OUTBURSTS AND RESEARCH NEEDS

WHAT IS AN OUTBURST?

Outbursts are expulsions of gas and coal from the working face of a coal mine. They have the potential to kill by mechanical force or by asphyxiation from the gas that they release. Outbursts may be small puffs of gas and coal to roaring great killers that throw out hundreds of tonnes of coal and rock and thousands of cubic metres of gas. Collinsville, 1954 (1), Leichhardt, 1978 (2), Akabira, 1958 (3) and Cynheidre, 1971 (4) are examples. The smaller event may apparently be benign but ignoring it may lead to a completely unexpected major outburst. Leichhardt, 1978 (2) is the classic example of this.

Outbursts usually occur whilst advancing into virgin ground during development. This is not however their only manifestation. Outbursts have been known on longwall faces Westcliff ( ).

Outbursts are more frequently associated with fault gouge material than with solid coal, however solid coal outbursts have occurred in a number of cases, Leichhardt, Cook, and Cynheidre.

Outbursts are usually not preceded by any particular noise though Cynheidre was a notable exception. There are many instances though where reduced gas make has been noted prior to an outburst occurring.

MECHANISTIC DESCRIPTION OF AN OUTBURST

The occurrence of an outburst is preceded by failure of the coal. Failure in itself is nothing remarkable as it occurs in mining all the time. The difference in an outburst is that the failed material is ejected with energy and with gas. Indeed the difference between a rockburst and an outburst is the gas that is emitted. The gas contributes in a major way to the expulsion of the coal and is generally thought to be the main contributor to total energy release in the majority of cases.

The Solid Coal Outburst

Failure of solid coal containing gas occurs with a combination of effective stresses that exceed the strength of the coal. By effective stress we mean the stresses due to total load minus fluid pressure. The fluid may be either liquid or gas. The existence of fluid pressures mean that tensile stresses may exist in the coal. Consider a mining face with stresses parallel to and no confining stress perpendicular to the face. In the presence of fluid pressure the situation of compressive effective stress parallel to the face and tensile effective stress perpendicular to the face may occur. Coal does not resist tensile stress well and in the presence of compressive and tensile stresses many shear planes may develop.

The way in which the coal breaks up is governed by the structure of the coal. In numerical terms it is governed by the toughness of the material. Toughness is not usually a concept applied to coal rock mechanics, however it is extremely important. Tough materials require energy to propagate a failure whilst brittle materials require little energy. The concept of crack tip propagation has however been pursed to great length with reference to hard rocks. This knowledge has come from theoretical work developed in gaining an understanding of rock breakage by blasting or mechanical splitting.

As the coal breaks up and expands outwards the rock stresses become less important and internal fluid pressure is the dominant stress that leads to the further fragmentation of the coal. This gas driven splitting has parallels with explosive driven splitting. The degree to which fragmentation occurs is vital to the outburst process.

The broken material is carried from the outburst cavity by energy. This energy may be to some degree supplied by the closure of the cavity as a rock mechanics effect but more importantly in outbursting it is carried out by gas. The gas is derived from the broken material. The transport of the material outward from the outburst cavity is at first the effect of gas expanding behind the coal, as a piston. Then as the fragments separate the process becomes more one of fluidized transport. Here the particles are being carried by the gas in turbulent flow. The combined effects may represent a large release of energy.

The rate of release of gas from the coal particles is overwhelmingly influenced by the size of the particles and to a lesser degree by the diffusion coefficient of the coal. The particle size is of critical importance because the surface area increases as the square of the particle size.

The transport of the coal particles by the gas released is also strongly influenced by particle size. Here the susceptibility of the particle to transport increases with surface area and decreases with weight. This is an inverse function of diameter.

Taking into account the gas release rate and the transport behaviour one could expect the severity of an outburst to increase to the third order of inverse particle size and proportionally to the diffusion coefficient. Particle size is therefore critical. The finer the material the more dangerous the outburst.

The solid coal outburst usually ceases to occur as the cavity from which the coal is being ejected reduces in size with depth. The cessation of the outburst is a function of confinement. As the outburst proceeds back into the coal the unconfined face reduces in size. In addition the outburst may choke itself off from the front if there either is not enough gas to expel the particles.

Outbursts from Gouge Material

The entire outburst process is facilitated greatly if the coal is already fine particles. A Mass of pre-ground fine particles will fail much more readily than solid coal. Once the mass is opened up and fluid pressure between particles drops then desorption will occur very rapidly. Because the material is fine it is easily transported and an outburst can ensue.

The size of the outburst is usually governed by the extent of the gouge material. Additional rock and coal may however be broken and swept out with the gouge material.

Different Gases

Outbursts occur with both carbon dioxide and methane alone or combined as the seam gasses. At the moment in Australia it is considered that carbon dioxide poses more of a problem. The occurrence of outbursts with both gasses is well recognised internationally (5).

Laboratory testing for sorption isotherms shows that coal absorbs more carbon dioxide than methane for a given pressure. If coal is subjected to a mixture of the two gasses then absorption of both gases occurs in a predictable combination. However if the seam gas is originally CH4 and a sweep of C02 in gaseous or hydrothermal form occurs then the gas left may be of any variation.

It would normally be expected that CO2 would desorb more slowly than CH4 because it is more preferentially bonded to the coal. In fact this is not usually the case where mixed gases exist as CO2 is released first. This means that the diffusion coefficient of the coal to CO2 in this state is higher than that of CH4. The reasons for this are unknown. It is possible however to hypothesise that the CH4 occupies the more preferential molecular sites on the coal and that these are not displaced by the later entry of CO2. Therefore the CO2 is attached to less preferential bonding sites and is given up more readily. The concept of varying diffusion coefficients not only applies to different gases. It is quite common to see two diffusion coefficients in the normal desorption of a coal core. The change from one to the other may be quite abrupt.

Carbon dioxide has a much higher molecular weight (44) than methane (16) and hence has a higher density. The ability of moving gas to entrain particles is directly proportional to its density. As a consequence CO2 may be expected apply 2.75 times more drag to a particle of coal for a given velocity than methane.

WHERE DO OUTBURSTS OCCUR?

Outbursts occur in gassy coal (or rock). They more frequently occur in gouge material from faults than they do from solid coal. Not all coal with high gas content or pressure however exhibit outbursts from solid coal. This can be because failure at the face does not occur. Alternatively it may be because the failure is not associated with fragmentation and rapid desorption.

The lack of gas pressure in front of the face is another reason why failure and an outburst does not occur. Variations in drainage characteristics do occur in coal seams and may lead to the changing gas pressure levels near the face. The variations in permeability may be brought about by raised confining stress or alternatively by filling of cleats. Local impervious zones such as dykes may also prevent drainage.

OUTBURST PREVENTION FALLACIES

Reliance should never be placed on the prevention of failure as a means to avoid an outburst. Conditions will change and failure will be initiated at some point in the mining cycle.

If solid coal outbursts have not occurred then it is not safe to assume that outbursts from gouge materials will not occur.

The occurrence of small gas events or mini outbursts does not mean that a large event will not occur. This especially applies if the small events have been from solid coal. The potential that conditions will change enough to cause a change from a mini event to a major event is too high to risk.

The use of shotfiring is not a guarantee that outbursting will not occur (5). Fatal outburst have occurred when mining crews have returned to the face.

PRACTICAL METHODS TO HELP DETERMINE WHETHER AN OUTBURST WILL OCCUR

As far as I am aware there has never been an instance of an outburst in coal that has been successfully cored. Coring constitutes making an opening in the coal seam and extracting coal with significant vibration and a sudden pressure drop. Core loss or breakage may not indicate that the coal is outburst prone but it provides a first level check as to whether problems exist. Combined with gas content measurement it provides a good indicator of conditions. To make it reliable coring needs to be conducted continuously and adequate gas content measurements are required. The deficiency of any drilling technique is that it takes a sample only in the line of the borehole.

Low permeability may be detectable by simply taking core and sucking on it. If it is not possible to draw air through the core then it is a good indicator that the coal is tight and will not readily drain. Tight coals do however exist that this test will not work for. These are coals that exhibit high permeability without confinement but when stressed in seam have the cleats close up. The softer the coal the more prone it is to closing up under stress.

The failure of coal to drain at the face is a major concern. If gas is not being emitted in what is normally a gassy seam then it is an indicator that conditions have changed.

OUTBURST RESEARCH NEEDS

What is not needed in outburst research is extensive numerical modelling of what conditions will constitute failure of the material around an opening. Failure will occur at some time in the mining cycle. Neither is modelling of some theoretical dyke or high permeability zone required. Dykes are known to be impermeable barriers at times and must therefore restrict drainage. Faults may well act as channels supplying gas. No miner will ever know how permeable these are nor will they ever care. Numerically modelling what may be catastrophic events as a method of prediction is too fraught with possibilities of error to countenance.

What is needed is some fundamental research into the toughness of coal and how it will influence outbursting. This should be followed up by a practical means to assess toughness of coal in the mine.

Coal toughness testing can probably be combined with an assessment of diffusion coefficient. If a core is taken and broken by some test process to determine its toughness then it is quite easy to conceive of a test system that will measure the gas release rate on breakage and to be able to relate this to a diffusion coefficient.

The next most important aspect of research is to be able to detect gouge and its character. The detection of a fault is possible through drilling. I have no doubt that a torque and thrust sensor (6) on a rotary drill will detect faults. To be really useful though it is necessary to know what is in the fault. It is quite possible for a fault to hold either no gouge material or to hold coarse tough particles that would never lead to an outburst. Thus a fault itself is not necessarily a risk. Once again the torque and thrust sensor can probably enable the detection of fault infill.

The detection of the gas release characteristics of the gouge is also important. This can be achieved by examining the gas release characteristics of coal particles as they come out of the hole. Ideally this would be done by collecting such particles using the borehole pressurisation tool developed for this purpose (7).

Alternatively though simple chip collection and desorption rate measurement has proved useful in the past. The Hargraves' emission meter was a useful tool in predicting outburst risk at Collinsville No 2 mine and Metropolitan for many years.

The problem with such techniques is that they provide a measurement that is a combination of parameters. To obtain a clear assessment of a situation it is highly desirable to know individual parameters. For example is a high chip desorption rate due to high gas content, high diffusion coefficient or small particle size. Some rethink of these techniques using proper experimental and mathematical techniques would probably be rewarding.

The detection of low permeability zones must be considered a must for further work. I have for some time considered that it is possible to build a straddle packer system that could be used to detect low flow zones at any depth. The same system could also be used for pressure build up testing. Building it properly will cost money and testing time. Its use will also take time, at least as long as the hole originally took to drill.

Obviously remote systems to detect faults, gouge, high gas pressures, material strength and other factors would be highly desirable. The reality is though that it is extremely unlikely that any of these will reveal a sufficient information to predict outburst conditions within a five year development time.

It is almost as important to be able to deal with an outburst prone location as it is to know that it is a difficult area. Two methods deserve some development. The first is the continued development of hydrofracturing for degassing and the second is the use of remote mining techniques such as large (approximately 1 m) auger drills to get through difficult ground. Such tools can both de-stress and de-gas.

Hydrofracture has potential to degas. Not all coal is however amenable to hydrofracturing as it is may not possible to either seat the packers used to provide a seal or to avoid destroying packers.

THE WILL TO SUCCEED

Outbursting is a problem that has been in existence as long as coal mining. In Australia the early 1980's was a period of significant development of understanding of what constituted an outburst and of methods to deal with them. These methods basically centred around gas drainage. Since this period the main focus on outburst control has been in management strategies and improved drilling capability.

We have now reached deeper coals where the management approach will in itself not bring economic answers to the problem of outbursting. More research, development and implementation is needed. Much of this research need is well understood and could be completed quickly, the development is probably partly done but what is overwhelmingly lacking is implementation.

The coal industry is a small market for any development to be used in. The technology it requires is substantially unique. Coal mines are difficult places in which to work both physically and from a legislative viewpoint. Obtaining intrinsic safety approval is painful. Therefore why would any commercial equipment developer dream of entering this difficult market? The only answer is that it will pay him to do so. Not in possible sales in the distant future but now in terms of payment for effort.

Payment is not the only issue. Why build or do research for an industry which has no interest in using the outcome of the endeavours? This is especially an ACARP problem. Connecting projects to real desires of the industry as well as real needs is very important. It does not matter what need can be addressed, nor how successful the potential outcome might be if the industry is not committed to see that development through.

There is nothing more depressing than working to overcome all of the issues related to underground mining only to find that the industry will not even consider providing a real testing ground let alone a market. The question here is why work for the coal industry when there are other much bigger markets to be served?

Does the coal industry want to improve its capability to deal with outbursts?

REFERENCES

1. Biggam, F B, Robinson, B and Ham B, (1980). Outbursts ad Collinsville - A Case Study. The Occurrence, Prediction and Control of Outbursts in Coal Mines Symposium, September 1980. The AusIMM.

2. Moore, Rodney D, and Hanes John, (1980). Outbursts ad Collinsville - A Case Study.

3. Gray, Ian (1980). Overseas Study of Japanese Methane Gas Drainage Practice and Visits to Coal Research Centres, June - August 1980. Australian Coal Industry Research Laboratories Ltd, Published Report 80-15.

4. Davies, A W, (1980). Available Defences Against Outbursts in the United Kingdom in 1980. The Occurrence, Prediction and Control of Outbursts in Coal Mines Symposium, September 1980. The AusIMM.

5. Suchodolski, Zhigniew and Hardygora, Monika (1995). Characteristics of Coal, Rock and Gas Outburst Hazards in Polish Coal Mines. Int. Symp. cum Workshop on Management and Control of High Gas Emission and Outbursts, Wollongong, 20 -24 March, 1995.

6. Gray, Ian (1997). Development of a Torque Thrust and RPM Sensor. ACARP Project C3070.

7. Gray, Ian (1998). Borehole Pressurisation System. ACARP Project C3072