Applied Energy 131 (2014) 67–78

Contents lists available at ScienceDirect

Applied Energy

journal homepage: www.elsevier .com/locate /apenergy

A sequential approach to control gas for the extraction of multi-gassycoal seams from traditional gas well drainage to mining-induced stressrelief

http://dx.doi.org/10.1016/j.apenergy.2014.06.0150306-2619/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: National Engineering Research Center for Coal & GasControl, China University of Mining & Technology, Xuzhou 221008, China. Tel.: +86516 83885948; fax: +86 516 83995097.

E-mail address: chengyuanpingcumt@gmail.com (Y. Cheng).

Shengli Kong a,b, Yuanping Cheng a,b,⇑, Ting Ren c, Hongyong Liu a,b

a National Engineering Research Center for Coal & Gas Control, China University of Mining & Technology, Xuzhou 221008, Chinab Faculty of Safety Engineering, China University of Mining & Technology, Xuzhou, Jiangsu 221116, Chinac School of Civil, Mining & Environmental Engineering, University of Wollongong, Wollongong, NSW 2522, Australia

h i g h l i g h t s

� The gas reservoirs characteristics are measured and analyzed.� A sequential approach to control gas of multi-gassy coal seams is proposed.� The design of gas drainage wells has been improved.� The utilization ways of different concentrations of gas production are shown.

a r t i c l e i n f o

Article history:Received 26 November 2013Received in revised form 19 May 2014Accepted 13 June 2014

Keywords:Coalbed methaneGassy multi-seamOutburst riskGas control

a b s t r a c t

As coal resources become exhausted in shallow mines, mining operations will inevitably progress fromshallow depth to deep and gassy seams due to increased demands for more coal products. However, dur-ing the extraction process of deeper and gassier coal seams, new challenges to current gas control meth-ods have emerged, these include the conflict between the coal mine safety and the economic benefits, thedifficulties in reservoirs improvement, as well as the imbalance between pre-gas drainage, roadwaydevelopment and coal mining. To solve these problems, a sequential approach is introduced in this paper.Three fundamental principles are proposed: the mining-induced stress relief effect of the first-minedcoalbed should be sufficient to improve the permeability of the others; the coal resource of the first-mined seams must be abundant to guarantee the economic benefits; the arrangement of the verticalwells must fit the underground mining panel. Tunlan coal mine is taken as a typical example to demon-strate the effectiveness of this approach. The approach of integrating surface coalbed methane (CBM)exploitation with underground gas control technologies brings three major benefits: the improvementof underground coal mining safety, the implementation of CBM extraction, and the reduction of green-house gas emissions. This practice could be used as a valuable example for other coal mines having sim-ilar geological conditions.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The strong dependence of Chinese economic development onenergy has led to the increasing use of coal products [1,2], whichaccounts for 70% of the nation’s total energy supply [3]. Industrialcoal consumption from 2008 to 2035 is expected to grow by 67%[4]. As a result, the mining depth of coalbeds has increased annu-

ally by 10–50 m from shallow to deep deposits. The recoverablereserves of coal resources, which is deeper than 1000 m, accountfor 2.95 trillion tons, or nearly 53% of the total reserves [5]. Basedon the demands for coal and the abundance of deep resources, theshift of mining depths from shallow to deep will be inevitable.

With the increase of the mining depth, more and more lowgassy mines will be replaced by high gassy outburst-prone mines.The protective seam mining and underground gas pre-drainage arethe primary measures to reduce the gas content and to control theoutburst [6]. Protective seam measure, through mining the lowgassy coalbed (called protective seam) for releasing the gas in adja-cent high gassy coalbeds (called protected seam) to ensure the

68 S. Kong et al. / Applied Energy 131 (2014) 67–78

safety in mining process, has been widely used in the coal fields ofHuainan, Huaibei, Yangquan, Shenyang and other regions [7]. Theapplication of these measures in different geological conditions –short-interburden protective seam [8], distant-interburden protec-tive coal seams [9], and extra-thin protective seam [10] – havebeen studied by many scholars. The regional pre-drainagemeasure, i.e., degassing the in situ coalbed through undergroundboreholes [11], has been used in many coal fields without theuse of the protective seam mining measure.

However, for gas control in a deep and gassy multi-coalbedlocation, a new challenge exists not only in the protective seammining but also in the regional gas pre-drainage. For protectiveseam mining, it is difficult to find the appropriate coalbed to befirst-mined as the protective seam, because almost all the coalbedshave the risk of an outburst, while the less risky coalbeds are eithertoo thin or high ash content with no economic value to mine. Forgas pre-drainage, it requires adequate roadways in strata and bore-holes to ensure the effectiveness of drainage. The maintenancecosts associated with roadways in deep strata are unavoidable. Ina multiple-coalbed formations, a large volume of pressure-reliefgas from adjacent coalbeds will migrate into active mining panelsas a result of mining excavations. This situation causes the gasemissions of coal mines to increase dramatically to hundreds ofcubic meters per minute. The output of coal production is conse-quently limited by the prevailing ventilation system, due to theexcessive high gas emission.

Known as a hazard to mining safety and a powerful greenhousegas, coalbed methane is also a form of clean and efficient energy[12]. The CBM industry are well-developed in America, Canada,Australia, Poland [13–17] and also in China [18,19]. Surface verticalwells have been introduced to the field of gas control in under-ground coal mines. The functions of vertical wells can be classifiedinto three major categories: firstly, to conduct hydro-fracturingand improve coalbed gas reservoirs for CBM recovery from surface;secondly, to drain the pressure relief CBM resulting from miningactivities [20]; thirdly, to perform gob gas drainage after the coal-bed was mined [21,22]. But in most cases, the CBM projects whichhave been run as a standalone system are not related to the miningprocess of the coal mine.

Aiming at the problems mentioned above, a concept of gas con-trol was proposed. In general, it can be divided into three stages inchronological order: stage one, the first-mined coalbed reservoir isselected and improved by surface vertical wells and hydro-fractur-ing to increase the reservoir permeability and eliminate the out-burst risk. Stage two, the gas content will get lower by theenhanced underground gas drainage to ensure the efficient andsafe mining of the first-mined coalbed. Stage three, during the min-ing process of the first-mined coalbed the permeability of otheradjacent coalbeds will enhanced by the pressure-relief effect.Through the pressure-relief gas drainage by the surface verticalwells and underground boreholes, the outburst risk of all the coal-beds will be eliminated. By integrating the present coal mine gascontrol method with surface CBM extraction technologies, weimprove the selection of the first-mined coalbed standard andmaximize the usage of vertical well to achieve high-efficiency co-exploitation of coal and methane. The concept is implemented inpractice in the Tunlan coal mine. This case study sets a good exam-ple for other coal mines with similar geological conditions.

2. Reservoir characteristics in Tunlan

The selection of Tunlan coal mine for a case study was signifi-cant because of following typical reasons: Firstly, it has character-istics of multi-coalbed formations and is generally representativeof most of the coalfields in China; Secondly, as more coal mines

are aging and have to explore coalbeds in greater depth, gas prob-lems have become a bottleneck for resource exploitation. From2004 to 2009, Tunlan mine exhibited a record of zero fatality inthe metric of DRPMT (death rate per million tons), however, on22nd February 2009, an extremely serious gas explosion occurred,causing 78 deaths and 114 injuries due to insufficient gas drainage[23]. As a key state-owned coal mine, Tunlan mine has receivedwidespread attention and played an important role in gas controland mining safety.

2.1. Geology

As shown in Fig. 1, Tunlan coal mine is located in the middlepart of the Xishan coalfield, which is one of the six major coalfieldsin Shanxi province at the northern rim of the Qingshui basin. Themine started operation in 2002 with a designed annual capacityof 4 million tons. The main coal bearing strata are the 100 m thickTaiyuan group in the upper series of the Permian and the 60 mthick Shanxi group in the lower series of the Carboniferous. Themain minable coalbeds for economic production are the No. 2,No. 4, No. 8 and No. 9 seam, which have an average dip of 7� andare to be extracted by longwall mining method. The permeabilityof the main minable coalbeds is less than 0.01 md, which is unfa-vorable for gas drainage in virgin coalbeds. The sequence of thecoalbeds is illustrated in Fig. 2.

2.2. Characteristics of gas reservoirs

2.2.1. Coal samples and preparationThe characteristics of the gas reservoirs depend on various

deposition environments [24]. The variations of the overburdenthickness over each coalbed may influence the sealing effect duringmethane migration in the hydrocarbon generation period. In theYanshanian, the temperature field of the coal-bearing strata waschanged by the magma intrusion, which resulted in different vol-umes of thermogenic gas [25,26]. To obtain better understandingof the reservoir characteristics, a series of tests were performed.One coal sample of each coalbed was taken from the tunneling ormining workface (Table 1). The samples were sealed in the packagefrom the site and prepared for the proximate analysis, petrographicanalysis, pore size distribution tests and methane adsorption tests.

2.2.2. Method and testsBy using the Automatic Proximate Analyzer 5E-6600, the prox-

imate analysis was performed according to the ISO 17246:2010[27] test method with particle sizes of 0.074–0.2 mm. The maceralgroup composition was determined in accordance with the ISO7403-3:2009 [28]. The pore size distribution was determined byfollowing the ISO 15901-1:2005 [29], using the mercury intrusionmethod with the AUTOPORE IV 9500 porosimeter and the adsorp-tion of CO2 method with a gas sorption instrument AUTOSORB-1.Following the GB/T19560-2008 [30] test method, the methaneadsorption isotherm was tested at 303 K under standard atmo-spheric pressure. The samples were all air dried ash-free basis withparticle sizes of 0.2�0.25 mm.

2.2.3. Results and discussionsThe proximate and petrographic analysis results, as shown in

Table 2, indicates that the four samples have similar content ofmoist, ash and volatile matter. All samples are low-moisture andwith moderately ash content. The percentage of volatile and fixedcarbon is approximately 20% and 60%, respectively.

Limited by the intrusion pressure, the mercury porosimeter canonly be used to measure the distribution of pores larger than 3 nm.The total pore volume, total pore area and porosity of samples aregiven in Table 3.

Fig. 1. Location of Tunlan coal mine.

Fig. 2. General stratigraphy of coal measures at Tunlan coal mine.

Table 1Coal samples locations.

Sample number Coalbed Site Depth (m)

TL-2 No. 2 coalbed 12,407 Workface 360TL-4 No. 4 coalbed 12,401 Drainage roadway 340TL-8 No. 8 coalbed 28,107 Transport roadway 290TL-9 No. 9 coalbed 18,203 Drainage roadway 400

S. Kong et al. / Applied Energy 131 (2014) 67–78 69

The Dubinin–Radushkevich (D–R) [31] method was establishedfor describing the adsorption acts of micropores in coal. Using theD–R method, provided by the AUTOSORB-1 program, we calculated

the micropore volume and the surface area (Table 3). The microp-ores, especially those pores smaller than 2 nm [32], are the maincontributors to the capacity of adsorption. For all the samples,the average micropore width ranged from 1.00 to 1.45 nm.

The descending order of Langmuir volume (VL) of samples isTL-8, TL-9, TL-2 and TL-4, from 24.75 to 19.22 mL/g (Fig. 3). TheVL represents the adsorption capacity of coal. The sequence in themicropore volume and the surface area of the samples indicate asimilar trend. The same trend was observed by An, who tested11 coal samples from the Huaibei coalfield [33]. From the results,the samples obtained from the Taiyuan group (TL-8, TL-9) exhibitlarger VL than the samples from the Shanxi group (TL-2, TL-4),which indicates that the adsorption capacity in lower coalbeds ishigher than the overlying seams.

2.3. Coalbed methane content and outburst risk analysis

In accordance with the China’s national standard [34], threeindicators are given for the judgment of the outburst risk in a coal-bed: (i) the actual dynamic phenomenon in mining; (ii) the quan-tity of gas emission per ton of casted coal; (iii) the indices ofoutburst risk such as gas pressure and gas content. If any of theseconditions is satisfied, the outburst risk of a coalbed is confirmed.

The two main indicators of outburst risk are tested in the mainminable coalbeds at Tunlan coal mine. Gas data exceeding the crit-ical value [34] is shown in Fig. 4. According to the measured data,the variations of the gas pressure with the mining depth are pre-dicted as shown in Fig. 5, by using the safety line method proposedby Wang [35]. The gradient of No. 2 coalbed (in upper group) is0.43 MPa/hm, much lower than the No. 8 coalbed (0.81 MPa/hm).

Table 2The results of the proximate analysis and petrographic analysis.

Sample Moist (wt.%, ad) Ash (wt.%, ad) Volatile (wt.%, ad) Fixed carbon (wt.%, ad) Vitrinite (%) Inertinite (%) Clay (%) Pyrite (%)

TL-2 0.83 18.23 19.85 64.87 92.05 1.50 5.80 0.65TL-4 0.76 23.17 22.81 58.72 93.50 1.45 4.65 0.40TL-8 1.09 20.27 21.83 61.47 87.40 1.65 3.35 7.60TL-9 1.18 24.03 18.84 60.70 88.65 2.30 7.60 1.45

Table 3Pore characteristics of samples.

Sample Total intrusionvolume (mL/g)

Total pore area(m2/g)

Porosity (%) Microporevolume (mL/g)

Micropore surfacearea (m2/g)

Average microporewidth (nm)

TL-2 0.0334 16.997 4.10 0.0014 4.06 1.09TL-4 0.0246 12.637 3.22 0.0007 1.93 1.00TL-8 0.0392 17.729 4.96 0.0019 5.53 1.45TL-9 0.0266 14.071 3.51 0.0019 5.43 1.15

Fig. 3. Methane adsorption isotherm of coal samples at STP.

Fig. 4. Gas pressure and gas content measured in the main minable coalbeds.

70 S. Kong et al. / Applied Energy 131 (2014) 67–78

With the results of proximate analysis and the methane adsorptionisotherm, the gas contents of these coalbeds were calculated.When the depth is over 400 m, both the gas pressure and the gascontent will exceed the critical value, i.e., the risk of outburst willbe very high in all coalbeds. Since the commencement of coal pro-duction, four small outburst phenomena have been observed in theSouth 5th mining panel (Table 4).

3. Reservoir improvement and integrated surface-undergroundgas drainage method in high gassy coalbeds

On the basis of analyses, the different gradients of gas pressureand contents and the outburst risk are confirmed. However, thelow permeability and high gas content in all minable coalbedsare the hindrance to the traditional gas drainage measures. Forthe CBM development, the different pressure systems have nega-tive effect on exploration of the entire coalbed group [36,37].Therefore, we designed an integrated way to control the gas and

Fig. 5. Variations of gas pressure and content in the No. 2 and No. 8 coalbeds.

Table 4Details of the observed dynamic activities.

Time Location Sea level (m) Overburdendepth (m)

Quantity ofcoal (t)

Quantity of gasemission (m3)

Quantity of gasemission per ton (m3/t)

February 4, 2008 12,503 Return airway@840 m +720 350 0.89 205 230February 8, 2008 12,503 Return airway@848 m +720 350 0.41 82 200February 21, 2008 12,503 Drawing roadway @ 842 m +720 350 3.41 117 34June 19, 2008 12,503 Return airway@1090 m +610 360 0.57 22 39

Fig. 6. Technological steps.

S. Kong et al. / Applied Energy 131 (2014) 67–78 71

exploit the coalbed methane resource. The technological steps areshown in Fig. 6.

3.1. Selection of the first-mined coalbed

3.1.1. Model developmentThe selection of a reasonable first-mined coalbed is the first step

and is thus the foundation of the entire approach. There are threefundamental principles of this approach. First, the mining-inducedstress relief effect caused by the mining process of the first-minedcoalbed must be sufficient to improve the permeability of theupper and lower adjacent reservoirs. That will activate the gasdrainage in them. Second, the coal resource of the first-mined coal-bed must be abundant to guarantee the economic benefits of themining. Third, the arrangement of the vertical wells must takethe layout of underground longwall panels into consideration.

In order to investigate the stress relief effect of each coalbed,FLAC3D [38,39] was used for calculating the stress and displace-ment distribution in different cases.

Based on the geological condition at Tunlan coal mine, themodel was developed to include all the major coalbeds in its coalmeasures formations. The model covers an area of500 m � 500 m and 202 m in height. To simulate the load of the450 m overburden, a compressive stress of 10 MPa was imposedon the top of the model. Fig. 7 shows the 3D model geometryand boundary conditions. The Mohr–Coulomb failure criterionwas selected for the simulation. The initial properties of the rockmass are listed in Table 5 according to the typical values of theTunlan coal mine.

3.1.2. Case studiesThere are five minable coalbeds in the Tunlan mine: No. 2, No. 4,

No. 7, No. 8 and No. 9. However, the vertical distance of the No. 2

and No. 4 coalbeds is too close, with a interburden of even less than10 m in some areas. The same situation also occurs between theNo. 8 and No. 9 coalbeds. The height of the caved zone is generally3–6 times the thickness of the mined coalbed [40]. If the No. 4 orNo. 9 coalbeds were selected as the first-mined coalbed, the No.2 and No. 8 coalbeds would be in the caved zone. Thus, in this

Fig. 7. Geometry and boundaries of the numerical simulation model.

Table 5Rock mass initial properties.

Lithology Density (kg m3) Bulk modulus (GPa) Shear modulus (GPa) Friction angle (�) Cohesion (MPa) Tensile strength (MPa)

Coal 1450 1.5 1.2 18 1.0 1.0Mudstone 2250 2.8 2.1 25 1.5 1.5Sandstone 2500 2.1 1.5 22 1.1 1.1Fine-grained sandstone 2600 3.4 2.5 30 2.2 2.2Sandy mudstone 2800 3.9 3.5 33 2.5 2.5Limestone 2800 8.5 5.5 43 4.5 4.5

Table 6Cases of the simulations.

Case First-mined coalbed Range of excavation X:Y:Z (m) Coalbed Relative position Distance to the first-mined coalbed (m)

Case A No. 2 180:100:3.3 No. 4 Lower pressure relief seam 9No. 8 Lower pressure relief seam 77No. 9 Lower pressure relief seam 92

Case B No. 7 180:100:0.8 No. 2 Upper pressure relief seam 9No. 4 Upper pressure relief seam 60No. 8 Lower pressure relief seam 17No. 9 Lower pressure relief seam 32

Case C No. 8 180:100:3.4 No. 2 Upper pressure relief seam 77No. 4 Upper pressure relief seam 68No. 9 Upper pressure relief seam 15

72 S. Kong et al. / Applied Energy 131 (2014) 67–78

study, three Cases were selected for the numerical simulation:Case A: the No. 2 coalbed was chosen as the first-mined coalbed;Case B: the No. 7 coalbed was chosen as the first-mined coalbed,and finally the No. 8 coalbed was chosen as the first-mined coalbedfor Case C (Table 6).

3.1.3. Results and discussionsThe stress field will redistribute when the first-mined coalbed is

excavated and the change of the vertical stress would indicate theextent of de-stress effect. Fig. 8 showed the redistributions of thevertical stress from the numerical simulations for the above threecases.

In the vertical projection of the first-mined coalbed’s excavatedarea, its adjacent coalbeds are de-stressed to varying magnitude.The coalbed further away from the first-mined coalbed tendsto exhibit less de-stress effect. The vertical stress increasesslightly under the effect of the abutment stress, which extends

approximately 5–40 m beyond the vertical projection both in x-and y-direction. For example, in Case A, the peak vertical stressin the No. 4 coalbed, which is only 9 m away from the first-minedcoalbed, increases to 18.1 MPa, almost 1.43 times higher than itsoriginal value. In comparison, in the No. 8 and No. 9 coalbeds,the vertical stresses only exhibit an increase of 2% and 0.7%,respectively.

To determine whether a significant outburst risk exists, thestress rule was applied. The critical stress relief value can be calcu-lated by the following equation [41,42]:

jrzcj 6 ðcos2 aþ k sin2 aÞcH ð1Þ

where rzc is the stress perpendicular to the coalbed, a is the coalseam dip, k is the lateral pressure coefficient, c is the bulk densityand H is the overburden depth of the initial outburst site. Accordingto the conditions of the Tunlan coal mine, a is 0�, c is 25,000 kg/

Fig. 8. Balanced redistribution of the vertical stress after the first-mined coalbed was excavated in each case.

Fig. 9. The re-established vertical stresses of each coalbed in different cases.

S. Kong et al. / Applied Energy 131 (2014) 67–78 73

m2 s2 and H is about 360 m (Table 4), which produces a verticalstress rzc of 9 MPa.

As illustrated in Fig. 9, the performances on stress relief of coal-beds are various in different cases. In Case A the re-established ver-tical stresses in No. 8 and No. 9 coalbed is still maintained at a highlevel. That is attributes to the distance of the coalbeds. Generally,the vertical scope of effective pressure relief cause by the uppercoalbed is no more than 60 m [7]. In Case B and Case C, the re-established vertical stresses in other minable coalbeds are lowerthan the critical value, which indicates that both of the No. 7 andNo. 8 could be chosen as the first-mined coalbed.

If only the stress reduction performance was taken as the prin-ciples for selecting the first-mined coalbed, clearly, the No. 7 coal-bed, as a traditional ‘‘protective seam’’, would be the best choice.However, the economic benefits cannot be neglected in the engi-neering design. The thickness of the No. 7 coalbed is only 0.8 m,so another 0.4 m of surrounding rock is required to be excavatedto match the minimum mechanized mining height in the Tunlancoal mine. The ash of raw coal excavated will account for over40% and the calorific value will be less than 4200 kcal/kg. Theunprofitable feature of the first-mined coalbed will reduce the out-put of the coal mine, thus impacting the development of the coalmine.

With all these factors taken into consideration, it is proposed todiscard the traditional descending mining sequence method andestablish a new selecting standard for the protective seam (alsocalled the first-mined coalbed). In this study, the No. 8 coalbedwas selected as the first-mined coalbed.

3.2. Integrated surface-underground gas drainage method

As shown in Fig. 6, there are three core technologies in thismethod: gas reservoir improvement and gas drainage in the first-mined coalbed by surface vertical wells, enhanced undergroundgas drainage, and reformation and integration of surface andunderground gas drainage in adjacent coalbeds.

3.2.1. Reservoir improvement and gas drainage in the first-minedcoalbed by surface vertical wells

As illustrated in Fig. 5, the gas content of No. 8 coalbed is about10–14 m3/t in deep area, but the permeability is less than 0.01 md.To maintain high gas flow rate in the vertical drainage wells and toimprove underground borehole gas drainage at a late stage, hydro-fracturing is needed to improve the permeability of the first-minedcoalbed.

The arrangement of vertical wells, one of the most importantfactors in CBM exploitation, directly determines the results of frac-turing and gas drainage performance. In traditional CBM program,it is often determined by the geological condition, the surface land-scape and logistic condition [36]. In this study, in addition to allthese conditions, the vertical wells must be arranged with due con-sideration of the layout of underground longwall panels as areference.

Detected by the microseismograph, the fracture could extend toover 150 m. The effective radius of vertical well extraction reached150–200 m, but the width of the longwall panel was 200–300 m.To achieve more effective drainage, the vertical wells can bedesigned to interconnect with the adjacent longwalls (Fig. 10).

At this stage, the emphasis is on the improvement and transfor-mation of the gas reservoir as well as the pre-drainage of thefirst-mined coalbed. The goal is to make the gas content of thefirst-mined coalbed down to 8 m3/t and eliminate the outburstrisk, thereby provide a favorable condition for underground gasdrainage at later stages.

Fig. 10. A schematic view of the arrangement of surface vertical wells.

Fig. 11. Schematic view of the underground intensified gas drainage.

74 S. Kong et al. / Applied Energy 131 (2014) 67–78

3.2.2. Enhanced underground gas drainage in first-mined coalbedAfter the stage one, the outburst risk of first-mined coalbed

could be diminished. However, due to the great drainage radiusand the uneven distribution of the coalbed, outburst risk still existsin some areas. In order to keep lowing the gas content and avoidany possible blind areas, where the gas is not extracted sufficiently,enhanced underground gas drainage is necessary to achieve effi-cient driving and mining.

With the development of directional drilling equipment andtechnologies, the underground directional longhole needs to beemployed to meet the demands of intensified gas drainage [43–45]. To pre-drain seam gas ahead of roadway drivage, the longholeshould cover the area of at least 15 m beyond the ribs of the road-way and 60 m ahead of the drivage direction. To ensure efficiencyand reduce drainage lead time, the longhole spacing should be keptwithin the range of 3–5 m. As the roadway is being developed, the

Fig. 12. Schematic view of the surface-underground combined pressure relief gas drainage.

S. Kong et al. / Applied Energy 131 (2014) 67–78 75

longholes are drilled into the longwall panel. Due to the pre-drain-age by the vertical wells, these holes are drilled at intervals of 8–10 m (Fig. 11).

The extraction by underground longholes could be more effi-cient to the vertical wells, because of the lower negative pressureof drainage and the larger contact area in the coalbeds. After a per-iod of enhanced underground gas drainage, we expect that the gascontent will be lower than 5 m3/t to ensure the high yield inworkface.

3.2.3. Reformation and integration of surface and underground gasdrainage in adjacent coalbeds

When the first-mined coalbed is being mined, the permeabilityof other coalbeds is enhanced by the stress relief effect. The des-orbed gas from the roof and floor coalbeds will migrate into thefirst-mined coalbed’s workface and will be captured by the verticalwells and the longholes (Fig. 12).

According to its source, the desorbed gas can be divided intothree parts: a. from the upper No. 2 and No. 4 coalbeds; b. accumu-lated in the gob of the first-mined coalbed; c. from the floor coal-beds. The first two parts are mostly drained by the vertical wells,and the last part is controlled by the longholes drilled in the No.

Table 7Gas drainage volumes by surface vertical wells.

Site code Number of wells Drainage start date Accumulative volume (m3) S

T1 6 2011/6/15 2,287,040 TT6 16 2011/6/15 4,485,254 TT86 9 2011/9/27 3,136,312 TT37 7 2012/9/10 903,795 TT44 3 2012/9/10 155,425 TT20 5 2012/9/10 489,998 TT49 7 2012/9/10 487,191 TT54 4 2012/9/12 196,354 TT66 5 2012/9/10 195,132 T

9 coalbed. The surface-underground combined pressure relief gasdrainage controls the gas emission in the first-mined coalbed min-ing workface and degasses other coalbeds to remove the potentialof outburst risk for ensuring mining safety.

4. Field application

Faced with the problem of extraction in deep high-gassy coal-beds, the traditional gas drainage measures are incorporated effec-tively in this concept. But it’s needs several years to realize thewhole project. In order to evaluate the effect in short-term, indus-trial tests of core technologies were carried out.

4.1. Reservoir improvement and gas drainage by surface vertical wells

Since June 2011, surface vertical wells have been drilled forexperiment in the deep mining panel at Tunlan coal mine. By theend of 2012, a total of 112 wells were implemented to reformationthe deep coalbed reservoirs.

According to the locations of the surface wells, 18 gas drainageplants were built. The volume of drainage gas flows were moni-tored continuously by real time monitors. The accumulative

ite code Number of wells Drainage start date Accumulative volume (m3)

68 5 2012/9/12 195,48876 6 2012/9/10 507,987B03 7 2012/10/28 96,037165 7 2012/10/28 103,170173 4 2012/10/28 56,704175 3 2012/10/28 60,874176 2 2012/10/28 25,468188 9 2012/10/28 137,817080 7 2012/10/28 59,969

Fig. 13. Gas flow rate and concentration from drainage boreholes in 18,207workface (the No. 8 coalbed).

Fig. 14. Average gas flow rate and concentration of the pressure relief gas drainagein the adjacent coalbed (the No. 9 coalbed).

76 S. Kong et al. / Applied Energy 131 (2014) 67–78

volumes of surface drainage gas since the date of commissioningare listed in Table 7. The average pure methane flow rate per wellcould reach 619 m3/d, and the methane concentration remains rel-atively stable, at a level of over 80%.

Taking South 5th mining panel for instance, the depth of No. 8coalbed is in the range of 450–600 m. According to the predictionin Fig. 5, the average gas content could reach 12 m3/t. In this panel,the recoverable reserves of coal is approximately 15.0 million tons.For the 12,501 and 12,503 workfaces have been completed, theresidual reserves is about 12.6 million tons. In order to low thegas content to 8 m3/t, 4.41 million cubic meters gas need to bedrained by the vertical well. It needs 8.2 years to achieve it.

4.2. Enhanced gas drainage in the first-mined coalbed

For the long period in the first stage, we take the experiment ofenhanced gas drainage in a relatively shallow mining panel. Takingthe 18,207 longwall workface in the No. 8 coalbed for example, 405longholes were drilled from the drainage roadway. These long-holes, perpendicular to the roadside, are 113 mm in diameterand 180 m in length, with a total length of 72,900 m. The flow rateand gas concentration are shown in Fig. 13.

In the first four months, the gas flow rate was approximately5 m3/min. As the drainage continued, the flow rate began to fall.

To maintain the efficiency, the drainage suction pressure wasadjusted from 30 kPa up to 50 kPa. With the adjustment, the flowrate increased to 7 m3/min. During the entire drainage period, theaverage pure gas flow rate was about 3.39 m3/min, and the gasconcentration was basically above 30%.

The gas content of 18,207 workface ranged from 8 to 11 m3/twith a median of 10 m3/t and the resource reserve was estimatedat 0.96 million tons. After extraction of 29 months, 4.3 millioncubic meters of gas had been drained. That means the gas contentreduced approximately to 5.5 m3/t. During the mining process, themaximum of gas concentration in airflow reached only 0.47% withan average of 0.34%. The highly gas extraction rate, 54% on average,effectively guaranteed the mining yield and the security of theworkface.

4.3. Underground gas drainage in the adjacent coalbeds

The experiment of pressure relief gas drainage was conductedduring the mining of the 18,207 longwall panel. A test roadwayin the No. 9 coalbed was developed before the start of longwallextraction under the 18,207 mining area. A total of 823 longholeswith a diameter of 113 mm were drilled into both sides of theroadway at an interval of 3 m. The entire vertical projection ofthe 18,207 excavated area in the No. 9 coalbed was covered by76,813 m of longholes.

The pressure relief boreholes started to drain gas when the18,207 longwall was being mined. The mining activities in theNo. 8 coalbed improved the permeability of the No. 9 coalbedand consequently increased the gas flow rate to 5.79 m3/min onaverage (Fig. 14), which is almost twice the rate of enhanced gasdrainage in the No. 8 coalbed. From the experiment, the pressurerelief gas drainage could decrease the gas content of the No. 9 coal-bed and eliminate the outburst risk; in this manner, the problem ofmassive gas emission in the active longwall face is solved.

4.4. Technology integration

The core technologies involved in this study were indepen-dently tested in the field and proved to be effective. Currently,the mining activities have not yet fully extended to the deep area,therefore the total integration of the technologies has not beencompletely implemented. Recently, reservoir reconstruction andgas drainage by surface vertical wells have been conducted in thedeep mining panels at Tunlan Mine. The complete integration oftechnologies is scheduled to be completed in the next 8–10 years.

4.5. The utilization of gas production

The integrated gas drainage method has been applied success-fully in Tunlan coalmine. The gas extraction volume has reached60 million cubic meters per year. The 3.7 million tons of coal’s out-burst risk is eliminated, safety reserves outstrip the designed pro-duction. The gas extraction rate of the mine has risen to 56.6%. Thenumber of CH4 overrunning accidents fell to 1 last year from 59 in2010. The safety situation has been improved dramatically andguarantees the mine to obtain better economic efficiency.

As a clean, efficient energy, gas has been used widely in Tunlancoalmine. It has two drainage systems for different concentrationof the gas. The gas drained by the vertical wells, has been collectedin the high concentration system and sent to the urban gas pipelinein Taiyuan city to meets the requirements for residents’ daily life.The gas drained by the underground longholes has been gatheredin low concentration system for the gas boiler, the peat drier andthe power generation. The 10-tonnes-class gas boiler project takesabout 13 million cubic meters gas per year. Two of the power gen-erators with capacity of 4 � 3000 kW h consume almost 30 million

Fig. 15. The schematic of the utilization of the gas products.

S. Kong et al. / Applied Energy 131 (2014) 67–78 77

cubic meters gas every year. The average rate of gas consumptionreaches 95%, and the utilization ratio reaches 90%. The schematic ofthe utilization is shown in Fig. 15.

On present trends, the output of the coal mine could reach 5million tons per year by 2020. The usable volume of gas couldreaches 0.35 billion cubic meters per year which would reducegreenhouse emissions equivalent to 6 million tons of carbondioxide.

5. Conclusions

With the increase of the mining depth, new challenges havegradually emerged. Taking the Tunlan coal mine as a typical casestudy, a concept of gas control in the deep gassy coalbeds was pro-posed in this paper. The main conclusions can be summarized as:

(1) The coalbeds exhibit differences in their reservoir charac-teristics. The micropore volume and the surface area ofthe coalbeds in lower group are higher than those inupper group. The methane adsorption capacities and thegas contents show the same tendency. Based on the actualdynamic phenomena and the indices, the outburst risk isconfirmed in all of the reservoirs. Traditional gas controlmethods are less effective in these deep and high gassycoalbeds.

(2) A concept of gas control was proposed and implemented inpractice at Tunlan coal mine as a typical case study. Based onnumerical simulations and mining economic analysis, thehigh gassy No. 8 coalbed was selected as the first-minedcoalbed and hydro-fractured to increase the reservoir per-meability. Surface vertical wells and underground longholeswere used to effectively reduce the gas content of the first-mined coalbed, thereby eliminating the outburst risk of thefirst-mined coalbed and ensuring the safety of mining activ-ities. During the mining process of the first-mined coalbed,the permeability of the other coalbeds (No. 2, No. 4 andNo. 9) was enhanced by the pressure-relief effect. The

outburst risk of the upper No. 2 and No. 4 coalbeds was con-trolled by the gas drainage of the vertical wells. In the No. 9coalbed, the outburst risk was eliminated by undergroundlongholes drainage. Finally, the efficient co-exploitation ofcoal-methane was achieved.

(3) The core technologies of this approach were independentlytested in the field and proven to be effective. The averagepure methane flow rate per well could reach 619 m3/d, andthe methane concentration could remain at a level over80%. It is expected the gas content of the deep area will fallto 8 m3/t and the outburst risk will be eliminated in 8 years.The enhanced gas drainage and pressure relief gas drainageby longholes were found to be effective in the 18,207 long-wall workface. Currently the complete integration of thegas drainage technologies is being implemented at Tunlancoal mine and will be completed in 8–10 years. The promis-ing results provide a valuable demonstration and referencefor other coalbed reservoirs with similar geologicalconditions.

(4) The integrated gas drainage method has been applied suc-cessfully in Tunlan coalmine. The gas extraction volumeincreased greatly, meanwhile, the mining safety situationis improved and the greenhouse gas emission is reduced.This case study could be used as a valuable example forother coal mines having similar geological conditions.

Acknowledgements

The authors are grateful to the National Basic Research Pro-grams of China (No. 2011CB201204), the National Natural ScienceFoundation of China (No. 51074160), the Fundamental ResearchFunds for the Central Universities (No. 2010QNA03) and ProjectFunded by the Priority Academic Program Development of JiangsuHigher Education Institutions for their support for this project. Weare also grateful to Wei Li, Shouqing Lu, Qingquan Liu and Chao Xufor improving the manuscript.

78 S. Kong et al. / Applied Energy 131 (2014) 67–78

References

[1] Fu F, Liu H, Polenske KR, Li Z. Measuring the energy consumption of China’sdomestic investment from 1992 to 2007. Appl Energy 2013;102:1267–74.

[2] Wang K, Wei Y-M, Zhang X. Energy and emissions efficiency patterns ofChinese regions: a multi-directional efficiency analysis. Appl Energy2013;104:105–16.

[3] BP. BP Statistical Review of World Energy. British Petroleum, plc; 2010.[4] Chandran Govindaraju VGR, Tang CF. The dynamic links between CO2

emissions, economic growth and coal consumption in China and India. ApplEnergy 2013;104:310–8.

[5] Peng S. Present study and development trend of the deepen coal resourcedistribution and mining geologic evaluation. Coal 2008;17:1–11.

[6] Fu J, Cheng Y. Situation of coal and gas outburst in China and controlcountermeasures. J Min Saf Eng 2007;3:000.

[7] Liu H, Cheng Y, Zhao C, Zhou H, Zhang Z. Classification and judgment methodof the protective layers. J Min Saf Eng 2010;4:006.

[8] Wang H, Cheng Y, Wu D. Gas emission and parameter optimization of gasextraction in mining face of short distance protective seam. J China Coal Soc2010;35:590–4.

[9] Wang L, Cheng Y, Li F, Wang H, Liu H. Fracture evolution and pressure relief gasdrainage from distant protected coal seams under an extremely thick keystratum. J China Univ Min Technol 2008;18:182–6.

[10] Liu H, Cheng Y, Song J, Shang Z, Wang L. Pressure relief, gas drainage anddeformation effects on an overlying coal seam induced by drilling an extra-thin protective coal seam. Min Sci Technol (China) 2009;19:724–9.

[11] Diamond WP. Methane control for underground coal mines. US Department ofthe Interior, Bureau of Mines; 1994.

[12] Cheng Y, Wang L, Zhang X. Environmental impact of coal mine methaneemissions and responding strategies in China. Int J Greenhouse Gas Control2011;5:157–66.

[13] Cockbill J, Finn C, Krawiec M. Economics of Mannville CBM development:drilling & production innovation at Corbett Creek. In: Canadian internationalpetroleum conference 2008.

[14] Ham YS, Kantzas A. Development of coalbed methane in Australia: uniqueapproaches and tools. In: CIPC/SPE gas technology symposium 2008 jointconference 2008.

[15] Zuber M, Kuuskraa V. Optimizing well spacing and hydraulic-fracture designfor economic recovery of coalbed methane. SPE Formation Eval1990;5:98–102.

[16] Warlick D. Gas shale and CBM development in North America. Oil Gas Financ J2006;3:1–5.

[17] Gatnar K, Tor A. Drainage and economic utilization of methane from coalseams in the Jastrzebie mining-field. Appl Energy 2003;74:331–41.

[18] Lv Y, Tang D, Xu H, Luo H. Production characteristics and the key factors inhigh-rank coalbed methane fields: a case study on the Fanzhuang Block,Southern Qinshui Basin, China. Int J Coal Geol 2012;96–97:93–108.

[19] Liu H, Sang S, Formolo M, Li M, Liu S, Xu H, et al. Production characteristics anddrainage optimization of coalbed methane wells a case study from low-permeability anthracite hosted reservoirs in southern Qinshui Basin China.Energy Sust Dev 2013.

[20] Sang S, Xu H, Fang L, Li G, Huang H. Stress relief coalbed methane drainage bysurface vertical wells in China. Int J Coal Geol 2010;82:196–203.

[21] Moore TD, Zabetakis MG. Effect of a surface borehole on longwall gobdegasification (Pocahontas No. 3 coalbed). US Department of the Interior,Bureau of Mines; 1972.

[22] Karacan CÖ, Ruiz FA, Cotè M, Phipps S. Coal mine methane: a review of captureand utilization practices with benefits to mining safety and to greenhouse gasreduction. Int J Coal Geol 2011;86:121–56.

[23] Sun J. The causes and lessons of ‘‘2� 22’’ gas explosion disaster at Tunlan coalmine. J China Coal Soc 2010;35:72–5.

[24] Jin Z, Zhang X, Zhao K, Liu R, Wang C. Control of late Carboniferous-earlyPermian sea level fluctuation on coal reservoir heterogeneity, Taiyuan, Shanxi.Pet Explor Dev 2004;5:011.

[25] Jiang J, Cheng Y, Wang L, Li W, Wang L. Petrographic and geochemical effects ofsill intrusions on coal and their implications for gas outbursts in the WolonghuMine, Huaibei Coalfield, China. Int J Coal Geol 2011;88:55–66.

[26] Wang L, Cheng Y, Xu C, An F, Jin K, Zhang X. The controlling effect of thick-hardigneous rock on pressure relief gas drainage and dynamic disasters in outburstcoal seams. Nat Hazards 2013;66:1221–41.

[27] ISO. ISO 17246–2010, coal – proximate analysis. ISO InternationalOrganization for Standardization; 2010. p. 9.

[28] ISO. ISO 7404-3-2009, Methods for the petrographic analysis of coals – Part 3:method of determining maceral group composition. ISO InternationalOrganization for Standardization; 2009. p. 13.

[29] ISO. ISO 15901–1:2005, Pore-size distribution and porosimetry of solidmaterials by mercury porosimetry and gas adsorption – Part 1: mercuryporosimetry. ISO International Organization for Standardization; 2005. p. 25.

[30] CNSMC. GB/T 19560–2008 Experimental method of high-pressure isothermaladsorption to coal. Beijing (China): Standards Press of China; 2008.

[31] Dubinin M. The potential theory of adsorption of gases and vapors foradsorbents with energetically nonuniform surfaces. Chem Rev 1960;60:235–41.

[32] Clarkson C, Bustin R. The effect of pore structure and gas pressure upon thetransport properties of coal: a laboratory and modeling study. 1. Isothermsand pore volume distributions. Fuel 1999;78:1333–44.

[33] An F, Cheng Y, Wu D, Wang L. The effect of small micropores on methaneadsorption of coals from Northern China. Adsorption 2012;19:83–90.

[34] SAWS. AQ 1024–2006 specification for identification of coal and gas outburstmine. Beijing (China): Standards Press of China; 2006.

[35] Wang L, Cheng Y, Wang L, Guo P, Li W. Safety line method for the prediction ofdeep coal-seam gas pressure and its application in coal mines. Saf Sci 2012;50:523–9.

[36] Li G, Hou Q. Dynamic process and difference of coalbed methane wellsproduction in southern Qinshui Basin. J China Coal Soc 2012;5:018.

[37] Li G, Li G. Study on the differences and main controlling factors of the coalbedmethane wells between single layer and multi-layer drainage. Meitan Xuebao/J China Coal Soc 2012;37:1354–8.

[38] Coulthard M. Applications of numerical modeling in underground mining andconstruction. Geotech Geol Eng 1999;17:373–85.

[39] Itasca F. Fast Lagrangian analysis of continua, version 4.0 user’sguide. Thrasher Square East: Itasca Consulting Group, Inc; 2002. p. 708.

[40] Palchik V. Formation of fractured zones in overburden due to longwall mining.Environ Geol 2003;44:28–38.

[41] Hu G, Wang H, Li X, Fan X, Yuan Z. Numerical simulation of protection range inexploiting the upper protective layer with a bow pseudo-incline technique.Min Sci Technol (China) 2009;19:58–64.

[42] Yang W, Lin B, Qu Y, Li Z, Zhai C, Jia L, et al. Stress evolution with time andspace during mining of a coal seam. Int J Rock Mech Min Sci 2011;48:1145–52.

[43] Karacan CÖ, Diamond WP, Schatzel SJ. Numerical analysis of the influence ofin-seam horizontal methane drainage boreholes on longwall face emissionrates. Int J Coal Geol 2007;72:15–32.

[44] Keim SA, Luxbacher KD, Karmis M. A numerical study on optimization ofmultilateral horizontal wellbore patterns for coalbed methane production inSouthern Shanxi Province, China. Int J Coal Geol 2011;86:306–17.

[45] Wang F, Ren TX, Hungerford F, Tu S, Aziz N. Advanced directional drillingtechnology for gas drainage and exploration in Australian coal mines. ProcediaEng 2011;26:25–36.