A visual experimental study on proppants transport in rough vertical fractures

https://doi.org/10.1016/j.ijrmms.2020.104446Get rights and content

Abstract

The roughness of fractures may play an important role in affecting the migration and placement of proppants during hydraulic fracturing operations. A series of proppant flow experiments were conducted in attempt to compare the proppant transport mechanisms in rough fractures against those in smooth fractures. We examine the migration of proppants in rough and vertical fractures and then quantitatively reveal the effect of roughness on the instantaneous proppants transport and final proppants placement. The proppants-transport behavior in the rough and vertical fractures was observed to be totally different from that in the smooth and vertical fractures. The proppants in a rough vertical fracture do not progress like the regular sand bank that commonly occurs in the smooth fracture, but rather as an irregular-shape sand cluster with fractal characteristics. In the rough and vertical fracture, the phenomenon of proppants bridging is visually observed. The roughness of the fracture model not only affects how much of the fracture area is being occupied by the proppants, but also affects how tightly the proppants fill up the fracture.

Introduction

Hydraulic fracturing has become one of the treatment techniques frequently used for tapping hydrocarbon resources from tight reservoirs. Creating a massive fracture network with high conductivity is one of the major challenges faced by the industry when making efforts to maximize the production efficiency from tight reservoirs. Once a fracture is created in the reservoir, proppants need to fill said fracture as completely as possible, which is essential to achieve high fracture conductivity after the hydraulic fracturing treatment.

A large number of experimental and theoretical studies have been conducted to better understand the migration of fluid and proppants in fractures. Experimentally, a vast body of researches has also been reported in the literature. Liu and Sharma1 conducted an experimental study to examine the effect of fracture width and fluid rheology on proppants settling and retardation in vertical fractures. Both static settling and dynamic settling of proppants were considered in the experiments, and they also used rough fractures to study the effect of roughness on proppants transport. Important findings include: 1) the settling velocity of proppants decreases as the ratio of particle diameter to fracture width approaches one; 2) wall roughness can dramatically retard the horizontal velocity of the proppants along the fracture; 3) high-viscosity fluid can form thin layers around the proppants, which hinders the settling of the proppants carried by the low-viscosity fluid. However, their study did not investigate how the local fracture roughness itself affects the settling of proppants and the areal distribution of proppants in the fracture as the flow continues. Roy et al.2 used 3D printing technology to create transparent flow cells containing reconstructions of natural shale fracture surfaces and conducted proppants settling experiments in such flow cells. They reported that when a cluster of proppants particles with a diameter of about 500 μm are naturally settling down in the vertical flow cell with an aperture of about 700 μm, the settling velocity of the proppants is significantly lower than the Stokes velocity. They attributed the hindered settling of proppants to the increased particle-particle interactions, which could have been enhanced by the roughness of the cell walls.

The underlying mechanisms that control the transport and settlement of proppants in rough fractures are very complex. But, in general, the major influential factors include inter-proppant interaction, fluid-proppant interaction, proppant-rock interaction, and fluid-rock interaction. The inter-particle interactions were studied by Sandeep and Senetakis3; while the shale-rock-particle interactions were studied by He and Senetakis4 and He et al.5 These studies revealed that the tribological inter-particle interactions and shale-rock-particle interactions were generally governed by factors including particle properties and rock properties (including surface morphology, elastic property, and material hardness). Xiao et al.6 studied the tribological properties of sliding shale rock–alumina contact under dry and wet conditions, revealing that water can perform as a lubricant between particle and shale rock. Later, Zhang et al.7 studied the tribological properties of sliding quartz sand particle and shale rock contact under water and guar gum aqueous solution, finding that under the lubrication effect generated by guar gum aqueous solution, the friction of the rock-particle contact varies with guar gum concentration.

Theoretically, based on the experimental and field observations, many models have been developed to simulate how proppants-laden fluid transports in the fracture. Novotny8 conducted a pioneer study that found that the presence of vertical fracture walls near the particle hinders the settling of the particle. A dimensionless wall effect parameter, which is a function of particle diameter and fracture width, was introduced to take into account the effect of the presence of fracture walls on the particle's settling. This study also concluded that the roughness of the carbonate fractures does not affect the settling velocity of the proppants, but it should be noted that such conclusion is true for fractures with a width of at least 1/8 in (about 3.2 mm). Tsang9 conducted a numerical study to investigate the effect of fracture roughness and flow path tortuosity on fluid flow through a single fracture and found that such effect varies with the roughness characteristics of the fracture. Brown10 established a numerical model to study the effect of surface roughness on the fluid flow through rock joints. Gadde et al.11 developed new models for proppants transport and settling in vertical and smooth hydraulic fractures and implemented this new model in a 3D hydraulic fracturing code. However, again, their study was limited to the scope of smooth and vertical fractures. Tomac and Gutierrez12 conducted a numerical simulation study to understand the micro-mechanical mechanisms of horizontal proppants flow and transport in a narrow smooth hydraulic fracture. Note that the above-mentioned studies are applicable to smooth fractures. Utilizing a coupled discrete element method and computational fluid dynamics code, Tomac and Gutierrez13 further studied the micromechanics of proppants agglomerations in narrow and rough channels; both inter-particle and particle-wall interactions are considered in their modeling study. It was found that the previous relationships, such as Novotny8; are inadequate in describing the settling rates of proppants in a rough and narrow hydraulic fracture; more importantly, they observed that the settling of proppants does not always follow the direction of gravity in a rough fracture, which can help to form particle agglomerates that are settling faster and even clogging the fracture. A recent article by Osiptsov14 gives a thorough review of the fluid mechanics of hydraulic fracturing, touching on the vast body of theoretical and experimental works in this field. Nonetheless, their study is only dedicated to the fluid-proppants transport in smooth vertical fractures. More recently, Zhang et al.15,16 used a coupled CFD-FEM approach to simulate the proppants-laden fluid flow in fractures. Later using a similar approach, Zhang and Prodanovic17 simulated proppant transport in rough fractures with varied aperture sizes. The simulation results showed that the proppant-settlement characteristics in rough fractures appear to be drastically different from those in smooth fractures. There are some simulators that have been recently developed to simulate how proppants-laden fluid transports in the fracture, such as the spatially heterogeneous aperture code18,19 and CFRAC simulator.20,21 McLennan et al.22 pointed out in their speculations that the bridging of proppants may occur at locations in the primary fracture due to a rough fracture wall, and the roughness measurements on the fracture surfaces could be valuable inputs in hydraulic fracturing simulators. However, currently, there is still a lack of comprehensive experimental studies that focus on the proppants transport in rough vertical fractures,1 which hampers the progress made in improving the predictive capabilities of numerical simulators by considering the roughness characteristics of real fractures.

Recently, we have conducted a series of visualized experiments to clarify the roughness effect on the fluid flow and proppants transport.23,24, 25, 26, 27, 28 Using transparent replicas of fractures obtained from different types of rocks, Raimbay et al.27 studied the effects of roughness on the proppants transport in horizontal fractures. It was found that transport and placement of proppants in fractures are controlled by both transport ability of fracturing fluids and the roughness characteristics of the rock surface. They also found that high-viscosity polymer fluid is capable of distributing the sand in fractures much better than water. In addition, surface roughness not only controls the flow path of fluid but also the placement stability of proppants; in a smooth fracture, proppants are distributed more uniformly and packed as multilayers, while in a rough fracture, proppants are distributed non-uniformly. The reason for this non-uniform distribution in rough fractures can be sought in the tortuous nature of flow caused by surface roughness.29 Herein, it should be emphasized that roughness can cause the formation of different flow paths within the fracture that exhibit channel-type flow (channeled flow) properties of different tortuosity degrees.30,31 On the other hand, while roughness controls the characteristics of flow, roughness character (i.e., roughness degree) is controlled by the type of rock lithology25 as well as the fracturing conditions.32 More importantly, Raimbay et al.27 also found that three roughness parameters (i.e., fractal dimension as obtained from variogram analysis, fractal dimension as obtained from the triangular prism, and the ratio between total surface area and planar surface area) can be used to estimate the distribution of proppants in the fracture. Using a similar experimental approach, Raimbay et al.28 further studied the proppants behavior in the rough joint- and shear-type fractures. The surface roughness of fracture walls is found to significantly affect the transport and placement of fracturing fluid. Proppants behavior in joint- and shear-type fractures is different; However, it is worthwhile to mention that all of these previous experiments were conducted in horizontal fractures. Also in the experiments by Raimbay et al.,27,28 the proppants are placed at the fracture entrance and flushed into the fracture by chasing fluids, which might not be representative of the field operation conditions where the fracturing fluid carries the proppants into the fracture. Thus, many of these findings might not apply to the actual proppants transport in rough vertical fractures. Vertical fractures are more often created in tight/shale reservoirs to increase the exposure of reservoir to the wellbore. It is thereby of utmost importance to use rough vertical fractures in the proppants transport experiments to better honor the reality.

The roughness of fractures is recognized as playing an important role in affecting the transport and placement of proppants during hydraulic fracturing operations. Previous studies focused on investigating the proppants transport in either smooth vertical fractures or rough horizontal fractures. To our knowledge, only very limited experimental efforts have been devoted to investigating the proppants transport in rough vertical fractures, although, in actual field operations, normally proppants transport happens in rough vertical fractures. In this research, we use a novel experimental apparatus to understand how the fracture roughness affects the migration and settling of proppants along vertical fractures. Such a study will provide important experimental data that can eventually be used to develop more accurate simulators for modeling proppant transport in real vertical fractures.

Section snippets

Materials

Most of the model fractures used in this study are the same as those from our previous studies.25, 26, 27, 28 They were produced from artificially created tensile fractures of lithological-different rocks. The reason for the consideration of diversity in lithology was to obtain fractures with different surface roughness characteristics depending on petrological properties. According to thin section observations of these rocks under a polarized microscope, sample Fr4 and 4s are coarse grained

Dimensional analysis

We have applied the Buckingham π theorem in the study in an attempt to bridge the findings from the lab to the field. In this paper, the basic parameters include the following: pressure measured at the inlet of the fracture model (p), actual density of proppants (ρ), viscosity (μ), velocity (ν), time (t), fracture width (w) (fracture height or fracture length), and flow rate (Q). Density, width, and velocity are taken as the fundamental variables. By applying the π theorem, we find the

Conclusions

In this study, we conducted a series of experiments to reveal the effect of fracture roughness on the fluid flow and proppants transport in vertical fractures. On the basis of the experimental results that were obtained under experimental conditions in this study, the following conclusions can be drawn:

  • 1)

    The interface between polymer solution and air in a flat vertical fracture always maintains a concave shape, while the counterpart in a rough vertical fracture exhibits a convex shape in the

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This research was conducted under T. Babadagli's NSERC Industrial Research Chair in Unconventional Oil Recovery (industrial partners are APEX Eng., Devon, Husky Energy, Petroleum Development Oman, Saudi Aramco, SIGNa Oilfield Canada, Total E&P Recherché Développement) and NSERC Discovery Grants (No: RES0011227 and NSPIN 05394) to T. Babadagli and H. Li, respectively. H. Huang is also grateful for the financial supports provided by the National Natural Science Foundation of China (No. 51874240,

References (40)

  • Z. Chen et al.

    An experimental investigation of hydraulic behavior of fractures and joints in granitic rock

    Int J Rock Mech Min Sci

    (2000)
  • T. Babadagli et al.

    Fractal characteristics of rocks fractured under tension

    Theor Appl Fract Mech

    (2003)
  • H. Huang et al.

    Effect of injection parameters on proppant transport in rough vertical fractures: an experimental analysis on visual models

    J Pet Sci Eng

    (2019)
  • J. Zeng et al.

    Numerical simulation of proppants transport in hydraulic fracture with the upscaling CFD-DEM method

    J Nat Gas Sci Eng

    (2016)
  • Y. Liu et al.

    Effect of fracture width and fluid rheology on proppants settling and retardation: an experimental study

  • P. Roy et al.

    Proppants transport at the fracture scale: simulation and experiment

  • H. He et al.

    A micromechanical study of shale rock-proppant composite interface

    J Pet Sci Eng

    (2019)
  • H. Xiao et al.

    Tribological properties of sliding shale rock–alumina contact in hydraulic fracturing

    Tribol Lett

    (2016)
  • E.J. Novotny

    Proppants transport

  • Y.W. Tsang

    The effect of tortuosity on fluid flow through a single fracture

    Water Resour Res

    (1984)
  • Cited by (43)

    View all citing articles on Scopus
    1

    This paper was written while the second author and the third author were residing at Xi'an Shiyou University as guest professors in 2019.

    View full text