IPTC-16580-MS-P

IPTC 16580 Rock Mechanical Properties of Shale Gas Reservoir and Their Influences on Hydraulic Fracture Qinghui Li, Mian Chen, Yan Jin, Yu Zhou, China University of Petroleum in Beijing, Fred P. Wang, University of Texas at Austin, and Rixing Zhang, CNPC Tarim Oilfield Company

Copyright 2013, International Petroleum Technology Conference This paper was prepared for presentation at the International Petroleum Technology Conference held in Beijing, China, 26–28 March 2013. This paper was selected for presentation by an IPTC Programme Committee following review of information contained in an abstract submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the International Petroleum Technology Conference and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the International Petroleum Technology Conference, its officers, or members. Papers presented at IPTC are subject to publication review by Sponsor Society Committees of IPTC. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the International Petroleum Technology Conference is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, IPTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax +1-972-952-9435

Abstract Each shale gas play is unique and the stimulation and completion method should be determined based on individual attributes. Selecting the completion style for an emerging shale play should begin in the laboratory rather than copy successful cases. An understanding of the rock mechanical properties is essential to help understand how the shale reservoir should be completed. We conducted triaxial compression tests for specimens from North America and South China. The failure modes and mechanical behaviors are analyzed. A new method combines rock mechanical properties and mineralogy to quantify the brittleness was derived and the practical use is demonstrated afterwards. Natural fracture analysis from core and borehole image are conducted and potentially active fractures, which could be most important hydraulic conduits and another index of the fracability, are identified through a three-dimensional Mohr representational method. The results of this study revealed that black shales are usually more brittle than imagined. Splitting and splitting-shear hybrid failure modes are the main failure modes under low confining pressures, while shear failure modes predominate under higher pressures. Types of shale, burial depth, coring direction and confining pressures are all affecting the mechanical behavior of specimens. The brittleness characterization of shale samples from China is similar to that from America. Compared with North America, shale from South China has higher strength and Young’s modulus, while lower Poisson’s ratio. A systematic approach of designing completion and stimulation is proposed based on brittleness, closure stress and hydraulic fracture width. The analyzing results from two shale gas wells verified the value of this method. Brittleness calculating results tallied well with initial production of these wells, indicating good application prospects. Calculation of strength properties, closure stress and fracture width can be practical useful in selecting and designing stimulation measures. Introductions Large-scale hydraulic fracturing treatments are usually essential in shale gas development for its low porosity and permeability. Besides the natural fractured characteristic, Fracability is another factor should be studied before stimulation (Dunn, 1973; Luffel, 1992; Soeder, 1988; Jarvie, 2007; Li, 2011; Li, 2012, et al). It is found that the brittleness index is an important evaluation parameter in the selection of high quality shale (Wang, 2009, 2010; Li, 2011; Li, 2012). Scholars and oil companies attach great importance to reservoirs mechanical characteristics evaluation. Laboratories also rely on independent equipment to conduct researches in addition to their own research teams. Application of new technologies and Efficiency exploiting are affected due to the unpublished research results. Systematical results related to mechanical properties of gas shales have not been seen yet, and studies involving gas shale brittleness are rarely reported. Some scholars started from conventional fracturing experimental evaluation, and evaluated the brittleness and core sensitivity of reservoirs; however, they didn’t involve the essential question of mechanical mechanism of brittleness fracture (Li, 2012). Thus what we can see is that this field is still lack of systematic analysis and experimental studies (Wang, 2009; Li, 2012). High-pressure servo test system was used to test and analyze the failure modes and mechanical behavior for key shale gas reservoirs in North America and Longmaxi black shale from South China in differential loading conditions. We analyzed brittleness characteristics of the shales, through the consideration of mechanical mechanism and micro feature of brittleness fracture, investigating the mechanical definition of the brittleness performance. By analyzing field case, a comprehensive brittleness evaluation method is investigated based on the double factors of elastic deformation feature and mineral composition. The calculation and analysis process is introduced and verified through two field cases.

2

IPTC 16580

1 Rock mechanical properties of shale 1.1 Experimental equipment and sampling method High-pressure servo test system built by rock mechanics laboratory of China University of Petroleum (Beijing) was used to conduct stress-strain tests on standard specimens under uniaxial and different confining pressure. Specimens were cut into standard cylinders with length of 50mm and diameter of 25mm, and its two ends were polished to keep them smooth and parallel with each other but vertical to the axis. Samples are mainly from Barnett, Haynesville, and Eagle ford Shale from North America and lower Silurian Longmaxi black shale from South China. Coring depth is 2133.6-4164.0m. Coring angle differ from0°, 45°, 90° horizon and vertical direction. Confining pressures are 10-90MPa unequally. 1.2 Experimental results and discussion Experimental results suggest that rock strength and elastic parameters ranged largely with compressive strength 14.8391.3MPa, elastic modulus 4.9-70.4MPa, and Poisson’s ratio 0.19-0.95. Average compressive strength sorted from high to low is Barnett Shale, Longmaxi Shale, Haynesville Shale, Eagle Ford Shale; elastic modulus is Barnett Shale, Longmaxi Shale, Haynesville Shale, Eagle Ford Shale; Poisson’s ratio is Barnett Shale, Haynesville Shale, Eagle Ford Shale, and Longmaxi Shale. Longmaxi Shale is familiar with Haynesville Shale in some aspects like compressive strength, Young’s modulus and Poisson’s ratio, while its heterogeneity feature is higher than that of Haynesville Shale. Bedding plane has a significance effect on the mechanical properties of the shale, especially for Haynesville Shale, for which disk-shaped cores are very common and the average thickness is 2.3cm. By analyzing the experimental results we conclude that shale are usually very brittle and the mechanical properties are affected by many different factors (Hoshino, 1974; Handin, 1958; Li, 2012). Main conclusions can be summarized as follows: (1) Mechanical behaviors of shales are affected by shale type, coring depth, coring direction and laboratory loading conditions. There are large difference between mechanical parameters such as Young’s modulus, Poisson’s ratio and compressive strength for different kinds of shale under similar testing conditions (Fig.1).Samples from the same shale but different coring depth can be mechanical different largely (Fig.2).Samples from the same shale with similar coring depth but different drilling direction show out notable difference in mechanical properties (Fig.3). Samples with all same coring parameters but different confining pressure show out notable difference in mechanical properties, too.

Fig.1 Strength difference of diverse shales

Fig.2 Influence of depth on shale strength (Hanesville shale)

Fig.3 Influence of drilling direction on shale strength (Eagle Ford shale, 4164m)

Fig.4 Influence of confining pressure on shale strength (Haynesville shale)

(2) Higher confining pressure generally causes less micro-crack and results a higher ultimate compressive strength, and the failure mode is primarily single shear (Fig.5). Under lower confining pressure slip and slip-shear failure modes are dominate for the longitudinal expansion of cracks is repressed with increasing confining pressure. (3) Failure modes and mechanical properties tend to becoming similar under confining pressure over 50MPa. That’s because under higher confining pressure the discontinuities, such as joints, bedding plane, flaws and natural faults, tend to

IPTC 16580

3

close and performance more homogeneity than under free conditions, which reduces energy consumption in micro-crack and concentrates more energy on the major fracture expansion. (4) Anisotropic behavior of shale samples are mainly affected by mineral compositions, sedimentary orientation, compacting environment and geologic structural history. The difference between factors mentioned above is showing out by coring depth, coring direction and experimental loading conditions. These mechanical differences affect shale gas drilling and completion engineering by providing optimized drilling direction and building angle. These factors will affect the borehole integrity and optimized stimulation method. 1.3 Failure mode and brittleness characterization analysis There are mainly three failure modes for shale samples under confining pressures, which is single-shear failure (Fig.5B1,B2), double-shear failure (Fig.5E3) and splitting-shear failure (Fig.5H4,H5,H6,H7,E8,E9,H10,H11,H12). The failure mode of gas shale under confining pressure less than 30MPa is mainly splitting modes rather than double- or single-shear modes. Bedding plane and micro-crack is developed in shale and failure modes are affected by these natural weak planes, which will become new initiation point of shear-slip cracks under compression environment and finally cause the shale failed. Considering the number of cracks, double-shear and splitting-shear failure modes can produce more failure zone and show out more brittle features and in volume-strain aspects, the former shows much more significant dilatation than the latter. For the failure results, split failure generates more cracks and fails more complete than the other two modes. To express the brittle feature of the shale, rock samples should be more elastically strong and low surface energized. That’s because the in coordinate deformation inside the sample increases under external loading conditions and the micro-crack causes loading off and effects the coordinate propagation of multi-fractures and thus minimizes the brittle failure cracks. While in isotropy homogeneity shale samples, multi-cracks’ initiation and propagation probability are almost similar which finally caused a more completive brittle failure with more macro failure surfaces as indicators.

B1(40MPa)

B2(50MPa)

E3(40MPa)

H4(30MPa)

H5(20MPa)

H6(20MPa)

H7(10MPa)

E8(10MPa)

E9(10MPa)

H10(10MPa)

H11(0MPa)

H12(0MPa)

Fig.5 Failure mode sketches of shale specimens under different confining pressure, Note: Capital letters before the number represents the type of shale, for instance, H, E and L represent Barnett, Haynesville, Eagle Ford and Longmaxi Shale respectively; Figures inside brackets are confining pressure.

2 Brittleness evaluation of shale gas well Brittleness evaluation is a key mechanical problem which is now being studied by lots of oil companies and research institutions. Brittleness is commonly expressed by using Young’s modulus and Poisson’s ratio. The former is believed an indicator of failure capability and can reflect the capability of stably maintaining crack propagation. The latter is an indicator of stress changing ability after crack produced in shale. Rickman and Mullen recently suggested that shale samples with low Poisson’s ratio and high Young’s modulus can be hydraulic fractured more easily and tend to brittle failure majorly (Rickman, 2008). Fig.6 describes this concept using experimental data from Barnett Shale; the direction illustrated by the arrow represents brittleness increasing. Test results of Haynesville, Eagle Ford Shale in North America and Longmaxi Shale in South China were projected to the graph. Projected results show that brittleness of black shale from South China is moderate degree, fracture initiation capability is ordinary, and brittleness feature of outcrops is better than that of downhole cores. In this paper, we recommended a method using Young’s modulus, Poison’s ratio and mineral composition feature as parameters to calculate brittleness index. That’s because the relationship between brittleness performance and mineral type is obvious. Shale with high brittle mineral content has better fracture initiation capacity and is already being used for brittleness calculation separately for a long time, thus its importance could not be ignored (Jarvie, 2007; Wang, 2010; Li, 2012). By using static Young’s modulus, static Poison’s ratio and percentage of brittle mineral as basal parameters to calculate the comprehensive brittleness index. Data processing procedures for a whole well can be summarized as follows:

4

IPTC 16580

(1) Calculating dynamic Young’s modulus and dynamic Poison’s ratio using conventional logging data, and getting static Young’s modulus and static Poison’s ratio (Rickman, 2008). (2) Analyzing brittle mineral content from Elemental Capture Spectroscopy (ECS) or radioactivity spectrum data. In some cases, these results can be calibrated by laboratory tests results again. (3) Static Young’s modulus, static Poison’s ratio and mass fraction of brittle mineral are normalized separately according to regional feature. (4) Calculating brittle profile of the whole wellbore, thereby predicting the brittleness of the target layer within a certain depth. This approach considers the effect of mineral composition on the mechanical behavior and can be used to distinguish layers with similar elastic parameters.

Fig.6 Relationship between brittleness, Young’s modulus and Poison’s ratio

A field case from south China was analyzed and used to illustrate the process for brittleness index calculation and target layer selection. From interpretation results showed in Fig.7 we can see that the brittleness feature of Longmaxi black shale from1480m to 1540m varies largely, which may have great impact on the effect of fracture initiation and stimulating results. According to previous experience (Li, 2012a; Li, 2012b), high brittleness tends to form fracture network and result in high production .The effect of brittleness on fracturing results is confirmed in practice by fracture tracing and micro-seismic monitoring. Hydraulic fractures in brittle intervals initiate and propagate efficiently (Li, 2012a; Li, 2012e),and will have more branch off and bigger steering angle when encounters natural fracture, so it has better stimulation results than plastic intervals(Li, 2012a; Li, 2012c).

Fig.7 Brittleness analysis results of Longmaxi Shale from well W

IPTC 16580

5

According to the brittle profile, the target layer from well W is divided into seven intervals as A, B, C, D, E, F, G, whose comprehensive brittleness index are all above 50. After further analysis these intervals can be divided into three categories: (1) Interval A, B: thin and medium high brittleness index without gas show during drilling. Although has good brittleness feature, this type of reservoir has low potential of gas generation and production, the hydraulic fracture can penetrate the reservoir easily and enter adjacent formations for the less thickness, so it need to be chosen carefully before treatment construction because its economic value needs to be further assessed. (2) Interval C, D, E: high organic content and brittleness index with good gas show. Interval E is significantly better than C and D for its greater thickness (about 13m). Noting that the distance between interval E and F,G is less than 2m and hydraulic fracture can penetrate into interval F and G easily so additional precautions needs to be taken for fracture propagation controlling. (3) Interval F, G: medium thickness and high brittleness index without gas show: Since there wasn’t any gas show during drilling, better geochemical evaluations are needed for target layer determining, although there is a higher brittleness index. The fracturing test of this well is divided into three steps, perforating intervals concentrated upon9.0m. Mini-fracturing and main fracturing was used, in which injected fracturing fluid volume is 175.0m3 and 1978.0m3 respectively and 2153.0m3 totally. The weight of proppant is 163.4t (102.1m3).Specific fracturing parameters are shown in Table 1. Table 1 Hydraulic fracture date set from well W Target Layer Longmaxi

Perforation/m 1511.0-1514.0 1520.0-1523.0 1532.0-1535.0

Thickness/m 3.0 3.0 3.0

Liquid Volume/ m 2153.0

3

Sand Volume/ m 102.1

3

Pump Pressure/MPa

Pump Rate/L/min

36.1-40.0

10.1-10.2

The production is roughly the same with prediction. Total production is about 2000-2500m3 per day and is not stable. Downhole pressure monitoring shows the pressure is abnormal at the depth of 1523.0m (bottom of interval E in Fig.7). The hydraulic fractures seemed propagated much better than other intervals, because there was a higher proppant and fluid injection volume. Such results verified the effectiveness of comprehensive brittleness evaluation method in characterizing reservoir mechanical properties and treatment interval selection. 3 Potential critical fractures and their affection on reservoir stimulation Potential critical fractures are natural fracture that can be reactivated when the pressure changes, which plays an important role in shale gas production. The analysis results of critical potential fractures in Well W1 and W2 are shown in Fig.8. These two wells are in the same field, whose pay layers are both lower Silurian Longmaxi shale. The target formations are similar in burial depth, total content of organic carbon, clay and brittle mineral content, effective porosity and total natural gas content. Table 2 provides the detailed data of these two neighbor wells. Table 2 Contrast of shale gas reservoir in W1 and W2 Well No. W1 W2

Burial Depth /m 2310.0-2338m 2490-2531.3m

OC Content /% 0.13-8.37 0.06-7.01

Clay Content /% 2.66-46.35 4.10-51.20

Brittle Mineral Content /% 24.29-47.12 26.70-43.61

Effective Porosity /% 0.57-10.26 0.67-8.53

Gas Content 3 /m /t 0.53-8.72 0.55-6.13

These two exploitation wells are all drilled vertically and conducted completion after penetrating into limestone beneath the shale. Perforation and fractured works are all designed and conducted by the same services company. What differential is that W1 located far away from an anticline and W2 drilled much closer to the flank, which showed out several miniature faults near it? They were logged when got to the interval formations. Logging data includes gamma spectrum, gamma ray, neutron density, electric resistivity, Sonic Scanner, FMI (Formation Micro Scanner Image), ECS (Elemental Capture Spectroscopy). Analysis results show that the maximum horizontal stress orientated NW-SE in W1 and roughly E-W in W2, which are shown by black arrows in Fig.8. By analysis the fracture characterization and in-situ stress, we can see that there are more critical potential fractures in W2 than W1. Although W1 has more horizontal bedding fracture, they are not easily reactive. Stress distribution conditions in these two wells are calculated and analyzed by using a three-dimension Mohr stress circle. We can find from Fig.9 that there are totally 65 natural fractures in W2 and 48 in W1. Fractures in W2 can be reactivated more easily than in W1 and most natural fractures in W2 are high permeable while in W1 mainly parallel to the deposition direction and filled by quartz and calcite. Critical potential natural fractures (red dots in Fig.6) in W2 have a high linear density and are mainly critical shear-sliding fractures, while the closed and hard to reactivate natural fractures (black dots in Fig.6) are mainly in the bedding plane direction. There are more than 72.31% natural fractures are critical potential ones in W2 and 10.41% in W1. Fig.10 illustrates the filtration potential fractures which are usually semi-closed or totally opened. These fractures can be easily reactivated to open again by fluid filtration. There are totally 7 filtration potential fractures in W1 and 19 in W2, which accounting for 14.58% and 29.23% respectively to the total fractures.

6

IPTC 16580

Fig.8 Interpretation result of fracture and stress orientation of part intervals in well W1 (Left) and well W2 (Right)

Fig.9 Triaxial stress state of natural fracture around wellbore in well W1 (Left) and well W2 (Right)

Fig.10 Result of filtration activated of natural fracture around wellbore in well W1 (Left) and well W2 (Right)

Fig.11 Potential natural fracture reactivated sketch in stress field and hydraulic fracturing conditions

IPTC 16580

7

These two wells are all completed and produced no longer than two years, so the productivity caused by critical potential fractures is compared by analyzing the initial production of these two wells. There are all gas shows in W1 and W2 during wellbore drilling. A large number of gas bubbles appeared in the process of tripping out. Gas desorption properties measured by quick field tests equipment are similar in these two wells and the gas content measured are listed in Table 1. These two wells are all tested successfully after hydraulic fracturing treatment. During production period, W2 has an initial production of 2300m3 per day with a φ9mm choke and estimated gas production of 30600 m3, while W1 doesn’t have industrial production after successfully tests. What we can see from the tests and analysis above is that besides productivity of the reserve the geological structure, especially which related to engineering operations, has a significance effect on gas production in the shale gas reservoir. W2 locates in a place near the flank of a anticline and has more mini-faults and natural fractures than W1. These natural fractures are mostly critical potential fractures which can easily be reactivated by hydraulic fracturing and then improved gas permeability largely and caused a higher production. W1 is far away from the folded structure and natural fractures are mainly horizontal direction, which are difficult to be reactivated again during stimulation. Communication of natural fractures and hydraulic fractures is not so effective than W2 and the ultra-production is eventually lower than the latter. 4 Fracturing design based on mechanical test Analysis results of shale brittleness characteristic could be used as an index of fracturing design and fluid selection ruler. High brittle shale can easily be stimulated out complex fractures and have higher potential for fracture initiation and propagation in the reservoir. The strain energy released rate promoting fracture propagation can usually several times higher than the surface energy needed for a single fracture initiated and in this situation the formation tends to generated fracture branching and reorientation fractures, which are important factor for improving hydraulic fracturing results. Proppant plays a role of bridging and supporting in the high brittleness shale and providing efficient channel for the seepage of shale gas. On the contrary, double wing fracture is the main type of artificial fracture in high plasticity shale, stimulated volume distributes around the main fracture and requires more proppant for keeping fracture open to avoiding settlement. Following adjustment should be taken when stimulated interval changes from ductile formation to brittle intervals. (1) Fracturing fluid type can be changed from cross-linked, foamed, linear gum, compound to slick-water for the purpose of improving fracture generating ability and the efficiency of natural fracture reactivity. At the meantime, pump rate and fracturing pressure should also be averaged up to provide enough energy for network fracture initiation and propagation. (2) Total fracturing fluid volume should also be increased to guarantee fracture initiating and enhance fracture network propagating. (3) To avoid sand bridging, proppant with smaller size and higher strength should be selected gradually. Proppant size can be calculated based on minimum fracture width and closure pressure of the formation to ensure all fractures can be filled and provide enough permeability. (4) Proppant selection is mainly depended on strength. High compressive strength proppant is usually used in regions where burial depth and in-situ stress are high. Another aspect for selecting proppant is the cost and we need to balance the cost and stimulation needs. At the meantime, long period flow conductivity and proppant embedding capacity should be tested in the laboratory scale first before field site using.

Fig.12 Proppant selection based on the fracture closure pressure

A shale gas well from South China is considered as an example to illustrate hydraulic fracturing parameters selection process considering the mechanical properties. Mineral composition and mechanical property parameters profile are first got by well logging interpretation and the results are shown in Fig.13. Closure pressures and ideal width of hydraulic fractures are calculated too, and the results can be seen from the third and fourth columns in Fig.13. The brittleness index of the shale formation is calculated based on mineral composition and mechanical properties and the results are shown in the fifth column. According to the calculation results in Fig.13, we take intervals A to G as potential stimulation targets. These intervals have relative higher brittleness which could be easily hydraulic fractured into fracture networks. Further analysis and selection procedures are shown as follows:

8

IPTC 16580

(1) Brittleness indexes of interval A, C, E and G are above 50 and the fracture width are relatively high, while interval B, D, F have much higher closure pressures which means a harder treatment in stimulation. So interval A, C, E has potential fracturing option on second thoughts, while intervals B, D, F are reserved for other consideration. (2) Since interval A, C, E, and G has higher potential for fracture network generation, we here select slick-water system as fracturing fluid. Slick-water as fracturing fluids can promote much farther than others which is useful for maintain complex fracture open after treatments. (3) Closure pressures in these four intervals are no larger than 45MPa, so quartz sands can be selected as fracturing proppant. (4) Calculated fracture width is between 2.1 and 3.0mm, so the selected proppant size is 30/50meshes. Based the analysis and calculated results mentioned above, fracturing designs for each interval of these seven ones are listed specifically in Table 4. What should be noted is that the TOC content, maturity of organic matters, porosity, permeability and gas content results should also be considered comprehensively when making final choice for hydraulic fracturing.

Fig.13 Hydraulic fracture parameters profile based on mechanical analysis results

Table 3 Selecting results for main hydraulic fracture parameters Interval

Brittle Index

Thickness

Closure Pressures

/m

/MPa

Pay Layer or Inter Layer

Fracture Width /mm

A

49

2.7

40.4

Reservoir

2.2

B

47

1.8

42.2

Interlayer

0 2.1

C

51

6.1

40.4

Reservoir

D

43

1.8

42.2

Interlayer

0

E

64

17.3

42.3

Reservoir

2.7

F

52

2.0

43.2

Interlayer

0.02

G

71

4.2

42.5

Reservoir

3.0

Fracturing Fluid and Proppant Selection Results Fracturing Proppant Proppant Whether to Fluid System Size Type Stimulating SlickQuartz 30/50 Yes Water Sand -No SlickQuartz 30/50 Yes Water Sand No SlickQuartz 30/50 Yes Water Sand No SlickQuartz 30/50 Yes Water Sand

5 Conclusions The paper has presented a series of laboratory scale testing results of gas shale for hydraulic fracturing designing purpose. The experimental results can be useful data for understanding and applying in gas shale hydraulic fracturing treatment behavior. Shales are commonly anisotropy heterogeneity with generally high Young’s modulus and low Poisson’s ratio. Shale types, coring depth, coring direction and loading conditions are all factors affected rock mechanical properties of gas shale. It is suggested that for understanding specific mechanical properties, laboratory experimental tests should be conducted in environmental conditions simulating in-situ background. A significant brittle behavior was observed during systemically study. Failure modes of standard shale specimens are mainly splitting shear under uniaxial loading, while shear or double

IPTC 16580

9

shear failure under high confining pressure. Overall stress-strain curve shows that shales fail quickly after peak yielding and with very little plastic strain. A new method combines elastic parameters and mineral composition to quantify the brittleness was derived, and the evaluation of brittleness for single well was achieved through the calculation of logging data. The results of production show that this method can reflect the brittleness failure and capability of fracturing initiation well. It has great value of reference in the mechanical properties evaluation and hydraulic fracturing interval selection. Mechanical properties of shale formation have significant impacts on shale gas exploration. Designing and conducting hydraulic fracturing should be done base on systematically tests and analysis. Understanding of the failure modes and mechanical properties can be basic work in horizontally drilling and hydraulically fracturing, and these researches should be given sufficient attention in the future. Systematically analysis of FMI logging data from two shale gas wells illustrated a method in identifying critical potential fractures using a three-dimensional Mohr circle. Production analysis found that two types of natural fracture in critical stress can improve the permeability significantly. In selecting fracturing intervals procedures, intervals with developed critical potential natural fracture can be considered as target layer perforating and stimulating. Finally, using mechanical property testing results and considering their influences on hydraulic fractures a systematical treatment parameters selection approach was proposed. This method suggested rock mechanical properties study conducted first before field construction and combining geochemistry study and well logging simultaneously to design efficient hydraulic fracturing treatments. Acknowledgement The results of this paper are parts of the State Key Program of National Natural Science of China (Grant No.51234006). The authors wish to thank China University of Petroleum in Beijing and the University of Texas at Ausin for granting permission to publish the results of this study. References Dunn, D. E., L. J. LaFountain and R. E. Jackson, 1973, Porosity dependence and mechanisms of brittle fracture in sandstones: J. Geophysical Research, v. 78, no. 14, p. 2403-17. Hoshino, K., 1974, Effect of porosity on the strength of the clastic sedimentary rocks: in Advances in Rock Mechanics, Proceeding of the 3rd ISRM Congress, Denver, USA, p. 511-16. Handin, J., R. V. Hager, Jr., 1958, Experimental deformation of sedimentary rocks under confining pressure: tests at high temperature: AAPG Bulletin, v. 42, no. 12, p. 2892-2934. Jarvie, D., Hill, R. J., Ruble, T. E., and Pollastro, R. M., 2007, Unconventional shale-gas systems: the Mississippian Barnett Shale of north-central Texas as one model for thermogenic shale-gas assessment: AAPG Bulletin, v. 91, no. 4, p. 475–499. Li, Q., Mian, C., Yan, J., Fred, P. W., Bing, H., and Baowei, Z., 2012a, Indoor evaluation method for shale brittleness and improvement: Chinese Journal of Rock Mechanics and Engineering, v. 31, no. 8, p 1680-1685. Li, Q., Mian, C., Yan, J., Bing, H., and Jiazhen, Z., 2012b, Rock mechanical properties and brittleness evaluation of shale gas reservoir: Petroleum Drilling Techniques, v. 40, no. 4, p 17-22. Li, Q., Mian, C., Yan, J., and Fred, P. W., 2012c, Experimental research on failure modes and mechanical behaviors of gasbearing shale: Chinese Journal of Rock Mechanics and Engineering, v. 31, no. 263, p 3763-3771. Li, Q., Mian, C., Fred, P. W., Yan, J., and Zhimeng, L., 2012d, Influences of engineering factors on shale gas productivity: A case study from the Haynesville shale gas reservoir in North America: Journal of Natural Gas Industry, v. 32, no. 4, p 5459. Li, Q., Mian, C., Yan, J., Fred, W., and Mingyuan, W., 2012e, A calculation tool for hydraulic fracture network-propagation in layered shale formation: International Review on Computers and Software, v.7, no. 5, p 1776-1783. Li, Q., Mian, C., Yan, J., Mingyuan, W., and Hailong, J., 2012d, Application of new fracturing technologies in shale gas development: Journal of Special Oil & Gas Reservoirs, v. 19, no. 6, p 1-7. Li, X., C. Zou, Z. Qui, J. Li, G. Chen, D. Dong, L. Wang, S. Wang, Z. Lu, S. Wang and K. Cheng, 2011, Upper Ordovician/Lower Silurian shale gas reservoirs in Southwestern Sichuan Basin, China: American Association of Petroleum Geologists, Annual Convention, Abstracts Volume, vol. 20, p. 111-112. Luffel, D. L, and K. Girdry, 1992, New core analysis methods for measuring reservoir rock properties of Devonian Shale: J. of Pet. Tech. (November), p. 1184-1190. Rickman, R., Mike, M., Erik P., Bill, G., and Donald K., 2008, A practical use of shale petrophysical for stimulation design optimization: All shale plays are not clones of the Barnett shale: SPE paper no. 115258 presented at the 2008 annual technical conference and exhibition, Denver, Colorado, Sep. 21-24, 7 p. Soeder, D. J., 1988, Porosity and permeability of Eastern Devonian Gas Shale, in Society of Petroleum Engineers, SPE Formation Evaluation, March, p. 116-124. Wang, F. P., and R. M. Reed, 2009, Pore networks and fluid flow in gas shales: SPE paper no. 124253-PP presented at the 2009 annual technical conference and exhibition, New Orleans, LA, Oct. 4-6, 7 p. Wang, F. P., and U. Hammes, 2010, Key petrophysical factors affecting fluid flow and production in the geopressured Haynesville Shale: World Oil, Vol. 231, No. 6, 4 p.