Micro-scale fracturing mechanisms in coal induced by adsorption of supercritical CO2

https://doi.org/10.1016/j.coal.2017.04.002Get rights and content

Highlights

  • A new method is proposed for predicting micro-mechanical coal behaviour

  • Mechanistic evidence in terms of how swelling stress fractures the unswelling (mineral) phase in coal is provided

  • Microscale rock mechanical behaviour and internal swelling stress fields in coal were successfully predicted

Abstract

Coal bed methane production can be assisted by CO2 injection. However, CO2 adsorption in the coal matrix leads to a dramatic reduction in permeability and an associated change in microstructure caused by coal matrix swelling. Furthermore, it has been recently observed that the induced swelling stress fractures the unswelling (mineral) phase in laboratory investigations. However, the failure mechanisms are still not understood, and the way internal swelling stresses are generated is not clear. Thus, in this paper, we propose a new method which combines X-ray microtomography imaging, nanoindentation testing and DEM modelling with which we can predict the rock mechanical performance at micro scale. Indeed we successfully simulated such swelling processes inside a coal sample, including a simulation of the fracture mechanism of the mineral phase, and a quantification of the in-situ von Mises stresses generated by swelling. We conclude that our proposed method is an efficient way for analysis and prediction of coal microfracturing and the associated microscale rock mechanical behavior.

Introduction

Coal bed methane (CBM) is an unconventional energy resource, which exists in coal mines and deep unmineable coal seams (Hamawand et al., 2013). Recently, due to the decline in conventional energy resources coupled with a globally increasing energy demand (Lior, 2008), CBM has gained increasing popularity (Connell et al., 2011, Pillalamarry et al., 2011, Hamilton et al., 2015, Vishal et al., 2015). Furthermore, CBM can be enhanced (enhanced coal bed methane, ECBM), e.g. through CO2 injection, which efficiently displaces CH4 from the coal matrix (White et al., 2005, Saghafi, 2010). However, CO2 injection dramatically reduces the coal seam's permeability (Mazumder et al., 2006, Siriwardane et al., 2009, Anggara et al., 2016), which largely limits application of this technology. Mechanistically, cleats (the main flow conduits in coal) close due to coal matrix swelling induced by CO2 adsorption (Shi and Durucan, 2005, Wu et al., 2011, Zhang et al., 2016a, Liu and Rutqvist, 2010, Espinoza et al., 2014) and it has recently been discovered that the swelling stress in the coal matrix can fracture the unswelling phase (i.e. inorganic mineral), (Zhang et al., 2016a). However, the detailed failure mechanisms and swelling stress quantification are still poorly understood due to only limited theoretical understanding of the micro-scale rock mechanical performance. It is thus of vital importance to further understand these mechanical changes in the coal so that advanced ECBM techniques can be developed.

The mechanical properties of small areas (up to nanoscale) on a material's surface can now be obtained by nanoindentation measurements; such method has for instance been applied to natural rock samples including sandstone, limestone, shale and coal (Zhu et al., 2009, Bobko et al., 2011, Lebedev et al., 2014, Manjunath and Nair, 2015, Vialle and Lebedev, 2015, Liu et al., 2016). Thus nanoindentation gives us a way to identify the mechanical properties of heterogeneous coal (note that coal consists of the organic coal base matrix, inorganic minerals and pores). These mechanical properties are essential input data into numerical models, which can predict the mechanical behavior of the whole (heterogeneous) material. Earlier studies considered the coal matrix as a homogenous elastic continuum (e.g. Izadi et al., 2011), which obviously cannot capture the clearly heterogeneous character of the coal, and thus can only provide rather biased predictions. To overcome this serious limitation we use discrete element method (DEM) modelling (cp. Cundall and Strack, 1979, Wang et al., 2014, Zhang et al., 2016e, Bai et al., 2016), where each material – coal matrix, mineral and void are assigned their respective true and individual mechanical properties, and combine this with high resolution x-ray micro-computed tomography (microCT) imaging, which can provide the detailed 3D morphology of the coal (Zhang et al., 2016b, Zhang et al., 2016c, Zhang et al., 2016d, Jing et al., 2016, Mostaghimi et al., 2017). Thus, in this paper, using this approach, we were able to quantify the swelling stresses generated by supercritical CO2 injection into coal, and to identify the failure mechanisms occurring in the un-swelling phase.

Section snippets

Experimental work

A small cylindrical coal plug (5 mm diameter and 10 mm length) was cut from a heterogeneous subbituminous medium rank coal block obtained from a coal seam at ~ 650 m to 700 m depth (buried at Permian period) from Pingdingshan coal mine, China; the generalized stratigraphic column is shown in Fig. 1. The coal had a 54% (± 2%) carbon content and a 36% (± 1%) volatile matter content (measured by Chinese Standard GB/T 212-2008 and DL/T 1030-2006; Xu et al., 2016, Zhang et al., 2016d), additional

The stress-strain method

Initially we estimated the swelling stress via the traditional stress-strain method using the volume fractions measured (i.e. strains measured) on the microCT images, Table 2 (note that the volume strain ε = (the volume difference before and after CO2 flooding)/(original volume). Negative values represent compression, while positive values represent expansion.

What is more, in an elastic 3D coordinate system, ε has following relation with Young's modulus (E), stress (σ), and Poisson's ratio (υ),

Discrete element method (DEM) simulation

The Discrete Element Method (DEM) has become a powerful numerical tool for analyzing the dynamic mechanical behavior of complex objects (Cundall and Strack, 1979, Scholtès and Donzé, 2012). Precisely, DEM models objects as an assembly of interacting particles, and the key advantage of DEM is that specific attributes (features, bonds, contacts, frictions and boundary conditions) can be assigned to each particle (and thus each material) simultaneously. We used the popular DEM built-in software –

Nanoindentation testing

The IBIS nanoindentation system and Berkovich nano-indenter were chosen for the nanoindentation tests, Fig. 7. A cuboid coal sample (l × w × h = 5 mm × 5 mm × 2 mm) was cut and carefully polished, and mounted on the objective stage. Subsequently the penetration depth (h) – loading/unloading force (P) curves were measured for each test point. Specifically 625 data points on a symmetric 25 × 25 grid (240 μm × 240 μm spacing) were measured. The maxim loading force was set to 4 mN (which is smaller than the one used

Results and discussion

The DEM simulations successfully predicted the change in coal microstructure caused by coal matrix swelling, which again was induced by scCO2 injection (see Fig. 8, Fig. 9). Clearly cracks appeared in the mineral phase when the coal matrix volume increased by 1%, consistent with the experimental microCT observations (cp. Fig. 3). Most failures in the calcite mineral phase were identified as tensile failures (red colored cracks in Fig. 8, Fig. 9) these appeared during the coal matrix swelling.

Conclusions

CO2 can be injected into coal seams to enhance methane production (White et al., 2005, Saghafi, 2010); however, the resulting coal matrix swelling effect leads to coal cleat closure and a dramatic permeability reduction (Karacan, 2003, Zhang et al., 2016f); furthermore, it has been recently observed that the unswelling phase fractured due to the induced swelling stresses (Zhang et al., 2016b). However, how precisely such swelling stresses are generated and the associated failure mechanisms in

Acknowledgements

The measurements were performed using the microCT system courtesy of the National Geosequestration Laboratory (NGL) of Australia, funding for the facility was provided by the Australian Government. This work was also supported by resources provided by the Pawsey Supercomputing Centre with funding from the Australian Government and the Government of Western Australia.

References (57)

  • C.Ö. Karacan et al.

    Coal mine methane: a review of capture and utilization practices with benefits to mining safety and to greenhouse gas reduction

    Int. J. Coal Geol.

    (2011)
  • N. Lior

    Energy resources and use: the present situation and possible paths to the future

    Energy

    (2008)
  • K. Liu et al.

    Applications of nano-indentation methods to estimate nanoscale mechanical properties of shale reservoir rocks

    J. Nat. Gas Sci. Eng.

    (2016)
  • G.L. Manjunath et al.

    Implications of the 3D micro scale coal characteristics along with Raman stress mapping of the scratch tracks

    Int. J. Coal Geol.

    (2015)
  • M. Pillalamarry et al.

    Gas diffusion behavior of coal and its impact on production from coalbed methane reservoirs

    Int. J. Coal Geol.

    (2011)
  • D.O. Potyondy et al.

    A bonded-particle model for rock

    Int. J. Rock Mech. Min. Sci.

    (2004)
  • T. Rahman et al.

    Residual trapping of supercritical CO2 in oil-wet sandstone

    J. Colloid Interface Sci.

    (2016)
  • A. Saghafi

    Potential for ECBM and CO2 storage in mixed gas Australian coals

    Int. J. Coal Geol.

    (2010)
  • L.U.C. Scholtès et al.

    Modelling progressive failure in fractured rock masses using a 3D discrete element method

    Int. J. Rock Mech. Min. Sci.

    (2012)
  • P. Sheorey

    A theory for in situ stresses in isotropic and transverseley isotropic rock

    Int. J. Rock Mech. Min. Sci. Geomech. Abstr.

    (1994)
  • H. Siriwardane et al.

    Influence of carbon dioxide on coal permeability determined by pressure transient methods

    Int. J. Coal Geol.

    (2009)
  • K. Tajduś

    New method for determining the elastic parameters of rock mass layers in the region of underground mining influence

    Int. J. Rock Mech. Min. Sci.

    (2009)
  • V. Vishal et al.

    Influence of sorption time in CO2-ECBM process in Indian coals using coupled numerical simulation

    Fuel

    (2015)
  • T. Wang et al.

    Simulation of hydraulic fracturing using particle flow method and application in a coal mine

    Int. J. Coal Geol.

    (2014)
  • Y. Wu et al.

    Evolution of coal permeability: Contribution of heterogeneous swelling processes

    Int. J. Coal Geol.

    (2011)
  • X. Xu et al.

    Experimental study on physical structure properties and anisotropic cleat permeability estimation on coal cores from China

    J. Nat. Gas Sci. Eng.

    (2016)
  • Y. Zhang et al.

    Swelling effect on coal micro structure and associated permeability reduction

    Fuel

    (2016)
  • Y. Zhang et al.

    Multi-scale x-ray computed tomography analysis of coal microstructure and permeability changes as a function of effective stress

    Int. J. Coal Geol.

    (2016)
  • Cited by (80)

    • Numerical simulations of supercritical carbon dioxide fracturing: A review

      2023, Journal of Rock Mechanics and Geotechnical Engineering
      Citation Excerpt :

      Among the three types of BPs (horizontal BPs, vertical BPs, and orthogonal BPs), the first two types restrict the fracture propagation in the other direction (Zhang et al., 2019), so reservoirs with orthogonal BPs are more likely to form a complex fracture network (Luo et al., 2021b). In addition, the expansion stress induced by SC-CO2 adsorption on coalbed and shale can increase the chance of rock damage and facilitate the formation of a complex fracture network (Zhang et al., 2017b; Liu et al., 2018). In the case of shallow coalbed, Yan et al. (2019, 2020, 2021) revealed that the phase changed fracturing stage after the SC-CO2 fracturing stage, and the fracture propagation mainly occurred in the first stage, accounting for 80%–90% of the fracture length, while the contribution to the fracture width was nearly the same in the two stages.

    View all citing articles on Scopus
    View full text