Tao Wang,Yaodong Jiang,Shaojian Zhan,Chen Wang
aSchool of Civil and Environmental Engineering,University of Science and Technology Beijing,Beijing 100083,China
bSchool of Mechanics and Civil Engineering,China University of Mining and Technology,Beijing 100083,China
Frictional sliding tests on combined coal-rock samples
Tao Wanga,b,*,Yaodong Jiangb,Shaojian Zhanb,Chen Wangb
aSchool of Civil and Environmental Engineering,University of Science and Technology Beijing,Beijing 100083,China
bSchool of Mechanics and Civil Engineering,China University of Mining and Technology,Beijing 100083,China
A R T I C L E I N F O
Article history:
Received 28 February 2014
Received in revised form
10 March 2014
Accepted 8 April 2014
Available online 21 April 2014
Stick-slip
A test system was developed to understand the sliding mechanism of coal-rock structure.The test system was composed by a double-shear testing model and an acousto-optic monitoring system in association with a digital camera and an acoustic emission(AE)instrument.The tests can simulate the movement of activated faults and the sliding in coal-rock structure.In this regard,instable sliding conditions of coalrock samples,sliding types under different conditions,displacement evolution law,and AE characteristics during sliding process were investigated.Several sliding types were monitored in the tests, including unstable continuous sliding,unstable discontinuous sliding,and stable sliding.The sliding types have close relation with the axial loads and loading rates.Larger axial load and smaller loading rate mean that unstable sliding is less likely to occur.The peak shear stress was positively correlated with the axial load when sliding occurred,whereas the displacement induced by unstable sliding was uncorrelated with the axial load.A large number of AE events occurred before sliding,and the AE rate decreased after stable sliding.The results show that the tests can well simulate the process of structural instability in a coal bump,and are helpful in the understanding of fault activation and the physical processes during squeezing process of roof and f l oor.
?2014 Institute of Rock and Soil Mechanics,Chinese Academy of Sciences.Production and hosting by Elsevier B.V.All rights reserved.
Coal-rock is a kind of heterogeneous and anisotropic geomaterial associated with signi f i cant nonlinear behavior and with discontinuous geological interfaces,including strati f i cations,joints, schistosity,and fractures,which divide the coal-rock into various structural blocks.Geological interfaces and structural blocks constitute the structure of coal-rock(Xie and Pariseau,1993). Statistically,coal bumps usually occur in vicinity of fault zones,fold zones,or the areas that have experienced a major change in coal seam dip(Jiang et al.,2012;Zhao et al.,2013),thus the occurrence of coal bumps is related to the type of coal-rock structures in the area of interest.Also the dilatancy effect induced by compression of coal seams usually causes tensile stresses in the coal mass.When the coal mass is disturbed under quasi-static loading induced by mining,the coal-rock will slide out in blocks or as a whole,which is called coal bump of structure unstable type that is induced by coaland-rock sliding(Jiang et al.,2009;Zhang et al.,2012).
Friction phenomenon exists in various scales of geological movement.The relative movement of the two walls of a fault is similar to stick-slip in form of friction,and stick-slip can be regarded as a factor causing tectonic earthquakes(Brace and Byerlee, 1966;Byerlee and Brace,1968;Byerlee,1970).In fact,a coal bump is usually caused by mining disturbance that induces fault activation(Ruina,1983;Farmer,1985).This unstable mode can be attributed to a stiffness difference between the coal seam,roof and fl oor,or by geological factors.In this case,it can be classi fi ed as a structural unstable coal bump.Research on the friction associated with a fault and the mining is important for understanding the mechanism of structural unstable coal bump.
In this context double-shear tests are important to understand the sliding mechanism of coal-rock structure.In the double-shear friction test,the contact area basically remained constant during the tests,making it easy to determine the stress state of the sliding surface.Itshould be noted that the test method is suitable when the normal stress is relatively low.However,a relatively large sliding displacement is allowed and can be directly measured.So the double-shear test method is chosen in this study(Price and Cosgrove,1990).
The process of stick-slip,to a certain extent,is related to fault activation.As a result,many seismologists studied the mechanismof earthquakes by conducting friction tests on rocks(Brace and Byerlee,1966;Ma et al.,2007;Song et al.,2012a;Passelègue et al.,2013).These tests were frequently observed in the f i eld of seismicity,but the one associated with coal mining was rarely reported.In this paper,the friction-sliding experiments were conducted on coal-rock samples to simulate mining-induced fault activation and structural unstable coal bump,especially the squeezing of the roof and f l oor.
2.1.Experimental device and rock samples
The double-shear friction test was conducted imposing biaxial loads by an independent hydro-cylinder and a loading device in vertical and horizontal directions.The independent loads on two orthogonal directions can be applied.The rock samples include granite,sandstone,and coal(Fig.1).
The testing model is shown in Fig.1a and testing samples in Fig.1b.The vertical loading direction was de fi ned as the axial direction,and the horizontal loading direction as the shear direction. A constant axial load was speci fi ed,and the lateral load was increased stepwise till sliding of the samples.The loading steps were considered as follows: fi rst,the axial stress was applied and increased to the desired value gradually.Once the desired value was reached,the axial stress was kept constant with the loading control method;then the shear stress was applied,with displacements control,to observe the occurrence of friction sliding.If any of the following conditions was reached,the test can be fi nished:(1) the sample was damaged;(2)unstable sliding occurred;or(3)the shear displacement was out of the limit of the testing apparatus. Before testing,the uniaxial compressive strength(UCS)was determined on rock samples with different lithologies.The sample parameters and basic experimental conditions are shown in Table 1.

Fig.1.Testing model and samples.(a)Testing model and(b)Testing samples.
2.2.Monitoring system
The experimental system and the displacement observation system are shown in Fig.2.Digital speckle measuring points were arranged on the front surface to measure the sliding displacement, and two AE sensors were mounted on the rear surface of the sample to collect the information obtained by the AE sensors during sliding.Before test,the timing system was calibrated to ensure consistent measurement of different monitoring systems.
Digital photographic technology was used in the test to acquire visual information about the specimen surface,and the digital speckle correlation method was used to analyze the slip surface by examining the speckle displacement regularity and sliding characteristics.A 1351μm Daheng-type industrial camera was used, and it was attached on a computer system for real-time image acquisition at a frequency of 15 fps.
An SWAES multichannel AE detector with full waveforms was used in testing.Its frequency ranges from 50 kHz to 400 kHz,and the resonance frequency is 150 kHz.The peak sensitivity is greater than 65 dB,and the pre-amplitude gain is 40 dB.The bandwidth range is 2-10 MHz,and the highest sampling rate of the acquisition card is 20 MHz.A number of analyses were performed based on ringing count and energy count(Lockner,1993;Mansurov,1994; Arasteh et al.,1997;Builo,2000;Shkuratnik et al.,2004;Zhao and Jiang,2010).
Eight groups of rock specimens with four different lithologies were considered in the sliding friction tests.Unstable continuous sliding,unstable discontinuous sliding,and stable sliding were observed in the tests.The sliding rule was analyzed based on different sliding mechanisms.
3.1.Stress and AE characteristics during the process of unstable continuous sliding
An axial load of 20 MPa was applied on the granite specimen 1-5, and the shear loading rate was 0.25 mm/min.Unstable continuous sliding occurred several times in the form of regular stick-slip, followed by a rhythmic“click”sound.The shear stress curve is shown in Fig.3,with the change in loading observed.
When the shear stress reached 6.05 MPa,the f i rst unstable sliding event was recorded in the sample with a stress drop of 0.72 MPa.The changes of peak stress and stress drops are shown in Fig.4,in association with the change in displacement.
It is obvious from Fig.4 that the peak stresses of different unstable sliding events were linearly arranged.The friction strength of the rock samples increased with the increasing displacement.The peak stresses were increased by approximately 0.06 MPa from one unstable period to the next.The stress drops increase linearly at the beginning,but with small f l uctuations in growth.
For the specimens under high stresses,small fractures occurred in the friction surface during the process of unstable sliding,which makes the friction surface roughly.The lubrication function of fragments made the stress drop curve f l uctuate irregularly.This phenomenon is similar to that of a coal bump during coal mining.If several working faces cross the same fault and coal bumps occur at one working face,coal bumps with the same intensity are likely to occur at one of the other working faces at approximately the same position.After that,the chance of coal bump decreases at other working faces.
The shear stress curve and AE characteristics during unstable sliding are shown in Fig.5,with which we can analyze the relationship between the sudden release of shear stress and AE count.

Table 1Sample parameters and basic experimental conditions.
In Fig.5,it can be seen that AE events of different energy values occurred at the initial stage of shear stress loading,because fractures were closed under low stresses and local rough areas became damaged.These phenomena are consistent with the AE law that governs the loading process in rock compression tests(Ohnaka and Mogi,1982;Yoshikawa and Mogi,1989;Eberhardt et al.,1999; Lacidogna et al.,2011).At the initial loading stage,the AE characteristics showed that the AE count was small,the energy and the amplitude were low during closure of the fractures and joints.As the load increased,the fractures and the contact surfaces were consolidated,and the AE signals became weak.The AE count,energy value,and amplitude also decreased.After that,the deformation of the contact surfaces increased and damage occurred.In the unstable sliding phase,the AE signals became stronger,and the AE count also increased,thus the energy and amplitude were higher.In the processes of rock compression and friction,the AE phenomenon was the same as that in the initial loading stage.
The AE events increased stepwise from the onset of unstable sliding,and the time interval of the large AE events was nearly the same.It should be noted that the AE events are correlated with friction strength and stress drop associated with the period of unstable sliding.
In Fig.5,numbers 1-14 indicate the 14 unstable sliding events, and letters A-N indicate the high-energy AE events that occurred with the unstable sliding events.Numbers 1-14 correspond to the AE events A-N,respectively.
From Fig.5,some conclusions can be drawn:

Fig.2.Schematic diagram of loading and monitoring systems.
(1)As the stress peak values of unstable sliding increased,the corresponding AE energy values increased generally.
(2)In the initial stage of unstable sliding,there were fewer lowenergy AE events,as the events A-G shown in Fig.5.In the later stage of unstable sliding,there were more low-energy AE events,such as the events K-N.
(3)Most of the unstable sliding events were accompanied with AE events.However,there were no large AE events in the 3rd,9th, and 11th unstable sliding events,and relatively more AE events occurred before or after low-energy sliding.

Fig.3.Shear stress-displacement curve of granite specimen 1-5.
(4)The AE events were a combination of advancing or hysteretic events relative to the stress drop.Eleven unstable slidingevents were recorded with powerful AE energies,f i ve of which were advancing and six were hysteretic.However,time differences between unstable sliding and stress drop were not the same.

Fig.4.Stress curves of specimen 1-5 during the process of unstable sliding.
The AE is a micro-seismic impulse occurring when the coal-rock is damaged,re f l ecting the development and expansion of microfractures in the coal-rock(Ma et al.,2012).The accumulation of AE energy is directly associated with the damage degree to the coalrock.As a result,AE can be used to study the damage characteristics of coal-rock and to monitor and forecast coal bumps.The test shows that coal bumps are associated with large energy releases,and AE may occur where no signi f i cant AE events occur before.Thus,it is not appropriate to forecast this type of coal bump only using AE data,and other measurements should also be used.
3.2.Evolution of displacementf ield in unstable continuous sliding
The evolutions of stress and displacement of the specimen 1-5 are shown in Fig.6.PointsA,B,C,andDwere selected as feature points on the loading curve in the second unstable sliding process. The shear stresses at these four points were 5.3 MPa,5.9 MPa, 6.1 MPa,and 5.2 MPa,respectively.
PointAis at the end of the f i rst unstable sliding event,and it is at the beginning of the interval before the second unstable sliding event.The speckle image at this moment is reference for analyzing the displacement f i eld,based on which the displacement f i elds of pointsB,C,andDare calculated and shown in Fig.7.
As seen in Fig.7a and b,the spatial deformation of specimen was heterogeneous because of the heterogeneity of the specimen and the characteristics of the contact surface.The deformation difference between the two sides of the contact surface directly re f l ects the heterogeneous distribution of deformation energy on the contact surface,making it more complex in sliding process.When sliding suddenly occurred in the sample,the deformation became homogeneous,and the deformation energy was released on the contact surface.The deformation was recovered to some extent, and the displacement isolines were nearly parallel,as shown in Fig.7d.
Based on the concept of energy dissipation(Song et al.,2012b) and the analysis of the displacement f i elds of speckle images, evolution of a coal bump is resultant fromenergy accumulation and dissipation.When a disturbance is applied,the state of energy balance is disrupted,thus the stored energy is suddenly released and a coal bump occurs.
The deformation of a coal-rock structure is a process of stress or strain energy accumulation.The free energy in rocks surrounding the coal increases,and the system goes into a quasi-steady state (Fukui et al.,2005).Rib spalling and coal burst are the manifestations at the macro-level,and the development,convergence, nucleation,and stable expansion of fractures are the manifestations at the micro-level.These phenomena correspond to the stress and displacement f i elds of pointB,as shown in Figs.6 and 7.As energy accumulates,the rocks surrounding the coal are in a critical state when the free energy reaches a critical value.In this point,the accumulation of stress or strain energy is the largest(this corresponds to the state of the stress and displacement f i eld at pointCin Figs.6 and 7).When a mining disturbance occurs,the system quickly goes into an unstable state,and the coal-rock structure fails. This is observed in a buckling failure of coal pillar or in stick-slip between the coal and rock at the macro-level.This process is quite short,corresponding to the transition from pointCtoDin Figs.6 and 7.After that,another stable state is reached,corresponding to the stress and displacement f i eld of pointDin Figs.6 and 7.

Fig.5.Acoustic characteristics during the process of unstable sliding.

Fig.6.Evolutions of stress and displacement during unstable sliding.
A structural unstable coal bump can be roughly caused by the instability of the coal-rock structure.The damage characteristics ofunstable sliding in coal-rock are similar to those of stick-slip:the coal seams burst out from the roof and f l oor,but the roof and f l oor remain undamaged.The friction sliding process of a coal-rock structure is unstable;essentially instant sliding is basically accompanied by energy release.The structural unstable coal bump is usually induced by the loading or unloading of the roof and f l oor on the coal mass.Detailed studies on the friction of coal-rock structure are signi f i cantly important for mechanism analysis of a structural unstable coal bump.
On the eight groups of tests,unstable continuous sliding,unstable discontinuous sliding,and stable sliding occurred.In this regard,the shear stress should be stopped at the moment when the axial load was applied.The shear stress,stress drop,sliding,and other characteristics for the f i rst sliding event are summarized in Table 2.The friction strength curves for different groups of specimens are shown in Fig.8.
In Fig.8,the friction coef f i cient of a sandstone-coal structure is 0.305,whereas the one of granite is 0.177.In fact,there are two reasons that the sliding characteristics of coal-rock structure are different from that of granite.One is the difference in initial stress conditions,and the other is the change in characteristics and friction coef f i cient at the contact surface,which changes the friction sliding properties.The following conclusions are drawn based on the different types of friction sliding observed:
(1)Relationship between axial stress and friction sliding
Friction tests were conducted on rocks of different lithologies under various axial stress levels on specimens.Due to the limit number of specimens,each group of specimens with the same lithology was only tested under four stress levels.Comparisons of axial stress applied and compression strength of specimens show that for a given lithology,the applied axial stress is usually 20% more than the compression strength for the specimens at failure. Thus,stable sliding was more likely to occur.Comparisons of axial stress and uniaxial compression strength between coal specimens show that the above-mentioned regularity can be also true for sandstone-coal structures.
(2)Relationship between loading rate and friction sliding
The loading rate is an important index in the mechanical test, as well as in the friction test.In the friction sliding test,continuous sliding was likely to occur when the loading rate was low,and the possibility of unstable sliding occurring at small magnitude was greater when the loading rate was high.Taking the red granite as an example,when the axial load was 20 MPa,many unstable continuous sliding events were recorded when the shear loading rate was 0.25 mm/min.Under these conditions,however,only two unstable discontinuous sliding events were recorded when the shear loading rate was 1 mm/min.

Fig.7.Evolution of displacement f i eld during unstable sliding(unit:mm).

Table 2Sliding characteristics of the specimens.

Fig.8.Friction strength of specimens.
Double-shear friction tests were conducted on coal-rock specimens,and the instability of the coal-rock structure was studied.The digital speckle correlation method was employed to analyze the displacement characteristics of the sliding process.It can be noted that unstable continuous sliding,unstable discontinuous sliding, and stable sliding were observed in the tests.The regularity of the friction strength of the sandstone-coal specimen was studied in terms of statistical data.
The following conclusions were drawn based on the results.The sliding type observed in the sample was correlated with the axial load and loading rate.The smaller the axial load and the loading rate are,the more likely the unstable sliding occurs.No signi f i cant displacement occurred before unstable sliding.After unstable sliding,the sliding displacement increased rapidly.The shear stress peak was positively correlated with the axial load after sliding,but the displacement observed under unstable sliding was not correlated with the axial load.Some intense AE events occurred before the sliding,and the AE events were reduced after stable sliding.
We wish to con fi rm that there are no known con fl icts of interest associated with this publication and there has been no signi fi cant fi nancialsupportfor this work thatcouldhave in fl uencedits outcome.
This work is f i nancially supported by the Major State Basic Research Development Program Fund(2010CB226801),the China Postdoctoral Science Foundation(2013M530770),the State Key Laboratory of Coal Resources and Safe Mining Open Research Fund (SKLCRSM11KFB08).
Arasteh MM,Chudnovsky A,Dudley II JW,Glaser S,Ma J.Broadband acoustic emission observations during fracture propagation in rock-like material. International Journal of Rock Mechanics and Mining Sciences 1997;34(3-4): 83.e1-83.e13.
Brace WF,Byerlee JD.Stick-slip as a mechanism for earthquakes.Science 1966;153(3739):990-2.
Byerlee JD,Brace WF.Stick slip,stable sliding,and earthquakes-effect of rock type, pressure,strain rate,and stiffness.Journal of Geophysical Research 1968;73(18): 6031-7.
Byerlee JD.The mechanics of stick-slip.Tectonophysics 1970;9(5):475-86.
Builo SI.Acoustic-emission testing and diagnostic of hazardous dynamic processes in a coal seam.Russian Journal of Nondestructive Testing 2000;36(4):282-9.
Eberhardt E,Stead D,Stimpson B.Quantifying progressive pre-peak brittle fracture damage in rock during uniaxial compression.International Journal of Rock Mechanics and Mining Sciences 1999;36(3):361-80.
Farmer IW.Bursts and unstable.In:Coal Mine Structures.Springer;1985. pp.259-87.
Fukui K,Okubo S,Terashima T.Electromagnetic radiation from rock during uniaxial compression testing:the effects of rock characteristics and test conditions.Rock Mechanics and Rock Engineering 2005;38(5):411-23.
Jiang YD,Zhao YX,Liu WG,Zhu J.Investigation on the mechanism of coal bumps and relating experiments.Beijing:Science Press;2009.pp.81-104(in Chinese).
Jiang YD,Wang HW,Xue S,Zhao YX,Zhu J,Pang XF.Assessment and mitigation of coal bump risk during extraction of an island longwall panel.International Journal of Coal Geology 2012;95(1):20-33.
Lacidogna G,Manuello A,Carpinteri A,Niccolini G,Agosto A,Durin G.Acoustic and electromagnetic emissions in rocks under compression.In:Experimental Mechanics on Emerging Energy Systems and Materials,vol.5;2011.pp.57-64. Indianapolis,USA.
Lockner D.The role of acoustic emission in the study of rock fracture.International Journal of Rock Mechanics and Mining Sciences and Geomechanics Abstracts 1993;30(7):883-99.
Ma J,Liu LQ,Liu PX,Ma SL.Thermal precursory pattern of fault unstable sliding:an experimental study of en echelon faults.Chinese Journal of Geophysics 2007;50(4):1141-9(in Chinese).
Ma YK,Wang EY,Xiao D,Li ZH,Liu J,Gan LJ.Acoustic emission generated during the gas sorption-desorption process in coal.International Journal of Mining Science and Technology 2012;22(3):391-7.
Mansurov AV.Acoustic emission from failure rock behaviour.Rock Mechanics and Rock Engineering 1994;27(3):173-82.
Ohnaka M,Mogi K.Frequency characteristics of acoustic emission in rocks under uniaxial compression and its relation to the fracturing process to failure.Journal of Geophysical Research:Solid Earth(1978-2012)1982;87(B5):3873-84.
Passelègue FX,Schubnel A,Nielsen S,Bhat HS,Madariaga R.From sub-Rayleigh to supershear ruptures during stick-slip experiments on crustal rocks.Science 2013;340(6137):1208-11.
Price NJ,Cosgrove JW.Analysis of geological structures.Cambridge University Press; 1990.
Ruina A.Slip unstable and state variable friction laws.Journal of Geophysical Research 1983;88(B12):10359-70.
Shkuratnik VL,Filimonov YL,Kuchurin SV.Experimental investigations into acoustic emission in coal samples under uniaxial loading.Journal of Mining Science 2004;40(5):458-64.
Song YM,Ma XP,Yang XB,Guo WJ,Shi HJ.Experimental study on the dynamic displacement evolution of fault in stick-slip process.Chinese Journal of Geophysics 2012a;55(1):171-9(in Chinese).
Song DZ,Wang EY,Liu J.Relationship between EMR and dissipated energy of coal rock mass during cyclic loading process.Safety Science 2012b;50(4):751-60.
Xie H,Pariseau WG.Fractal character and mechanism of rock bursts.International Journal of Rock Mechanics and Mining Sciences and Geomechanics Abstracts 1993;30(4):343-50.
Yoshikawa S,Mogi K.Experimental studies on the effect of stress history on acoustic emission activity-a possibility for estimation of rock stress.Journal of Acoustic Emission 1989;8(4):113-23.
Zhao YX,Jiang YD.Acoustic emission and thermal infrared precursors associated with bump-prone coal failure.International Journal of Coal Geology 2010;83(1): 11-20.
Zhang CQ,Feng XT,Zhou H,Qiu SL,Wu WP.Case histories of four extremely intense rockbursts in deep tunnels.Rock Mechanics and Rock Engineering 2012;45(3): 275-88.
Zhao YX,Zhao GF,Jiang YD.Experimental and numerical modelling investigation on fracturing in coal under impact loads.International Journal of Fracture 2013;183(1):63-80.

Dr.Tao Wang,is a lecturer of University of Science and Technology Beijing,and a postdoctoral fellow in China University of Mining and Technology(Beijing).He is now PI of two projects,which are funded by China Postdoctoral Science Foundation Grant and State Key Laboratory for Coal Resources and Safe Mining.Dr.Wang has published more than f i fteen journal papers.
*Corresponding author.Tel.:+86 10 62339851.
E-mail address:tao.w@139.com(T.Wang).
Peer review under responsibility of Institute of Rock and Soil Mechanics,Chinese Academy of Sciences.
Acoustic emission(AE)
Coal bump
Activated faults
Journal of Rock Mechanics and Geotechnical Engineering2014年3期