120 centigrade and 7 MPa. (a) Pre-etched, (b) Post-etched.
apparatus is insufficient to draw the conclusion that acid is improper to obtain sufficient conductivity. A pair of sheared rock samples without carbonate cemented fractures from well K2-7 were sheared in order to evaluate the acid etched fracture conductivity with acidizing core flood apparatus at 120 centigrade and 12 MPa. The conductivity experiment indicated that the acid would etch the sheared scratches to generate fracture conductivity, and acid etching could be a viable alternative approach to proppant to gain network fracture conductivity. The acid etched fracture conductivity of Kuqa tight sand was about 1.5 mm2-cm, which is much smaller than the 30e60 mm2-cm for dolomite or limestone etched by hydrochloric acid, as Fig. 5 shows (see Fig. 6). A core penetrated with carbonate cemented fractures from well K5-1 with a diameter of 2.478 cm and a length of 2.711 cm was used to test the effects of acid stimulation at 130 centigrade. There was no fluid at the outlet after 3 h with displacing pressure of 18 MPa for base fluid composed of 4 wt% NH4Cl. Then, chelation acid with a pH of 1.7 was injected, and the displacing pressure decreased sharply from 15.992 MPa. Moreover, the fluid started to flow out after injection acid at a time of 25min, while the flow rate was found to increase obviously and up to 1 ml/min at 31.5min and 80min, respectively. Eventually, the base fluid composed of 4 wt% NH4Cl was injected at a pumping rate of 1 ml/min and a displacing
Fig. 5. Acid etched fracture conductivity of sheared core without carbonate cemented fracture from well K2-7 and well K2-5.
pressure of 4e6 MPa. The pre- and post-acid photos of this carbonate cemented fractured core indicated that some wormholelikely flow channels could be formed with acid. The etching experiment with the rotating disk, as well as the acidizing core flood tests for artificial sheared fractures and
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
6
Z. Luo et al. / Petroleum xxx (2017) 1e8
carbonate cemented fracture, demonstrated that acid could be used to generate fracture conductivity in networks. Because shear slippage and acid etching can create complex fracture networks, it can be concluded that network acid fracturing could be a viable alternative stimulation strategy in certain tight gas reservoirs.
4. Pilot test PetroChina Tarim Oil Company has been introducing fiber assisted re-orientation SRV to its wells. Thus far, this technique has been applied in 7 wells, including 2 wells with microseismic fracture mapping [15] in Kuqa tight gas since 2013. The degradable fiber enhanced the performance of SRV with average post-frac open-flow capacity about 320 104 m3/d, but the technique is very costly. Fiber-ball and fiber-grain assisted multistage network acid fracturing could be an alternative stimulation strategy for fiber assisted water fracs. In order to verify the feasibility of network acid fracturing, 17 wells have been selected to be tested with such a stimulation strategy. The post-frac production performances were shown to be comparable to fiber assisted SRV. Well K2-18 is located in the K2 structure, within the Cretaceous tight sandstone range from 6676 to 6858 m. The well has 12 zones with a total thickness of 13.5 m and porosity between 3 and 5%, and 47 zones with thickness of 107 m and porosity above 5%. The in-situ stress regime is strike-slip faulting with high horizontal stress anisotropy, where the maximum and minimum horizontal stress gradients and vertical stress gradient are about 2.68, 2.18, and 2.45 MPa/100 m, respectively. Imaging logging showed 47 pre-existing natural fractures with apertures larger than 1 mm in the targeted interval, and oriented nearly EW with a density of 0.258 per meter. Core sample observation showed that almost all of these fractures were either partially or completely cemented with carbonate. The azimuth of natural fractures indicated that the intersection angles between natural fractures and hydraulic fracture, which was parallel to SHmax oriented SW-NE, range from 40 to 50 . Most of the intersected pre-existing natural fractures will dilate as bottom hole pressure gradient is increased to 2.15 MPa/100 m, as shown in Fig. 7. The stimulation strategy is intended to link natural fractures with hydraulic fractures to form complex fracture networks and generate fracture conductivity with sequential injection of hydrochloric and hydrofluoric contained acid to dissolve carbonate cement and drilling mud solids. Fig. 8 shows the pressure response of the acid treatment of well K2-18. The tubing pressure decreased sharply at the pumping rate when formation rock came into contact with pad acid, which is composed of 9 wt% hydrochloric and 3 wt% acetic. This suggests that the carbonate cement was dissolved by acid, and fracture conductivity was greatly enhanced. Degradable fiber grains and fiber powder were used as diverters to improve zone acid coverage. The well head tubing pressure increased by 4.6 MPa at a rate of 1.2 m3/min in the first diversion, and 7.6 MPa at a rate of 4.2 m3/min in the second diversion. After the successful execution of network acid fracturing, the well was flowed back for cleanup and rate assessment. The well initially flowed 43.64 104 m3/d with 84.8 MPa of flowing wellhead pressure. This initial result is comparable to the flow-back results of offset wells in the area that were stimulated with fiber assisted SRV. Table 2 displays and compares the net-pay thickness, porosity
Fig. 6. Acidizing core flood experiment for sample with carbonate cemented fractures from well K5-1. (a) Displacing pressure versus time, (b) lateral faces of pre-test core, (c) lateral faces of post-test core from almost the same view of (b), (d) pre- and post-test inlet face, and (e) pre- and post-test outlet face.
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
Z. Luo et al. / Petroleum xxx (2017) 1e8
7
Fig. 7. Coulomb failure function of natural fractures in well K2-18 with increasing pore fluid pressure, white dots represent activated natural fractures: (a) 1.95 SG, (b) 2.05 SG, and (c) 2.15 SG.
Fig. 8. Acid treatment of well K2-18. The dotted circle shows the increasing of well head tubing pressure for injection of fiber grains with powder.
Table 2 Comparison of network acid fracturing and fiber assisted SRV. SS
Well Name
NPT, m
(4H)5, m-%
(4H)3, m-%
NFD, 1/m
q,
AOF, 104 m3/d
Fiber Assisted SRV
K1-1 K1-8 K1-14 K2-8 K2-12 K2-14 K5-1 K1-6 K2-3 K2-18 K8-1 K8-3
114 112 104.5 178.5 140 148.5 157 167.5 129.5 140.5 150.5 117.5
606.9 618.15 604.95 706.65 654.75 739.2 383.25 586.8 323.85 615.2 520.85 665.45
864.1 833.9 733.3 1253.25 933.35 1103.85 871.95 1067.25 795.35 943.9 949.6 863
0.21 0.24 0.67 0.63 0.7 0.291 0.43 0.67 0.1 0.258 0.10 0.31
60 35 26 33 20 30 15 11 70 40 15 15
55 223 208 686 407 447 217 550 60 465 350 570
Network Acid Fracturing
multiplied by thickness with porosity higher than 3% and 5%, natural fracture density, intersection angle between hydraulic fracture and pre-existing natural fracture, and post-stimulation open flow capacity of wells located in K1-K2 and K8 structure. The overall post stimulation open flow capacity of wells with network acid fracturing is comparable to those stimulated with fiber assisted SRV.
Furthermore, the adjacent wells of K2-14 and K2-18 located in the K1-K2 structure with similar reservoir properties obtained almost the same stimulation effect. These pilot wells demonstrated that the network acid fracturing strategy is an effective technology for Kuqa tight gas reservoir. However, either SRV or acid fracturing performed disappointingly for wells with improper natural fracture
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002
8
Z. Luo et al. / Petroleum xxx (2017) 1e8
orientation, such as well K1-1 and K2-3, suggesting higher intersection angles between hydraulic fracture and natural fractures in these wells. SS: stimulation strategy; NPT: Net-pay thickness; (4H)n: porosity multiplied by thickness with porosity higher than n%, n ¼ 3, 5; NFD: natural fracture density; q: intersection angle between pre-existing natural fracture and hydraulic fracture; AOF: absolutely open flow capacity. 5. Results and discussion For high horizontal stress anisotropy reservoirs, interface slipping of natural fractures is profitable for stimulations to create complex network fracture, but natural fractures with intersection angles higher than 60 are difficult to dilate and contribute little to stimulation effects. For carbonate-cemented natural fractures, acid can be used as an alternative approach to proppant for obtaining fracture conductivity. A novel and economic stimulation strategy, network acid fracturing, has been proposed and implemented in carbonate-cemented fractured tight gas reservoirs. Poststimulation production performances of wells stimulated with network acid fracturing are comparable to those treated with fiber assisted SRV. The fiber assisted SRV is more suitable for poor formation dominated reservoirs, while network acid fracturing results in higher production for reservoirs with more desirable properties. The dilation of natural fractures with shear mode at high stress anisotropy was concluded only from intersection criteria in this paper, and should be verified with further laboratory experiments. The present designs of volume of acid and slick water for network acid fracturing were determined empirically rather than theoretically. The simulation of network fracture propagation with acidrock reaction for optimized design is in progress and will be published in a coming paper. Acknowledgments The authors thank financial support from the National Natural Science Fund(51404207) and the management of the PetroChina Tarim Oil Company for permission to publish this article. Special thanks to the technical teams in the Oil & Gas Engineering Institute of PetroChina Tarim Oil Company for their assistance in providing and evaluating data, assessing well performance, and implementation of new technology in the field. References [1] Fisher, M.K., C.A. Wright, and B.M. Davidson et al., Integrating Fracture Mapping Technologies to Optimize Stimulations in the Barnett Shale. Paper SPE 77441 Presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, 29 Sept.-2 Oct. 2002. [2] Maxwell, S.C., T.I. Urbancik, and N.P. Steinsberger et al., Microseismic Imaging of Hydraulic Fracture Complexity in the Barnett Shale. Paper SPE77440 Presented at the SPE Annual Technology Conference and Exhibition, San Antonio, Texas, USA, 29 Sept.-2 Oct. 2002. [3] M.K. Fisher, J.R. Heinze, C.D. Davidson, et al., Optimizing horizontal completion techniques in the Barnett shale using microseismic fracture mapping, in: Paper SPE 90051 Presented at the SPE Annual Technical Conference and Exhibition, Houston, Sept. 2004, pp. 26e29. [4] M.J. Mayerhofer, E.P. Lolon, N.R. Warpinski, et al., What is stimulated reservoir volume? SPE Prod. Oper. 25 (1) (2010) 89e98. [5] J.F. Gale, R.M. Reed, J. Holder, Natural fractures in the Barnett shale and their importance for hydraulic fracture treatments, AAPG Bull. 91 (4) (2007) 603e622. [6] K.A. Bowker, Recent development of the Barnett shale play, Fort Worth Basin
West Texas Geol. Soc. Bull. 42 (6) (2003) 1e11. [7] W. Weng, O. Kresse, C. Cohen, et al., Modeling of hydraulic-fracture-network propagation in a naturally fractured formation, SPE Prod. Oper. 26 (04) (2011) 368e380. [8] H. Gu, X. Weng, J. Lund, et al., Hydraulic fracture crossing natural fracture at nonorthogonal angles: a criterion and its validation, SPE Prod. Oper. 27 (01) (2012) 20e26. [9] Gu, H., and X. Weng, Criterion for fractures crossing frictional interfaces at non-orthogonal angles. Paper ARMA 10e198 Presented at the 44th U.S. Rock Mechanics Symposium and 5th U.S. Canada Rock Mechanics Symposium, Salt Lake City, Utah, USA, 27-30 June 2010. [10] Olson, J.E. and A. Dahi-Taleghani. Modeling simultaneous growth of multiple hydraulic fractures and their interaction with natural fractures. Paper SPE 119739 presented at the SPE Hydraulic Fracturing Technology Conference, The Woodlands, Texas, USA, 19e21 Jan. 2009. [11] A. Dahi-Taleghani, J.E. Olson, Numerical modeling of multistranded-hydraulicfracture propagation: accounting for the interaction between induced and natural fractures, SPE J. 16 (03) (2011) 575e581. [12] G. Schein, The Application and Technology of Slickwater Fracturing, 2005. [13] Stegent, N.A., L.M. Leotaud, and W. Prospere et al. Cement technology improves fracture initiation and leads to successful treatments in the Eagle Ford shale. SPE 137441 Presented at the SPE Tight Gas Completions Conference Held in San Antonio, Texas, USA, Nov.2010. [14] N. Li, J. Dai, C. Liu, et al., Feasibility study on application of volume acid fracturing technology to tight gas carbonate reservoir development, Petroleum 1 (3) (2015) 206e216. [15] F. Gui, K. Rahman, D. Moos, et al., Optimizing hydraulic fracturing treatment integrating geomechanical analysis and reservoir simulation for a fractured tight gas reservoir, Tarim Basin, China, in: ISRM International Conference for Effective and Sustainable Hydraulic Fracturing, International Society for Rock Mechanics, 2013. [16] Xuefang Yuan, Fei Liu, et al., Pilot tests of degradable fiber assisted reorientation stimulated reservoir volume (in Chinese), in: Conference for Unconventional Tight Oil & Gas Exploration Technologies, Xinjiang Petroleum Institute, 2013. [17] Sondergeld, C.H., and K.E. Newsham, et al. Petrophysical considerations in evaluating and producing shale gas resources. SPE-131768-MS presented at the SPE unconventional Gas Conferences held in Pittsburgh, Pennsylvania, USA, Feb.2010. [18] Rickman, R., and M. Mullen, et al. A practical use of shale petrophysics for stimulation design optimization: all shale plays are Not clones of the Barnett shale. SPE-115258-MS presented at the SPE Annual Technical Conference and Exhibition held in Denver, Colorado, USA, Sept.2008. [19] M.D. Zoback, Reservoir Geomechanics, Cambridge University Press, 2007. [20] R.M. Pollastro, R.J. Hill, D.M. Jarvie, et al., Assessing Undiscovered Resources of the Barnett-paleozoic Total Petroleum System, Bend Arch-Fort Worth Basin Province, Texas, 2003. [21] M. Ostadhassan, Geomechanics and Elastic Anisotropy of the Bakken Formation, Doctor Dissertation, University of North Dakota, Williston Basin, 2013. [22] J.K. Pitman, L.C. Price, et al., Diagenesis and Fracture Development in the Bakken Formation, Williston Basin: Implications for Reservoir Quality in the Middle Member, US Department of the Interior, US Geological Survey, 2001. [23] J. Havens, Mechanical Properties of the Bakken Formation, Master Dissertation, Colorado School of Mines, 2012. [24] Stegent, N.A., and A.L. Wagner et al. Engineering a successful fracturestimulation treatment in the Eagle Ford shale. Paper SPE-136183-MS presented at the SPE Tight Gas Completions Conference held in San Antonio, Texas, USA, Nov.2010. [25] Offenberger, R., and N. Ball et al. Integration of natural and hydraulic fracture network modeling with reservoir simulation for an Eagle Ford well. Paper SPE 168683 /URTeC1563066 Presented at the Unconventional Resources Technology Conference held in Denver, Colorado, USA, Aug. 2013. [26] C.A. Barton, M.D. Zoback, et al., Fluid flow along potentially active faults in crystalline rock, Geology 23 (8) (1995), 683e683. [27] M.W. [26] McClure, R.N. Horne, Investigation of injection-induced seismicity using a coupled fluid flow and rate/state friction model, Geophysics 76 (6) (2011). WC181-WC198. [28] N.R. Warpinski, L.W. Teufel, Influence of geologic discontinuities on hydraulic fracture propagation (includes associated papers 17011 and 17074), J. Petrol. Technol. 39 (2) (1987) 209e220. [29] T.L. Blanton, Propagation of hydraulically and dynamically induced fractures in naturally fractured reservoirs, in: SPE Unconventional Gas Technology Symposium, SPE15261, 1986. [30] C. Malagon, M. Poumik, A.D. Hill, The texture of acidized fracture surfaces: implications for acid fracture conductivity, SPE Prod. Oper. 23 (3) (2008) 343e352. [31] C. Dong, D. Zhu, A.D. Hill, Modeling of the acidizing process in naturally fractured carbonates, SPE J. 7 (04) (2002) 400e408.
Please cite this article in press as: Z. Luo, et al., A novel stimulation strategy for developing tight fractured gas reservoir, Petroleum (2017), https://doi.org/10.1016/j.petlm.2017.12.002