Characterisation of creep in coal and its impact on permeability: An experimental study
Introduction
Coal permeability is a critical parameter for the prediction and evaluation of Coalbed Methane (CBM) production. Coal is a dual-porosity medium composed of cleats and matrices, which are the main conduit for gas migration and storage site, respectively. The change of coal permeability results from variation of effective stress and shrinkage and swelling of coal matrix owing to desorption and sorption of the gases in the reservoir. Matrix shrinkage and cleat compression mechanisms have inverse effects on permeability during CBM production. Whereas matrix shrinkage leads to dilation of coal cleat and an increase in permeability, cleat compression results in a decrease in permeability. The effects of the two mechanisms on coal permeability have been extensively studied (Levine, 1996, Palmer and Mansoori, 1996, Palmer and Mansoori, 1998, Seidle and Huitt, 1995, Shi and Durucan, 2004).
Alteration in porosity due to deformation of cleat-matrix assemblage leads to change of permeability. Deformation in coal can occur much faster due to being much softer than adjacent rocks (roof and floor rocks) (Brantut et al., 2013, Kaiser and Morgenstein, 1981). The relative softness of coal is due to large macromolecular organic networks in coal that do not possess strong bonds (Espinoza et al., 2016). Coal deformation process occurs at very slow rates during coalification and formation of overlying sedimentary rocks over geologic time scales. However, the deformation process (elastic and/or inelastic) may accelerate due to increasing effective stress during extraction of fluids in the reservoir (Schatz and Carroll, 1981). The impact of elastic deformation on coal permeability has been considered by some researchers (Liu and Rutqvist, 2010, Pan and Connell, 2011, Shi and Durucan, 2004). Nevertheless, investigations on the effect of inelastic deformation on coal permeability have not been carried out rigorously. Inelastic deformation of coal may occur during CBM production and well shut-in at static pore pressure. Compaction of coal reservoir occurs due to pressure depletion under uniaxial strain condition (the reservoir is confined laterally), which causes reduction in permeability and hence production rate (Wang et al., 2012). The mechanically induced compaction of coal due to increased effective stress is generally called primary consolidation, which is an inelastic deformation. When effective stress is constant, the compaction is known as secondary consolidation or creep, which is also an inelastic deformation (Barden, 1968, Bjørlykke et al., 2010). However, it is sometimes difficult to differentiate between genuine creep and consolidation effects (Fjær et al., 2008).
Creep is a mechanical and/or chemical process that is initiated by microstructure deterioration or restructuring of rocks. It is affected by parameters such as temperature, stress, and time. Creep can occur through four mechanisms namely: (1) Cataclasis: a delayed deterioration of microstructure that its dependency on time is relatively negligible and also generates a finite stress-dependent deformation (Fabre and Pellet, 2006, Frayne et al., 1990); (2) Pressure solution: solubility of the solids immersed in liquids change with stress (Yost and Aronson, 1987), or in other words, stress induces dissolution-precipitation. This type of creep may be dominant when water-gas two phase flow exists in coal reservoir; (3) Granular creep or particulate sliding: the imposition of grains rearrangement by frictional sliding as well as pressure solution in order to accommodate grain shape alteration throughout compaction process (Frayne et al., 1990); and (4) Adsorption-diffusion: a temporary compaction deformation induced by adsorption or diffusion which is different from that of permanent deformation of solid phase (Hol et al., 2013, Sone and Zoback, 2010). The dominant creep mechanism is determined by material properties such as moisture content, grain size, and strength as well as in situ conditions such as stress and strain rates (Frayne et al., 1990). Considering single-phase flow in coal, the dominant creep mechanisms are cataclasis and particulate sliding. In cataclastic flow regime, permeability and porosity are affected by the development of microstructure during compressive cataclastic failure (Zhu and Wong, 1997). Development of compaction is influenced by initial stress state and the stress path in the reservoir during drainage (Settari, 2002).
The impact of mechanical properties and rank on coal deformation and permeability has been extensively studied. Uniaxial compressive strength and Young's modulus increase with coal rank. This is due to less microporous structure of higher rank coal (Pan et al., 2013). Also, studies show higher permeability with pore pressure depletion for the coal with higher lateral Young's modulus (parallel to bedding) (Danesh et al., 2016, Pan and Connell, 2011). Higher rank coals such as anthracite do not creep and generally break explosively in uniaxial compression tests (Pomeroy, 1956). This is because coal matrix and cleat systems are generally stiffer (or denser) in higher rank coals, during the loading process under equal conditions, so that the matrix and cleat systems accommodate less creep deformation compared to lower rank coals. Hence, permeability change for higher rank coals is expected to be less.
Triaxial compression tests have been conducted for simulation of in-situ conditions for coal in order to measure coal geomechanical properties. In such tests, axial and hydrostatic stresses are applied to the coal core that is saturated with a specific gas (e.g. CH4, CO2, N2). Some studies have been carried out on coal characteristics under triaxial compression (Hobbs, 1964, Lin, 2010, Pan et al., 2010) as well as when high pressure gas is involved (Alexeev et al., 2012, Ujihira et al., 1985). In addition, creep behaviour of coal saturated with gas has been studied (Wang et al., 2011, Yang and Zoback, 2011, Yin et al., 2008, Yin et al., 2009, Zhu et al., 2011). Different types of gases have diverse adsorption capacity and consequently show dissimilar creep behaviours. The more the adsorption capacity of the gas is, the softer the coal becomes due to greater degree of coal matrix swelling. Yang and Zoback (2011) studied the effect of gas type on visco-plastic behaviour. They utilised He, N2, CH4, and CO2 in their experiments to examine the impact of adsorption of various gases on the mechanical and flow properties of coal. Their results show that the gas with higher adsorption capacity induces greater matrix swelling and therefore weakening of the coal mass that causes more creep in the coal than the gas with lower adsorption capacity. More recently, Danesh et al. (2016) developed a new stress-strain model to reflect the impact of inelastic deformation on coal permeability, and compared the change in gas production with and without inelastic deformation.
As aforementioned, creep behaviour of the coal saturated with gas has been widely studied. However, experimental studies that explicitly consider the influence of consolidation and creep (i.e. primary consolidation and secondary consolidation) on permeability of the coal saturated with different gases under varying stress conditions have yet to be conducted. In this study, a bituminous coal sample saturated with the gas (helium for the first test and methane for the second test), which was accommodated in a triaxial rig was used. For the case of helium, creep was induced by applying an axial load to the coal sample after the sample has reached equilibrium under hydrostatic stress. For the case of methane, gas production process was simulated by pore pressure depletion due to gas desorption from the same sample under constant hydrostatic and axial stress conditions. Permeability measurements were performed before and after the change in the effective stress.
Section snippets
Creep strain in coal
Fig. 1 shows a typical creep curve of coal under deviatoric stress condition. It consists of three stages: (1) primary or decelerating creep; (2) secondary or steady-state creep; and (3) tertiary or accelerating creep. Strain rate decreases with time in primary creep; then, it becomes zero in steady-state creep; and finally, it increases with time in tertiary creep.
Creep is normally modelled through exponential (Bai et al., 2012, Brantut et al., 2013, Li et al., 2011, Li et al., 2013,
Experimental set-up
A triaxial gas rig was used to study the impact of creep on coal permeability (Fig. 3). The rig was equipped with two axial displacement transducers and two radial strain gauges used for measuring axial displacements and radial strains. It was also capable of applying hydrostatic and axial loads separately. The two radial strain gauges were installed on the sample perpendicularly. The sample used for the experiments was a high-volatile bituminous coal excavated from Bowen Basin, Australia (see
Model validation
To extend the application of the experimental findings, the permeability model introduced by Danesh et al. (2016) along with a permeability model developed for creep under constant effective stress were employed to validate their suitability by fitting the experimental coal permeability data. The validation aims to apply the permeability model to better predict CBM production performance.
The permeability model (Danesh et al., 2016) incorporating the impact of viscoelastic deformation
Conclusions
This study presented a series of experimental results for a coal sample saturated with helium first and then methane under triaxial condition. The impact of time-dependent deformation on coal permeability and gas production was investigated in detail. For the case of helium, the irrecoverable deformation and the loss of permeability due to creep are substantial. RDR of 14.1% was observed in this test. In addition, a significant PLR of 71% due to residual deformation was observed. The
Acknowledgements
The authors would like to acknowledge School of Mechanical and Mining Engineering of the University of Queensland for provision of the scholarship, the partial support by ARC Discovery Project (DP150103467), and State Key Laboratory of Coal Resources and Safe Mining at China University of Mining and Technology (Project No.: SKLCRSM16KFA03). These sources of support are gratefully acknowledged. Furthermore, authors would like to acknowledge the contribution of CSIRO lab technicians, Mr. David
References (45)
- et al.
Time-dependent cracking and brittle creep in crustal rocks: a review
J. Struct. Geol.
(2013) - et al.
A technique for estimating the complete creep curve of a sub-bituminous coal under uniaxial compression
Int. J. Rock Mech. Min. Sci.
(1987) - et al.
Impact of creep on the evolution of coal permeability and gas drainage performance
J. Nat. Gas Sci. Eng.
(2016) - et al.
Adsorptive-mechanical properties of reconstituted granular coal: Experimental characterization and poromechanical modeling
Int. J. Coal Geol.
(2016) - et al.
Creep and time-dependent damage in argillaceous rocks
Int. J. Rock Mech. Min. Sci.
(2006) - et al.
Developments in Petroleum Science, Petroleum Related Rock Mechanics
(2008) - et al.
Time-dependent deformation of small tunnels-I. Experimental facilities
Int. J. Rock Mech. Min. Sci. Geomech.
(1981) - et al.
An analytic solution describing the visco-elastic deformation of coal pillars in room and pillar mine
Min. Sci. Technol. (China)
(2011) - et al.
Modelling of anisotropic coal swelling and its impact on permeability behaviour for primary and enhanced coalbed methane recovery
Int. J. Coal Geol.
(2011) - et al.
Laboratory characterisation of coal reservoir permeability for primary and enhanced coalbed methane recovery
Int. J. Coal Geol.
(2010)