Abstract

The unconventional resources from an ultradeep tight gas reservoir have received significant attention in recent decades. Hydraulic fracturing is the main method for tight gas reservoir development because of its extremely low permeability and porosity. During hydraulic fracturing, high hydraulic fracturing fluid (HFF) that invaded the zone near the fracture face may reduce gas relative permeability significantly and impede gas production. The sources of this damage can be the high capillary pressure (HCP) and the presence of water-sensitive clays (PWC). For tight rock, it is usually infeasible to identify the primary damage mechanism using the traditional steady-state measurement method due to long measurement time and gauge accuracy. In this paper, we present a new experimental approach to identify the primary mechanism of the fracture face damage (FFD) through the application of the pressure transmission method and pressure decay method. Both rock matrix and naturally fractured tight samples (depth 18,000 ft, Tarim field, China) were tested. The experimental results showed that the average high capillary pressure damage indexes () of rock matrix cores and naturally fractured cores are 94.9% and 92.4%, respectively, indicating severe damage caused by HCP. The average clay-swelling and mobilization (CSM) damage indexes () of rock matrix cores and naturally fractured cores are 29.6% and 38.4%, respectively, indicating that the damage caused by CSM is lighter than that by HCP. HCP is the primary damage mechanism for the tight sandstone. And the damage degree of the rock matrix cores is higher than that of the naturally fractured core. The proposed procedures can be applied to identify the FFD mechanism of other tight and shale formation and provide insightful fundamental data for HFF optimization.

1. Introduction

The unconventional resources from an unconventional tight/shale reservoir have received significant attention in the past decades. Tight or shale gas plays have specific characteristics, such as tight reservoir rock with micro-nanodarcy permeability, high temperature (up to 200°C), and high-stress deviation [13]. During hydraulic fracturing, the viscosity of hydraulic fracturing fluid (HFF) should be carefully optimized considering the proppant carrying ability and filtrate loss control [46]. For conventional resources such as sandstone or carbonate formation, the viscosity of HFF should be higher to achieve the proppant carrying capacity and fluid loss control purpose. However, for the tight/shale play, the permeability of the rock matrix range is from 10−3 to 10−8μm2; the HFF could not invade to the rock matrix of the fracture surface in a deep distance due to the micrometer-to-the-nanometer pore size of tight sandstone or shale and the presence of water sensitive clay in tight or shale plays. Because the permeability of the tight reservoir is extremely low (less than 0.1 mD) and the pore size of tight cores is in micron or nanometer scale, tight and shale rocks usually have some clay. Clay is composed of fine slices of silicate aluminate minerals. When immersed in fracture fluid, water molecules will enter the crystal lattice of clay minerals, which is easy to cause volume expansion, thus blocking the pores of the fracture surface [710].

The fluid saturation near the fracture face zone (1–2 cm) may reduce gas relative permeability significantly thus impeding gas production [11]. Holditch [12] concluded that during fracturing fluid injection, when the water saturation is constant, the capillary pressure is inversely proportional to the square root of permeability, resulting in the reservoir being damaged and capillary force being increased. Therefore, the capillary force has a great influence on the reservoir recovery rate. Lan et al. [13] pointed out that the water absorption rate of shale/tight samples largely depended on the clay content, because a large amount of water in the injection fluid can be adsorbed by the clay flakes. Makhanov et al. [14] concluded that the fracturing fluid with high viscosity can only partially reduce the water absorption and the low recovery rate of water-base fracturing fluid is partly due to the spontaneous imbibition of water into the shale matrix. Therefore, the damage of fracturing fluid to the reservoir matrix should be paid more attention to and cannot be ignored. The specific permeability reduction mechanism can be as follows: (1) pore fluid is blocking due to the micro-nanopore size. A smaller pore size may cause higher capillary pressure (HCP) and increase the fluid saturation at the rock matrix near the fracture surface, thus reducing the gas relative permeability significantly and (2) pore size reduction due to the high clay content, especially for swelling clay minerals such as smectite and easy mobilization or pore blocking minerals such as illite.

For tight or shale cores, it is usually infeasible to identify which is the primary damage factor by the traditional steady-state measurement core flooding apparatus due to long measurement time (hours or even days) and the gauge accuracy (pressure gauge and flowmeter) [15]. According to experimental theory, permeability measurements in the laboratory can be divided into two categories: the steady-state method and unsteady-state method [16]. While the research on fracture fluid damage evaluation of tight reservoirs is limited and conventional evaluation methods are still widely used, core samples with 1-inch length and 2 inch-diameter are usually used to measure permeability damage in the conventional steady-state method [17]. The conventional steady-state method needs to test permeability many times under different pore pressures in order to adjust slippage effect and then get Klinkenberg permeability independent of test pressure through Klinkenberg equation matching [16, 18]. Due to the extremely low permeability of tight core samples, conventional core permeability measurements need a long time to stabilize the flow rate and pressure, and the experimental error is largely caused by temperature changes during the experiment. Therefore, the accuracy of experimental results is affected [17, 1921].

Brace et al. [22] proposed the pressure pulse method to solve the deficiency of conventional core permeability measurements. The pressure pulse method is different from the traditional constant pressure method and constant flow rate method, which is a transient pressure pulse method based on one-dimensional unsteady flow theory. When testing permeability, the different displacing velocity and pressure at the exit need not to be recorded, while only the attenuation of the pressure pulse needs to be recorded. This method can help in precisely controlling the flow rate and guarantee the measurement accuracy. The permeability could be calculated according to the testing results combined with the corresponding theoretical formula [23]. For tight reservoirs, it is difficult to measure the flow rate when measuring the liquid permeability of the rock sample. However, measuring the change of the pressure with time as the fluid passes through the core can be a good way to evaluate the permeability [17, 24, 20].

In this study, the pressure transmitter was designed and developed independently for matrix damage evaluation of the low-permeability tight Tarim reservoir based on the pressure pulse method. A new method for matrix damage evaluation of the low-permeability tight reservoir called the pressure transmission method for testing low permeability was established. The primary formation damage mechanism was identified through the application of an integrated unsteady-state experimental apparatus.

2. Methods and Procedures

2.1. Theory and Methods

A novel unsteady-state formation damage evaluation method based on pressure transmission and pressure decay theories was used to investigate the FFD mechanism. The previous researchers well documented the pressure decay theory and pressure transmission theory as well as the experimental procedures ([2534]). The permeability measurement theory of the pressure decay method can be described in equations (1) and (2).

Pressure difference:

Effective gas permeability:

The permeability measurement theory of the pressure transmission method can be described in equation (3).

2.2. The Experimental Apparatus

The PDP200 (pressure decay method) apparatus was used to measure the degree of fracture face damage caused by HCP, while the PTM (pressure transmission method) device was used to measure the degree of fracture face damage due to CSM. Figures 1 and 2 showed the PDP200 and PTM apparatus, respectively. The PDP200 can measure the permeability in a 10−3 to 10−8μm2 range, and the PTM permeability measurement range is also from 10−3 to 10−8μm2.

2.3. Experimental Procedures

The specific procedures of the HCP damage test are shown as follows: (1)Dry the core in the oven at 60°C for twenty-four hours or until the mass of the core is unchanged(2)Measure the initial permeability of the dried core (1 in diameter and 2 in length) by PDP200 apparatus(3)Load the core in a vacuum holder for twenty-four hours or until the pressure gauge showed −15 kPa and is unchanged for two hours and then fill the vacuum holder with simulated formation water (the same ion type and concentration as formation water) with 2% clay-stabilizing agent for twelve hours(4)Measure the permeability of the soaked core by the PDP200 apparatus again(5)Calculate the HCP damage index by equation (6) as follows:

The objective of adding a clay-stabilizing agent is to avoid the clay swelling during the measurement. The advantage of the pressure decay method is that it would not induce “fluid force” that can cause the clay mobilization. The indicates the degree of permeability impediment by fluid saturation change or by “fluid block.”

The experimental flow chart of the CSM damage test is shown in Figure 3.

The experimental procedures of the CSM damage test are shown as follows: (1)Mix epoxy resin and hardener at 1 : 1 ratio by weight, pour the mixture inside a polycarbonate tubing (2.5 in diameter, 4 in length), place the core in the center of the pipe, and cure for twenty hours. Slice the core in 0.25 in thickness (Figure 4), name as core slice, and load to a vacuum holder(2)Close all control valves (8)–(17) and open valves (12) and (13) to vacuum(3)Close control valves (8)–(17) and open valves (10) and (11). The formation water is injected into the downstream of the rock sample at the specific pressure. After the downstream pressure stabilized, it is recorded as the initial downstream pressure , which should be the same as the original pore pressure of the rock sample(4)Close control valves (8)–(17) and open ISCO pump (1), intermediate vessel (4), special core gripper (7), back pressure valve (23), gas pressure-regulating valve (22), nitrogen gas bottle (3), and valves (8), (9), (14), and (17). The formation water flows continuously on the surface of the rock sample with flow pressure and back pressure . Continuously monitor the change of the downstream pressure until it reaches the upstream flow pressure (5), , , and equations (4) and (5) were used to obtain the liquid permeability of the rock sample(6)Close valves (8)–(17) and open special core gripper (7), intermediate vessel (6), gas pressure regulating valve (22), nitrogen gas bottle (3), and valves (15) and (16). Adjust gas pressure regulating valve (22), setup the pressure, and push the working fluid into the surface of the rock sample for approximately 5 hours(7)Close valves (8)–(17) and repeat procedures (2)–(5). The measured liquid permeability of the damaged rock sample is (8)The damage rate of working fluid to the core was calculated through equation (7) as follows:

The indicates the degree of permeability impediment by clay swelling and mobilization. Finally, by comparing the and , the primary formation damage mechanism can be identified.

3. Results and Discussion

Nine tight sandstone cores of an ultradeep tight gas formation in the Tarim field (China) at a depth of around 18000 ft including two rock matrix cores and seven naturally fractured cores with different fracture widths were used for the HFF evaluation and optimization.

Figure 5 is a typical rock matrix core, and Figure 6 showed a naturally fractured core. All diameter (1 in) and length (2 in) of the cores were the same; Table 1 listed the detailed properties of the samples. The porosity of the cores is from 0.95% to 6.09%. For the naturally fractured cores, the fracture width is ranging from 0.020 in to 0.048 in and the fractures were filled by calcite based on XRD analysis. The average formation water salinity of the Tarim tight gas formation is 247025 mg/L (Table 2), and 2% KCl was also added as a clay-stabilizing agent.

3.1. The HCP Damage Test Result

Table 3 summarized the experimental data for the HCP damage test. The is from 88.4% to 99.7%, indicating severe damage caused by HCP. For the naturally fractured cores, it did not show a strong correlation between the and natural fracture width (Figure 7). The average of rock matrix cores and naturally fractured cores are 94.9% and 92.4%, respectively.

3.2. The CSM Damage Test Result

According to the XRD results, the tested tight core samples contain high content of clay minerals, between 7% and 9% (see Table 4). When immersed in water-based fracture fluid, water molecules will enter the crystal lattice of clay minerals, which easily causes volume expansion, leading to water blocking and water sensitivity. Table 5 summarized the experimental data for the CSM damage test. The is from 28.6% to 43.9%, indicating that the damage caused by CSM is lighter than that by HCP. For the naturally fractured cores, the relationship between the and natural fracture width is also not stable (Figure 8). The average of rock matrix cores and naturally fractured cores are 29.6% and 38.4%, respectively.

Results showed that the range is from 88.4% to 99.7% and the range is from 28.6% to 43.9%. HCP damage is identified as the primary damage source of the tight gas formation. The HFF is needed to be optimized, and we modified the HFF by adding 0.5% wt anti-water-blocking agent (surfactant) to reduce the interfacial tension of gas and water. Two cores (one rock matrix core no.17 and a naturally fractured core no.18) were used to evaluate the optimized HFF through the HCP test and CSM test at the same condition as the previous test (see Figures 9 and 10). In the HCP damage test, we also added the same 0.5% wt anti-water-blocking agent (surfactant) to the simulated formation water. Table 6 showed that the is reduced for the two cores (45.3% and 50.2% for the naturally fractured core and rock matrix core, respectively). The is stabilized and within the range of the previous CSM test. The HFF can be further optimized by using more effective surfactant or increase the content of surfactant.

4. Conclusions

In this paper, we introduce a novel experimental approach for tight gas formation damage mechanism identification. The pressure transmission method and pressure decay method were integrated to evaluate the degree of fracture face damage (FFD) caused by HCP and CSM. This method can be further applied to identify the FFD mechanism of other tight or shale formations and provide insightful fundamental data for HFF optimization. The experimental results showed the following specific conclusions: (1)The average of rock matrix cores and naturally fractured cores are 94.9% and 92.4%, respectively, indicating severe damage caused by HCP(2)The average of rock matrix cores and naturally fractured cores are 29.6% and 38.4%, respectively, indicating that the damage caused by CSM is lighter than that by HCP(3)The fluid block due to HCP is the primary damage mechanism for Tarim tight sandstone. And the degree of matrix fluid block is higher than that of natural fractures(4)After adding 0.5% wt surfactant, the is reduced from 94.9% to 50.2% for the rock matrix, which showed the effectiveness of the proposed experimental procedures

Nomenclature

:Pressure of the upper holder (MPa)
:Pressure of the lower holder (MPa)
:The pressure difference between the upper and the lower holders (MPa)
:Unit change factor
:The slope of curve
:Gas viscosity (MPa·s)
:Core length (cm)
:Calibration factor of gas compressibility
:Calibration factor of mass flow rate
:Core cross-section area (cm2)
:Average pore pressure (MPa)
:The volume of the upper holder (cm3)
:The volume of the lower holder (cm3)
:The rock pressure-transmitting coefficient
:Rock porosity (%)
:Fluid compressibility (1/MPa)
:Initial pressure of the upper holder
:Initial pressure of the lower holder
:The volume of the lower holder
:The slop of and .

Data Availability

The data used to support the findings of this study are included within the article.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This research was supported by the National Science Foundation of China (no. 51834005), by the Innovation Fund of China National Petroleum Corporation (2019D-5007-0201), and by the Science Foundation of China University of Petroleum, Beijing (no. 2462017YJRC041). The authors appreciate the 53rd U.S. Rock Mechanics/Geomechanics Symposium for the support to publish this paper.