Simulation of flow behaviour through fractured unconventional gas reservoirs considering the formation damage caused by water-based fracturing fluids

https://doi.org/10.1016/j.jngse.2018.06.039Get rights and content

Highlights

  • The flow behaviour of gas with non-linear properties in siltstone has been simulated using Discrete Fracture Model (DFM).

  • A field-scale model is built to investigate the effect of fracture aperture variation on shale gas production.

  • The effect of formation damage caused by water flooding on shale gas production has been studied.

Abstract

Hydraulic fracturing is essential for commercial-scale gas production from many unconventional gas reservoirs. While the effectiveness of the fractures created is associated with the stress created during the fracturing process, the use of water in the hydraulic fracturing process has been found to significantly reduce fracturing efficiency. In particular, the formation damage caused by water imbibition may have a significant negative impact on the flow capacity through both created fractures and rock matrix. These effects can be minimised by using non-water based fracking fluid such as CO2. The intention of this study is to investigate the formation damage caused by water invasion and multi-cycle confinement on the gas production of fractured reservoirs. A laboratory-scale discrete fracture model (DFM) was developed based on the experimental results of a series of permeability tests conducted on intact and fractured siltstone samples under steady-state conditions at room temperature using gaseous CO2 and water as the injection fluids. The developed model shows the ability to simulate the flow behaviour of fractured samples. Based on the laboratory-scale model, an expanded DFM model of an assumed fractured reservoir with a horizontal well was then built to quantitatively investigate the influences of multi-cycle confinement and water invasion damage on gas production from gas reservoir.

Based on the results of the expanded scale simulation, when the effect of water invasion damage on matrix permeability is not considered and only the change of fracture aperture is considered, the ratio of the rate of gas production from fractured formation without any formation damage, fractured formation suffering from the effect of multi-cycle confinement, fractured formation suffering from the combination effect of multi-cycle confinement and water invasion damage, and unfractured formation is around 15.15, 5.14, 2.23, 1 respectively, and the corresponding ratio of total gas production is around 17.62, 4.86, 2.13, 1 respectively, over a 10-year period. If the effects of water invasion damage on matrix permeability and fracture permeability are considered at the same time, the ratio of the rate of gas production rate from fractured formation without any damage on matrix permeability in the damage zone, fractured formation with 30% of initial permeability in the damage zone, fractured formation with 3% of initial permeability in the damage zone, and unfractured formation is around 2.23, 1.69, 0.94, 1 respectively, and the ratio of total gas production is 2.13, 1.54, 0.92, 1 respectively. This indicates that formation damage caused by multi-cycle confinement and fracturing water invasion can greatly impair gas production from unconventional gas reservoirs.

Introduction

Although unconventional gas reservoirs present a great opportunity to overcome the world energy crisis, the ultra-low permeability of these reservoirs is a challenge for commercial gas production from them (Conti et al., 2014). For example, the matrix permeability of a tight gas reservoir is usually less than 0.01mD, and the value can be several orders of magnitude less for shale gas reservoirs (Heller et al., 2014; Ghanizadeh et al., 2015; Bhandari et al., 2015). The ultra-low permeability of unconventional reservoirs greatly obstructs gas transport through the rock matrix to the production well (Lin et al., 2017), which indicates the importance of productivity enhancement techniques. Unconventional gas reservoirs are a kind of dual porosity system, most natural gas is stored in matrix pores and flowed by fractures, and natural gas production greatly depends on the fracture flow capacity and fracture density. Hydraulic fracturing has recently been used to create more fractures to enhance formation permeability by increasing the connections between matrix pores and production well (Daigle and Screaton, 2015; Kumar et al., 2015; Ma et al., 2016; Sun et al., 2016; He et al., 2017). During the fracturing process, the fractures can be kept open by injecting proppants with the fracturing fluid. However, conventionally-used fracturing fluids with low viscosity have very poor proppant-carrying capacity, sometimes resulting in no proppants in some fractures located far from the horizontal well (Warpinski, 2009). However, the shear displacement of fracture walls and the loose particles or debris created during the fracturing process can keep the fractures open, significantly increasing fracture flow capacity (Fredd et al., 2001). The flow capacity of the fracture system without proppants is therefore critical for gas production. However, the fracture system created is quite unstable and sensitive to many factors, especially to formation damage caused by stress and water-softening effects, and the variation of fracture flow capacity can directly determine the success or failure of hydraulic fracturing (Xu et al., 2016b; Zhang et al., 2015a, 2017b).

During the hydraulic fracturing process, a pressurized fluid, generally a water-based fracking fluid, is injected into the gas reservoir to fracture the rock matrix, which may result in a great change in the in-situ stresses and thus potentially creates seismic activities (Skoumal et al., 2015; Yang et al., 2015). Further, the redistribution of surrounding rock stress during gas production also increases the stress instability of the created fractures (Davies et al., 2013; Skoumal et al., 2015). This unstable stress development, which can be much higher than previous stable stress, may crush the contact areas between fracture surfaces and such irreversible deformations result in the reduction of fracture apertures. According to Zhang et al. (2017b), the effect of unstable stress on the fracture apertures was investigated by applying multi-cycle confining pressure varying between 10 MPa and 40 MPa, and the fracture aperture at 10 MPa was decreased by 60% after suffering from 40 MPa confining pressure, leading to a great reduction in fracture flow capacity. The fracture flow capacity during gas production life cycle is therefore greatly dependent on the residual fracture aperture suffered from higher stress in stress-sensitive fractured reservoirs.

The presence of water can induce serious formation damages during well drilling, completion, hydraulic fracturing and production, including physical and chemical damages (Bahrami et al., 2012; Xu et al., 2016a, 2017). In the fractured formation with dual porosity system, the fracture permeability and matrix permeability can be significantly decreased by formation damage caused by the complex physico-mechanical reaction between rock matrix and residual fracturing water (Bahrami et al., 2012; Xu et al., 2016a, 2016b, 2017; Lin et al., 2017). Studies have shown that water saturation can greatly reduce rock strength through the softening effect (Wong, 1998; Das et al., 2014; Zhang et al., 2017c), especially in clay-abundant shale formations, because the swelling of clay minerals induced by clay-water interaction can reduce the bonding between mineral particles and even create micro-fractures (Erguler and Ulusay, 2009; Zhang et al., 2017a). The softening effect decreases the brittleness of fracture surfaces, and the stress concentration between fracture surfaces or between the proppant and surfaces can crush the contact points (Pagels et al., 2013). This process can greatly decrease the fracture apertures and broken fine particles can also occupy the free space for gas flow. For example, according to Zhang et al. (2015a and 2015b), the decreased conductivity of propped fractures due to water invasion damage can cause around 90% reduction of gas flow rate. Moreover, the residual fracturing water can also significantly reduce matrix permeability through the water blocking and phase trapping damage. Studies have shown that the recovery of fracturing water is as low as around 10%–30% during flow-back and most residual water leaks into the rock matrix (Bostrom et al., 2014). This water retention basically happens due to the high capillary pressure and the low relative permeability (Holditch, 1979). With the increase of water content in the water sensitive formations, the damage on matrix permeability can hinder gas flow from matrix pores to fractures through water blockage and phase trapping damage in the tiny pores. Swelling of clay minerals can further reduce the pore porosity and close the pore throat, which greatly aggravate the damage on matrix permeability (Zhang et al., 2017a). The fluid invasion damage of water blocking and phase trapping is not only present in tiny pores, and even in narrow fractures it cannot be ignored (Reitsma and Kueper, 1994).

However, there is little published research focusing on the combined effect of water invasion damage and multi-cycle stress on the gas flow capacity of fractured formation without proppants, and the effects on gas productivity from unconventional gas reservoirs are therefore unclear. Although many simulation studies have been conducted using DFM, which is solved using the finite element analysis software COMSOL Multiphysics, to investigate the effect of fractures on gas production (Mi et al., 2014, 2016; Liang et al., 2016a, 2016b), most have adopted water as a flow fluid with nearly constant density under different pore pressures. This is not applicable for most gases due to their pressure-dependent density. Moreover, most current models are 2-D, which is also inconsistent with the reality that the gas produced in horizontal wells is collected from the whole 3-D gas reservoir volume instead of from a 2-D area (Sun and Schechter, 2014). This kind of limitation of current models has caused many inaccuracies in evaluating the efficiency of fracture systems.

The aim of the study was therefore to investigate the effect of formation damage on gas production from a fractured reservoir by conducting a simulation study. Based on the experimental data in Zhang et al. (2016 and 2017b), who investigated flow behaviour through intact and fractured siltstone samples considering the effects of multi-cycle confining pressure and water invasion damage on the fracture flow capacity of gas, a laboratory-scale 3-D model of cylinder sample was first developed using DFM and then expanded to a greater scale model of a fractured gas reservoir with a horizontal well.

Section snippets

Experimental results

Permeability tests were conducted using intact and fractured siltstone samples under steady-state conditions at room temperature (22 °C), where the pressure and flow rate along the sample will remain constant with time, the constant injection pressure at upstream was maintained by a 260D syringe pump and downstream pressure was open to atmosphere (0.1 MPa). The experimental results have been provided in previous publications (Zhang et al., 2016, 2017b). The influences of different factors

Laboratory-scale simulation

Due to the huge geometrical size difference between the whole sample and the fracture width, a dense mesh with a great number of tiny elements is necessary if a long and narrow fracture domain is used along the fracture, which is not realistic for the simulation. In the DFM model, the fracture is represented as an interior boundary instead of a separate domain and the model can efficiently and accurately simulate the fracture and matrix flow in a fractured block. Darcy's law governs the fluid

Expanded scale simulation of an assumed fractured reservoir

According to Section 3, the effect of water invasion damage on gas flow behaviour is obvious, for even laboratory-scale samples and therefore should not be ignored for gas production from field-scale reservoirs. In addition, the feasibility of DFM simulation of the flow behaviour of gas flow through intact and fractured samples has been verified in Section 3. Therefore, for a better understanding of the effects of multi-cycle confinement and water invasion damage on reservoir productivity, an

Conclusions

Based on an experimental study of the effect of formation damage caused by multi-cycle confinement and water invasion on the productivity of deep unconventional gas reservoirs in previous studies, a laboratory-scale DFM model was developed to simulate flow behaviour through intact and fractured 38 mm diameter and 76 mm high siltstone samples. An expanded scale DFM model based on the laboratory-scale model was then developed to investigate the reservoir-scale gas productivity behaviour under the

References (47)

  • C. Xu et al.

    Review on formation damage mechanisms and processes in shale gas reservoir: known and to be known

    J. Nat. Gas Sci. Eng.

    (2016)
  • C. Zhang et al.

    A novel approach to precise evaluation of carbon dioxide flow behaviour in siltstone under tri-axial drained conditions

    J. Nat. Gas Sci. Eng.

    (2016)
  • J. Zhang et al.

    Development of new testing procedures to measure propped fracture conductivity considering water damage in clay-rich shale reservoirs: an example of the Barnett Shale

    J. Petrol. Sci. Eng.

    (2015)
  • J. Zhang et al.

    Experimental and numerical studies of reduced fracture conductivity due to proppant embedment in the shale reservoir

    J. Petrol. Sci. Eng.

    (2015)
  • Z. Zhou et al.

    Experimental investigation of the effect of imbibition on shale permeability during hydraulic fracturing

    J. Nat. Gas Sci. Eng.

    (2016)
  • H. Bahrami et al.

    Phase trapping damage in use of water-based and oil-based drilling fluids in tight gas reservoirs

  • A.R. Bhandari et al.

    Anisotropy and stress dependence of permeability in the Barnett shale

    Transport Porous Media

    (2015)
  • N. Bostrom et al.

    The time-dependent permeability damage caused by fracture fluid

  • J. Conti et al.

    Annual Energy Outlook 2014 with Projections to 2040

    (2014)
  • H. Daigle et al.

    Evolution of sediment permeability during burial and subduction

    Geofluids

    (2015)
  • P. Das et al.

    Impact of formation softening and rock mechanical properties on selection of shale stimulation fluid: laboratory evaluation

  • C. Fredd et al.

    Experimental study of fracture conductivity for water-fracturing and conventional fracturing applications

    SPE J.

    (2001)
  • Y. He et al.

    A semianalytical methodology to diagnose the locations of underperforming hydraulic fractures through pressure-transient analysis in tight gas reservoir

    SPE J.

    (2017)
  • Cited by (9)

    • Change mechanism of pore structure and filling efficiency during injection production of sandstone underground gas storage

      2022, Journal of Natural Gas Science and Engineering
      Citation Excerpt :

      Most of them were based on petrophysical experiments (thin rock section microscopic observation, particle size analysis, X-diffraction analysis, scanning electron microscope analysis, physical property analysis, pore structure analysis, etc.) to determine the changes in physical parameters before and after different fluids were injected into the rock, describing the variation law of reservoir pore structure parameters (Fig. 2a), to clarify the degree of influence of the reservoir by external factors and clarify the reservoir damage mechanism in the process of oil and gas development (Sun et al., 2012; Hao et al., 2013; Tian et al., 2017; Liu et al., 2020). Scholars have also conducted much research on the changes in the stress sensitivity of reservoir rocks, sorted out various factors causing changes in reservoir sensitivity, such as injection mode, injection speed, and injection fluid type, and discussed the changes in reservoir pore structure and fluid flow resistance caused by these factors (Zhang et al., 2018; Kong et al., 2021) (Fig. 2b). In recent years, with the continuous updating of various new testing technologies, significant progress has been made in reservoir change mechanisms during injection production, and quantitative research has been gradually realized.

    • Combined micro-proppant and supercritical carbon dioxide (SC-CO<inf>2</inf>) fracturing in shale gas reservoirs: A review

      2021, Fuel
      Citation Excerpt :

      The addition of coated silicon nanoparticles 5 nm in diameter decreased the CO2 fingering effect and the leak-off rate (see Fig. 11(b)), and the action of nanoparticles not only increased the CO2 pressure to initiate hydraulic fractures but also improved the drainage area, implying more fractures [90]. Compared with nano-sized pores, micro-sized fractures are more likely to be the leakage channels [163]. For example, according to Hoelscher et al. [62], matrix pore sizes ranging from 3 to 100 nm and comparable fracture widths of 5–45 µm were observed for Marcellus shale cores, and the measured permeability was in the millidarcy to darcy and nanodarcies for the tested shale with and without small fractures, respectively.

    • Comparison of fracturing unconventional gas reservoirs using CO<inf>2</inf> and water: An experimental study

      2021, Journal of Petroleum Science and Engineering
      Citation Excerpt :

      The application of horizontal drilling and hydraulic fracturing has created a revolution in shale gas exploration in North America (Conti et al., 2014; Hayati-Jafarbeigi et al., 2020; Asadi et al., 2020). However, there are many issues in WBF fluids, including long clean-up time, serious formation damage, excessive water consumption and wastewater generation, which have limited its large-scale application in unconventional gas production (Middleton et al., 2015; Lyu et al., 2018; Sun et al., 2019; Zhang et al., 2018, 2020a). According to the research of Kondash et al. (2018), they adopted the statistical methods and found that the water consumption for hydraulic fracturing per well increased up to 770% from 2011 to 2016, and the flowback and produced water volumes generated in the first year for production increased up to 550%.

    • Evaluating the potential for oil recovery by imbibition and time-delay effect in tight reservoirs during shut-in

      2020, Journal of Petroleum Science and Engineering
      Citation Excerpt :

      The role of the type of clay minerals cannot be ignored due to the complexity of the microscopic pore structure involved. A formation with excessive clay content can be damaged after the fracturing fluid entering the reservoir (Yuan et al., 2017; Yuan and Wood, 2018; Zhang et al., 2018; Alvarado and Manrique, 2010). Fig. 14a illustrates an example of an oil distribution before clay swelling (Red blocks represent particles and blue part represents the distribution of oil).

    View all citing articles on Scopus
    View full text