i ? ^ "'A * ^* - s nap : >/. Bureau of Mines Information Circular/1987 A Review of the Mechanisms of Gas Outbursts in Coal By David M. Hyman UNITED STATES DEPARTMENT OF THE INTERIOR Information Circular 9155 A Review of the Mechanisms of Gas Outbursts in Coal By David M. Hyman - UNITED STATES DEPARTMENT OF THE INTERIOR Donald Paul Hodel, Secretary BUREAU OF MINES David S. Brown, Acting Director Library of Congress Cataloging in Publication Data: Hyman, D. A review M. (David M.) of the mechanisms of gas outbursts in coal. (Information circular; 9155) Bibliography p.lO-11. Supt. of Docs no.: I 28.27: 9155. 1. Coal mines and mining. 2. Gas bursts. I. Title. II. Series: Information circular (United States. Bureau of Mines); 9155. TN295.U4 [TN313] 622 s [622 .8] 87-600180 CONTENTS Page Abstract 1 Introduction 2 Coal-gas sorption-desorption methods 3 Borehole prediction method 6 Mitigation of outburst events 7 Summary and conclusions 9 References 10 ILLUSTRATIONS 1« Methane emissions from mining events 3 2. Comparison of theoretical coal chip desorption 4 3. Volumes of gas released by outburst events 6 UNIT OF MEASURE ABBREVIATIONS USED IN THIS REPORT cm centimeter m meter cm /g cubic centimeter gram per m 3 min cubic meter minute cm /(kg 'min 2 ) cubic centimeter kilogram square minute per mm m 3 /mt millimeter cubic meter per g gram metric ton h hour MPa megapascal kPa kilopascal pet percent L liter s second L/min liter per minute A REVIEW OF THE MECHANISMS OF GAS OUTBURSTS IN COAL By David M. Hyman 1 ABSTRACT Outbursts are sudden and violent releases of gas and coal that result from a complex function of geology, stress regime, and gas pressure and content. The Bureau of Mines has reviewed methods for prediction and mitigation of such outbursts in use worldwide, as an aid in selecting the proper techniques for use in specific mine environments. Outburst- prone coal may be distinguished from normal coal by its sorption- desorption velocity. Three types of methods used to characterize the kinetics of sorption-desorption are described; all are based on the ability of outburst-prone coal to release, through desorption, methane or carbon dioxide much more rapidly than normal coals. Other prediction methods, based on borehole samples, are also described. Various mitigation methods described and evaluated include (1) working the least stressed, less disturbed, lowest gas content seam in multiple- seam areas; (2) mine opening geometry; (3) inducer shot firing; (4) water infusion; (5) localized stress relief, using boreholes or by cut- ting a reliever slot in the longwall face; and (6) other gas drainage methods. 1 Geologist, Pittsburgh Research Center, Bureau of Mines, Pittsburgh, PA. INTRODUCTION An outburst is defined as a violent, simultaneous release of gas(es) and com- minuted rock material into a working face or the interior of a borehole. In gen- eral, an outburst event has the following phases (1_) : 2 1. A stressed volume of rock contain- ing gas(es) is exposed to a rapid change of confining stress. This rock volume has been highly fractured as a result either of some preexisting geologic dis- turbance (such as a fault) or of mining- induced stress concentration. 2. Gas(es) adsorbed in or contained in sandstone or evaporite rocks are rapidly released into the fractures, which al- ready contain free gas. When more gas enters the fracture space than can be transported away through the less per- meable rock body, a state of stress due to gas pressure may be reached where the rock body cannot contain the increasingly stressed fractured rock volume. 3. When the rock body can no longer contain the stressed and fractured rock volume, containment ceases and the frac- tured rock mass and gas(es) undergo move- ment as they are driven by the gas into a pressure sink, e.g., a mine opening or borehole. 4. After the movement of the fractured rock and gas(es), there may be continued gas flow from the fractured but in-place rock that forms the outburst cavity. This gas flow generally decreases over time. Two major theories — the "pocket" and the "dynamic" theories — can describe the basis of the coal outburst mechanism. The pocket theory holds that there exist cer- tain volumes of "soft" or crushed coal enclosed by "harder" or less fractured coal that form reservoirs of gas con- tained in the fracture void space. These crushed coal volumes are associated with faulted or sheared zones and with in- tensely folded strata. This comminuted 2 Underlined numbers in parentheses re- fer to items in the list of references at the end of this report. coal has little unconfined compressive strength and is separated from the mine opening by an intact zone of coal under sufficient stress to become a "permeabil- ity dam." When mine development ap- proaches a "soft coal" region, an out- burst can result if the region is not sufficiently drained of free gas and/or the stresses in the region are not dissipated (2). The dynamic theory holds that a volume of relatively gassy coal, which is highly stressed and penetrated by mining-induced fractures, is outburst prone. When a mine opening and induced stresses ap- proach such a coal volume, the coal frac- tures, releasing high-pressure desorbed gas, and the coal face fails, resulting in an outburst (2)» Common to both theories is high-gas- content fractured coal that is able to desorb gas rapidly upon release of con- fining pressure. This rapid desorption feature of out burst -prone coal is the ba- sis for a rather extensive set of predic- tive methods, which are detailed later in this report. Other aspects of out burst -prone coal include low in situ strength due to fis- suring, high free-gas pressure, and asso- ciation with geologic structures such as fracture zones and igneous dikes. These aspects are also the basis of a variety of predictive methods (2_). Outbursts in coal mines represent con- siderable hazards. The most immediate hazard is the unexpected inundation of the ventilation systems with asphyxiating volumes of gas. When methane is the re- leased gas, an explosive hazard can be created, possibly exacerbated by ejected coal dust. The force of the released gas and displaced material can be sufficient not only to disrupt mine ventilation but to debilitate stoppings and ground con- trol structures such as arches and posts, and to injure or kill mine personnel. Additionally, an outburst zone presents a ground control problem due to the fissile nature of the rock that forms the remain- ing outburst cavity. Furthermore, gas may continue to be emitted, and without appropriate ventilation can accumulate in the outburst cavity. While the scientist and researcher would prefer to describe the mechanics of coal-gas outburst in very exact quantita- tive terms, the mining geologist and engineer need to reliably foresee the preconditions and precursors. The body of literature concerning coal-gas out- bursts has abundant works (1-8) that represent overviews of the outburst phenomena at both national and inter- national levels. Case studies of out- bursts are extensive, and the bibliog- raphies of the aforementioned references contain numerous examples. An overview of some of the more commonly practiced coal-gas outburst prediction and preven- tion methods used was compiled as a re- sult of Bureau of Mines research. COAL-GAS SORPTION-DESORPTION METHODS A fundamental component of a coal-gas outburst is the ability of coal, whether in a fractured or relatively solid state, to release sorbed gas fast enough and in a large enough volume to overcome con- fining stresses and drive the outburst process. Whether one subscribes to the "pocket" theory of outburst mechanism or the "dynamic" theory and its mathematical description (9) , the desorption kinetics are at the heart of the outburst mechan- ism. Studies in West Germany ( 10 ) and Wales (11) suggest that outburst-prone coal has essentially the same gas content and capacity as normal coal. Figure 1 shows this as well. What distinguishes the coals in terms of outburst potential is their sorption-desorption velocities. A popular theory holds that outburst- prone coal is much more extensively mi- crofissured than normal (non-outburst- prone) coal. This permits a shorter diffusion path and a higher surface-to- volume ratio. To compare the desorption rates between outburst-prone and normal coal, a gas emission equation and some of its con- stants from the literature (2, 11) were used in the following analysis. The emission equation used is Airey's empiri- cal relationship for gas emission from coal lumps (12) : V(t) = A(l-e( (t/t o )n ) = A(l-exp( ( - t/t o )n ) and where A = V (k L P) L l+k L P V(t) = volume of methane desorbed after time t, min, ROCK MINED OR OUTBURSTED, mt FIGURE 1.— Methane emissions from mining events. A = to = n = V, = ki = equilibrium sorption capacity of coal at gas pressure, P, kPa, time constant min related to coal chip size, min, constant related to coal type or rank, maximum Langmuir sorptive capacity of coal sample, cm 3 /g, Langmuir strength of attraction for gas to sorb, kPa -1 . The value of t is defined as the time required for the subject coal sample of a given effective chip size to desorb 63 pet of its gas content. A calculated t range of about 5 to 15 min (11 ) is con- sistent with microfissure densities with corresponding t values for outburst- prone coals (2^ 12). The coal constant, n, had been found empirically to be about 0.5 for anthracite, 0.33 for bituminous coals, and 0.25 or less for outburst- prone coal. As studies have indicated that some outburst-prone coals do not have a markedly different gas content than that of normal coal in the same coalbed, the same in situ equilibrium gas content, A, will be used in the analysis. Assuming that coal chips have been ob- tained from both a normal coal volume and an outburst-prone zone, the relative (with respect to a constant) desorption curves for normal bituminous (n = 0.33, to = 60 min), anthracite (n = 0.5, t = 60 min), and outburst-prone (n = 0.25, to = 15 min) coal chip samples are calcu- lated and presented as figure 2. As shown in figure 2, the outburst-prone coals initially desorb at a faster rate than normal coals. The highest contrast in desorption rates occurs within ap- proximately the first 10 min of desorp- tion time. It is apparent from this 1.00 > Q LU CD or o en LU Q LU < X o > < o < .50 KEY Outburst-prone coal Bituminous coal Anthracite 50 100 TIME, min 150 200 pressurized cham- the chips from degradation over FIGURE 2.— Comparison of theoretical coal chip desorption. simplistic illustration that in order for desorption indices to differentiate be- tween normal and outburst-prone coals, they must be determined within this short span of time. It also follows that ex- traordinary care must be exercisec in quickly obtaining and preserving (if necessary) coal chip samples for these predictive index determinations. One should also recognize that preserving coal chip samples in a ber does not prevent undergoing structural time (_5)« Three basic classes of tests character- ize sorption-desorption kinetics (_5)« The first class comprises volume-desorbed methods, or so-called AV methods. One such method practiced on a working- mine-section scale is the V30 index (1, 13). The volume of methane emitted within the first 30 min after shot firing is measured and normalized to the mass of coal broken by the shot. This value is divided by the desorbable gas content (qd), which is determined by a coal chip desorption test described later in this section. Normal coals have a V30 value ranging between 0.10 and 0.17, compared with about 0.40 for outburst-prone coals and >0.60 for outburst coals. This index is used in the Federal Republic of Ger- many as part of a hierarchy of tests to assess the risk of encountering an outburst. Another volume-type index is used pri- marily in Australia to predict the risk of outbursts in advance of mining. This test is mainly used for carbon dioxide- coal outbursts (fO« The Hargraves AV index uses a 4-g sample of drill cuttings sized from 0.6 to 1.2 mm. The cuttings are obtained from a drill hole and sealed in a desorption meter within 1 min of be- ing cut by the drill bit. The desorbed gas volume is measured from 1 to 6 min after drilling. When this value exceeds 1.2 enr/g, the subject coal is considered outburst prone. This threshold value is gas specific and colliery specific. The use of a slightly modified AV index with a shorter observation period was at- tempted in Belgium with some success. If the volume of gas (V1) desorbed between 35 and 70 s after drilling is greater than 0. 1 cm 3 /g, there may be an outburst risk; a Vi value greater than 0.2 cm 3 /g indicates a serious outburst risk (]_). A rather large class of outburst pre- diction indices are the popularly known AP indices. These indices are based on pressure changes during either desorption or sorption tests performed on coal chip samples generally within the 0.25- to 0.50-mm size range. The basic AP index is the APq-60 of Soviet origin (1_, _7). This is a desorption type of test. Originally, different coal chip sizes were used for different ranks of coal, but the 0.25- to 0.50-mm chip size range has become standard through practice. A 3.5-g coal chip sample is placed into a 6. 5-cm chamber that has a free gas space of 4 era . Other workers have used 3- to 10-g samples in a chamber with 4 to 10 era of free gas space. The chamber is then evacuated to a negative pressure of about 100 kPa for 90 min to degas the sample. The chamber is then pressurized to about 100 kPa with helium and evacu- ated; then the pressure change is moni- tored. A resultant pressure rise repre- sents the baseline condition for the test procedure, as the helium is not sorbed by the coal. The sample chamber is then evacuated before being pressurized with methane at about 100 kPa for 90 min so as to saturate the coal sample. After saturation, the sample chamber is con- nected to an evacuated chamber to reduce the pressure in the sample chamber to a negative 100-kPa pressure very rapidly. The pressure rise is then measured 10 to 60 s after this chamber pressure reduc- tion. The APo- 60 index is equal to the pressure rise at 60 s minus the baseline pressure rise for the assumed inert (with respect to sorption by coal) helium. The APio-60 index is the difference between pressure rises measured at the 10- and 60-s time periods and indicates outburst- prone conditions when it is greater than about 1.3 kPa. When APo- 60 is greater than 2 kPa, the sample is considered to represent outburst-prone conditions. The primary disadvantages of the APo-60 test are that it is a laboratory-based deter- mination and requires 6 to 8 h to perform (1). A variation on the APo-60 method, de- veloped by Lama (_5 ) , uses shorter obser- vation times to evaluate the sorption kinetics of coal samples. This method is known as the AP express method and cor- relates rather well with the APq-60 method (_5_). For the AP express method, a 50-g sample of coal chips within the 0.25- to 0.50-mm size range is sealed in a 250-cm chamber. This chamber is evacuated for 5 min at a negative pres- sure of 100 kPa. After degassing, the sample chamber is pressurized with meth- ane at about 200 kPa. The pressure drop due to adsorption is then measured for a 10-min period. While the AP eX pr ess index has a good correlation with the APo-60 index, its usefulness as an outburst pre- diction has not been demonstrated. If the pressure drop curve due to adsorption for the APexp ress method is examined for the 5- to 10-min time interval, and this pressure drop value (expressed in kPa) is divided by 300 s, the Li index results. This L] index is a measure of sorption rate and has had some success in detect- ing shear zones of coal. A third class of indices for predicting outburst-prone coal measures the rate of change of desorption rates or desorption deceleration. In the Federal Republic of Germany a series of calculations is used that is based upon the desorption decel- eration of borehole cuttings collected and sealed in a desorption meter within 1 min of cutting (1_, J^0« About 10 g of coal cuttings in the 0.4- to 0.63-mm size range are collected, and the desorption rates are measured over a 5- to 10-min period. For this time period, a power law relationship for desorption rate over time is assumed: dV(l) = dV(t) ( k v dt dt V ' where time (t) is in minutes. If this power law relationship is obeyed, a plot of logarithmic desorption rate versus logarithmic time will yield a straight line of slope k. The intercept at time = 1 min is the desorption rate at 1 min. When this intercept value is multiplied by a time constant, which is a function of coal chip size, the absorb- able gas content, qd, is obtained. A value of qd greater than 9 m /mt indi- cates a suspected outburst condition. This qd, or desorbable gas content value, is the scale for the V30 index described earlier. Note that the 9-m /mt threshold for outburst-prone coal is very close to the smallest specific emission or out- burst gas content for coal outburst pre- sented in figure 3. A value of k greater than 0.75 cm 3 /(kg*min 2 ) indicates that the coal sample is from an outburst-prone area. Normal coals have a k value of about 0.65 cm /(kg'min ). The time con- 106 stant, A, used to calculate qd is min for the 0.4- to 0. 63-mm coal size range and about 25 min for the conventional 0.25- to-0.5-mm coal size range. What these methods, and the scribed in the literature have is a means to differentiate prone coals from normal coals. 29.4 chip more chip others de- in common outburst- This is based upon the ability of the former to release, through desorption, methane and/or carbon dioxide much more rapidly than normal coals. Such is the basis of the British desorption ratio, where a sample's desorbable gas content for a certain time period is divided into a mine-specific representative desorption volume. When this ratio exceeds 4, then the sample in question is considered to represent an outburst-prone zone (.!_)• This property of very rapid desorption is a fundamental precondition to the ro I0 5 E q" LlI ••/ / Jy / / KEY N / • Coal outbursts (CH4) ^/ A Sandstone outburst (CH 4 ) X / \?/ <>,♦ CO2 dominant outburst / ° Evaporite outburst 1 1 1 10' I0 6 I0 2 I0 3 I0 4 I0 5 ROCK EJECTED, mt FIGURE 3. — Volumes of gas released by outburst events. development of gas and coal outbursts. Given coal chips of a particular size range (e.g., 0.25 to 0.50 mm) from outburst-prone coal and normal coal, one would expect the outburst-prone coal chips to have a smaller effective size due to some partitioning feature (such as microf issuring) in their structure that would help explain their faster desorp- tion kinetics. One might also expect this smaller effective size to contribute to increased friability and lower strength in comparison to normal coal in the same coalbed (10-11). Other types of outburst-prediction methods also take advantage of these aspects. BOREHOLE PREDICTION METHOD During the drilling of boreholes into outburst-prone zones, drillers often note gas "kicks," increased gas flows, and disproportionately large volumes of drill cuttings (10). In the Federal Republic of Germany, a drill-cuttings-to-hole volume ratio greater than about 3:1 to 7:1 is indicative of outburst conditions in the coal penetrated by the drill hole. The borehole diameters ranged from 50 to 140 mm in the work reported (10). A French study did not establish an out- burst risk threshold for drill cutting volumes from 43-mm-diameter drill holes even though 2 to about 130 times more cuttings were encountered than could be accounted for by hole volume (J L )» Since volume measurements of drill cuttings are not accurately reproducible owing to dif- ferences in bulking between samples, a gravimetric method would be more useful. A more direct and hopefully more useful prediction method based on the sheared and/or low-strength qualities of outburst-prone coal is presented by Kidybinski (15-16), whose two papers de- scribe the use of a borehole penetrometer with a conical tip to determine faulted areas and adjacent zones of relatively soft coal, as an outburst prediction tool. This use of this tool yields a coal and/or rock strength index, Z, that is equal to the applied thrust on the cone divided by the penetration distance. In tests at three mines, the coal strength was found to be related to in situ gas pressures. Some preliminary empirical relationships were developed, but more research is required to develop a more definitive relationship. MITIGATION OF OUTBURST EVENTS The ultimate method for preventing an outburst event during mining is to pre- dict potential zones of outbursts and avoid them or at least reduce their out- burst potential. As the outburst mechan- ism is a complex relationship between geologic structure, mining-induced stres- ses, and gas content and pressure, the removal or mitigation of one or more of these elements can possibly reduce out- burst potential. Geologic structures such as faults or sheared zones can be avoided to a certain extent. Stresses can be reduced by changes in mining rates, methods, and geometry. Gas con- tents and pressures can be reduced by drainage. Most of the methods practiced throughout the world involve relieving stress concentrations and/or in situ gas pressure. A whole-seam stress reduction method applicable in areas where several minable coal seams occur in close proximity is known as "working the protective seam" (1_). The idea behind this method is to mine the least stressed, and/or lowest gas content, and/or least disturbed coal- bed of those in a multiple-seam configur- ation. It is preferred to mine as a pro- tective seam an overlying one, instead of an underlying one, unless the outburst problems are more severe than the ground control problems due to subsidence from undermining. If conditions permit, the pillars could be superimposed for the protective and outburst-prone seams so as not to defeat the stress-relief aspect of this protective method. Mining the pro- tective seam can also induce some fissur- ing and thus potentially drain some por- tion of the gas in the outburst-prone seam. Not only does the gas drainage ef- fect tend to help lower the overall gas content of the outburst-prone seam, it also can help lessen the magnitude of the desorption rate by allowing the desorption process to begin prior to min- ing. The overall effect of mining a pro- tective seam is to reduce the stress and gas dynamic potential fields of the outburst-prone seam (17). Thus, it is no surprise that this method is one of the most effective methods of reducing out- bursting probability. Unfortunately, this method is applicable only in multi- seam configurations. In single-seam con- figurations, other methods must be employed to reach the same end of reduc- ing the stress and gas dynamic potential fields of an outburst-prone coalbed. Given an outburst-prone coalbed, a variety of methods have been practiced to relieve stress concentrations. Mine opening geometry control is a relatively effective method. Stress concentrations are greater at the face of a single heading or roadway than along a longwall face. Retreat longwalls are less prone to outbursts than advancing longwalls. Pillar extraction is less prone to out- bursting than retreating longwalls, but this may be due to degassing more than to stress relief. The most outburst-prone mining operation is when a heading or tunnel moves from one coal seam through a rock interburden to an adjacent coal seam (4)» In general, longwall mining methods with gate roads not developed more than 2 m ahead of an advancing longwall face are less liable to trigger outbursts than are room-and-pillar mining methods (_1_)» Besides mine opening geometry, several stress-relief measures are practiced at the mine face level. Inducer shot firing is employed in several countries to relieve stress accumulations and to trig- ger outbursts in a relatively controlled fashion. One form that inducer shot fir- ing takes is destressing at the ends or gate road faces of a longwall panel. This precautionary shot firing involves drilling holes outby the gate road faces, usually to a depth of about 3.7 to 4.6 m, although in Turkey holes up to 8.3 m deep are used (8). These holes are charged with explosives. When an outburst-prone zone is predicted within the longwall face area, shot firing is performed by detonating explosives in boreholes across the longwall face on both sides of the suspected zone (I) . The explosive charges can be detonated either simul- taneously (camouflet) for maximum shock loading, or with short delays to facili- tate mucking operations. A variation on this shot-firing theme is termed pulsed infusion shot firing. For this method boreholes 4 to 9m deep are drilled, charged with submarine explosives, filled and pressurized with water, and deto- nated. Although shot firing has been used with some success in controlling the occurrence of outbursts and in mitigating to some extent the severity of further outbursts in the treated area, it is an inherently hazardous practice; also the deleterious effects of expelled coal and gas due to outbursting still occur and must be considered when using these techniques. To a limited extent, the erection of a barricade 8 to 15 m from the face can mitigate these deleterious effects. Unfortunately, such barricades introduce problems for postshot face ventilation gas checks (4^« A less energetic method of destressing and fracturing an outburst-prone zone in a coalbed is through water infusion. Water infusion for destressing is a modi- fied form of water infusion for dust and gas emission control. It is performed where prediction methods such as prox- imity to a fault, high Po-6 values, or discharge of a large relative volume of drill cuttings indicate an outburst-prone zone. The infusion method has been implemented in the Federal Republic of Germany (19) by drilling 50-mm-diameter boreholes at a spacing of three to four gate road widths in the suspect zone. The boreholes are drilled into the sus- pect zone and generally are less than 10 m from the face; depending on the nature and extent of the suspected outburst- prone zone, they can be up to 60 m from the face. Water is pumped into the holes at about 70 to 100 L/min, at pressures up to 40 MPa but averaging 11 to 23 MPa. Each hole pattern is infused sequentially until either cracking or separation of the coalbed occurs. The infused water both redistributes stress and forces free gas in the fractures away from the face. Water infusion becomes hydraulic fractur- ing when a series of pressure pulses are applied to the water in the hole. A major drawback to water infusion is that its degassing or gas-displacing effects are relatively short-lived; consequently, it must be performed frequently, poten- tially interfering with production. Localized stress relief can be accom- plished through the use of boreholes in terms of relieving both local stress and gas pressure. These stress-relief bore- holes have diameters of about 82 to 300 mm. An effective borehole diameter is found by drilling holes of progressively larger diameters until an excessive vo- lume of cuttings is discharged, indicat- ing a miniature outburst event in the hole. The relief hole spacing is found by reducing borehole spacing until no audible stress readjustments are heard (4) as the holes are bored. The length of these boreholes is 10 to 25 m in ad- vance of the face. Additionally, cavi- ties can be excavated at some depth in the borehole to trigger an outburst in the borehole to further relieve stress and gas pressure. This practice is known as "perforation" in Hungary and has also been used in the U.S.S.R. and Australia. A considerable drawback to perforation is controlling the coal dust and gas ejected from the hole. A stress-relief technique practiced in the U.S.S.R. is the cutting of a reliever slot in the longwall face. A cutting cable saw is used to cut an 80-mm slot about 3 to 5 m deep, effectively under- cutting the face. Different slot orien- tations and cutting depths are used, depending on local conditions (17). Relief of excess stress removes, through redistribution, a portion of the triggering energy for an outburst. The reduction of gas pressure as well as gas content also removes potential energy from an outburst-prone zone. An outburst-mitigation method therefore re- duces gas pressure in a gas drainage borehole, as well as reducing stress con- centrations to a limited extent. In general, gas drainage boreholes have smaller diameters (40 to 100 mm) than stress-relief boreholes. The strategy in gas drainage is to degas volumes of rock in advance of mining with lateral and vertical holes in gate roads and faces (4_). It has been found that open hole or free-flow drainage is not as effective or efficient as applying a negative pressure of 7 to 40 kPa. Studies (1_, IT, 1_9) have shown that application of a slight nega- tive pressure to increase the borehole pressure sink increased gas output from the drainage holes by factors of 2 to 4 times compared with free-flow drainage. What determines the spacing of boreholes in a gas drainage plan is a complex func- tion of coalbed permeability, gas pres- sure, pressure gradient, degree of water saturation, gas dynamics of the coal, and duration of drainage. A Chinese experi- ment (1_) involved the drilling of 22 gas drainage holes of 75-mm diam into an outburst-prone coalbed. These holes were monitored for 19 months. When mining operations penetrated the drained zone, no outburst events occurred. A 6- to 7-m radius of influence was calculated, indicating that a 10- to 15-m gas drain- age hole spacing would be effective for this mine. The Japanese (20) performed a gas drainage study in which 65- to 90-mm boreholes were drilled at 10- to 20-m spacings. The gas drainage was aided by suction. During drainage, test holes were bored and gas pressure and flow mea- surements were made. Mining was started when the test boreholes showed that gas pressures and flows had been lowered. This gas reduction occurred after 1 to 2 months. SUMMARY AND A gas drainage experiment was conducted in an Australian coalbed containing shear zones that were outburst prone (2_0. The experiment consisted of drilling three 100-mm-diameter holes parallel to each other and separated by about 10 m. The holes were drilled to 23-, 45-, and about 60-m depths. The longest hole penetrated a shear zone. Gas pressure and flows were monitored for about 140 days. The shortest hole produced 0.35 L/min per meter of length, the 45-m hole produced about 2 L/min per meter, and the hole that penetrated the shear zone produced 63 L/min per meter. The gas pressure in the shear zone dropped from an initial 410 kPa to about 200 kPa after 30 days and to about 100 kPa after 80 days. The depth of the coalbed was about 500 m, and the holes were drained without the ap- plication of suction. When mining inter- cepted this drained shear zone, no out- bursts occurred. In another experiment in the same mine, 40-kPa suction was ap- plied to boreholes, yielding 200- to 400- pct increases in flow rates. The range of influence for drainage holes in this mine was estimated to be about 30 m. While not all gas drainage exercises reported in the literature have prevented outbursts, they have been somewhat ef- fective in draining gas and reducing gas pressure in advance of mining. Gas drainage works best in outburst zones where the coal is crushed or comminuted, as in a fault or shear zone. Outburst-prone zones where the coal is relatively intact are not very amenable to gas drainage without the application of borehole stimulation techniques. CONCLUSIONS The body of scientific literature that describes outburst mechanics, precondi- tions, prediction techniques, and pre- vention or defensive measures is exten- sive. Effective management of outburst zones requires that they be predicted in advances of mining by geological in- vestigations to delineate fault or shear zones and igneous intrusions. A variety of physical testing methods is available to predict outbursts, centering on rela- tively rapid (on the order of minutes) desorption tests. Borehole prediction methods rely on the instability and fis- sile nature of outburst-prone coal or gas and comminuted coal expulsions from bore- holes. As outbursts are a complex func- tion of geology, stress regimes, and 10 gas contents, available defenses attempt to relieve stress and gas pressures once an outburst-prone zone has and geologically mapped. been defined REFERENCES 1. Baker, A. Outbursts in Coal Mines. IEA Coal Res., 1984, 55 pp. 2. Shepard, J. , L. K. Rixon, and L. Griffiths. Outbursts and Geological Structures in Coal Mines: A Review. Int. J. Rock Mech. , Min. Sci. & Geomech. Abstr. , v. 18, 1981, pp. 267-283. 3. Hargraves, A. J. Particular Gas Problems of Australian Deep Coal Mining. Paper in Third International Mine Venti- lation Congress (Harrogate, England, June 13-19, 1984). Inst. Min. and Metall. , 1984, pp. 127-133. 4. . Instantaneous Outbursts of Coal and Gas - A Review. Proc. Austral- asian Inst. Min. and Metall. , No. 258, Mar. 1983, pp. 1-37. 5. Lama, R. D. Adsorption and De- sorption Techniques in Predicting Out- burst of Gas and Coal. Paper in Sympo- sium on the Occurrence, Prediction, and Control of Outbursts in Coal Mines. Australasian Inst. Min. and Metall. , 1980, pp. 173-191. 6. Patching, T. H. , and J. C. Bolt- ham. Occurrence, Research, and Control of Sudden Outbursts of Coal and Gas in Canada. Dep. Energy, Mines and Re- sources, Mines Branch, Ottawa, Canada, Reprint Series RS28, 1967, 29 pp. 7. Vandeloise, R. Survey of New Methods Used in Belgium. Paper in Sym- posium of Coal and Gas Outbursts (Nimes, France, Nov. 25-27, 1964), United Nations, New York, 1967, pp. 46-71. 8. Saltoglu, S. The Presentation, Evaluation, and Fighting Procedures of the Sudden Gas and Coal Outbursts in the Zonguldok Coal Field of Turkey. Paper in Ninth World Mining Congress, Federal Re- public of Germany, May 11-21, 1976, pp. 1-10; available from D. M. Hyman, Bu- Mines, Pittsburgh, PA. 9. Litwiniszyn, J. A Model for the Initiating of Coal-Gas Outbursts. Int. J. Rock Mech. , Min. Sci. & Geomech. Abst. , v. 22, No. 1, 1985, pp. 39-46. 10. Paul, K. Forewarning and Pre- diction of Gas Outbursts in a West German Coal Mine. Paper in Symposium on the Occurrence, Prediction, and Control of Outbursts in Coal Mines (Supplement). Australasian Inst. Min. and Metall. 1980, pp. 1-21. 11. Brown, K. M. , N. Rigby, and G. R. Barker-Read. Gas Emission and Outburst Predictions. Paper in Third Inter- national Congress on Mine Ventilation (Harrogate, England, June 13-19, 1984). Inst. Min. and Metall. , 1984, pp. 151- 155. 12. Airey, E. M. Gas Emission From Broken Coal, An Experimental and Theo- retical Investigation. Int. J. Rock Mech. , Min. Sci. & Geomech. Abstr. , v. 5, 1986, pp. 475-494. 13. Noack, K. , K. Paul, and F. Poer- tge. Present Stage in the Prevention of Outbursts of Gas and Coal in the West German Bituminous Coal Mines. Paper in Proceedings of 20th International Con- ference of Safety in Mines Research Institute, (Sheffield, England, Oct. 3-7, 1983). pp. B3, 1-20; available upon re- quest from D. M. Hyman, BuMines, Pittsburgh, PA. 14. Janas, H. Improved Method for As- sessing the Risk of Gas/Coal Outbursts. Paper in Second International Mine Venti- lation Congress (Reno, NV, Nov. 4-8, 1979). Soc Metall. Eng. AIME, 1980, pp. 372-377. 15. Kidybinski, A. Experiment With Hard Rock Penetrometers Used for Mine Rock Stability Predictions. Paper in Proceedings of the Fourth Congress of the International Society for Rock Mechanics (Montreux, Switzerland, Sept. 2, 1979). A. A. Balkema, 1979, pp. 293-301. 16. . Significance of In Situ Strength Measurements for Prediction of Outburst Hazard in Coal Mines of Lower Silesia. Paper in Symposium on the Oc- currence, Prediction, and Control of Out- bursts in Coal Mines. Australasian Inst. Min. and Metall. , 1980, pp. 193-201. 17. Ayruni, A. T. Theory and Practice of Mine Gas Control at Deep Levels. Nedra Press, 1981, 335 pp. 11 18. Grobkrenz , H. De-stressing Infu- sion in Out burst-Prone Steep Faces. Gluckauf, v. 120, No. 5, 1984, pp. 268- 274. 19. Lunarzewski, L. , and S. Battino. Application of Gas Pre-Drainage for Pre- vention of C02 Gas Outburst. Paper in Third International Congress of Mine Ventilation (Harrogate, England, June 13- 19, 1984). Inst. Min. and Metall. , 1984, pp. 159-165. 20. Oda, N. , T. Isobe, and M. Umezu. On the Strata Disturbance and Gas Outbursts. Paper in 19th International Conference of Safety in Mines Research (Katowice, Poland). 1981, pp. 524-530; available upon request from D. M. Hyman, BuMines, Pittsburgh, PA). 21. Lama, R. D. , P. Marshall, and L. Griffiths. Methane Drainage Investiga- tions as a Method of Control of Outbursts at West Cliff Colliery. Paper in Sympo- sium on the Occurrence, Prediction, and Control of Outbursts in Coal Mines. Australasian Inst. Min. and Metall. , 1980, pp. 223-239. U.S. GOVERNMENT PRINTING OFFICE 1987 - 605-017/601 10 INT.-BU.0F MINES,PGH.,PA. 28566 U.S. Department of the Interior Bureau of Mines— Prod, and Distr. Cochrans Mill Road P.O. Box 18070 Pittsburgh, Pa. 15236 OFFICIAL BUSINESS PENALTY FOR PRIVATE USE. S300 ] Do not wi sh to recei ve thi s material, please remove from your mailing list* "2 Address change* Please correct as indicated* AN EQUAL OPPORTUNITY EMPLOYER , H EC KM AN RINDFRY INH BINDERY INC. SEP 92 KI UAKI^UCCTC N. MANCHESTER, INDIANA 46962