Experimental study of seepage characteristics of single rock fracture based on stress states and stress history

ZHANG Chi, ZHANG Yanjun, LI Zhengwei, ZHANG Tong, LIU Tong and XIE Yangyang

College of Construction Engineering, Jilin University, Changchun 130012, China



10.3969/j.issn.1673-9736.2016.03.06

Experimental study of seepage characteristics of single rock fracture based on stress states and stress history

ZHANG Chi, ZHANG Yanjun, LI Zhengwei, ZHANG Tong, LIU Tong and XIE Yangyang

CollegeofConstructionEngineering,JilinUniversity,Changchun130012,China

Through seepage tests under different loading and unloading confining pressures and different hydraulic gradients, the authors studied the effects of stress states and stress history on fracture permeability evolution for single granite fracture and sandstone fracture. The results show that there exists a linear relationship between the seepage discharge and osmotic pressure in sandstone fissure under each level of confining pressure. With the increasing in the confining pressure, the permeability of the fracture decreases, but the decreasing rate is changeing. During the unloading process, the fracture seepage velocity cannot be fully recovered to the size of the loading process. Therefore, in the unloading process of the confining pressure, the recovery of fracture permeability shows obvious hysteresis effects. The flow rate of the fracture remains unchanged during five cycles of loading and unloading processes of the confining pressure. In each cycle, the evolution character of the flow rate with the confining pressure remains unchanged. These experiments show that the seepage characteristics of sandstone and granite fractures are not the same under the same stress state.

rock single fracture; rock seepage; stress states; stress history

1　Introduction

Understanding the evolution of rock seepage characteristics with stress state is important in study of rock mechanics and rock engineering (Zhietal., 2012), such as in the geological burial of nuclear waste, CO2geological sequestration, and exploitation of oil and geothermal resources (Zengetal., 2012). In naturally fractured rock masses, permeability of rock is mainly contributed by fracture seepage. The spatial distribution and interconnectivity of fractures, as well as the properties of fluid flow through a single fracture, essentially determine the overall permeability and flow distribution of rock masses (Juetal., 2013). Thus, an accurate knowledge and description of the behavior of fluid flow and the interaction mechanism of stress and fluid flow are of great significance to solving these engineering problems (Xiongetal., 2009).

Since Snow (1968) proposed cube theorem, the relationship between stress states [normal stress (Shenetal., 2010), shear stress (Xu & Xie, 2009), and complex stress (Houetal., 2011)] and the properties of fluid flow through a single fracture has been studied. However, the literature reveals that no mature theory explains the relationship between stress and single fracture seepage. Nonetheless, the relationship between stress history and single fracture seepage has also been studied (Braceetal., 1968). However, the study of the relationship between fracture seepage characteristics and stress state is still in the exploratory stage (Yangetal., 2013). To improve the exploration in this field, the relationship between stress and single fracture flow is studied in the present work.

In this study, we first introduce test methods, including sample preparation, test equipment, and test processes. Secondly, laboratory experimental data are used to study the relationship among the seepage characteristics of rock fracture and stress states, stress history, and lithology. Finally, the conclusions of this experimental study are summarized, and suggestions for future research are proposed.

2　Laboratory experiment scheme

2.1Preparation of rock sample

Two types of lithology are chosen: sandstone and granite. The mineral composition and quality percent content of the sandstone are as follows: quartz, 54%-59%; alkaline feldspar, 12%-17%; plagioclase, 13%-17%; and others 7%-18%. The mineral composition and quality percent content of the granite are as follows: quartz, 51%-56%; alkaline feldspar, 17%-22%; plagioclase, 21%-27%; and others 1%-3%. All the rock samples measure φ50 mm×H100 mm.

A straight smooth fracture is first cut through each sample in the direction parallel to the axis of the sample (Fig.1).

2.2Laboratory experiment

The experiments in this study are carried out on the fracture flow simulation system, which was developed by the deep geothermal and hot dry rock research team of Jilin University. The main parameters of this experiment system are as follows: confining pre-ssure, 0-40 MPa; osmotic pressure, 0-50 MPa; and temperature, 0-200℃. The diameter of the test sample is required to be around 50 mm, and the height can range from 100 mm to 140 mm. All flow pipelines of the system are made of stainless steel to avoid the corrosion effects of high temperature and pressure environments on the system. An illustrative diagram of the experiment system is shown in Fig.2.

Fig.1　Rock samples

Fig.2　Structure diagram of rock fracture seepage test system

Working fluids are injected into the rock fracture with an injection pump. Different levels of confining pressure can be applied to the rock sample.

2.3Experimental procedure

2.3.1Test under different stress conditions

The artificially made rock sample is first placed into the experiment system. Fracture flow rates are tested while the confining pressure and osmotic pressure are maintained at 5 MPa and 1 MPa, respectively. At the same level of confining pressure, osmotic pressure is gradually increased to the ultimate value of 70% of the confining pressure. Thereafter, the confining pressure is increased to 10, 15, 20, 25, and 30 MPa. The experiments are repeated as mentioned above, under different confining pressures.

2.3.2Test under different stress histories

The artificially made rock sample is first placed into the experiment system. The fracture flow rates are tested while the confining pressure and osmotic pressure are maintained at 5and 1 MPa, respectively. Gradually, the confining pressure is increased to 10, 15, 20, 25, and 30 MPa, and the evolution of the fracture flow rates is recorded. Thereafter, the confining pressure is gradually decreased to 25, 20, 15, 10, and 5 MPa, and the evolution of the fracture flow rates is recorded. The loading and unloading processes are repeated in five cycles.

3　Results of laboratory experiment

3.1Relationship between stress state and fracture seepage flow

As shown in Fig.3, there exists a linear relationship between the seepage discharge and osmotic pressure in sandstone fissure under each level of confining pressure. With the increase in the confining pressure, the permeability of the fracture decreases, and the decreasing rate changes. As shown in Fig.4, the fracture seepage flow rate with the increase in confining pressure decreases rapidly at the stage with low confining pressure (10-20 MPa). When the confining pressure is increased from 25 MPa to 30 MPa, the flow rate of the fracture under different osmotic pressures is slowly decreased and stabilized at 0.18-0.35 mL/min.

Fig.3　Relationship between osmotic pressure and flow rate of sandstone fracture

Fig.4　Relationship between confining pressure and flow rate of sandstone fracture

3.2Relationship between stress history and fracture seepage flow

There exists a nonlinear relationship between the confining pressure and fracture flow rate in the first cycle of the loading and unloading processes (Fig.5). In the unloading process, the fracture flow rate cannot be fully recovered in comparison with the value in the loading process. Hence, the recovery of fracture permeability shows obvious hysteresis effects in the unloading process of the confining pressure. Such effects can be explained by the fact that the deformation of the fracture surface cannot recover in time. Thus, the permeability of the same rock fracture can vary in significant ranges even under the same stress conditions.

Fig.5　 Relationship between confining pressure and seepage flow rate of sandstone fracture in process of single loading and unloading

Fig.6　Relationship between confining pressure and flow rate of sandstone fracture during five cycles of loading and unloading processes

Fig.6 illustrates that the flow rate of the fracture remains unchanged during the five cycles of loading and unloading processes of the confining pressure. In each cycle, the evolution character of the flow rate with the confining pressure remains unchanged, thus indicating that the deformation is elastic and recoverable.

3.3Relationship between lithology and fracture seepage flow

Fig.7　Relationship between osmotic pressure and flow rate of granite fracture

Fig.7 shows the seepage characteristics of granite. Compared with the experiment results for sandstone (Figs.3,7), the following conclusions can be drawn from the results for granite. There exists a linear relationship between the flow rate and seepage pressure of the granite fracture at each level of the confining pressure. The magnitude of the seepage flow rate of the single sandstone fracture is 10 times that of the granite fracture under the same confining pressure.This result can be explained by the rougher surface of the sandstone fracture and the inadequate closure of the fracture surface. With the increase of confining pressure, the permeability variation of the granite fracture is smaller than that of the sandstone fracture. Therefore, the seepage characteristics of rock fracture obviously vary with different lithologies.

4　Conclusions

(1) The flow rates and osmotic pressures of sandstone and granite fractures indicate a linear relationship. Moreover, confining pressure can significantly reduce fracture permeability.

(2) In the unloading process of confining pressure, the recovery of fracture permeability exhibits obvious hysteresis effects, which indicate that the effects of stress history cannot be ignored.

(3) These experiments show that the seepage characteristics of sandstone and granite fractures are not the same under the same stress state. Therefore, future studies need to conduct extensive experiments to further explore the seepage characteristics of rock fractures.

Brace W F, Walsh J B, Frangos W T. 1968. Permeability of granite under high pressure.JournalofGeophysicalResearch, 73(6): 2225--2236.

Hou Z F, Ji H G, Wang J A,etal. 2011. Experiment research on the laws of seepage in sandstone within single fracture under triaxiality stresses.ChinaMiningMagazine, 20(5): 107--110. (in Chinese with English abstract)

Ju Y, Zhang Q G, Yang Y M,etal. 2013. An experimental investigation on the mechanism of fluid flow through single rough fracture of rock.ScienceChinaTechnologicalSciences, 56(8): 2070--2080.

Shen L F, Feng X T, Pan P Z,etal. 2010. Experimental research on mechano-hydro-chemical coupling of granite with single fracture.ChineseJournalofRockMechanicsandEngineering, 29(7): 1379--1388. (in Chinese with English abstract)

Snow D T. 1968. Rock fracture spacings, openings, and porosities.JournalofSoilMechanics&FoundationsDivision, 94: 73--91.

Xiong X B, Zhang C H, Wang E Z. 2009. A review of steady state seepage in a single fracture of rock.ChineseJournalofRockMechanicsandEngineering, 28(9): 1839--1847. (in Chinese with English abstract)

Xu L H, Xie N. 2009. Test and theoretical research on permeability characteristics of shear fracture in rock mass.ChineseJournalofRockMechanicsandEngineering, 28(11): 2249--2257. (in Chinese with English abstract)

Yang J B, Feng X T, Pan P Z,etal. 2013. Experimental study of permeability characteristics of single rock fracture considering stress history.RockandSoilMechanics, 34(6): 1629--1635. (in Chinese with English abstract)

Zeng Y C, Su Z, Wu N Y,etal. 2012. Temperature distribution characteristics of bedrock fracture groundwater system at Zhangzhou geothermal.JournalofJilinUniversity:EarthScienceEdition, 42(3): 814--820. (in Chinese with English abstract)

Zhi Q G, Zhao Y Z, Cheng Y F,etal. 2012. Finite element simulation on percolation of low permeability reservoir with permeability anisotropy.GlobalGeology, 15(4): 265--269.

19 January 2016, accepted 1 March 2016

Supported by National Natural Science Foundation of China (No.41372239); Graduate Innovation Fund of Jilin University (No.2015096)

Article ID: 1673-9736(2016)03-0177-05