Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Experimental investigation of shale imbibition capacity and the factors influencing loss of hydraulic fracturing fluids

Experimental investigation of shale imbibition capacity and the factors influencing loss of... Pet. Sci. (2015) 12:636–650 DOI 10.1007/s12182-015-0049-2 ORIGINAL PAPER Experimental investigation of shale imbibition capacity and the factors influencing loss of hydraulic fracturing fluids 1 1 1 1 1 • • • • • Hong-Kui Ge Liu Yang Ying-Hao Shen Kai Ren Fan-Bao Meng 1 1 Wen-Ming Ji Shan Wu Received: 11 May 2015 / Published online: 21 September 2015 The Author(s) 2015. This article is published with open access at Springerlink.com Abstract Spontaneous imbibition of water-based frac- water greater than their measured pore volume. The aver- turing fluids into the shale matrix is considered to be the age ratio of the imbibed water volume to the pore volume main mechanism responsible for the high volume of water is approximately 1.1 in the Niutitang shale, 1.9 in the loss during the flowback period. Understanding the matrix Lujiaping shale, 2.8 in the Longmaxi shale, and 4.0 in the imbibition capacity and rate helps to determine the frac- Yingcheng volcanic rock, and this ratio can be regarded as turing fluid volume, optimize the flowback design, and to a parameter that indicates the influence of clay. In addition, analyze the influences on the production of shale gas. surfactants can change the imbibition capacity due to Imbibition experiments were conducted on shale samples alteration of the capillary pressure and wettability. A 10 from the Sichuan Basin, and some tight sandstone samples wt% KCl solution can inhibit clay absorption to reduce the from the Ordos Basin. Tight volcanic samples from the imbibition capacity. Songliao Basin were also investigated for comparison. The effects of porosity, clay minerals, surfactants, and KCl Keywords Imbibition  Shale  Fracturing fluid solutions on the matrix imbibition capacity and rate were Capillary pressure  Clay systematically investigated. The results show that the imbibition characteristic of tight rocks can be characterized by the imbibition curve shape, the imbibition capacity, the 1 Introduction imbibition rate, and the diffusion rate. The driving forces of water imbibition are the capillary pressure and the clay Multistage hydraulic fracturing is a critical technology for absorption force. For the tight rocks with low clay contents, economic production from shale reservoirs. Large amounts the imbibition capacity and rate are positively correlated of water-based fracturing fluids are pumped into forma- with the porosity. For tight rocks with high clay content, tions, generating extensive fracture networks and stimu- the type and content of clay minerals are the most impor- lating low-permeability formations. Field operations have tant factors affecting the imbibition capacity. The imbibed demonstrated that large volumes of injected fluids are water volume normalized by the porosity increases with an retained in shale formations, with a flowback efficiency of increasing total clay content. Smectite and illite/smectite lower than 30 % (Makhanov et al. 2014). In the U.S. tend to greatly enhance the water imbibition capacity. Haynesville shale formation, the flowback rate is even Furthermore, clay-rich tight rocks can imbibe a volume of lower than 5 % after fracturing operations (Penny et al. 2006). Besides possibly causing a series of environmental problems, the retention of fracturing fluids in shale for- & Liu Yang mations can greatly enhance the water saturation near shidayangliu@126.com fracture surfaces and influence two-phase fluid flow, thus State Key Laboratory of Petroleum Resources and further inhibiting the production of shale gas (Sharma and Prospecting, China University of Petroleum, Beijing 102249, Agrawal 2013). Furthermore, intense interaction between China fluid and shale can dramatically change rock properties and impact on the generation of fracture networks during Edited by Yan-Hua Sun 123 Pet. Sci. (2015) 12:636–650 637 fracturing (Yuan et al. 2014). Therefore, studying the This paper focuses on the imbibition capacity and the imbibition capacity and its main controlling factors is influence of the mineral composition and physical proper- essential to understanding reservoir performance and ties of tight rocks. Samples include gas shales from the optimizing fracturing operations. Sichuan Basin, tight sandstones from the Ordos Basin, and It is generally believed that spontaneous imbibition of tight volcanic rocks from the Songliao Basin. Experiments fracturing fluids into the shale matrix plays an important can be divided into three groups. In group 1, the imbibition role in water loss. Many researchers have focused on the capacity and rate of deionized water uptake are investi- mechanism of fracturing fluid imbibition. Makhanov et al. gated systematically. In group 2, each sample is immersed (2012) found that imbibition rates perpendicular and par- repeatedly in deionized water several times to address the allel to the bedding plane are different, and the latter is water sensitivity of different rocks. In group 3, comparative higher. Hu et al. (2012) considered that the Barnett shale experiments are conducted to explore the effects of dif- has a poor connectivity, which greatly influences the flow ferent fluids on the imbibition capacity. and diffusion of fluid. Roychaudhuri et al. (2013) deter- mined that a surfactant can effectively reduce the imbibi- tion rate of fracturing fluids, and the driving force of 2 Experimental imbibition is the capillary pressure. Dehghanpour et al. (2013) mentioned that the amount of imbibition in shale is 2.1 Rock samples and fluids positively related to mineral composition and physical properties. Fakcharoenphol et al. (2014) investigated the Sixty-six shale and tight gas rock samples from the Ordos effects of salinity on water imbibition and found that the Basin, Songliao Basin, and the Sichuan Basin were used to osmotic pressure can act as the driving force for water conduct comparative imbibition experiments, and reservoir intake. Currently, it is well known that the imbibition of rock properties are presented in Table 1. The mineral fracturing fluids is mainly controlled by the capillary composition (in wt%) of the shale and tight gas rock pressure, while the effects of clay absorption have not been samples and the relative abundance of clay minerals are studied thoroughly. The imbibition capacity, imbibition listed in Table 2. The samples were neither cleaned nor rate, and other influencing factors in shale reservoirs have exposed to air beforehand. According to the observed not been investigated systematically. brittleness of rocks, the samples were machined into Table 1 Tight reservoir Label Formation Lithology Depth, m Source Geological age properties in this study S Shihezi Tight sandstone 2120 Erdos Basin Early Permian H Huoshiling Tight volcanic 2523 Songliao Basin Lower Jurassic UY Upper Yingcheng Tight volcanic 3524 Songliao Basin Lower Jurassic LY Lower Yingcheng Tight volcanic 3557 Songliao Basin Lower Cretaceous L Lujiaping Shale 1235 Sichuan Basin Lower Cambrian LM Longmaxi Shale 786 Sichuan Basin Lower Silurian N Niutitang Shale 895 Sichuan Basin Lower Cambrian Table 2 Results of XRD mineralogy analysis Label Mineral composition, wt% Relative abundance, % TOC, wt% Quartz Feldspar Calcite Dolomite Clay Smectite Illite I/S Chlorite Kaolinite S 32.2 26.4 5.1 25.8 10.3 0 100.0 0 0 0 0 H 1.3 61.5 3 0 34.2 0 10.5 0 89.5 0 0 UY 13.2 11.9 0 0 74.9 0 16.8 74.9 8.3 0 1.1 LY 40.6 11.6 0 0 47.8 0 7.9 78.0 11.1 2.9 1.2 L 29.4 7.2 24.7 14.9 23.7 7.6 23.6 53.2 8.0 7.6 3.1 LM 40.3 8.8 7.5 6.5 36.9 4.3 15.9 62.3 8.7 8.7 3.6 N 31.2 15.8 11.5 18.2 23.3 3.4 5.2 78.9 12.4 0 2.5 I/S is Illite/smectite mixed-layer 123 638 Pet. Sci. (2015) 12:636–650 cylindrical or prismatic core plugs, as shown in Tables 3, 4, after the imbibition experiments. The experimental and 5. The effects of the sample size and shape can be data were normalized by the scaling method, which normalized by the scaling method. is described in the other sections. Deionized water, 10 wt% KCl solution, and an aqueous The experimental procedure of the group 2 experiments solution of cationic surfactant were used as the imbibing is the same as that of the group 1 experiments. Each sample fluids. Properties of the test fluids are listed in Table 6. The was immersed in deionized water and dried repeatedly most commonly used boundary conditions for imbibition several times. The basic sample data are presented in are one-end-open (OEO), all-faces-open (AFO), and two- Table 4. ends-open (TEO). Considering the effect of lamination on In group 3, samples of the same formation were the imbibition rate, for 1-D imbibition (OEO and TEO), the acquired from the same core to reduce the influences of open face is parallel to the bedding plane to maintain the heterogeneity. A total of 18 samples were dried for 24 h same experimental conditions. and submerged into different fluids until there was no further change in weight. This process lasted approxi- mately 7 days. The basic information about the samples is 2.2 Experimental apparatus and procedure shown in Table 5. In group 1 experiments, the experimental results are sen- sitive to test environments and the instrumental error due to 3 Experimental data and analysis the relatively low imbibition rate in shale and tight rock samples. Therefore, a series of measurements are required 3.1 Scaling method for experimental data to improve the measurement accuracy: normalization (1) All of the samples were weighted by an analytical balance (Mettler XPE205) with an accuracy of The samples used had different sizes and shapes. Charac- 0.00001 g. terization methods need to be developed to normalize the (2) Impermeable and nonelastic strings used to suspend effects of size and shape and represent the imbibition core slugs had a diameter of approximately capacity and rate. 0.13 mm, which could avoid the error caused by (1) The imbibition capacity can be determined based on the reduction in the liquid volume. the curve of water volume gain per pore volume (3) The experimental device was placed in a chamber versus time. with constant temperature and humidity to lower the (2) Handy (1960) established a famous gas–water imbi- effect of variable external temperature and humidity. bition model, which is given by the following The experimental device is shown in Fig. 1. equation: (4) The whole apparatus was placed in the basement of a sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi building to minimize the vibration from the ground pffiffi 2P /KS c wf surface. V =A ¼ t; ð1Þ imb c The experimental procedure is as follows: where V is the volume of imbibed water, cm ; P imb c (1) The initial dimensions and mass of the core slug is the capillary pressure, MPa; / is the porosity, %; were measured before experiment. K is the permeability, mD; S is the water satura- wf (2) Impermeable epoxy was used to satisfy TEO and tion, %; A is the imbibition cross-sectional area, OEO, imbibition boundary conditions. cm ; l is the fluid viscosity, mPa s; and t is the (3) The core slug was dried in an oven at 105 C until imbibition time, s. there was no further change in weight. (4) After the core slug cooled down, the sample was The slope of the volume of imbibed fluid per sectional suspended on the analytical balance. The sample was area versus the square of time, A , can be used to represent totally submerged into water by adjusting the liquid the imbibition rate, which can be obtained from experi- level. mental data (Makhanov et al. 2014). (5) The variation of the sample mass with time was According to Eq. (1), the imbibition rate can be given as measured and then recorded on a computer as water follows: sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi was spontaneously imbibed into the sample. 2P /KS c wf (6) The mass of the imbibed water was calculated by A ¼ : subtracting the initial mass from the mass recorded 123 Pet. Sci. (2015) 12:636–650 639 Table 3 Basic properties of core plugs used in group 1 a b c No. Shape Cross-sectional area Length Permeability Porosity Boundary condition Imbibition rate Imbibed volume per 2 0.5 d A ,cm L, cm K,mD U,% A , cm/h sample volume c i C,% S-1 Cylinder 5.1 5.1 2.1 12.3 OEO 0.1069 8.6 S-2 Cylinder 5.0 5.1 2.1 13.0 TEO 0.1123 8.9 S-3 Cylinder 5.1 5.0 2.2 12.8 OEO 0.1173 9.5 H-1 Cylinder 4.9 1.1 0.0028 12.5 TEO 0.0382 11.4 H-2 Cylinder 5.1 1.9 0.0045 8.4 OEO 0.0467 8.0 H-3 Cylinder 5.0 0.9 0.0031 9.6 TEO 0.0403 9.6 H-4 Cylinder 4.9 0.8 0.0069 9.7 TEO 0.0356 8.6 H-5 Cylinder 5.0 1.6 0.0083 13.6 TEO 0.0368 11.1 H-6 Cylinder 5.0 1.6 0.0034 14.1 TEO 0.0466 12.0 H-7 Cylinder 4.9 1.0 0.0096 10.8 TEO 0.0375 10.3 H-8 Cylinder 5.1 0.9 0.0069 10.1 TEO 0.0381 9.1 UY-1 Cylinder 4.9 0.6 0.0012 0.3 TEO 0.0017 2.7 UY-2 Cylinder 4.9 0.6 0.0013 0.4 TEO 0.0099 2.8 UY-3 Prism 9.0 0.5 0.0023 0.4 TEO 0.0021 3.2 UY-4 Cylinder 4.9 0.6 0.0012 0.5 TEO 0.0060 2.8 LY-1 Cylinder 4.9 0.5 0.0032 3.3 TEO 0.0044 4.7 LY-2 Cylinder 4.9 0.5 0.0025 1.6 TEO 0.0055 4.5 LY-3 Cylinder 4.9 0.8 0.0007 2.2 TEO 0.0053 4.7 LY-4 Cylinder 4.9 0.8 0.0012 2.1 TEO 0.0029 3.9 LY-5 Cylinder 4.9 0.5 0.0007 1.9 TEO 0.0037 3.4 LY-6 Cylinder 4.9 0.5 0.0031 0.9 TEO 0.0016 2.6 LY-7 Prism 10.7 0.5 0.0011 0.6 TEO 0.0008 1.8 L-1 Cylinder 40.4 0.8 0.0021 1.29 TEO 0.0015 2.7 L-2 Cylinder 40.4 1.0 0.0035 1.5 TEO 0.0017 3.5 L-3 Cylinder 40.4 1.1 0.0042 2.0 TEO 0.0020 4.1 L-4 Cylinder 29.2 1.1 0.0014 1.8 TEO 0.0015 2.2 L-5 Cylinder 29.0 0.9 0.0027 0.9 TEO 0.0016 1.9 L-6 Cylinder 29.4 1.0 0.0032 2.4 TEO 0.0018 4.8 L-7 Cylinder 29.0 1.1 0.0018 1.1 TEO 0.0012 2.0 L-8 Cylinder 29.5 0.6 0.0021 2.1 TEO 0.0012 3.8 LM-1 Prism 6.2 0.5 0.0046 2.0 TEO 0.0037 5.7 LM-2 Cylinder 30.1 0.7 0.0053 1.4 TEO 0.0035 1.4 LM-3 Cylinder 31.3 0.9 0.0038 1.5 TEO 0.0025 4.5 LM-4 Cylinder 31.3 0.9 0.0062 2.3 TEO 0.0020 6.5 LM-5 Prism 6.1 0.7 0.0071 2.2 TEO 0.0028 5.0 LM-6 Prism 6.0 0.4 0.0031 1.9 TEO 0.0041 6.1 N-1 Cylinder 29.8 0.7 0.0026 2.1 TEO 0.0008 2.0 N-2 Cylinder 29.9 0.7 0.0038 1.4 TEO 0.0006 1.3 N-3 Cylinder 29.9 0.7 0.0035 1.9 TEO 0.0007 1.7 N-4 Cylinder 29.3 0.6 0.0025 2.4 TEO 0.0008 2.9 N-5 Cylinder 29.0 0.6 0.0019 2.7 TEO 0.0010 3.5 N-6 Cylinder 29.4 0.6 0.0029 3.1 TEO 0.0011 3.2 Permeability was obtained using a nitrogen pressure pulse decay permeability porosimeter Porosity was measured in a helium porosimeter Imbibition rate was obtained from the experimental data Imbibed volume per sample volume was obtained from the experimental data 123 640 Pet. Sci. (2015) 12:636–650 3.2 Imbibition curve characteristics The log–log relationship can be given by the following equation: The imbibition characteristics of tight rocks can be char- 1 2P KS 1 c wf acterized by the imbibition capacity, the imbibition rate, log A ¼ log þ log /; ð2Þ 2 l 2 w and the diffusion rate. Figure 2 shows plots of the volume normalized by the 1 2P /S 1 c wf log A ¼ log þ log K: ð3Þ pore volume versus time. The cumulative imbibed volume 2 l 2 increases with time. However, the rate of water intake The effective imbibition driving force can be given as obviously slows down with increasing time, and the rate follows: approximately reaches zero, which represents the equilib- rium condition (Sun et al. 2015). However, some curves A l l i w w P ¼ ¼ a ; ð4Þ may have ‘‘upward tails,’’ which demonstrate an obvious 2K/S 2S wf wf diffusion effect and may be related to the complex pore where a is the driving force coefficient, 1/s. a reflects the structure in these rock samples, as shown in Fig. 2f. effect of difference between imbibition driving force and Though the size and shape of core samples vary signifi- friction resistance, which can be obtained by experiments. cantly, the water volume normalized by the pore volume Equations (2)–(4) can be used to analyze contributing can represent the imbibition capacity well. It is worth factors to the imbibition rate, which is discussed in the noting that the highest points in the curves may not always following sections. remain at the same value, which may be explained by the heterogeneity of core samples as shown in Fig. 2c, d, e, f. Table 4 Basic properties of core slugs used in group 2 In addition, the volume gain fluctuates in some samples, which may be due to the large amount of water-sensitive No. Shape Dry mass, g Boundary condition I/S mixed-layer in this tight rock, as shown in Fig. 2c, e, f. S-4 Cylinder 69.5 AFO This effect will be discussed in detail later. H-9 Cylinder 13.9 AFO Figure 3 shows plots of the imbibed volume normalized UY-5 Cylinder 11.5 AFO by the sectional area versus the square root of time. The L-9 Cylinder 14.0 AFO effect of the sectional area is normalized well. The mea- LM-7 Prism 15.0 AFO surements tend to sit close to a smooth curve that repre- N-7 Cylinder 57.5 AFO sents the imbibition rate as shown in Fig. 3a, b, d, e. In Table 5 Basic properties of No. Shape Cross-sectional area Length Imbibing fluid Boundary condition core slugs used in group 3 2 A cm L,cm c, S-5 Cylinder 5.1 5.0 Deionized water AFO S-6 Cylinder 5.1 5.1 2.5 wt% surfactant AFO S-7 Cylinder 4.9 5.1 10 wt% KCl brine AFO H-10 Cylinder 4.9 1.2 Deionized water AFO H-11 Cylinder 4.9 1.2 2.5 wt% surfactant AFO H-12 Cylinder 4.9 1.0 10 wt% KCl brine AFO UY-6 Cylinder 5.1 1.1 Deionized water AFO UY-7 Cylinder 5.0 1.0 2.5 wt% surfactant AFO UY-8 Cylinder 5.1 1.1 10 wt% KCl brine AFO L-10 Cylinder 40.4 1.2 Deionized water AFO L-11 Cylinder 40.4 1.1 2.5 wt% surfactant AFO L-12 Cylinder 40.3 1.0 10 wt% KCl brine AFO LM-8 Prism 6.1 0.9 Deionized water AFO LM-9 Prism 6.0 1.0 2.5 wt% surfactant AFO LM-10 Prism 6.1 0.9 10 wt% KCl brine AFO N-8 Cylinder 29.4 1.1 Deionized water AFO N-9 Cylinder 29.4 1.0 2.5 wt% surfactant AFO N-10 Cylinder 29.3 1.1 10 wt% KCl brine AFO 123 Pet. Sci. (2015) 12:636–650 641 Table 6 Properties of fluids at 3 Fluids Density, g/cm Viscosity, cP Surface tension, N/m 25 C used in group 3 Deionized water 0.998 1 0.072 10 wt% KCl 1.06 0.88 0.074 2.5 wt% surfactant 0.96 1.1 0.058 Blance String (0.13 mm) Computer Liquid Fig. 1 Schematic for imbibition experiments in the OEO condition Fig. 3c, e, the departure from a smooth curve can be consistent with the results of Zhou et al. (2014). There is a explained by the strong heterogeneity in these tight rocks. significant difference between these results and the con- The imbibition curves generally behave similarly. Each ventional opinion that the water saturation should be 0–1. profile is divided into three regions: the initial linear Therefore, the capacity for fracturing fluid intake in tight imbibition region (Region 1), the transition region (Region reservoirs may greatly exceed the conventional estimation. 2), and the diffusion region (Region 3), as shown in Fig. 4 The relative relationship of R is UY tight volcanic[ LM (Lan et al. 2014). However, the curve characteristics of shale [ LY tight volcanic [ L shale[ N shale [ H tight different tight rocks vary significantly, which may be volcanic [ S tight sandstone. attributed to the pore size distribution and pore connec- Figure 5b presents the distribution of the imbibition tivity. The slope in Region 1 represents the imbibition rate rate. The imbibition rates in the S and H formations are 0.5 (A ), where the capillary pressure may be the primary approximately 0.05–0.1 cm/h , which is obviously higher driving force. As the water saturation increases, the capil- than the rates for the LY, LM, N, and L formations. The lary pressure decreases, and the water intake process begins relative relationship of the imbibition rate is S tight sand- to enter Region 2. In Region 3, diffusion is the primary stone [ H tight volcanic[ UY tight volcanic [ LY tight flow driving force compared with capillary imbibition in volcanic [ LM shale [ L shale [ N shale. Region 1. In tight reservoir rocks, the diffusion effect is It is worth noting that the imbibed volume exceeds the more obvious than that in conventional reservoir rocks. It pore volume in rocks, so the water presumably entered into can be represented by the diffusion rate (A ). However, a a space where the gas used to measure the porosity could study of the diffusion rate is not included in this paper. not enter. Another explanation is that absorption of water by the clay can induce cracks and increase the pore vol- 3.3 Imbibition capacity and rate ume. The high water imbibition capacity in tight rocks may be related to clay minerals. This is consistent with the well- The ratios of the maximum imbibed volume to the pore known phenomenon of wellbore instability in well drilling. volume, R exceed 1 in the UY, LY, L, LM, and N for- The experiments in group 2 were conducted to explore the mations, as shown in Fig. 5a. In particular, for the effects of clay minerals on the imbibition process. A single UY formation, R is 6–8. In other words, the imbibed vol- sample was tested several times repeatedly. ume of the rock in the tight reservoir can exceed the pore The samples in the S, H, L, and N formations have good volume of the rock by a factor of 6–8. This result is reproducibility, which illustrates that the inner pore volume Temperature and humidity chamber 642 Pet. Sci. (2015) 12:636–650 0.9 1.2 0.8 1.0 0.7 0.6 0.8 H-1 S-1 0.5 H-2 0.6 H-3 S-2 0.4 H-4 0.3 H-5 S-3 0.4 H-6 0.2 H-7 0.2 H-8 0.1 0 0 010 20 30 40 50 024 6 8 10 12 14 16 18 Time, h Time, h (a) (b) 9 2.5 UY-1 UY-2 2.0 L-1 UY-3 L-2 6 UY-4 L-3 Clay-adsorption 1.5 Clay-adsorption LY-1 L-4 LY-2 L-5 LY-3 L-6 1.0 LY-4 L-7 LY-5 L-8 LY-6 0.5 LY-7 05 10 15 20 25 30 35 40 45 020 40 60 80 100 120 140 160 180 200 Time, h Time, h (c) (d) 1.4 4.0 Clay-adsorption 3.5 1.2 3.0 0.1 2.5 LM-1 0.8 N-1 2.0 LM-2 N-2 Clay-adsorption 0.6 LM-3 N-3 1.5 LM-4 N-4 0.4 1.0 N-5 LM-5 0.5 N-6 0.2 LM-6 010 20 30 40 50 60 70 80 90 020 40 60 80 100 120 140 160 180 Time, h Time, h (e) (f) Fig. 2 Normalized cumulative imbibed volume versus time in group 1. a S formation; b H formation; c UY/LY formations; d L formation; e LM formation; f N formation of the rock does not obviously change after water imbibi- disintegrates during the third imbibition, as shown in tion, as shown in Fig. 6a, b, d, f. However, the imbibed Fig. 6a. The rock from the LM formation generates micro- water volume of the sample from the UY and LM forma- fractures, as shown in Fig. 6b. The expansion of clay tions obviously increased after the first immersion, as minerals with water imbibition and the generation of shown in Fig. 6c, e. The rock from the UY formation micro-fractures could increase the pore volume, achieving Imbibed volume/pore volume Imbibed volume/pore volume Imbibed volume/pore volume Imbibed volume/pore volume Imbibed volume/pore volume Imbibed volume/pore volume Pet. Sci. (2015) 12:636–650 643 0.50 0.14 0.45 0.12 0.40 H-1 0.35 0.10 H-2 0.30 H-3 0.08 0.25 S-1 H-4 0.06 0.20 H-5 S-2 0.15 H-6 S-3 0.04 H-7 0.10 H-8 0.02 0.05 0 0 01 2 3 4 567 8 0 1 2 3 4 5 0.5 0.5 Sqrt(t), h Sqrt(t), h (a) (b) 0.020 0.014 0.018 UY-1 0.012 0.016 UY-2 UY-3 0.014 0.010 UY-4 0.012 LY-1 0.008 L-1 0.010 LY-2 L-2 0.006 0.008 L-3 LY-3 L-4 LY-4 0.006 0.004 L-5 LY-5 0.004 L-6 LY-6 0.002 L-7 0.002 LY-7 L-8 0 0 0 2 4 6 8 10 12 14 16 0.5 0.5 Sqrt(t), h Sqrt(t), h (c) (d) 0.007 0.006 0.006 0.005 LM-1 0.005 0.004 LM-2 N-1 0.004 LM-3 0.003 N-2 LM-4 0.003 LM-5 N-3 0.002 LM-6 N-4 0.002 N-5 0.001 0.001 N-6 01234567 8 9 0 5 10 15 20 0.5 0.5 Sqrt(t), h Sqrt(t), h (e) (f) Fig. 3 Imbibed volume per sectional area versus the square of time in group 1. a S formation; b H formation; c UY/LY formations; d L formation; e LM formation; f N formation an additional imbibition capacity beyond the initial pore water could enter a space that gas could not enter. There- volume. In addition, the R values of shales in the L and N fore, the water imbibed into the tight rocks exits in the formations are higher than 1, though these formations have space that can be divided into two parts: the pore space and good imbibition reproducibilities. This suggests that the the clay crystal lattice space. In addition, the imbibed water Imbibed volume/surface area, cm Imbibed volume/surface area, cm Imbibed volume/surface area, cm Imbibed volume/surface area, cm Imbibed volume/surface area, cm Imbibed volume/surface area, cm 644 Pet. Sci. (2015) 12:636–650 volume per unit pore volume, R can be used as a parameter can be used to determine the fracturing fluid volume intake. to indicate the clay content and type. A plot of the imbibition capacity versus the porosity is shown in Fig. 7a. The general trend is that the imbibition 3.4 Influencing factors capacity increases with increasing porosity. The green line in Fig. 7a shows when the imbibition capacity is equal to In this section, the authors try to address the factors influ- the porosity. In other words, the imbibed volume is equal to encing the imbibition capacity and rate, including the poros- the pore volume. Obviously, when the rocks have high ity, permeability, clay minerals, and the fluid component. porosities, the points tend to be below the line. One explanation is that the rocks with high porosities have low 3.4.1 Porosity water saturation. When the rocks have low porosities, the points tend to be above the line. Therefore, the plot can be For a ratio of imbibition volume to pore volume of more divided into three regions. In Region I, the low-porosity than 1, the conventional definition of water location is not region, the points are above the line, and R tends to exceed applicable for tight formations. The imbibition capacity 1. The imbibition capacity is mainly controlled by the clay can be defined as the imbibition volume per unit dry mineral content. In Region II, the moderate-porosity sample volume. The parameter is broadly applicable and region, the points are on the line, and R tends to be 1. The imbibition capacity is mainly controlled by the clay min- eral content and porosity. In Region III, the high-porosity region, the points are below the line, and R tends to be lower than 1. In this region, the rocks with high porosities A A tend to have low clay contents. Therefore, the imbibition capacity is mainly controlled by the porosity. In Fig. 7b, the imbibition rate is positively correlated Region 3 with the porosity, which is consistent with the prediction of Region 2 3 Eq. (2). This result means that porosity is the main con- trolling factor for the imbibition rate. 3.4.2 Permeability Region 1 In Fig. 8, the imbibition capacity and rate are positively correlated with permeability. However, the correlation is not strong. In addition, there is an obvious departure from 0 5 10 15 20 25 30 35 40 the prediction of Eq. (3). This result means that perme- 0.5 Sqrt(t), h ability is not the main controlling factor for the imbibition Fig. 4 Schematic of imbibition curve behavior rate. 9 1 UY/LY UY/LY LM N 0.1 LM N UY 0.01 0.001 0 0.0001 02468 10 12 Sample number Sample number (a) (b) Fig. 5 Distribution of the imbibition capacity normalized by porosity and imbibition rate. a Imbibition capacity by porosity; b Imbibition rate Imbibed volume/surface area, cm Imbibed volume/pore volume 0.5 Imbibition rate A , cm/h i Pet. Sci. (2015) 12:636–650 645 3.5 3.5 3.0 3.0 2.5 2.5 2.0 2.0 1st immersion (H-9) 1st immersion (S-4) 1.5 1.5 2nd immersion (H-9) 2nd immersion (S-4) 1.0 1.0 3rd immersion (H-9) 0.5 0.5 0 10 20 30 40 50 60 70 0 5 10 15 20 25 Time, h Time, h (a) (b) 1.2 0.9 0.8 1.0 0.7 0.8 0.6 0.5 0.6 1st immersion (L-9) 0.4 1st immersion (UY-5) 2nd immersion (L-9) 0.4 0.3 2nd immersion (UY-5) 3rd immersion (L-9) 0.2 0.2 3rd immersion (UY-5) 0.1 0 10203040506070 80 90 100 010 20 30 40 50 Time, h Time, h (c) (d) 2.0 0.8 1.8 0.7 1.6 0.6 1.4 0.5 1.2 0.4 1.0 1st immersion (LM-7) 1st immersion (N-7) 0.8 0.3 2nd immersion (LM-7) 0.6 0.2 2nd immersion (N-7) 3rd immersion (LM-7) 0.4 0.1 0.2 01 23 4 56 7 89 020 40 60 80 100 120 140 160 180 Time, h Time, h (e) (f) Fig. 6 Repeated immersion tests in group 2. a S formation; b H formation; c UY formation; d L formation; e LM formation; f N formation 3.4.3 Clay minerals porosity is normalized. Then the water saturation and the driving force coefficient are used to represent the imbibi- In order to explore the effect of the clay content and type tion capacity and rate, respectively. on the imbibition capacity and rate, the effect of the Enhancement Imbibed mass/dry mass, % Imbibed mass/dry mass, % Imbibed mass/dry mass, % Imbibed mass/dry mass, % Imbibed mass/dry mass, % Imbibed mass/dry mass, % 646 Pet. Sci. (2015) 12:636–650 UY/LY 0.1 LM N Region II 0.01 Region I Region III S H 0.001 UY/LY L LM N 0.0001 0.1 1 10 100 0 3 6 9 12 15 Porosity, % Porosity, % (a) (b) Fig. 7 Imbibition capacity and rate versus porosity. a Imbibition capacity; b Imbibition rate 0.1 0.01 0.001 UY/LY UY/LY LM N LM N 0.0001 0.0001 0.001 0.01 0.1 1 10 0.0001 0.001 0.01 0.1 1 10 Permeability, mD Permeability, mD (a) (b) Fig. 8 Imbibition capacity and rate versus permeability. a Imbibition capacity; b Imbibition rate The imbibition capacity is positively related to the total the LY formation does not contain smectite. The specific clay content, the I/S, and illite concentrations, as shown in surface area of smectite is much higher than that of I/S, Figs. 9a, 10a, and 11a. The imbibition rate is also posi- which is probably the main reason for the deviation from tively related to the total clay content and the I/S concen- the rule in the LY formation. This proves that the imbi- tration, as shown in Figs. 9b and 10b. However, the bition capacity is related to not only the clay content but correlation between the imbibition rate and the illite con- also the clay type. centration is not strong. In Fig. 9a, the imbibition capacity is closely related to 3.4.4 Fluids the total clay concentration. However, the H and LY formations deviate from the rule. The H and LY forma- In group 3 experiments, a surfactant solution and a KCl tions have relatively high clay contents, but their imbi- solution changed the two driving forces and were used to bition capacities are lower than those of the L and LM explore the effect of the fracturing fluid component on the formations. The clay type in the H formation is mainly imbibition capacity. chlorite, while the clay of the L formation is mainly I/S. The contact angle on the surface of LM-8 rock was I/S has a relatively high specific surface area that can measured with an imaging method, as shown in Fig. 12. absorb larger amounts of water. Unlike the LM formation, The contact angle is 14 before cationic surfactant Imbibed volume/dry sample volume, % Imbibed volume/dry sample volume, % 0.5 Imbibition rate A , cm/h 0.5 Imbibition rate A , cm/h i Pet. Sci. (2015) 12:636–650 647 S H S H UY (74.9, 7.5) LY LY UY N 6 0.1 UY N LM LM 0.01 LM (36.9, 2.9) L (23.7, 1.9) LY (47.8, 2.3) 0.001 S (10.3, 0.7) H (31.2, 0.9) N (23.3, 1.1) 0.0001 0 20 40 60 80 0 10 100 Total clay concentration, wt% Total clay concentration, wt% (a) (b) Fig. 9 Imbibition capacity normalized by porosity and driving force coefficient versus total clay concentration. a Imbibition capacity normalized by porosity; b Driving force coefficient S H LY LY L UY N LM UY N 0.1 LM 0.01 0.001 0 0.0001 0 10 20 30 40 50 60 0 10 100 I/S concentration, wt% I/S concentration, wt% Fig. 10 Imbibition capacity normalized by porosity and driving force coefficient versus I/S concentration. a Imbibition capacity normalized by porosity; b Driving force coefficient S H S H LY LY UY N 6 0.1 UY N LM LM 0.01 0.001 0.0001 02468 10 12 14 0.1 1 10 100 Illite concentration, wt% Illite concentration, wt% Fig. 11 Imbibition capacity normalized by porosity and rate versus illite concentration. a Imbibition capacity normalized by porosity; b Driving force coefficient Average imbibed volume/pore volume Average imbibed volume/pore volume Average imbibed volume/pore volume Driving force coefficient A /K Driving force coefficient A /Kφ 2 i Driving force coefficient A /Kφ i 648 Pet. Sci. (2015) 12:636–650 Fig. 12 Contact angle on sample LM-8 with a water drop. a Before surfactant treatment (14); b After surfactant treatment (51) treatment and 51 after treatment. The surface wettability is 70.0 65.0 altered and tends to be intermediately wet, which may lead DI water 60.0 to a lower imbibition capacity. 10 wt% KCl 52.3 The capillary suction time (CST) tests are quick and 50.0 47.4 45.0 easy and can provide information about inhibition charac- 42.5 38.2 40.0 teristics of an additive (Berry et al. 2008). The authors tried 34.0 to understand the inhibitive ability of a 10 wt% KCl 30.2 29.1 28.6 30.0 23.8 solution on clay expansion and explored its effect on the 22.5 20.0 imbibition capacity. The CST apparatus measures the time required for the fluid to travel a given distance, as shown in 10.0 Fig. 13. A short CST time interval reflects a poor clay dispersibility. The results of CST tests are shown in 0.0 SN LM L UY H Fig. 14. The relative relationship of the CST time in tight formations is UY [ LM [ N [ L [ H [ S, which is Fig. 14 CST time in different formations with two solutions positively related to the I/S concentration. The 10 wt% KCl solution inhibits the clay expansion in tight formations, the imbibition capacity. For the UY and LM formations, especially for the UY and LM formations. This effect may the reduced imbibition capacity is more obvious. Figure 16 lead to a lower imbibition capacity. shows pictures of the UY and LM formation samples after The results from group 3 experiments are presented in exposure to the surfactant and 10 wt% KCl solutions. Fig. 15. The surfactant and 10 wt% KCl solutions reduce When exposed to deionized water, the sample from the UY 9.5 9.1 8.8 8.4 8.3 DI water 7.6 Surfactant 7 10 wt% KCl 6 5.7 4.6 4.2 3.5 3.1 2.9 2.8 2.4 2.2 2.1 1.9 1.8 SH UY L LM N Fig. 15 The effect of imbibed fluids on the imbibition capacity in Fig. 13 CST apparatus group 3 Imbibed volume/dry sample volume, % CST time, s Pet. Sci. (2015) 12:636–650 649 imbibition capacity normalized by the porosity increases with increasing total clay, I/S, and illite concentrations. Smectite and I/S with a high specific area tend to lead to strong water imbibition. (4) The water volume imbibed into the clay-rich tight rocks is much greater than the pore volume mea- sured by helium. The imbibed volume per unit pore volume is a parameter that can evaluate the effect of the clay content and type. (5) Surfactants can change the imbibition capacity of tight rocks by altering the interfacial tension and wettability. A 10 wt% KCl solution can inhibit the clay expansion, which also reduces the imbibition capacity. Acknowledgments This research program was financially sup- ported by the National Basic Research Program of China (973 Pro- gram) Granted No. 2015CB250903 and the National Natural Science Foundation of China Granted No. 51490652. The Chongqing Institute of Geology and Mineral Resources supported this field work. Open Access This article is distributed under the terms of the Fig. 16 Pictures of tight rock samples after exposure to different Creative Commons Attribution 4.0 International License (http://crea fluids. a UY formation; b LM formation tivecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a formation disintegrates completely, and the sample from link to the Creative Commons license, and indicate if changes were the LM formation generates some micro-fractures along made. the bedding plane. The 10 wt% KCl solution has a good inhibitive effect on disintegration and micro-fractures, References which is consistent with the results of CST tests, as shown in Fig. 16. Berry SL, Boles JL, Brannon HD, et al. Performance evaluation of ionic liquids as a clay stabilizer and shale inhibitor. In: SPE international symposium and exhibition on formation damage 4 Conclusions control, 13–15 February, Lafayette, Louisiana, USA. 2008. doi:10.2118/112540-MS. Dehghanpour H, Lan Q, Saeed Y, et al. Spontaneous imbibition of A series of spontaneous imbibition experiments were car- brine and oil in gas shales: effect of water adsorption and ried out, and the following conclusions were reached: resulting microfractures. Energy Fuels. 2013;27(6):3039–49. doi:10.1021/ef4002814. (1) Tight rock imbibition can be characterized by the Fakcharoenphol P, Kurtoglu B, Kazemi H, et al. The effect of osmotic imbibition capacity, the imbibition rate, and the pressure on improve oil recovery from fractured shale forma- diffusion rate. The imbibition capacity and rate are tions. In: SPE unconventional resources conference, 1–3 April, The Woodlands, Texas, USA. 2014. doi:10.2118/168998-MS. positively correlated with porosity. The imbibition Handy LL. Determination of effective capillary pressures for porous curves can typically be divided into three regions: media from imbibition data. Trans AIME. 1960;219:75–80. the initial linear imbibition region, the transition Hu Q, Ewing RP, Dultz S. Low pore connectivity in natural rock. region, and the diffusion region. J Contam Hydrol. 2012;133:76–83. doi:10.1016/j.jconhyd.2012. 03.006. (2) The water-holding space in tight rocks consists of Lan Q, Dehghanpour H, Wood J, et al. Wettability of the Montney two parts: the pore space and the clay crystal lattice tight gas formation. In: SPE/CSUR unconventional resources space. The driving forces of spontaneous imbibition conference—Canada, 30 September–2 October, Calgary, in tight rocks are the capillary force and the clay Alberta, Canada. 2014. doi:10.2118/171620-MS. Makhanov K, Dehghanpour H, Kuru E. An experimental study of absorption force. A new parameter, the effective spontaneous imbibition in Horn River shales. In: SPE Canadian driving force of imbibition, is defined to describe the unconventional resources conference, 30 October–1 November, effect of clay absorption. Calgary, Alberta, Canada. 2012. doi:10.2118/162650-MS. (3) The clay mineral content significantly influences the Makhanov K, Habibi A, Dehghanpour H, et al. Liquid uptake of gas shales: a workflow to estimate water loss during shut-in periods imbibition capacity for clay-rich tight rocks. The 123 650 Pet. Sci. (2015) 12:636–650 after fracturing operations. J Unconv Oil Gas Resour. Sun Y, Bai B, Wei M. Microfracture and surfactant impact on linear 2014;7:22–32. doi:10.1016/j.juogr.2014.04.001. cocurrent brine imbibition in gas-saturated shale. Energy Fuels. Penny GS, Dobkins TA, Pursley JT. Field study of completion fluids 2015;29(3):1438–46. doi:10.1021/ef5025559. to enhance gas production in the Barnett Shale. In: SPE gas Yuan W, Li X, Pan Z, et al. Experimental investigation of interactions technology symposium, 15–17 May, Calgary, Alberta, Canada. between water and a lower Silurian Chinese shale. Energy Fuels. 2006. doi:10.2118/100434-MS. 2014;28(8):4925–33. doi:10.1021/ef500915k. Roychaudhuri B, Tsotsis TT, Jessen K. An experimental investigation Zhou Z, Hoffman BT, Bearinger D, et al. Experimental and numerical of spontaneous imbibition in gas shales. J Pet Sci Eng. study on spontaneous imbibition of fracturing fluids in shale gas 2013;111:87–97. doi:10.1016/j.petrol.2013.10.002. formation. In: SPE/CSUR unconventional resources confer- Sharma M, Agrawal S. Impact of liquid loading in hydraulic fractures ence—Canada, 30 September–2 October, Calgary, Alberta, on well productivity. In: SPE hydraulic fracturing technology Canada. 2014. doi:10.2118/171600-MS. conference, 4–6 February, The Woodlands, Texas, USA. 2013. doi:10.2118/163837-MS. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Petroleum Science Springer Journals

Experimental investigation of shale imbibition capacity and the factors influencing loss of hydraulic fracturing fluids

Download PDF
15 pages

Loading next page...
 
/lp/springer-journals/experimental-investigation-of-shale-imbibition-capacity-and-the-ZKN5fU5jV7

References (15)

Publisher
Springer Journals
Copyright
Copyright © 2015 by The Author(s)
Subject
Earth Sciences; Mineral Resources; Industrial Chemistry/Chemical Engineering; Industrial and Production Engineering; Energy Economics
ISSN
1672-5107
eISSN
1995-8226
DOI
10.1007/s12182-015-0049-2
Publisher site
See Article on Publisher Site

Abstract

Pet. Sci. (2015) 12:636–650 DOI 10.1007/s12182-015-0049-2 ORIGINAL PAPER Experimental investigation of shale imbibition capacity and the factors influencing loss of hydraulic fracturing fluids 1 1 1 1 1 • • • • • Hong-Kui Ge Liu Yang Ying-Hao Shen Kai Ren Fan-Bao Meng 1 1 Wen-Ming Ji Shan Wu Received: 11 May 2015 / Published online: 21 September 2015 The Author(s) 2015. This article is published with open access at Springerlink.com Abstract Spontaneous imbibition of water-based frac- water greater than their measured pore volume. The aver- turing fluids into the shale matrix is considered to be the age ratio of the imbibed water volume to the pore volume main mechanism responsible for the high volume of water is approximately 1.1 in the Niutitang shale, 1.9 in the loss during the flowback period. Understanding the matrix Lujiaping shale, 2.8 in the Longmaxi shale, and 4.0 in the imbibition capacity and rate helps to determine the frac- Yingcheng volcanic rock, and this ratio can be regarded as turing fluid volume, optimize the flowback design, and to a parameter that indicates the influence of clay. In addition, analyze the influences on the production of shale gas. surfactants can change the imbibition capacity due to Imbibition experiments were conducted on shale samples alteration of the capillary pressure and wettability. A 10 from the Sichuan Basin, and some tight sandstone samples wt% KCl solution can inhibit clay absorption to reduce the from the Ordos Basin. Tight volcanic samples from the imbibition capacity. Songliao Basin were also investigated for comparison. The effects of porosity, clay minerals, surfactants, and KCl Keywords Imbibition  Shale  Fracturing fluid solutions on the matrix imbibition capacity and rate were Capillary pressure  Clay systematically investigated. The results show that the imbibition characteristic of tight rocks can be characterized by the imbibition curve shape, the imbibition capacity, the 1 Introduction imbibition rate, and the diffusion rate. The driving forces of water imbibition are the capillary pressure and the clay Multistage hydraulic fracturing is a critical technology for absorption force. For the tight rocks with low clay contents, economic production from shale reservoirs. Large amounts the imbibition capacity and rate are positively correlated of water-based fracturing fluids are pumped into forma- with the porosity. For tight rocks with high clay content, tions, generating extensive fracture networks and stimu- the type and content of clay minerals are the most impor- lating low-permeability formations. Field operations have tant factors affecting the imbibition capacity. The imbibed demonstrated that large volumes of injected fluids are water volume normalized by the porosity increases with an retained in shale formations, with a flowback efficiency of increasing total clay content. Smectite and illite/smectite lower than 30 % (Makhanov et al. 2014). In the U.S. tend to greatly enhance the water imbibition capacity. Haynesville shale formation, the flowback rate is even Furthermore, clay-rich tight rocks can imbibe a volume of lower than 5 % after fracturing operations (Penny et al. 2006). Besides possibly causing a series of environmental problems, the retention of fracturing fluids in shale for- & Liu Yang mations can greatly enhance the water saturation near shidayangliu@126.com fracture surfaces and influence two-phase fluid flow, thus State Key Laboratory of Petroleum Resources and further inhibiting the production of shale gas (Sharma and Prospecting, China University of Petroleum, Beijing 102249, Agrawal 2013). Furthermore, intense interaction between China fluid and shale can dramatically change rock properties and impact on the generation of fracture networks during Edited by Yan-Hua Sun 123 Pet. Sci. (2015) 12:636–650 637 fracturing (Yuan et al. 2014). Therefore, studying the This paper focuses on the imbibition capacity and the imbibition capacity and its main controlling factors is influence of the mineral composition and physical proper- essential to understanding reservoir performance and ties of tight rocks. Samples include gas shales from the optimizing fracturing operations. Sichuan Basin, tight sandstones from the Ordos Basin, and It is generally believed that spontaneous imbibition of tight volcanic rocks from the Songliao Basin. Experiments fracturing fluids into the shale matrix plays an important can be divided into three groups. In group 1, the imbibition role in water loss. Many researchers have focused on the capacity and rate of deionized water uptake are investi- mechanism of fracturing fluid imbibition. Makhanov et al. gated systematically. In group 2, each sample is immersed (2012) found that imbibition rates perpendicular and par- repeatedly in deionized water several times to address the allel to the bedding plane are different, and the latter is water sensitivity of different rocks. In group 3, comparative higher. Hu et al. (2012) considered that the Barnett shale experiments are conducted to explore the effects of dif- has a poor connectivity, which greatly influences the flow ferent fluids on the imbibition capacity. and diffusion of fluid. Roychaudhuri et al. (2013) deter- mined that a surfactant can effectively reduce the imbibi- tion rate of fracturing fluids, and the driving force of 2 Experimental imbibition is the capillary pressure. Dehghanpour et al. (2013) mentioned that the amount of imbibition in shale is 2.1 Rock samples and fluids positively related to mineral composition and physical properties. Fakcharoenphol et al. (2014) investigated the Sixty-six shale and tight gas rock samples from the Ordos effects of salinity on water imbibition and found that the Basin, Songliao Basin, and the Sichuan Basin were used to osmotic pressure can act as the driving force for water conduct comparative imbibition experiments, and reservoir intake. Currently, it is well known that the imbibition of rock properties are presented in Table 1. The mineral fracturing fluids is mainly controlled by the capillary composition (in wt%) of the shale and tight gas rock pressure, while the effects of clay absorption have not been samples and the relative abundance of clay minerals are studied thoroughly. The imbibition capacity, imbibition listed in Table 2. The samples were neither cleaned nor rate, and other influencing factors in shale reservoirs have exposed to air beforehand. According to the observed not been investigated systematically. brittleness of rocks, the samples were machined into Table 1 Tight reservoir Label Formation Lithology Depth, m Source Geological age properties in this study S Shihezi Tight sandstone 2120 Erdos Basin Early Permian H Huoshiling Tight volcanic 2523 Songliao Basin Lower Jurassic UY Upper Yingcheng Tight volcanic 3524 Songliao Basin Lower Jurassic LY Lower Yingcheng Tight volcanic 3557 Songliao Basin Lower Cretaceous L Lujiaping Shale 1235 Sichuan Basin Lower Cambrian LM Longmaxi Shale 786 Sichuan Basin Lower Silurian N Niutitang Shale 895 Sichuan Basin Lower Cambrian Table 2 Results of XRD mineralogy analysis Label Mineral composition, wt% Relative abundance, % TOC, wt% Quartz Feldspar Calcite Dolomite Clay Smectite Illite I/S Chlorite Kaolinite S 32.2 26.4 5.1 25.8 10.3 0 100.0 0 0 0 0 H 1.3 61.5 3 0 34.2 0 10.5 0 89.5 0 0 UY 13.2 11.9 0 0 74.9 0 16.8 74.9 8.3 0 1.1 LY 40.6 11.6 0 0 47.8 0 7.9 78.0 11.1 2.9 1.2 L 29.4 7.2 24.7 14.9 23.7 7.6 23.6 53.2 8.0 7.6 3.1 LM 40.3 8.8 7.5 6.5 36.9 4.3 15.9 62.3 8.7 8.7 3.6 N 31.2 15.8 11.5 18.2 23.3 3.4 5.2 78.9 12.4 0 2.5 I/S is Illite/smectite mixed-layer 123 638 Pet. Sci. (2015) 12:636–650 cylindrical or prismatic core plugs, as shown in Tables 3, 4, after the imbibition experiments. The experimental and 5. The effects of the sample size and shape can be data were normalized by the scaling method, which normalized by the scaling method. is described in the other sections. Deionized water, 10 wt% KCl solution, and an aqueous The experimental procedure of the group 2 experiments solution of cationic surfactant were used as the imbibing is the same as that of the group 1 experiments. Each sample fluids. Properties of the test fluids are listed in Table 6. The was immersed in deionized water and dried repeatedly most commonly used boundary conditions for imbibition several times. The basic sample data are presented in are one-end-open (OEO), all-faces-open (AFO), and two- Table 4. ends-open (TEO). Considering the effect of lamination on In group 3, samples of the same formation were the imbibition rate, for 1-D imbibition (OEO and TEO), the acquired from the same core to reduce the influences of open face is parallel to the bedding plane to maintain the heterogeneity. A total of 18 samples were dried for 24 h same experimental conditions. and submerged into different fluids until there was no further change in weight. This process lasted approxi- mately 7 days. The basic information about the samples is 2.2 Experimental apparatus and procedure shown in Table 5. In group 1 experiments, the experimental results are sen- sitive to test environments and the instrumental error due to 3 Experimental data and analysis the relatively low imbibition rate in shale and tight rock samples. Therefore, a series of measurements are required 3.1 Scaling method for experimental data to improve the measurement accuracy: normalization (1) All of the samples were weighted by an analytical balance (Mettler XPE205) with an accuracy of The samples used had different sizes and shapes. Charac- 0.00001 g. terization methods need to be developed to normalize the (2) Impermeable and nonelastic strings used to suspend effects of size and shape and represent the imbibition core slugs had a diameter of approximately capacity and rate. 0.13 mm, which could avoid the error caused by (1) The imbibition capacity can be determined based on the reduction in the liquid volume. the curve of water volume gain per pore volume (3) The experimental device was placed in a chamber versus time. with constant temperature and humidity to lower the (2) Handy (1960) established a famous gas–water imbi- effect of variable external temperature and humidity. bition model, which is given by the following The experimental device is shown in Fig. 1. equation: (4) The whole apparatus was placed in the basement of a sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi building to minimize the vibration from the ground pffiffi 2P /KS c wf surface. V =A ¼ t; ð1Þ imb c The experimental procedure is as follows: where V is the volume of imbibed water, cm ; P imb c (1) The initial dimensions and mass of the core slug is the capillary pressure, MPa; / is the porosity, %; were measured before experiment. K is the permeability, mD; S is the water satura- wf (2) Impermeable epoxy was used to satisfy TEO and tion, %; A is the imbibition cross-sectional area, OEO, imbibition boundary conditions. cm ; l is the fluid viscosity, mPa s; and t is the (3) The core slug was dried in an oven at 105 C until imbibition time, s. there was no further change in weight. (4) After the core slug cooled down, the sample was The slope of the volume of imbibed fluid per sectional suspended on the analytical balance. The sample was area versus the square of time, A , can be used to represent totally submerged into water by adjusting the liquid the imbibition rate, which can be obtained from experi- level. mental data (Makhanov et al. 2014). (5) The variation of the sample mass with time was According to Eq. (1), the imbibition rate can be given as measured and then recorded on a computer as water follows: sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi was spontaneously imbibed into the sample. 2P /KS c wf (6) The mass of the imbibed water was calculated by A ¼ : subtracting the initial mass from the mass recorded 123 Pet. Sci. (2015) 12:636–650 639 Table 3 Basic properties of core plugs used in group 1 a b c No. Shape Cross-sectional area Length Permeability Porosity Boundary condition Imbibition rate Imbibed volume per 2 0.5 d A ,cm L, cm K,mD U,% A , cm/h sample volume c i C,% S-1 Cylinder 5.1 5.1 2.1 12.3 OEO 0.1069 8.6 S-2 Cylinder 5.0 5.1 2.1 13.0 TEO 0.1123 8.9 S-3 Cylinder 5.1 5.0 2.2 12.8 OEO 0.1173 9.5 H-1 Cylinder 4.9 1.1 0.0028 12.5 TEO 0.0382 11.4 H-2 Cylinder 5.1 1.9 0.0045 8.4 OEO 0.0467 8.0 H-3 Cylinder 5.0 0.9 0.0031 9.6 TEO 0.0403 9.6 H-4 Cylinder 4.9 0.8 0.0069 9.7 TEO 0.0356 8.6 H-5 Cylinder 5.0 1.6 0.0083 13.6 TEO 0.0368 11.1 H-6 Cylinder 5.0 1.6 0.0034 14.1 TEO 0.0466 12.0 H-7 Cylinder 4.9 1.0 0.0096 10.8 TEO 0.0375 10.3 H-8 Cylinder 5.1 0.9 0.0069 10.1 TEO 0.0381 9.1 UY-1 Cylinder 4.9 0.6 0.0012 0.3 TEO 0.0017 2.7 UY-2 Cylinder 4.9 0.6 0.0013 0.4 TEO 0.0099 2.8 UY-3 Prism 9.0 0.5 0.0023 0.4 TEO 0.0021 3.2 UY-4 Cylinder 4.9 0.6 0.0012 0.5 TEO 0.0060 2.8 LY-1 Cylinder 4.9 0.5 0.0032 3.3 TEO 0.0044 4.7 LY-2 Cylinder 4.9 0.5 0.0025 1.6 TEO 0.0055 4.5 LY-3 Cylinder 4.9 0.8 0.0007 2.2 TEO 0.0053 4.7 LY-4 Cylinder 4.9 0.8 0.0012 2.1 TEO 0.0029 3.9 LY-5 Cylinder 4.9 0.5 0.0007 1.9 TEO 0.0037 3.4 LY-6 Cylinder 4.9 0.5 0.0031 0.9 TEO 0.0016 2.6 LY-7 Prism 10.7 0.5 0.0011 0.6 TEO 0.0008 1.8 L-1 Cylinder 40.4 0.8 0.0021 1.29 TEO 0.0015 2.7 L-2 Cylinder 40.4 1.0 0.0035 1.5 TEO 0.0017 3.5 L-3 Cylinder 40.4 1.1 0.0042 2.0 TEO 0.0020 4.1 L-4 Cylinder 29.2 1.1 0.0014 1.8 TEO 0.0015 2.2 L-5 Cylinder 29.0 0.9 0.0027 0.9 TEO 0.0016 1.9 L-6 Cylinder 29.4 1.0 0.0032 2.4 TEO 0.0018 4.8 L-7 Cylinder 29.0 1.1 0.0018 1.1 TEO 0.0012 2.0 L-8 Cylinder 29.5 0.6 0.0021 2.1 TEO 0.0012 3.8 LM-1 Prism 6.2 0.5 0.0046 2.0 TEO 0.0037 5.7 LM-2 Cylinder 30.1 0.7 0.0053 1.4 TEO 0.0035 1.4 LM-3 Cylinder 31.3 0.9 0.0038 1.5 TEO 0.0025 4.5 LM-4 Cylinder 31.3 0.9 0.0062 2.3 TEO 0.0020 6.5 LM-5 Prism 6.1 0.7 0.0071 2.2 TEO 0.0028 5.0 LM-6 Prism 6.0 0.4 0.0031 1.9 TEO 0.0041 6.1 N-1 Cylinder 29.8 0.7 0.0026 2.1 TEO 0.0008 2.0 N-2 Cylinder 29.9 0.7 0.0038 1.4 TEO 0.0006 1.3 N-3 Cylinder 29.9 0.7 0.0035 1.9 TEO 0.0007 1.7 N-4 Cylinder 29.3 0.6 0.0025 2.4 TEO 0.0008 2.9 N-5 Cylinder 29.0 0.6 0.0019 2.7 TEO 0.0010 3.5 N-6 Cylinder 29.4 0.6 0.0029 3.1 TEO 0.0011 3.2 Permeability was obtained using a nitrogen pressure pulse decay permeability porosimeter Porosity was measured in a helium porosimeter Imbibition rate was obtained from the experimental data Imbibed volume per sample volume was obtained from the experimental data 123 640 Pet. Sci. (2015) 12:636–650 3.2 Imbibition curve characteristics The log–log relationship can be given by the following equation: The imbibition characteristics of tight rocks can be char- 1 2P KS 1 c wf acterized by the imbibition capacity, the imbibition rate, log A ¼ log þ log /; ð2Þ 2 l 2 w and the diffusion rate. Figure 2 shows plots of the volume normalized by the 1 2P /S 1 c wf log A ¼ log þ log K: ð3Þ pore volume versus time. The cumulative imbibed volume 2 l 2 increases with time. However, the rate of water intake The effective imbibition driving force can be given as obviously slows down with increasing time, and the rate follows: approximately reaches zero, which represents the equilib- rium condition (Sun et al. 2015). However, some curves A l l i w w P ¼ ¼ a ; ð4Þ may have ‘‘upward tails,’’ which demonstrate an obvious 2K/S 2S wf wf diffusion effect and may be related to the complex pore where a is the driving force coefficient, 1/s. a reflects the structure in these rock samples, as shown in Fig. 2f. effect of difference between imbibition driving force and Though the size and shape of core samples vary signifi- friction resistance, which can be obtained by experiments. cantly, the water volume normalized by the pore volume Equations (2)–(4) can be used to analyze contributing can represent the imbibition capacity well. It is worth factors to the imbibition rate, which is discussed in the noting that the highest points in the curves may not always following sections. remain at the same value, which may be explained by the heterogeneity of core samples as shown in Fig. 2c, d, e, f. Table 4 Basic properties of core slugs used in group 2 In addition, the volume gain fluctuates in some samples, which may be due to the large amount of water-sensitive No. Shape Dry mass, g Boundary condition I/S mixed-layer in this tight rock, as shown in Fig. 2c, e, f. S-4 Cylinder 69.5 AFO This effect will be discussed in detail later. H-9 Cylinder 13.9 AFO Figure 3 shows plots of the imbibed volume normalized UY-5 Cylinder 11.5 AFO by the sectional area versus the square root of time. The L-9 Cylinder 14.0 AFO effect of the sectional area is normalized well. The mea- LM-7 Prism 15.0 AFO surements tend to sit close to a smooth curve that repre- N-7 Cylinder 57.5 AFO sents the imbibition rate as shown in Fig. 3a, b, d, e. In Table 5 Basic properties of No. Shape Cross-sectional area Length Imbibing fluid Boundary condition core slugs used in group 3 2 A cm L,cm c, S-5 Cylinder 5.1 5.0 Deionized water AFO S-6 Cylinder 5.1 5.1 2.5 wt% surfactant AFO S-7 Cylinder 4.9 5.1 10 wt% KCl brine AFO H-10 Cylinder 4.9 1.2 Deionized water AFO H-11 Cylinder 4.9 1.2 2.5 wt% surfactant AFO H-12 Cylinder 4.9 1.0 10 wt% KCl brine AFO UY-6 Cylinder 5.1 1.1 Deionized water AFO UY-7 Cylinder 5.0 1.0 2.5 wt% surfactant AFO UY-8 Cylinder 5.1 1.1 10 wt% KCl brine AFO L-10 Cylinder 40.4 1.2 Deionized water AFO L-11 Cylinder 40.4 1.1 2.5 wt% surfactant AFO L-12 Cylinder 40.3 1.0 10 wt% KCl brine AFO LM-8 Prism 6.1 0.9 Deionized water AFO LM-9 Prism 6.0 1.0 2.5 wt% surfactant AFO LM-10 Prism 6.1 0.9 10 wt% KCl brine AFO N-8 Cylinder 29.4 1.1 Deionized water AFO N-9 Cylinder 29.4 1.0 2.5 wt% surfactant AFO N-10 Cylinder 29.3 1.1 10 wt% KCl brine AFO 123 Pet. Sci. (2015) 12:636–650 641 Table 6 Properties of fluids at 3 Fluids Density, g/cm Viscosity, cP Surface tension, N/m 25 C used in group 3 Deionized water 0.998 1 0.072 10 wt% KCl 1.06 0.88 0.074 2.5 wt% surfactant 0.96 1.1 0.058 Blance String (0.13 mm) Computer Liquid Fig. 1 Schematic for imbibition experiments in the OEO condition Fig. 3c, e, the departure from a smooth curve can be consistent with the results of Zhou et al. (2014). There is a explained by the strong heterogeneity in these tight rocks. significant difference between these results and the con- The imbibition curves generally behave similarly. Each ventional opinion that the water saturation should be 0–1. profile is divided into three regions: the initial linear Therefore, the capacity for fracturing fluid intake in tight imbibition region (Region 1), the transition region (Region reservoirs may greatly exceed the conventional estimation. 2), and the diffusion region (Region 3), as shown in Fig. 4 The relative relationship of R is UY tight volcanic[ LM (Lan et al. 2014). However, the curve characteristics of shale [ LY tight volcanic [ L shale[ N shale [ H tight different tight rocks vary significantly, which may be volcanic [ S tight sandstone. attributed to the pore size distribution and pore connec- Figure 5b presents the distribution of the imbibition tivity. The slope in Region 1 represents the imbibition rate rate. The imbibition rates in the S and H formations are 0.5 (A ), where the capillary pressure may be the primary approximately 0.05–0.1 cm/h , which is obviously higher driving force. As the water saturation increases, the capil- than the rates for the LY, LM, N, and L formations. The lary pressure decreases, and the water intake process begins relative relationship of the imbibition rate is S tight sand- to enter Region 2. In Region 3, diffusion is the primary stone [ H tight volcanic[ UY tight volcanic [ LY tight flow driving force compared with capillary imbibition in volcanic [ LM shale [ L shale [ N shale. Region 1. In tight reservoir rocks, the diffusion effect is It is worth noting that the imbibed volume exceeds the more obvious than that in conventional reservoir rocks. It pore volume in rocks, so the water presumably entered into can be represented by the diffusion rate (A ). However, a a space where the gas used to measure the porosity could study of the diffusion rate is not included in this paper. not enter. Another explanation is that absorption of water by the clay can induce cracks and increase the pore vol- 3.3 Imbibition capacity and rate ume. The high water imbibition capacity in tight rocks may be related to clay minerals. This is consistent with the well- The ratios of the maximum imbibed volume to the pore known phenomenon of wellbore instability in well drilling. volume, R exceed 1 in the UY, LY, L, LM, and N for- The experiments in group 2 were conducted to explore the mations, as shown in Fig. 5a. In particular, for the effects of clay minerals on the imbibition process. A single UY formation, R is 6–8. In other words, the imbibed vol- sample was tested several times repeatedly. ume of the rock in the tight reservoir can exceed the pore The samples in the S, H, L, and N formations have good volume of the rock by a factor of 6–8. This result is reproducibility, which illustrates that the inner pore volume Temperature and humidity chamber 642 Pet. Sci. (2015) 12:636–650 0.9 1.2 0.8 1.0 0.7 0.6 0.8 H-1 S-1 0.5 H-2 0.6 H-3 S-2 0.4 H-4 0.3 H-5 S-3 0.4 H-6 0.2 H-7 0.2 H-8 0.1 0 0 010 20 30 40 50 024 6 8 10 12 14 16 18 Time, h Time, h (a) (b) 9 2.5 UY-1 UY-2 2.0 L-1 UY-3 L-2 6 UY-4 L-3 Clay-adsorption 1.5 Clay-adsorption LY-1 L-4 LY-2 L-5 LY-3 L-6 1.0 LY-4 L-7 LY-5 L-8 LY-6 0.5 LY-7 05 10 15 20 25 30 35 40 45 020 40 60 80 100 120 140 160 180 200 Time, h Time, h (c) (d) 1.4 4.0 Clay-adsorption 3.5 1.2 3.0 0.1 2.5 LM-1 0.8 N-1 2.0 LM-2 N-2 Clay-adsorption 0.6 LM-3 N-3 1.5 LM-4 N-4 0.4 1.0 N-5 LM-5 0.5 N-6 0.2 LM-6 010 20 30 40 50 60 70 80 90 020 40 60 80 100 120 140 160 180 Time, h Time, h (e) (f) Fig. 2 Normalized cumulative imbibed volume versus time in group 1. a S formation; b H formation; c UY/LY formations; d L formation; e LM formation; f N formation of the rock does not obviously change after water imbibi- disintegrates during the third imbibition, as shown in tion, as shown in Fig. 6a, b, d, f. However, the imbibed Fig. 6a. The rock from the LM formation generates micro- water volume of the sample from the UY and LM forma- fractures, as shown in Fig. 6b. The expansion of clay tions obviously increased after the first immersion, as minerals with water imbibition and the generation of shown in Fig. 6c, e. The rock from the UY formation micro-fractures could increase the pore volume, achieving Imbibed volume/pore volume Imbibed volume/pore volume Imbibed volume/pore volume Imbibed volume/pore volume Imbibed volume/pore volume Imbibed volume/pore volume Pet. Sci. (2015) 12:636–650 643 0.50 0.14 0.45 0.12 0.40 H-1 0.35 0.10 H-2 0.30 H-3 0.08 0.25 S-1 H-4 0.06 0.20 H-5 S-2 0.15 H-6 S-3 0.04 H-7 0.10 H-8 0.02 0.05 0 0 01 2 3 4 567 8 0 1 2 3 4 5 0.5 0.5 Sqrt(t), h Sqrt(t), h (a) (b) 0.020 0.014 0.018 UY-1 0.012 0.016 UY-2 UY-3 0.014 0.010 UY-4 0.012 LY-1 0.008 L-1 0.010 LY-2 L-2 0.006 0.008 L-3 LY-3 L-4 LY-4 0.006 0.004 L-5 LY-5 0.004 L-6 LY-6 0.002 L-7 0.002 LY-7 L-8 0 0 0 2 4 6 8 10 12 14 16 0.5 0.5 Sqrt(t), h Sqrt(t), h (c) (d) 0.007 0.006 0.006 0.005 LM-1 0.005 0.004 LM-2 N-1 0.004 LM-3 0.003 N-2 LM-4 0.003 LM-5 N-3 0.002 LM-6 N-4 0.002 N-5 0.001 0.001 N-6 01234567 8 9 0 5 10 15 20 0.5 0.5 Sqrt(t), h Sqrt(t), h (e) (f) Fig. 3 Imbibed volume per sectional area versus the square of time in group 1. a S formation; b H formation; c UY/LY formations; d L formation; e LM formation; f N formation an additional imbibition capacity beyond the initial pore water could enter a space that gas could not enter. There- volume. In addition, the R values of shales in the L and N fore, the water imbibed into the tight rocks exits in the formations are higher than 1, though these formations have space that can be divided into two parts: the pore space and good imbibition reproducibilities. This suggests that the the clay crystal lattice space. In addition, the imbibed water Imbibed volume/surface area, cm Imbibed volume/surface area, cm Imbibed volume/surface area, cm Imbibed volume/surface area, cm Imbibed volume/surface area, cm Imbibed volume/surface area, cm 644 Pet. Sci. (2015) 12:636–650 volume per unit pore volume, R can be used as a parameter can be used to determine the fracturing fluid volume intake. to indicate the clay content and type. A plot of the imbibition capacity versus the porosity is shown in Fig. 7a. The general trend is that the imbibition 3.4 Influencing factors capacity increases with increasing porosity. The green line in Fig. 7a shows when the imbibition capacity is equal to In this section, the authors try to address the factors influ- the porosity. In other words, the imbibed volume is equal to encing the imbibition capacity and rate, including the poros- the pore volume. Obviously, when the rocks have high ity, permeability, clay minerals, and the fluid component. porosities, the points tend to be below the line. One explanation is that the rocks with high porosities have low 3.4.1 Porosity water saturation. When the rocks have low porosities, the points tend to be above the line. Therefore, the plot can be For a ratio of imbibition volume to pore volume of more divided into three regions. In Region I, the low-porosity than 1, the conventional definition of water location is not region, the points are above the line, and R tends to exceed applicable for tight formations. The imbibition capacity 1. The imbibition capacity is mainly controlled by the clay can be defined as the imbibition volume per unit dry mineral content. In Region II, the moderate-porosity sample volume. The parameter is broadly applicable and region, the points are on the line, and R tends to be 1. The imbibition capacity is mainly controlled by the clay min- eral content and porosity. In Region III, the high-porosity region, the points are below the line, and R tends to be lower than 1. In this region, the rocks with high porosities A A tend to have low clay contents. Therefore, the imbibition capacity is mainly controlled by the porosity. In Fig. 7b, the imbibition rate is positively correlated Region 3 with the porosity, which is consistent with the prediction of Region 2 3 Eq. (2). This result means that porosity is the main con- trolling factor for the imbibition rate. 3.4.2 Permeability Region 1 In Fig. 8, the imbibition capacity and rate are positively correlated with permeability. However, the correlation is not strong. In addition, there is an obvious departure from 0 5 10 15 20 25 30 35 40 the prediction of Eq. (3). This result means that perme- 0.5 Sqrt(t), h ability is not the main controlling factor for the imbibition Fig. 4 Schematic of imbibition curve behavior rate. 9 1 UY/LY UY/LY LM N 0.1 LM N UY 0.01 0.001 0 0.0001 02468 10 12 Sample number Sample number (a) (b) Fig. 5 Distribution of the imbibition capacity normalized by porosity and imbibition rate. a Imbibition capacity by porosity; b Imbibition rate Imbibed volume/surface area, cm Imbibed volume/pore volume 0.5 Imbibition rate A , cm/h i Pet. Sci. (2015) 12:636–650 645 3.5 3.5 3.0 3.0 2.5 2.5 2.0 2.0 1st immersion (H-9) 1st immersion (S-4) 1.5 1.5 2nd immersion (H-9) 2nd immersion (S-4) 1.0 1.0 3rd immersion (H-9) 0.5 0.5 0 10 20 30 40 50 60 70 0 5 10 15 20 25 Time, h Time, h (a) (b) 1.2 0.9 0.8 1.0 0.7 0.8 0.6 0.5 0.6 1st immersion (L-9) 0.4 1st immersion (UY-5) 2nd immersion (L-9) 0.4 0.3 2nd immersion (UY-5) 3rd immersion (L-9) 0.2 0.2 3rd immersion (UY-5) 0.1 0 10203040506070 80 90 100 010 20 30 40 50 Time, h Time, h (c) (d) 2.0 0.8 1.8 0.7 1.6 0.6 1.4 0.5 1.2 0.4 1.0 1st immersion (LM-7) 1st immersion (N-7) 0.8 0.3 2nd immersion (LM-7) 0.6 0.2 2nd immersion (N-7) 3rd immersion (LM-7) 0.4 0.1 0.2 01 23 4 56 7 89 020 40 60 80 100 120 140 160 180 Time, h Time, h (e) (f) Fig. 6 Repeated immersion tests in group 2. a S formation; b H formation; c UY formation; d L formation; e LM formation; f N formation 3.4.3 Clay minerals porosity is normalized. Then the water saturation and the driving force coefficient are used to represent the imbibi- In order to explore the effect of the clay content and type tion capacity and rate, respectively. on the imbibition capacity and rate, the effect of the Enhancement Imbibed mass/dry mass, % Imbibed mass/dry mass, % Imbibed mass/dry mass, % Imbibed mass/dry mass, % Imbibed mass/dry mass, % Imbibed mass/dry mass, % 646 Pet. Sci. (2015) 12:636–650 UY/LY 0.1 LM N Region II 0.01 Region I Region III S H 0.001 UY/LY L LM N 0.0001 0.1 1 10 100 0 3 6 9 12 15 Porosity, % Porosity, % (a) (b) Fig. 7 Imbibition capacity and rate versus porosity. a Imbibition capacity; b Imbibition rate 0.1 0.01 0.001 UY/LY UY/LY LM N LM N 0.0001 0.0001 0.001 0.01 0.1 1 10 0.0001 0.001 0.01 0.1 1 10 Permeability, mD Permeability, mD (a) (b) Fig. 8 Imbibition capacity and rate versus permeability. a Imbibition capacity; b Imbibition rate The imbibition capacity is positively related to the total the LY formation does not contain smectite. The specific clay content, the I/S, and illite concentrations, as shown in surface area of smectite is much higher than that of I/S, Figs. 9a, 10a, and 11a. The imbibition rate is also posi- which is probably the main reason for the deviation from tively related to the total clay content and the I/S concen- the rule in the LY formation. This proves that the imbi- tration, as shown in Figs. 9b and 10b. However, the bition capacity is related to not only the clay content but correlation between the imbibition rate and the illite con- also the clay type. centration is not strong. In Fig. 9a, the imbibition capacity is closely related to 3.4.4 Fluids the total clay concentration. However, the H and LY formations deviate from the rule. The H and LY forma- In group 3 experiments, a surfactant solution and a KCl tions have relatively high clay contents, but their imbi- solution changed the two driving forces and were used to bition capacities are lower than those of the L and LM explore the effect of the fracturing fluid component on the formations. The clay type in the H formation is mainly imbibition capacity. chlorite, while the clay of the L formation is mainly I/S. The contact angle on the surface of LM-8 rock was I/S has a relatively high specific surface area that can measured with an imaging method, as shown in Fig. 12. absorb larger amounts of water. Unlike the LM formation, The contact angle is 14 before cationic surfactant Imbibed volume/dry sample volume, % Imbibed volume/dry sample volume, % 0.5 Imbibition rate A , cm/h 0.5 Imbibition rate A , cm/h i Pet. Sci. (2015) 12:636–650 647 S H S H UY (74.9, 7.5) LY LY UY N 6 0.1 UY N LM LM 0.01 LM (36.9, 2.9) L (23.7, 1.9) LY (47.8, 2.3) 0.001 S (10.3, 0.7) H (31.2, 0.9) N (23.3, 1.1) 0.0001 0 20 40 60 80 0 10 100 Total clay concentration, wt% Total clay concentration, wt% (a) (b) Fig. 9 Imbibition capacity normalized by porosity and driving force coefficient versus total clay concentration. a Imbibition capacity normalized by porosity; b Driving force coefficient S H LY LY L UY N LM UY N 0.1 LM 0.01 0.001 0 0.0001 0 10 20 30 40 50 60 0 10 100 I/S concentration, wt% I/S concentration, wt% Fig. 10 Imbibition capacity normalized by porosity and driving force coefficient versus I/S concentration. a Imbibition capacity normalized by porosity; b Driving force coefficient S H S H LY LY UY N 6 0.1 UY N LM LM 0.01 0.001 0.0001 02468 10 12 14 0.1 1 10 100 Illite concentration, wt% Illite concentration, wt% Fig. 11 Imbibition capacity normalized by porosity and rate versus illite concentration. a Imbibition capacity normalized by porosity; b Driving force coefficient Average imbibed volume/pore volume Average imbibed volume/pore volume Average imbibed volume/pore volume Driving force coefficient A /K Driving force coefficient A /Kφ 2 i Driving force coefficient A /Kφ i 648 Pet. Sci. (2015) 12:636–650 Fig. 12 Contact angle on sample LM-8 with a water drop. a Before surfactant treatment (14); b After surfactant treatment (51) treatment and 51 after treatment. The surface wettability is 70.0 65.0 altered and tends to be intermediately wet, which may lead DI water 60.0 to a lower imbibition capacity. 10 wt% KCl 52.3 The capillary suction time (CST) tests are quick and 50.0 47.4 45.0 easy and can provide information about inhibition charac- 42.5 38.2 40.0 teristics of an additive (Berry et al. 2008). The authors tried 34.0 to understand the inhibitive ability of a 10 wt% KCl 30.2 29.1 28.6 30.0 23.8 solution on clay expansion and explored its effect on the 22.5 20.0 imbibition capacity. The CST apparatus measures the time required for the fluid to travel a given distance, as shown in 10.0 Fig. 13. A short CST time interval reflects a poor clay dispersibility. The results of CST tests are shown in 0.0 SN LM L UY H Fig. 14. The relative relationship of the CST time in tight formations is UY [ LM [ N [ L [ H [ S, which is Fig. 14 CST time in different formations with two solutions positively related to the I/S concentration. The 10 wt% KCl solution inhibits the clay expansion in tight formations, the imbibition capacity. For the UY and LM formations, especially for the UY and LM formations. This effect may the reduced imbibition capacity is more obvious. Figure 16 lead to a lower imbibition capacity. shows pictures of the UY and LM formation samples after The results from group 3 experiments are presented in exposure to the surfactant and 10 wt% KCl solutions. Fig. 15. The surfactant and 10 wt% KCl solutions reduce When exposed to deionized water, the sample from the UY 9.5 9.1 8.8 8.4 8.3 DI water 7.6 Surfactant 7 10 wt% KCl 6 5.7 4.6 4.2 3.5 3.1 2.9 2.8 2.4 2.2 2.1 1.9 1.8 SH UY L LM N Fig. 15 The effect of imbibed fluids on the imbibition capacity in Fig. 13 CST apparatus group 3 Imbibed volume/dry sample volume, % CST time, s Pet. Sci. (2015) 12:636–650 649 imbibition capacity normalized by the porosity increases with increasing total clay, I/S, and illite concentrations. Smectite and I/S with a high specific area tend to lead to strong water imbibition. (4) The water volume imbibed into the clay-rich tight rocks is much greater than the pore volume mea- sured by helium. The imbibed volume per unit pore volume is a parameter that can evaluate the effect of the clay content and type. (5) Surfactants can change the imbibition capacity of tight rocks by altering the interfacial tension and wettability. A 10 wt% KCl solution can inhibit the clay expansion, which also reduces the imbibition capacity. Acknowledgments This research program was financially sup- ported by the National Basic Research Program of China (973 Pro- gram) Granted No. 2015CB250903 and the National Natural Science Foundation of China Granted No. 51490652. The Chongqing Institute of Geology and Mineral Resources supported this field work. Open Access This article is distributed under the terms of the Fig. 16 Pictures of tight rock samples after exposure to different Creative Commons Attribution 4.0 International License (http://crea fluids. a UY formation; b LM formation tivecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a formation disintegrates completely, and the sample from link to the Creative Commons license, and indicate if changes were the LM formation generates some micro-fractures along made. the bedding plane. The 10 wt% KCl solution has a good inhibitive effect on disintegration and micro-fractures, References which is consistent with the results of CST tests, as shown in Fig. 16. Berry SL, Boles JL, Brannon HD, et al. Performance evaluation of ionic liquids as a clay stabilizer and shale inhibitor. In: SPE international symposium and exhibition on formation damage 4 Conclusions control, 13–15 February, Lafayette, Louisiana, USA. 2008. doi:10.2118/112540-MS. Dehghanpour H, Lan Q, Saeed Y, et al. Spontaneous imbibition of A series of spontaneous imbibition experiments were car- brine and oil in gas shales: effect of water adsorption and ried out, and the following conclusions were reached: resulting microfractures. Energy Fuels. 2013;27(6):3039–49. doi:10.1021/ef4002814. (1) Tight rock imbibition can be characterized by the Fakcharoenphol P, Kurtoglu B, Kazemi H, et al. The effect of osmotic imbibition capacity, the imbibition rate, and the pressure on improve oil recovery from fractured shale forma- diffusion rate. The imbibition capacity and rate are tions. In: SPE unconventional resources conference, 1–3 April, The Woodlands, Texas, USA. 2014. doi:10.2118/168998-MS. positively correlated with porosity. The imbibition Handy LL. Determination of effective capillary pressures for porous curves can typically be divided into three regions: media from imbibition data. Trans AIME. 1960;219:75–80. the initial linear imbibition region, the transition Hu Q, Ewing RP, Dultz S. Low pore connectivity in natural rock. region, and the diffusion region. J Contam Hydrol. 2012;133:76–83. doi:10.1016/j.jconhyd.2012. 03.006. (2) The water-holding space in tight rocks consists of Lan Q, Dehghanpour H, Wood J, et al. Wettability of the Montney two parts: the pore space and the clay crystal lattice tight gas formation. In: SPE/CSUR unconventional resources space. The driving forces of spontaneous imbibition conference—Canada, 30 September–2 October, Calgary, in tight rocks are the capillary force and the clay Alberta, Canada. 2014. doi:10.2118/171620-MS. Makhanov K, Dehghanpour H, Kuru E. An experimental study of absorption force. A new parameter, the effective spontaneous imbibition in Horn River shales. In: SPE Canadian driving force of imbibition, is defined to describe the unconventional resources conference, 30 October–1 November, effect of clay absorption. Calgary, Alberta, Canada. 2012. doi:10.2118/162650-MS. (3) The clay mineral content significantly influences the Makhanov K, Habibi A, Dehghanpour H, et al. Liquid uptake of gas shales: a workflow to estimate water loss during shut-in periods imbibition capacity for clay-rich tight rocks. The 123 650 Pet. Sci. (2015) 12:636–650 after fracturing operations. J Unconv Oil Gas Resour. Sun Y, Bai B, Wei M. Microfracture and surfactant impact on linear 2014;7:22–32. doi:10.1016/j.juogr.2014.04.001. cocurrent brine imbibition in gas-saturated shale. Energy Fuels. Penny GS, Dobkins TA, Pursley JT. Field study of completion fluids 2015;29(3):1438–46. doi:10.1021/ef5025559. to enhance gas production in the Barnett Shale. In: SPE gas Yuan W, Li X, Pan Z, et al. Experimental investigation of interactions technology symposium, 15–17 May, Calgary, Alberta, Canada. between water and a lower Silurian Chinese shale. Energy Fuels. 2006. doi:10.2118/100434-MS. 2014;28(8):4925–33. doi:10.1021/ef500915k. Roychaudhuri B, Tsotsis TT, Jessen K. An experimental investigation Zhou Z, Hoffman BT, Bearinger D, et al. Experimental and numerical of spontaneous imbibition in gas shales. J Pet Sci Eng. study on spontaneous imbibition of fracturing fluids in shale gas 2013;111:87–97. doi:10.1016/j.petrol.2013.10.002. formation. In: SPE/CSUR unconventional resources confer- Sharma M, Agrawal S. Impact of liquid loading in hydraulic fractures ence—Canada, 30 September–2 October, Calgary, Alberta, on well productivity. In: SPE hydraulic fracturing technology Canada. 2014. doi:10.2118/171600-MS. conference, 4–6 February, The Woodlands, Texas, USA. 2013. doi:10.2118/163837-MS.

Journal

Petroleum Science – Springer Journals

Published: Sep 21, 2015

There are no references for this article.