Characteristics of Fracture Propagation Induced by Supercritical CO<sub>2</sub> in Inter-Salt-Shale Reservoir

Zhang, Yixiang; He, Jianming; Li, Fengxia; Fan, Xin; Li, Xiao

doi:https://doi.org/10.1155/2019/7132843

Geofluids

On this page

Abstract Introduction Results and Discussion Conclusions Data Availability Conflicts of Interest Acknowledgments References Copyright Related Articles

Research Article | Open Access

Volume 2019 | Article ID 7132843 | https://doi.org/10.1155/2019/7132843

Characteristics of Fracture Propagation Induced by Supercritical CO₂ in Inter-Salt-Shale Reservoir

Yixiang Zhang,^1,2,3Jianming He ,^1,2,3Fengxia Li,⁴Xin Fan,⁴and Xiao Li^1,2,3

Academic Editor: Julie K. Pearce

Received11 Sept 2018

Revised03 Jan 2019

Accepted10 Feb 2019

Published10 Apr 2019

Abstract

Hydraulic fracturing using freshwater is difficult for the commercial exploitation of a shale oil reservoir in Jianghan Basin with a continental saline lake basin sedimentary background. Supercritical CO₂ (SC-CO₂) is a promising fracturing fluid under consideration for the reservoir stimulation, especially in the case of the presence of water sensible salt layers. In this study, SC-CO₂ fracturing experiments on the inter-salt-shale and salt specimens, which were obtained from the drilling well, were carried out in the laboratory. The characteristics of fracture propagation, including morphology and width variation, were analyzed based on the observations of a stereoscopic microscope, X-ray micro-CT scanner, and 3D scanner. The existing weak planes in the shale can really impact the fracture propagation in SC-CO₂ fracturing. Deflection, branching, and approaching can occur during the process of fracture propagation. The average width value of the reactivated natural fracture is bigger than that of a newly created fracture. In addition, the fracturing results indicate the greater breakdown pressure of rock salt if compared with the inter-salt-shale. The induced fractures in the salt specimen are compact and smaller in average width than those in the shale specimen. The higher breakdown pressure and relatively smaller fracture width of rock salt are real challenges for the fracturing of an inter-salt-shale oil reservoir.

1. Introduction

Shale oil has recently been found in the Jianghan Basin, revealing a new field for exploring and developing shale oil [1–3]. Qianjiang Sag is located in the middle of the Jianghan Basin, which is the main salt-forming center, rich in the hydrocarbon source [4]. The oil is developed in the inter-salt-shale reservoir of the southern sag resulting from a unique continental saline lake basin sedimentary background. The inter-salt-shale reservoir, ranging from dozens of centimeters to a few meters in thickness, is distributed in the continental basin [5–8]. Hydraulic fracturing using freshwater as the fracturing fluid is difficult for the commercial exploitation of a shale oil reservoir due to the presence of mixed salt.

Supercritical CO₂ has great potential as a fracturing fluid to stimulate this special reservoir, which can cause wettability alteration on the low permeability reservoir and lower the surface tension and viscosity of the oil to improve the displacement [9–12]. The studies by Zhang et al. [13] indicate that SC-CO₂ is considered as an ideal nonaqueous fracturing fluid which can produce a denser crack network. In addition, SC-CO₂ has the traits of low viscosity like a gas and high density like a liquid [14, 15]. There are multiple aspects of research results on SC-CO₂ fracturing fluid in its viscous properties and superiorities. Fracturing experiments on shale and granite, using fluids of different viscosities, show that SC-CO₂ can induce a shear-dominant fracture with more branches and generate cracks extending more three dimensionally with a greater fractal dimension [16–19]. Li et al. [20] explored the usage of H₂O, CO₂, and N₂ as stimulants on Green River shale. The results show that CO₂ fracturing can create nominally the most complex fracturing patterns as well as the roughest fracture surface and the greatest apparent local damage. Zhou and Burbey [21] used numerical simulations to indicate that thin fluids, such as CO₂, would induce thinner and shorter fractures than fluids exhibiting properties similar to water. The results of an experiment and numerical simulation by Wang et al. [22] indicate that SC-CO₂-based fracturing has a lower breakdown pressure and may develop more complex fractures than those developed with water-based and oil-based fracturing. Supercritical CO₂ fracturing experiments on marine shale indicated that anisotropy has a significant impact on the mechanical behavior and fracture propagation of shale [23].

Comparing with shale in North America and other parts of China, the shale in Jianghan Basin has a mineral composition of low clay and quartz content and high carbonate content (Figure 1) [6]. The characteristics of transverse isotropy and salt layer mixture in the reservoir can lead to complex fracture propagation in fracturing stimulation. We performed SC-CO₂ fracturing experiments on the inter-salt-shale and salt specimens for study on the propagation of hydraulic fractures and fracture width variation. The rock specimens were obtained from the reservoir in Jianghan Basin. The fracture morphology and width were analyzed and failure patterns were summarized.

2. Experiment Method

The rock cores of inter-salt-shale were obtained from the drilling well in Qianjiang Formation in Qianjiang Sag of Jianghan Basin. The shale specimen composition with developed bedding structures is complex, and cracks can be observed with the naked eyes on the lateral surface (Figure 2). In this study, two kinds of specimens were prepared for the experiment. One is the cylindrical shale and salt specimens with dimensions of 100 mm in diameter and 120 mm in height, and the other is the cylindrical shale specimens with diameter of 25 mm and height of 50 mm. Among these specimens, the smaller specimens were obtained by drilling along (near-vertical) and perpendicular (near-horizontal) to the bedding orientation to consider the anisotropy. The blind boreholes were drilled with 10 mm and 4 mm in diameter along the central axis of the specimens for the injection of the fracturing fluid (Figure 2). Depth of the blind borehole was kept as half of the specimen height.

(a)

(b)

The SC-CO₂ fracturing experiments of the specimens were conducted on the TAW-2000 servo-controlled triaxial compression machine accompanied with a SC-CO₂ generator in the laboratory. Figure 3 shows the schematic diagram of the whole fracturing system. The triaxial compression machine can provide confining pressure and axial loading on the specimen with a precision of 0.05 MPa and 20 kN, respectively. The hydraulic pump and temperature controller can hold the CO₂ above its critical pressure and temperature to achieve the supercritical state for hydraulic fracturing. The hydraulic pump is an ISCO syringe pump with a maximum pressure of 60 MPa and the accuracy value is 0.3 MPa. A temperature transducer with a range of 0-100°C and an accuracy of 0.1°C was used to detect the inlet temperature of CO₂ injected into the specimens.

SC-CO₂ fracturing experiments were performed on the specimens under the axial stress of 25 MPa and confining pressure of 20 MPa, in which the injection rate of the fracturing fluid was set as 40 ml/min. After the installation of the rock specimens in the loading chamber, the values of the confining pressure and axial stress were increased to the target level. This was followed by injecting the fracturing fluid at a constant rate into the specimen through the borehole. The fractured specimen was retrieved after the experiment for the subsequent test. A high-precision X-ray micro-CT scanning device was employed to obtain the internal fracture distribution of the fractured specimen. CT technology allows us to virtually cut through the sample at any given location. In this study, X-ray CT scanning tests were performed at 140 keV and 72 mA for optimum image quality and contrast. The specimens were rotated continuously during the scanning with momentary stops to collect each projection and minimize the artifacts with an image voxel size of 52 μm. Every sample was scanned to obtain 1000 CT slices along the vertical direction from top to bottom. The Leica M205A microscope was employed to measure the hydraulic fractures on the lateral surface. It enables us to obtain micrographs with high definition and measure the fracture width on the surface of the specimen. In addition, a 3D scanner was used to measure the fracture surfaces of the fractured specimens to evaluate the fracturing results.

3. Results and Discussion

The mechanical parameters of the elastic modulus and Poisson’s ratio were obtained to calibrate the characteristics of the specimens [24]. The results from all the rock property tests and the mechanical strength parameters are given in Table 1.

The SC-CO₂ fracturing experiments on inter-salt-shale specimens with two different bedding orientations and a salt specimen were performed in this study. Table 2 summarizes the breakdown pressure and corresponding experimental conditions of these experiments. In this section, the failure patterns, fracture width, and surface roughness of the hydraulic fractures are characterized quantitatively.

3.1. Development of Pump Pressure Curves for SC-CO₂ Fracturing

The curves of the pump pressure versus time for SC-CO₂ fracturing of the inter-shale and salt specimens are shown in Figure 4. The specimens of Nos.1, 2, and 3 have the near-horizontal bedding orientations, while the specimens of Nos.4, 5, and 6 have the near-vertical bedding orientations. The specimen of No.7 is the relatively homogeneous rock salt. The stress conditions with axial stress MPa and confining pressure MPa were applied on the specimens, and the injection rate of the fracturing fluid was set as 40 ml/min. The compressibility of CO₂ decreases with the increase of pressure under the constant temperature condition. Obviously, the greater volume compressibility under low pressures illustrates the lower loading rate of the pump pressure at the initial stage of the fracturing experiment. The loading rate of the pump pressure increased gradually with the constant injection rate of SC-CO₂ before the occurrence of fracturing, which is closely related to the compressibility of CO₂ under different pressures. Comparing the breakdown pressures of the salt specimen (No.7) and the rest of the specimens (No.1-No.6), the salt specimen shows a slightly bigger value if compared with the intershale specimen. Therefore, shale is more likely to fracture if compared with the relatively homogeneous salt due to the presence of obvious weak planes.

3.2. Fracture of the Shale Specimens

Figure 5 shows the failure patterns of the inter-salt-shale specimens, in which the fractures of specimens were closely related to the sedimentary structures. The specimens of No.1, No.2, and No.3 were fractured into two halves along the near-horizontal bedding planes obviously. These shale specimens were easier to break along the preexisting weak planes under the hydraulic pressure. For the specimens of No.4 and No.5 with near-vertical bedding planes, approximately symmetrical fractures can be seen at the two ends of the specimens. However, there were only a few cracks visible to the naked eyes on the surface. These cracks developed along the sedimentary plane only near the hole. As for the fracture morphology in the internal structures, a high-precision X-ray micro-CT scanning device was employed to conduct the X-ray microtomography tests to obtain it in a nondestructive way. Figure 6 shows five typical CT images (slices 200, 400, 500, 600, and 800) of specimen No.5 with the internal morphology of fractures induced by SC-CO₂. The hydraulic fractures mainly propagate along the bedding orientation. More micro cracks were developed on the half section with the borehole of the specimen besides the two symmetrical main fractures. However, a few micro cracks were available on the other half of the specimen on the CT images.

$(a) The fractured specimens of Nos.1, 2, and 3 with near-horizontal bedding orientation$
(a) The fractured specimens of Nos.1, 2, and 3 with near-horizontal bedding orientation

$(b) The fractured specimens of Nos.4 and 5 with near-vertical bedding orientation$
(b) The fractured specimens of Nos.4 and 5 with near-vertical bedding orientation

Morphology of the fracture surfaces was obtained through the 3D scanner, and Figure 7 shows the reconstructed fracture surfaces based on the scanning data. The parameter of the Area Ratio (AR) was calculated to describe the roughness of fracture surfaces quantitatively, which is defined as the ratio of actual contact area to projected area of the main fracture surface [25]. The actual area of the fracture can be calculated by the reconstructed fracture surface, and the projected area is the cross-sectional area of each specimen core minus the borehole area. The AR values of Nos.1, 2, and 3 are 1.42, 1.38, and 1.41, respectively. Zhao et al. [26] had measured the AR values of continental shale from the Ordos Basin in China, which are much smaller than the AR values in this study. Shale of the Qianjiang Formation from Jianghan Basin shows the characteristics of low clay and high carbonate content [6], so the high content of brittle minerals can result in rougher fracture surfaces comparing with the continental shale.

3.3. Comparison of Fractures in Shale and Salt

The morphology of the fracture has significant effects on the fluid flow in the fractures, and it can change the hydraulic conductivity [27–29]. In this study, fracture propagation on the lateral surfaces of the shale and salt specimens was observed for illustration of the fracture morphology.

Figure 8 shows the photographs and schematic diagrams of fractured shale and salt specimens. In general, the fracture propagation induced by SC-CO₂ in the salt specimen mainly develops along the boundaries of crystal grains around the borehole, which can break the specimen in half. However, the natural fractures (e.g., bedding planes) really could influence the hydraulic fractures for the shale specimen, in which the hydraulic fractures mainly develop along the natural fractures. In addition, hydraulic fractures near the borehole were more developed in both the shale and salt specimens. Figure 9 shows several typical local fracture morphology patterns of shale with an amplification factor of 10. Deflection, branching, and approaching can occur during the fracture propagation, which are the foundation of a complex fracture network to enhance the reservoir permeability significantly.

(a)

(b)

The fractures induced by SC-CO₂ fracturing constitute the pathways of the exploitation of oil and gas resources. The width of hydraulic fractures plays an important role in the transportation efficiency of oil and gas resources. Figure 10 shows the morphology of the hydraulic fractures on the lateral surface of the shale specimen, including the newly created fractures and reactivated natural fractures. To illustrate the characteristics of the width variation along the fracture pathway, the fracture width on the lateral side of the specimen was measured through a stereomicroscope. In consideration of fracture tortuosity, the width values were measured using the segmental observation method, which was carried out once in a 1 mm long straight-line segment along the axial direction.

$(a) Main fracture induced by SC-CO2$
(a) Main fracture induced by SC-CO2

$(b) Opening natural fracture$
(b) Opening natural fracture

Figure 11 shows the variation of the fracture width along the axis direction of the specimen. It is indicated that the width values of the fracture in the half of the specimen with the borehole are generally greater if compared with the other half. In addition, the average width value of the natural fracture (Figure 11) is greater than that of the newly created fracture. This implies that the reactivated natural fracture is more conducive for oil transportation in shale if compared with the newly created fracture.

Figure 12 shows the hydraulic fractures on the lateral surface of the salt specimen and the local enlarged images. The fractures mainly propagate along the common boundaries of the crystal grains. If there is a small amount of mud in the salt, the weak intergranular cohesion is conducive to fracture expansion. Figure 13 shows the variation of the fracture width along the axis direction of the specimen. In general, the average fracture width values on the side surface are smaller than those of the shale specimen. For the salt with large crystal grains, the fracture surface with large roughness hinders its shear sliding. However, under the action of deviatoric stress, the shear slip occurs comparatively among the fractures of shale due to fine clay grains with small friction. Therefore, the hydraulic fractures of the salt specimen are compact and relatively narrow in width. In addition, when the hydraulic fracture extends encountering the mud mixture, it is easy to propagate along the boundary and tends to be of greater width.

4. Conclusions

The inter-salt-shale oil reservoir in Jianghan Basin has a continental saline lake basin sedimentary background. The shale matrix with the salt layer mixture is particularly sensitive to the presence of water and can lead to complex fracturing characteristics. In this study, a series of fracturing experiments were performed on the intersalt specimens to study the breakdown pressure and fracture propagation using SC-CO₂. The following conclusions can be obtained: (1)The inter-salt-shale with developed bedding structures is easier to be fractured along the bedding planes under SC-CO₂ fracturing. Such reactivated natural fractures are critical for the oil production, and they generally show greater width values than newly created fractures. For the salt layer mixture in shale, the weak intergranular cohesion is conducive to the fracture propagation(2)Affected by natural bedding planes, three patterns of fracture propagation occur in shale fracturing, including deflection, branching, and approaching. However, the hydraulic fractures mainly propagate along the common boundaries of crystal grains for the salt layer(3)The fracturing fluid of SC-CO₂ can avoid the damage of salt to protect the reservoir in the inter-salt-shale reservoir. The hydraulic fracture propagation of the inter-salt-shale reservoir is hindered by salt layers due to its higher breakdown pressure and relatively narrow fracture width during the SC-CO₂ fracturing

Data Availability

The data used to support the findings of this study are plotted within the article, and the raw data files are available by contacting the corresponding author at http://[email protected].

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work is supported by the National Natural Science Foundation of China (Grant Nos. 41572310 and 41877270) and the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant Nos. XDB10030301 and XDB10030304).

References

S. J. Liang, Z. L. Huang, B. Liu, L. C. Yan, H. M. Li, and J. Ma, “Formation mechanism and enrichment conditions of Lucaogou Formation shale oil from Malang sag, Santanghu Basin,” Acta Petrolei Sinica, vol. 33, no. 4, pp. 588–594, 2012.
View at: Google Scholar
K. P. Luo, Q. Qiu, and J. Z. Ye, “Petroleum accumulation regulations and exploration targets in Jiangling Sag, Jianghan Basin,” Petroleum Geology & Experiment, vol. 35, no. 2, pp. 127–132, 2013.
View at: Google Scholar
C. Zou, Z. Yang, J. Cui et al., “Formation mechanism, geological characteristics and development strategy of nonmarine shale oil in China,” Petroleum Exploration and Development, vol. 40, no. 1, pp. 15–27, 2013.
View at: Publisher Site | Google Scholar
Z.-X. Fang, “Hydrocarbon exploration signification of intersalt sediments in Qianjiang saline lake basin,” Acta Sedimentologica Sinica, vol. 20, no. 4, pp. 608–613, 2002.
View at: Google Scholar
X. G. Pu, Z. X.. Qi, X. L. Zheng, D. X. Han et al., “Basic petroleum geological characteristics and resource potential of inter-salt non-sandstone reservoir,” Petroleum Exploration & Development, vol. 29, no. 5, pp. 28–30, 2002.
View at: Google Scholar
W. Furong, H. Sheng, Z. Youheng, H. Yuguang, and W. Shiqiang, “Mineral composition and brittleness characteristics of the inter-salt shale oil reservoirs in the Qianjiang Formation, Qianjiang Sag,” Petroleum Geology & Experiment, vol. 38, no. 2, pp. 211–218, 2016.
View at: Google Scholar
Y. Zheng, “Exploration direction and strategy for lithologic accumulations in Qianjiang Formation of the Qianjiang Sag, Jianghan Basin,” Petroleum Geology & Experiment, vol. 32, no. 4, p. 330, 2010.
View at: Google Scholar
Y. S. Zhang, Y. Q. Yang, Z. X. Qi, Y. D. Qiao, H. R. Yuan et al., “Sedimentary characteristics and environments of the salt-bearing series of Qianjiang Formation of the paleogene in Qianjiang Sag of Jianghan Basin,” Journal of Palaeogeography, vol. 5, no. 1, pp. 29–35, 2013.
View at: Google Scholar
X. Sun, C. Dai, Y. Sun et al., “Wettability alteration study of supercritical CO₂ fracturing fluid on low permeability oil reservoir,” Energy & Fuels, vol. 31, no. 12, pp. 13364–13373, 2017.
View at: Publisher Site | Google Scholar
Z. Li, T. Lu, L. Tao, B. Li, J. Zhang, and J. Li, “CO₂ and viscosity breaker assisted steam huff and puff technology for horizontal wells in a super-heavy oil reservoir,” Petroleum Exploration and Development, vol. 38, no. 5, pp. 600–605, 2011.
View at: Publisher Site | Google Scholar
A. Shah, R. Fishwick, J. Wood, G. Leeke, S. Rigby, and M. Greaves, “A review of novel techniques for heavy oil and bitumen extraction and upgrading,” Energy & Environmental Science, vol. 3, no. 6, pp. 700–714, 2010.
View at: Publisher Site | Google Scholar
X. Li, D. T. Yang, and Z. Fan, “Phase behaviour and viscosity reduction of CO₂-heavy oil systems at high pressures and elevated temperatures,” in SPE Heavy Oil Conference-Canada, Calgary, Alberta, Canada, June 2014.
View at: Publisher Site | Google Scholar
X. Zhang, Y. Lu, J. Tang, Z. Zhou, and Y. Liao, “Experimental study on fracture initiation and propagation in shale using supercritical carbon dioxide fracturing,” Fuel, vol. 190, pp. 370–378, 2017.
View at: Publisher Site | Google Scholar
R. Middleton, H. Viswanathan, R. Currier, and R. Gupta, “CO₂, as a fracturing fluid: potential for commercial-scale shale gas production and CO₂, sequestration,” Energy Procedia, vol. 63, pp. 7780–7784, 2014.
View at: Publisher Site | Google Scholar
H. Wang, G. Li, and Z. Shen, “A feasibility analysis on shale gas exploitation with supercritical carbon dioxide,” Energy Sources, vol. 34, no. 15, pp. 1426–1435, 2012.
View at: Publisher Site | Google Scholar
T. Ishida, Q. Chen, Y. Mizuta, and J. C. Roegiers, “Influence of fluid viscosity on the hydraulic fracturing mechanism,” Journal of Energy Resources Technology, vol. 126, no. 3, pp. 190–200, 2014.
View at: Publisher Site | Google Scholar
T. Ishida, K. Aoyagi, T. Niwa et al., “Acoustic emission monitoring of hydraulic fracturing laboratory experiment with supercritical and liquid CO₂,” Geophysical Research Letters, vol. 39, no. 16, 2012.
View at: Publisher Site | Google Scholar
S. Inui, T. Ishida, Y. Nagaya, Y. Nara, Y. Chen, and Q. Chen, “AE monitoring of hydraulic fracturing experiments in granite blocks using supercritical CO2, water and viscous oil,” in 48th US Rock Mechanics/Geomechanics Symposium, American Rock Mechanics Association, 2014.
View at: Google Scholar
Y. Chen, Y. Nagaya, and T. Ishida, “Observations of fractures induced by hydraulic fracturing in anisotropic granite,” Rock Mechanics and Rock Engineering, vol. 48, no. 4, pp. 1455–1461, 2015.
View at: Publisher Site | Google Scholar
X. Li, Z. Feng, G. Han et al., “Breakdown pressure and fracture surface morphology of hydraulic fracturing in shale with H₂O, CO₂ and N₂,” Geomechanics and Geophysics for Geo-Energy and Geo-Resources, vol. 2, no. 2, pp. 63–76, 2016.
View at: Publisher Site | Google Scholar
X. Zhou and T. J. Burbey, “Fluid effect on hydraulic fracture propagation behavior: a comparison between water and supercritical CO₂–like fluid,” Geofluids, vol. 14, no. 2, 188 pages, 2014.
View at: Publisher Site | Google Scholar
J. Wang, D. Elsworth, Y. Wu, J. Liu, W. Zhu, and Y. Liu, “The influence of fracturing fluids on fracturing processes: a comparison between water, oil and SC-CO₂,” Rock Mechanics and Rock Engineering, vol. 51, no. 1, pp. 299–313, 2018.
View at: Publisher Site | Google Scholar
Y. Zhang, J. He, X. Li, and C. Lin, “Experimental study on the supercritical CO₂ fracturing of shale considering anisotropic effects,” Journal of Petroleum Science and Engineering, vol. 173, pp. 932–940, 2019.
View at: Publisher Site | Google Scholar
X. Fan, J. Z. Su, X. Chang et al., “Brittleness evaluation of the inter-salt shale oil reservoir in Jianghan Basin in China,” Marine and Petroleum Geology, vol. 102, pp. 109–115, 2019.
View at: Publisher Site | Google Scholar
Y. Wang, X. Li, J. He, Z. Zhao, and B. Zheng, “Investigation of fracturing network propagation in random naturally fractured and laminated block experiments,” Energies, vol. 9, no. 8, p. 588, 2016.
View at: Publisher Site | Google Scholar
Z. Zhao, X. Li, J. He, T. Mao, B. Zheng, and G. Li, “A laboratory investigation of fracture propagation induced by supercritical carbon dioxide fracturing in continental shale with interbeds,” Journal of Petroleum Science & Engineering, vol. 166, pp. 739–746, 2018.
View at: Publisher Site | Google Scholar
B. Li, R. C. K. Wong, and T. Milnes, “Anisotropy in capillary invasion and fluid flow through induced sandstone and shale fractures,” International Journal of Rock Mechanics and Mining Sciences, vol. 65, no. 1, pp. 129–140, 2014.
View at: Publisher Site | Google Scholar
S. Baraka-Lokmane and R. Liedl, “From micro to meso: an exercise in determining hydraulic conductivity of fractured sandstone cores from detailed characterization of the fractures,” Geophysical Journal International, vol. 166, no. 3, pp. 1440–1453, 2006.
View at: Publisher Site | Google Scholar
J. He, C. Lin, X. Li, Y. Zhang, and Y. Chen, “Initiation, propagation, closure and morphology of hydraulic fractures in sandstone cores,” Fuel, vol. 208, pp. 65–70, 2017.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2019 Yixiang Zhang et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

967

Downloads

1358

Citations