International Journal of Coal Geology 35 Ž1998.

27–55

Instantaneous outbursts in underground coal mines:

An overview and association with coal type
a,) b
B. Basil Beamish , Peter J. Crosdale
a
Department of Geology, The UniÕersity of Auckland, PriÕate Bag 92019, Auckland, New Zealand
b
Department of Earth Sciences, James Cook UniÕersity, TownsÕille, Qld 4811, Australia

Received 18 December 1996; accepted 16 May 1997

Abstract

Instantaneous outbursts in underground coal mines have occurred in at least 16 countries,

involving both methane ŽCH 4 . and carbon dioxide ŽCO 2 .. The precise mechanisms of an
instantaneous outburst are still unresolved but must consider the effects of stress, gas content and
physico-mechanical properties of the coal. Other factors such as mining methods Že.g., develop-
ment heading into the coal seam. and geological features Že.g., coal seam disruptions from
faulting. can combine to exacerbate the problem. Prediction techniques continue to be unreliable
and unexpected outburst incidents resulting in fatalities are a major concern for underground coal
operations. Gas content thresholds of 9 m3rt for CH 4 and 6 m3rt for CO 2 are used in the Sydney
Basin, to indicate outburst-prone conditions, but are reviewed on an individual mine basis and in
mixed gas situations. Data on the sorption behaviour of Bowen Basin coals from Australia have
provided an explanation for the conflicting results obtained by coal face desorption indices used
for outburst-proneness assessment. A key factor appears to be different desorption rates displayed
by banded coals, which is supported by both laboratory and mine-site investigations. Dull coal
bands with high fusinite and semifusinite contents tend to display rapid desorption from solid coal,
for a given pressure drop. The opposite is true for bright coal bands with high vitrinite contents
and dull coal bands with high inertodetrinite contents. Consequently, when face samples of dull,
fusinite- or semifusinite-rich coal of small particle size are taken for desorption testing, much gas
has already escaped and low readings result. The converse applies for samples taken from coal
bands with high vitrinite andror inertodetrinite contents. In terms of outburst potential, it is the
bright, vitrinite-rich and the dull, inertodetrinite-rich sections of a coal seam that appear to be
more outburst-prone. This is due to the ability of the solid coal to retain gas, even after pressure
reduction, creating a gas content gradient across the coal face sufficient to initiate an outburst.

)
Corresponding author.

0166-5162r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved.

PII S 0 1 6 6 - 5 1 6 2 Ž 9 7 . 0 0 0 3 6 - 0
28 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

Once the particle size of the coal is reduced, rapid gas desorption can then take place. q 1998
Elsevier Science B.V.

Keywords: instantaneous outbursts; coal mine; gas emission; coal lithotype; differential desorption; gas content
gradient

1. Introduction

Two gases are predominantly associated with coal seams, methane and carbon
dioxide. Methane is generated as a result of the coalification process ŽJuntgen and
Karweil, 1966a,b.. Carbon dioxide is often introduced to the coal seam as a result of the
presence of igneous intrusions ŽStutzer, 1936; Smith and Gould, 1980., which may or
may not be in contact with the coal seam. Both gases pose a hazard when encountered in
sufficient quantities in the coal seam. Methane mixtures are explosive in the range of
5–15% in air. Carbon dioxide is not a life supporting gas and at concentrations above
1–2% in air, it begins to have major detrimental physiological effects.
Fundamental investigations concerning the gases found in coal seams can be related
back to the earliest coal mining in Europe Žvon Meyer, 1872, 1873; Thomas, 1876;
Fischer et al., 1932.. The primary concern has always been safety, whether to assess gas
content and composition for: Ž1. ventilation purposes in order to reduce the hazard of
methane emissions and possible ignitions ŽCurl, 1978; Dunmore, 1981. or Ž2. indica-
tions of outburst-proneness ŽHargraves, 1958; Jackson, 1984.. This work is continuing
as high production operations must combat large quantities of gas into workings
ŽHargraves, 1993., and the complexity of the gasrcoal system is recognized ŽLevine,
1992..
The discharge of gas from coal may take place in three ways: Ž1. it may flow evenly
from the pores and fractures of the coal, Ž2. it may escape from the coal in the form of
more or less persistent ‘blowers’, issuing from some fractures or major seam disruptions
or Ž3. it may burst out suddenly in great quantities into the mine workings.
The two latter cases are rare, but the most dangerous to deal with in underground
workings and are generally referred to as outburst phenomena. The outburst problem
arises from the effects of three main factors: stress, gas content and physical and
mechanical properties of the coal. An unfavourable combination of these factors with
mining methods can lead to a recipe for disaster if not recognized at an early stage of
mine development.
The global threat that the outburst problem continues to pose to the future of
underground mining was acknowledged by the ‘‘symposium-cum-workshop on the
management and control of high gas emissions and outbursts in underground coal
mines’’ held in Wollongong in March 1995 ŽLama, 1995.. This paper will use much of
the Australian experience to provide a summary of old and new concepts on instanta-
neous outbursts in coal mines. In particular, due to the banded nature of the Bowen and
Sydney Basin coals of Australia, more emphasis will be placed on coal lithotype effects
where these have been recognized and documented.
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 29

2. Instantaneous outburst phenomena

Violent ejections of coal and gas from the working coal seam have plagued
underground mining operations for over a century. These phenomena are referred to as
instantaneous outbursts and have been reviewed by Hargraves Ž1958, 1980, 1983, 1993..
Such outbursts range in size from a few tonnes to thousands of tonnes of coal with
corresponding gas volumes from tens of cubic metres to hundreds of thousands of cubic
metres. Gas composition is predominantly methane, carbon dioxide or their mixtures.
Carbon dioxide outbursts tend to be more violent, although there is the added risk of a
subsequent explosion accompanying a methane outburst. Fatalities continue to occur
from instantaneous outbursts Že.g., West Cliff Colliery, New South Wales, 1994. and the
phenomenon is not totally understood with respect to the gasrcoal system.
A classical instantaneous outburst event resembles someone opening the top of a
shaken carbonated soft drink. The coal is often in pulverized form and appears to flow,
although some incidents are merely large face slumps or floor heaves and associated gas
release. A cavity results and in the vicinity of the outburst a gas haze is often seen
associated with temperature differential andror layering and refractive indices.
Noack et al. Ž1995. reported that: ‘‘The German guidelines of the Chief Mines
Inspectorate on gas outbursts differentiate between four types of occurrences, namely
outbursts of gas and coal, outbursts of gas and rock, heavy floor gas emissions, and
other sudden liberation of major amounts of gas’’.

3. Instantaneous outburst occurrence

3.1. Global experience

Jackson Ž1984. gives an historical account of the global incidence of instantaneous

outbursts. His tables have been modified and updated ŽTables 1 and 2., to reinforce the
extent of the problem in underground coal operations around the world.

3.2. Australian experience

In Australia there has been considerable research related to gas in coals. Several
symposia have presented Australian research results and related overseas experiences
with respect to this topic ŽLama, 1995.. Summaries of instantaneous outburst occur-
rences and consequent fatalities in Australian coal mines ŽTables 3 and 4. show both the
major coal producing Bowen and Sydney Basins have suffered from this phenomenon at
depth. However, instantaneous outbursts in the Bowen Basin have occurred at much
shallower depths ŽTable 3.. The violence often associated with carbon dioxide outbursts
is highlighted by the corresponding number of fatalities ŽTable 4., which continue to
occur despite many years of research into the problem.
Instantaneous outbursts in the Bowen Basin have been nonexistent since the early
1980’s, primarily due to the lack of mine development in gassy areas. However, as Peter
Allonby, general mine manager at Broken Hill Proprietary Co. Ltd. ŽBHP. Appin
30 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

Table 1
Outburst-prone areas of the world Žin part from Jackson, 1984.
Country Area Numbers Gas
of involved
outbursts
Australia Sydney and Bowen Basins over 650 to 1995 CO 2 rCH 4
Belgium Charleroi–Mons 1,190 from 1956 to 1963 CH 4
Bulgaria Balkans 105 to 1974 CH 4
Canada AlbertarBritish Columbia over 400 to 1995 CH 4
rNova Scotia
China widespread over 14,000 to 1995 CH 4
Commonwealth Donetz Basin over 3,500 to 1995 CH 4
of Independent States
Czechoslovakia Ostrava–Karvina 279 to 1974 CO 2 rCH 4
France Cevennes Basin 6,245 from 1899 to 1964 CO 2 rCH 4
Germany North Rhine–Westphalia 338 from 1970 to 1993 CH 4
Hungary Pecs Basin 565 to 1982 CH 4
Japan widespread 1,000 from 1925 to 1964; CH 4
21 from 1970 to 1980 CH 4
Poland Lower Silesia over 2,000 to 1995 CO 2 rCH 4
South Africa Karoo Basin several from 1993 to 1994 CH 4
Turkey Zonguldak Coalfield 57 from 1962 to 1993 CO 2 rCH 4
UK West Wales 250 from 1907 to 1981 CH 4
USA Colorado a few since mid-1970’s CH 4

Table 2
Major outburst incidents Žin part from Jackson, 1984 and Hargraves, 1980.
Country Year Area Mine Outburst Outburst Fatalities
coal Žt. gas
Žm3 .
Australia 1954 Bowen Basin Collinsville State 500 14,000 CO 2 7
Mine
Australia 1991 Sydney Basin South Bulli 120 3
Belgium 1879 Agrappe 340,000 CH 4 141
Canada 1904 British Columbia No. 1 Morrissey 3,500 140,000 CH 4 14
China Sichuan Province Tianfu 12,780
Commonwealth 1969 Donbass Yu.A. Gagarin 14,000 250,000
of Independent States
France 1938 Ricard 1,270 400,000 CH 4 2
Germany 1981 Westphalia Ibbenbueren 750 21,240 8
Hungary 1957 Istvan 1,400 273,000 CH 4
Japan 1981 Hokkaido Yubari–Shin 4,000 Žm3 . 600,000 CH 4 93
Poland 1930 Silesia Wenceslaus 5,000 28,000 CO 2 151
Poland Silesia Nowa Ruda 5,000 800,000 CO 2
Turkey 1992 Zonguldak Kozlu 263
Coalfield
UK 1971 West Wales Cynheidre 400 60,000 CH 4 6
rPentremawr
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 31

Table 3
Summary of Australian outburst occurrences Žin part from Hargraves, 1983 and Harvey, 1995.
Colliery YearŽs. Seam Depth Žm. Number Maximum size Žt. Gas
Sydney Basin
Appin 1969–1995 Bulli 520 11 88 CH 4
Brimstone 1995 Bulli 2 30 CO 2
Bulli 1972 Bulli 380 1 100 CH 4 & CO 2
Coal Cliff 1961 Bulli 450 2 2 CH 4 & CO 2
Corrimal ŽCordeaux. 1967–1995 Bulli 400 4 50 CH 4 & CO 2
Kemira 1980 Bulli 220 2 100 CO 2
Metropolitan 1895–1995 Bulli 400–450 40 300 CO 2 minor CH 4
North Bulli 1911 Bulli 370 1 1 CH 4
South Bulli 1991–1995 Bulli 400 7 120 CO 2
Tahmoor 1981–1985 Bulli 410 88 400 CO 2
Tower 1981–1995 Bulli 480 19 80 CH 4
West Cliff 1977–1995 Bulli 480 250 320 CH 4

Bowen Basin
Collinsville State Mine 1954–1961 Bowen 215–235 13 500 CO 2
Collinsville No.3 Mine 1972 Bowen 230 2 1 CO 2
Collinsville No.2 Mine 1978–1981 Bowen 250–265 7 35 CO 2
Leichhardt 1975–1982 Gemini 370 200q 500 CH 4
Moura No.4 Mine 1980q C upper 130 2q 20 CH 4

Colliery points out Žquoted in Roberts, 1995.: ‘‘Seams in Queensland are becoming
deeper and are known to be as gassy and have similar structures to those commonly
associated with the Bulli seam outbursts. Outbursts have been recorded at Leichhardt
Colliery and at Collinsville and as other mines in other areas become deeper they should
also consider the risk’’. This would appear sound advice given the history of instanta-
neous outbursts in Australia, and the temptation to become complacent once the
incidence of events diminishes.

Table 4
Summary of Australian fatal outbursts Žin part from Harvey, 1995.
Colliery Year Fatalities Size Žt. Gas
Sydney Basin
Metropolitan 1896 3 unknown CH 4
Metropolitan 1925 2 140 CO 2
Metropolitan 1954 2 90 CO 2
Tahmoor 1985 1 400 CO 2 and CH 4
South Bulli 1991 3 120 CO 2
West Cliff 1994 1 300 CO 2

Bowen Basin
Collinsville State Mine 1954 7 500 CO 2
Leichhardt 1978 2 350 CH 4
32 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

4. Instantaneous outburst mechanisms

Many models exist to explain the occurrence of instantaneous outbursts from coal.
Generally, consideration is given to components of gas content and flow, stress, and coal
failure. Kidybinski Ž1980. proposed the presence of three zones in the coal seam ahead

Fig. 1. Coal face conditions in an outburst zone Žfrom Williams and Weissmann, 1995..
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 33

of the mining operations starting at the coal face: Ž1. protectionrdegassed zone, Ž2. high
gas pressureractive zone and Ž3. abutment pressure zone.
Within this model, three fundamental conditions are assumed to be met for an
outburst to occur: Ž1. failure of the coal in compression within the active zone, Ž2.
penetration of a hole through the protection zone and Ž3. fluidized bed outflow of the
products from outburst cavern.
Gray Ž1980. considered two gas-initiated-failure mechanisms to exist, either tensile
failure of unconfined coal or piping of sheared material. Paterson Ž1986. took the
general view that when gas is released from coal there are body forces on the coal equal
to the pressure gradients of the flowing gas. His models therefore were based on the
fundamental assumption that an outburst is the structural failure of coal due to excess
stress resulting from these body forces.
A model proposed by Litwiniszyn Ž1985. was based on the gas existing in a
condensed state within the coal. When a shock wave passes through the coal a phase
transformation occurs of the liquid substance into a gaseous state. This sudden creation
of gas causes the skeleton of the medium to be destroyed and an outburst is initiated.
Support for this model is found in the following observations: Ž1. sometimes ‘bumps’
and instantaneous outbursts occur together, and some ‘bumps’ are regarded as initiation
of instantaneous outbursts, and Ž2. in hand-working, especially without noise of machin-
ery, successive knocks in the coal were often precursors to an instantaneous outburst
ŽHargraves pers. comm., 1997.. However, Paterson Ž1986. identified several flaws in
this model, in particular cause and effect; where do the shock waves originate?
Thermodynamic descriptions have also been proposed for outburst modelling ŽJagiełło
et al., 1992.. In this case the premise is liberation of gas, upon entering a coal seam, is
connected to a decrease in temperature of the system. As a result of the work performed
by the gas, the internal energy of the system decreases. The gas contained in the coal has
some ‘store’ of work, which can be used in an outburst process.
Williams and Weissmann Ž1995. used a schematic of an outburst in frequently
encountered Australian conditions ŽFig. 1. to discuss gas content thresholds for out-
bursts. They placed emphasis on a gas pressure gradient existing ahead of the working
face. However, they also believed ‘‘the most important parameter is gas desorption rate,
in conjunction with the gas pressure gradient ahead of the face’’.

5. Instantaneous outburst assessment

5.1. Emission rates and gas contents

The majority of gas emission research work has been applied to longwall mining with
methane as the dominant coal seam gas, although in recent times the presence of carbon
dioxide in Australian coal mines, particularly the Sydney Basin, has been an increasing
area of research interest. Gas make in longwall systems appears to be closely linked with
coal face production and face advance rate ŽAirey, 1971; Klebanov, 1971; Myszor,
1974; Cybulski and Myszor, 1974; Noack, 1976.. In bord and pillar workings, increase
in rate of advance also appears to correlate with increased gassiness in the working face
34 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

ŽHargraves et al., 1964. and conversely any period of face delay results in a decrease of
gassiness in the working face ŽBeamish, 1990.. The presence of any form of coal seam
gas predrainage similarly impacts to reduce the seam gassiness ŽBeamish, 1990..
However, caution is required when assessing the effects of these gas reduction measures
as unsurveyed boreholes can deviate markedly from their intended path, leaving areas of
undrained coal. In-seam gas drainage may not be uniform through the coal seam profile
or in lateral extent.
General gas assessment work in the Sydney and Bowen Basins of Australia is
routinely performed by measuring gas contents of all coal seams as input to underground
mine planning of ventilation. An Australian gas content measurement standard ŽAnon,
1991. identifies three components to evaluate the total desorbable gas content Ž Q TD .:
Q TD s Q1 q Q 2 q Q 3 , where Q1 is the lost gas content, Q2 the measurable gas content
and Q3 the residual gas content. Desorbable gas content Ž Q D . is the sum of lost gas
content Ž Q1 . and measurable gas content Ž Q2 ..
Regional assessment of outburst-prone areas has been conducted in Australia, but
projections have been confined to deep mines with high rank bituminous coal. The most
common parameter used on a regional level is the gas content of the coal. Originally,
threshold values were determined as 9 m3rt for methane and 5 m3rt for carbon dioxide,
based on empirical experience from operations in Australia and Germany ŽBeamish,
1984.. However, these limits are constantly under review, particularly for mixed gas
situations, and may vary from one coal mine to another ŽWilliams and Slater, 1995..
Truong and Williams Ž1989. highlighted the variations that occur in gas content and
composition of Australian coals. They concluded the processes controlling the eventual
in situ gas content of coal are highly complex and cannot be explained by simple
geologic factors.
The New South Wales Department of Mineral Resources has implemented a set of
standard practices for dealing with underground coal mining, which require gas contents
to be known in advance of any mining ŽHarvey, 1995.. At the forefront of these
procedures is the quick crush gas content method ŽBeamish and Vance, 1990; Williams
and Weissmann, 1995. that is a modified version of the Australian Standard ŽAnon,
1991.. In this procedure a fast and reliable result is obtained that is supplied to mining
operators to decide on a safe mining procedure for the conditions established.

5.2. Seam gas composition

Generally, coal adsorbs about twice as much pure carbon dioxide as it does pure
methane. A range of intermediate isotherms are obtained for mixed gas composition,
with a regular increase in total gas content occurring as the percentage of carbon dioxide
increases. Consequently, for a given coal seam gas pressure, larger volumes of carbon
dioxide can exist and emission problems therefore become more acute. Surface area
measurements also indicate that vitrinite-rich coals have a larger carbon dioxide sorption
capacity than their inertinite-rich equivalents ŽBeamish and O’Donnell, 1992.. Similar
results have been obtained for methane in recent studies ŽLamberson and Bustin, 1993..
Greaves et al. Ž1993. reported on a mixed gas sorption experiment using an initial gas
composition of 75% CH 4r25% CO 2 , which resulted in a final adsorbed gas composition
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 35

Fig. 2. Methane sorption history for an initial adsorption gas composition of 75% CH 4 r25% CO 2 . Adsorbed
gas composition at the final pressure of 7.1 MPa was 28% CH 4 r72% CO 2 Žfrom Greaves et al., 1993..

at a pressure of 7.1 MPa of 28% CH 4r72% CO 2 . Subsequent desorption resulted in the

following: Ž1. methane was released relatively quickly as pressure falls and the methane
content of the coal drops rapidly ŽFig. 2. and Ž2. significant desorption of carbon dioxide
did not occur until pressures were reduced to less than 0.7 MPa ŽFig. 3.. These results
support the observed strong affinity of coal for CO 2 compared to CH 4 . The delay in
release of the carbon dioxide until low pressures are achieved has implications for gas
emissions and outbursts in coal seams containing carbon dioxide.

Fig. 3. Carbon dioxide sorption history for an initial adsorption gas composition of 75% CH 4 r25% CO 2 .
Adsorbed gas composition at the final pressure of 7.1 MPa was 28% CH 4 r72% CO 2 Žfrom Greaves et al.,
1993..
36 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

5.3. Mining and geological factors

Mining methods are important considerations in assessing the risk of instantaneous
outbursts. These methods may be categorized in order of decreasing outburst-proneness
ŽHargraves, 1993.: Ž1. cross measuring into the seam, Ž2. development heading in the
seam, Ž3. longwall advancing, Ž4. longwall retreating and Ž5. pillar extraction.
Cross measuring into a coal seam is given the highest rating as mining is advancing
from a more competent material to a less competent material. Upon intersecting the coal
seam, the confining pressure of the surrounding rock is suddenly removed, and if gas is
present in the seam in sufficient quantity it can be released very rapidly.
Heading advance creates a situation of atmospheric conditions at the working face
with much higher virgin gas pressures only a short distance ahead. Encountering any
coal seam weakness or disruption therefore can be catastrophic, as again confinement of
the coal seam is seriously diminished. Longwall advance suffers in a similar manner to
development headings, although the longwall rate of advance is usually less and more
gas is able to drain away naturally.
Mining by longwall retreat is often performed through an area already drained by
natural drainage into the roadways used to delineate the longwall block and perhaps
in-seam drainage boreholes, thus reducing the risk of outbursts. Nevertheless, gas
emission into the coal face area can still pose a problem in longwall retreat due to the
high coal outputs achieved with this technique. Again, this can be adequately dealt with
by well planned in-seam drilling and drainage. Gas problems with pillar extraction are
nonexistent as the pillars have drained naturally into the surrounding roadways after
development.
Seam disruptions are a common feature associated with outburst phenomena ŽShepherd
et al., 1981., however, faulting is not a prerequisite for instantaneous outbursts
ŽHargraves, 1993.. Fault types may be categorized in order of proneness as strike-slip)
reverse ) normal ŽShepherd et al., 1981., although this concept has been refuted
ŽHargraves, 1993.. Other features such as palaeochannels and floaters in the coal seam
also cause disrupted Žsheared. coal to be present due to differential compaction. Dykes
cutting a seam may provide conditions resembling cross measuring according to
Hargraves Ž1993..
Only the high rank bituminous coals have been considered outburst-prone from
Australian experience Žhigh volatile bituminous A–low volatile bituminous.. However,
instantaneous outbursts are not impossible in low rank coals, with occurrences recorded
at Valenje Colliery, Yugoslavia ŽHargraves, 1983.. ‘Gas blowers’ have been reported for
New Zealand subbituminous to high volatile bituminous B coals ŽPatterson et al., 1967..
5.4. Summary of sorption properties of coal
Adsorption isotherms for two coals are used to illustrate the effects of rank, moisture
content, and coal lithotype ŽFig. 4.. Sample ERDC7 is a medium volatile bituminous
coal from the Bowen Basin ŽAustralia. and sample 52r085 is a subbituminous coal from
the Huntly Coalfield ŽNew Zealand.. On a raw coal basis, ERDC7 bright coal has the
highest sorption capacity in the dry state. Based on this result, a higher gas content could
be expected for bright coals in situ than dull coals.
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 37

Fig. 4. Methane isotherms for a Bowen Basin medium volatile bituminous coal ŽERDC7., and a Huntly
Coalfield subbituminous coal Ž52r085..

The most significant effect on sorption capacity is the level of moisture present in the
coal. For an increase in moisture from 0.3% to 1.1%, the ERDC7 dull coal sample has
its sorption capacity reduced from 25.1 ccrg to 19.5 ccrg at 5.1 MPa. Sample 52r085
has its sorption capacity reduced from 22.6 ccrg to 6.5 ccrg for an increase in moisture
from dry state to 11.6% at a similar pressure. Adsorption capacity is normally said to
increase with rank, but it is also strongly influenced by the moisture content. High rank
coals have low moisture and therefore there is less competition between gas and
moisture for sorption sites.
Desorption rates of bright and dull coal types in the same seam differ, with the
primary cause attributable to the maceral composition of each type ŽBeamish and
Crosdale, 1995; Fig. 5.. Bright coal bands rich in vitrinite are predominantly micro-
porous and their slow desorption rate is generally described by a unipore diffusion
model ŽSmith and Williams, 1984.. An example of this is ERDC5 bright coal in Fig. 5.
Dull coal bands rich in fusinite and semifusinite, with minor amounts of vitrinite have a

Fig. 5. Typical desorption data for Bowen Basin coal lithotypes with diffusion model curves superimposed.
38 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

large degree of macroporosity and their fast desorption rate is generally described by a
bidisperse pore diffusion model ŽRuckenstien et al., 1971.. An example of this is
ERDC1 dull coal in Fig. 5. Dull coal bands containing the inertinite maceral inertodetri-
nite behave similarly to bright coals. An example of this is ERDC5 dull coal in Fig. 5.
The different desorption rates displayed by dull and bright coal bands have serious
implications for face emissions and gas contents, in-seam gas drainage, and outburst-
proneness of a coal seam.

5.5. Moisture effects on gas content in the working face

Coal in the vicinity of an instantaneous outburst is often reported as being dry

ŽHargraves, 1993.. Isotherm data show, for a given pressure, coal adsorbs more gas
when in a low moisture state and consequently, gas migrating to the exposed, dry coal
face may become readsorbed in the immediate face zone. Gas contents may be thereby
elevated, helping to explain the larger than expected volumes of gas released in an
outburst. The existence of such a mechanism would be most important in coals of low
rank, which have high in situ moisture contents and which are known to dry out very
readily. A pressure lag effect exists along similar lines when draining the coal. Low rank
subbituminous coals with their high moisture content have high pressure gradients for a
given gas content. Consequently, gas content threshold limits for outbursts in high rank
coals do not directly transfer to low rank coals.

5.6. Regional effect of Õitrinite content on coal seam gassiness

Distinct relationships exist between the coal rank and instantaneous outbursts. At No.
2 Mine, Collinsville ŽBowen Basin., outbursts only occurred in coal with greater than
1.2% vitrinite reflectance ŽWilliams and Rogis, 1980.. Here it was noted that the
vitrinite content of the mining section Žthe top 2.5 m to 2.8 m of the coal seam.
increased from 28%, 200 m east of the Western Panels Fault ŽFig. 6. to 48% near the
fault. Towards the fault, vitrinite content increased to 63% in the lower half of the
working coal section but remained a constant 26% in the upper part.
Comparison of Hargraves emission value ŽEV; Hargraves, 1962. measurements along
a profile of the Western Panels area with vitrinite reflectance and vitrinite content ŽFig.
7., show a clear relationship between vitrinite content and gassiness of the coal. Such a
correlation would be expected since laboratory data suggest vitrinite-rich coal has a
greater propensity to retain its gas. Face cut-out cycle ŽFCC. emission profiles for
roadways approaching the Western Panels Fault ŽBeamish, 1990., show a response
similar to the EV readings. High gas emissions were recorded during face cutting of
vitrinite-rich coal sections. Additionally, the vitrinite-rich coal was often highly sheared,
allowing a much finer particle size to be produced during cutting and resulting in high
emissions as the vitrinite content increased approaching the fault. During prolonged
periods of face stoppages and slow advancement of roadways, the sheared coal drained
naturally, because of its fine particle size and significant interparticle porosity and
permeability.
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 39

Fig. 6. Mining layout for Western Panels area, No. 2 Mine, Collinsville Žfrom Beamish, 1990..

5.7. Coal lithotype effect on emission Õalues across a working face

EV readings from a full face exposure in the Western Panels area of No. 2 Mine,
Collinsville ŽBiggam et al., 1980. have been redrawn and contoured ŽFig. 8. to

Fig. 7. Profiles of vitrinite reflectance, vitrinite content and EV readings approaching the Western Panels Fault,
No. 2 Mine, Collinsville.
40 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

Fig. 8. Hargraves EV readings taken across a working coal face Žmodified from Biggam et al., 1980..

emphasize the significance of the results in relation to coal lithotype variations in the
face. EV values ranged from 0.43 to 1.18 ccrg, and displayed a preferential increase
from the top of the working coal section to the bottom of the working coal section. Coal
lithotypes in the upper half are dominated by dull coal and the lower half by bright coal,
which is reflected in their relative vitrinite contents ŽWilliams and Rogis, 1980.. The
upper and lower coal sections are separated by a bedding plane fault.
In a free face environment, according to the desorption results presented in Fig. 5,
rapid initial desorption would have occurred in the dull, inertinite-rich coal seam section
prior to retrieving the EV sample. Gas release after retrieving the EV sample would be
on the slow part of the desorption curve. In contrast, the bright, vitrinite-rich part of the
coal seam would have retained most of its gas and would rapidly release it on reducing
the particle size as is the case with the EV reading procedure. This same differential
desorption phenomenon was well recognized by the face crews at Collinsville, and it
was common knowledge that taking an EV reading in the dull coal produced a lower EV
than in the bright part of the seam. The significance of this ‘home truth’ was not fully
understood at the time and was often mistaken for implying the dull coal was
impermeable in comparison to the bright coal. Hargraves Ž1962., acknowledged that
coal lithotype may make a difference to EV readings. He recommended that EV
readings be taken from duro-clarain or in a consistent coal band for direct comparison.
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 41

A second trend is also observable in Fig. 8. The EV readings progressively decrease

away from the virgin rib side, which is again more noticeable in the upper part of the
working coal section. High bedding-plane permeability ŽBartosiewicz and Hargraves,
1985. is reflected by the EV readings.
High EV readings are also observed when nearing a hole-through to complete pillar
formation ŽBiggam et al., 1980.. High stress in this region reduces the coal’s permeabil-
ity and retards gas drainage. Hargraves Ž1969. showed increase of gas pressure in this
region due to increase of stress. As the hole-through is completed, removal of the
confining stress and particle size reduction result in larger than expected gas release.
Desorption measurements of lump samples taken from coal faces ŽCreedy, 1986;
Lama, 1986; Ashton, 1990. also display a spread of values, which has been previously
unexplained. Creedy Ž1986. has formulated a statistical method for analysing this data to
obtain a representative in situ gas content and he recommends using the bright coal
samples from a working face.
Gas emission rate into the ventilating system will also be affected by differential
desorption and is reflected in face monitoring ŽBeamish, 1990.. Emissions are more
erratic throughout a gas-drained panel compared to an undrained panel, which is
expected if differential desorption occurs in response to the in-seam drainage boreholes.
Localized changes in specific gas emissions may also result from a change in the coal
seam composition with respect to banding. The effect may be accentuated by the
differing storage capacities of the coal lithotypes. A predominantly dull coal seam rich
in fusinite and semifusinite with large open macropores would be expected to produce
high general body emissions of gas. A predominantly bright coal seam rich in vitrinite
would produce the opposite effect. However, the bright coal seam would produce high
emissions of gas during periods of face cutting as the particle size of the coal is reduced
enabling the retained gas to be liberated.

Fig. 9. Relationship between uniaxial compressive strength Žas determined from NCB cone indenter measure-
ments. and vitrinite reflectance of Bowen Basin coals.
42 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

5.8. Coal strength

A number of strength tests on small scale coal specimens from the Bowen Basin
using a uniaxial compression testing machine and a National Coal Board ŽNCB. cone
indenter were reported by Beamish et al. Ž1991.. Coal strengths of dull and bright coal
bands and the deformation behaviour they display under loading parallel and perpendic-
ular to the banding were investigated. Microstructures developed during uniaxial com-
pression testing were examined by scanning electron microscopy ŽSEM..
NCB cone indenter numbers have been converted to uniaxial compressive strength
and plotted against coal rank ŽFig. 9.. Limited bright coal data is due to the tendency for
the samples to break under point load in the tester, presumably along the finer
microcleats observed by SEM ŽGamson and Beamish, 1991.. Uniaxial compressive

Fig. 10. Axial deformation in dull and bright coals from the Bowen Basin.
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 43

strength decreases substantially as rank increases from high volatile bituminous A to low
volatile bituminous ŽFig. 9.. The bright coal sample stressed perpendicular to banding
appears to have a lower strength than its dull rank equivalent, whereas the opposite is
true when the sample is stressed parallel to banding. Of the high volatile bituminous
coals tested, the lowest strength is displayed by the sample rich in inertodetrinite
ŽERDC5 dull..
Uniaxial compression testing results ŽFig. 10. show dull coals loaded parallel and at
right angles to the banding failed at around 27 MPa. Similarly, bright coals loaded
parallel to the banding Žright angles to the cleat network. failed at 29 MPa. In
comparison, load applied to bright coals at right angles to the banding Žparallel to the
cleat network. created failure at significantly low pressures of 11 MPa. Typically, the
microfractures that have developed under stress in the bright coal bands do not
propagate into the surrounding dull coals or clays. This competency contrast has
influenced the way in which microfractures have propagated through coal bands in
response to an applied stress.
When the coal strength information is combined with the sorption behaviour of coal
macerals ŽCrosdale and Beamish, 1995., it can be seen that the outburst-proneness of
coals rich in vitrinite and inertodetrinite is greatly increased. Basically, these coal
lithotypes will retain their stored gas for long periods of time in the face environment
and when failure of the coal occurs, which is more likely in these coal lithotypes due to
their low strength, the small block sizes produced will release large quantities of gas
very rapidly.

6. Prediction and alleviation

6.1. Gas emission indices for outburst-proneness assessment

The D P index ŽEttinger et al., 1958., based on the initial rate of gas desorption from
coal, has been widely adopted in Europe and elsewhere. Coals with high initial
desorption rates are considered prone to instantaneous outbursts ŽLidin et al., 1954.. The
Australian experience indicates conflicting results, with mines having a history of
instantaneous outbursts giving low D P values. Striking differences were also found
between D P values for coals of the Sydney and Bowen Basins, which were otherwise
analytically identical ŽBartosiewicz and Hargraves, 1985.. This index makes no al-
lowance for differences in emission rates of inertinite- and vitrinite-rich coals, which
could account for the differences between Bowen and Sydney Basin coals. Therefore,
D P does not appear to be a reliable index of proneness to instantaneous outbursts for
Australian coals ŽBartosiewicz and Hargraves, 1985..
The L2 index ŽLama, 1980. is based on the gradient of the desorption curve between
the interval 30 and 600 s, using a log–log plot. This value is analogous to effective
diffusivity as defined by Smith and Williams Ž1984.. Solid dull coal gives the best
response, with little differentiation between crushed Žy0.5 q 0.25 mm. bright and dull
coals. It was inferred that the dull coal could be used in solid form to predict
outburst-prone conditions. However, Beamish Ž1996. suggests that the desorption ob-
44 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

served by Lama Ž1980. is more the norm for the dull coal, which contains the inertinite
macerals fusinite and semifusinite. Therefore, caution must be used when trying to
implement desorption indices for outburst prediction in Gondwanan coals. The role of
macerals should not be overlooked and the blanket use of fast desorption being more
outburst-prone needs to be taken into context with the size fraction of the sample used to
obtain the result.
Hargraves EV readings are a raw coal measurement, which greatly exaggerates the
difference between dull and bright coals sorption capacities. Even if the dull and bright
coals have not partially desorbed prior to measurement, bright coal would give a high
result for Bowen Basin coals. The fact that dull coal desorbs faster than bright coal
increases this difference. Nevertheless, a high Hargraves EV reading works in terms of
outburst-proneness as it indicates the gas content of the coal is still high, and that the
coal also desorbs rapidly when particle size is reduced. Assessment of bright-coal
fractions would appear to be the best indicator for this index, a fact well known to the
experienced coal miners at No. 2 Mine, Collinsville.
The FCC emission of Beamish Ž1990. works as it is a direct reading of the coal in the
mining face, as long as driveage is not interrupted. In gas drainage areas care must be
taken to establish if the coal seam has segregated intervals of bright and dull coal. Low
readings may be a false sense of security as they may be a result of little gas in the dull
coal, but a large proportion of the original gas content still remaining in the bright coal.
Therefore, on a tonnage mined basis, the overall FCC value may appear low, when
specific intervals of the coal face may still be prone to outbursts.

6.2. In-seam gas drainage

Incremental flow differences in a horizontal borehole will exist as a result of the

borehole intersecting different coal bands. Most documented histories of gas flows from
in-seam boreholes refer to high gas flow zones associated with cleat direction and
structural disturbances. However, a more simple explanation of these regions can be
attributed to the changing coal seam lithology, which may be occurring both through the
seam profile or laterally. Consequently, a uniform drainage pattern will not exist away
from the borehole due to differential desorption of the coal bands.
When using in-seam drilling for gas drainage it is therefore necessary to ensure good
coal seam coverage and to assess the borehole location with respect to the coal
lithotypes. This means surveying the boreholes and preferably using directional drilling
with in-hole motors to ensure the coal seam is criss-crossed and no ‘shadow zones’
exist.

6.3. Water infusion for outburst-proneness reduction

Water infusion has been used as a method of controlling instantaneous outbursts in

some countries ŽJackson, 1984.. It is believed that as humidity or moisture increases, the
capability of the coal to accumulate elastic strain energy decreases and the permanent
non-recoverable strain energy increases. In this way Peng Ž1978. proposes that the
energy index of liability to outburst Wet decreases. Hargraves Ž1983. points out that
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 45

Fig. 11. Effect of water infusion on sorption capacity.

when the infusion is stopped, the process reverses and so the effects of infusion are not
long lasting.
The influence of water infusion could equally be explained using sorption isotherm
results. With an increase in the coal moisture state at high pressure, the water molecules
would begin to compete with the methane molecules for sorption sites and subsequently
displace them, hence lowering the gas content of the coal ŽFig. 11.. Hargraves Ž1983.
reported on longwall faces in the Commonwealth of Independent States, where high
pressure infusion had the effect of driving the gas from the coal with a whistling sound.
Once the water infusion pressure is removed, natural drainage would recommence
including loss of moisture. As seen from the isotherm ŽFig. 11., the gas content would
again increase for a given pressure.

7. Case history examples showing coal lithotype effects on instantaneous outbursts

7.1. Outburst potential of coal from Leichhardt Colliery, Blackwater

Gray Ž1980. used material from Leichhardt Colliery in his paper on energy release
associated with instantaneous outbursts. His desorption tests were conducted on three
types of coal: Ž1. normal outburst coal, D heading east, straddling the 3 m parting, Ž2.
brecciated material from the ribs of the December 1st, 1978 outburst and Ž3. mylonitized
Žsheared. material from the December 1st, 1978 outburst. These samples were subjected
to methane repressurisation followed by rapid gas pressure release, and the amount of
gas liberated was monitored. The mylonitized coal showed the largest volume of initial
gas released. This is not surprising as the details of the coal samples ŽTable 5. clearly
46 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

Table 5
Leichhardt Colliery coal samples tested by Gray Ž1980.
Normal coal Brecciated coal Mylonitized coal
Vitrinite reflectance Ž%. 1.24 1.23 1.28

Maceral composition Ž%.

Vitrinite 35 51 56
Exinite 1 0 1
Inertinite 59 43 38
Mineral matter 5 6 5

Size analysis Ž% retained.

q12.7 mm 31.0 7.2 12.3
6.35–12.7 mm 26.6 20.0 14.2
3.18–6.35 mm 18.1 23.0 16.2
1.00–3.18 mm 15.0 25.4 21.2
0.50–1.00 mm 4.8 10.8 11.8
0.25–0.50 mm 2.6 6.6 10.8
0.125–0.25 mm 1.0 3.1 6.1
y0.125 mm 0.9 3.9 8.0

show two things: Ž1. the brecciated and mylonitized coals contain a large percentage of
fine material, 13.6% and 24.9% less than 500 m m respectively, as opposed to 4.5% from
the normal coal; therefore, desorption rate is greater and Ž2. the brecciated and
mylonitized coals contain a large percentage of vitrinite, 51% and 56% respectively, as
opposed to 35% for the normal coal, therefore, the amount of sorbed gas is greater.
The coal profile in the vicinity of the outburst cavity is shown in Fig. 12. The layers
of vitrinite-rich, myolinitized coal are clearly shown and presumably were the main coal
component responsible for the outburst material. Another notable feature of this profile
is the presence of a very dull band of coal. Hunt and Botz Ž1986. made particular
reference to this band ŽFig. 12. in their paper on petrographic factors in outbursts in
Australian coal mines. This coal band was subjected to desorption rate testing on
crushed material, using the method of Janas and Winter Ž1977.. The coal displayed the
highest adsorption and desorption rates of coal lithotypes from the Leichhardt Colliery.
The major maceral component of the dull band was inertodetrinite.

Fig. 12. View of eastern rib December 1, 1978, outburst, Leichhardt Colliery, showing mylonite banding and
inertodetrinite-rich dull coal band Žfrom Gray, 1980..
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 47

At first, this result would appear to be in conflict with the laboratory results presented
in Fig. 5. However, this apparent difference is simply a result of particle size in the
desorption test method. In fact the two results are entirely consistent with outburst-prone
coal conditions. Namely at large particle sizes Že.g., in situ in the coal face. the
desorption rate of coal rich in inertodetrinite is slow, therefore gas is retained at a high
level. Once the particle size is reduced however, the retained gas is rapidly released.
Consequently, the dual presence of vitrinite-rich mylonitized coal and an
inertodetrinite-rich coal contributes to a much greater potential for outbursts. Once the
mylonitized coal breaks from the coal face bringing the inertodetrinite-rich material with
it, and particle size reduction occurs, the gas release is almost instantaneous.

7.2. Instantaneous outburst occurrence at No. 2 Mine, CollinsÕille

At 5.27 p.m. on the 16 April 1981, an instantaneous outburst ŽFig. 13., at No. 2
Mine, Collinsville released approximately 35 tonnes of coal and 380 m3 of carbon
dioxide ŽBeamish, 1981., in the belt road 29 m inbye of 11A cut-through ŽFig. 6.. A

Fig. 13. Gas analyzing system chart record for afternoon shift 16 April 1981, showing carbon dioxide outburst,
No. 2 Mine, Collinsville.
48 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

Fig. 14. Plan view of continuous miner location at the time of outburst No. 22, No. 2 Mine, Collinsville, with
face cross-section locations indicated.

plan view of the outburst face is shown in Fig. 14 with corresponding cross-sections in
Figs. 15 and 16. The origin of the outburst appeared to be from the floor line in the
centre of the roadway, 4.5 m inbye of the coal face. A rectangular prism-shaped cavity
resulted. Drillhole data placed the intersection of the Western Panels Fault ŽFig. 6. with
the mining section floor, approximately 5 m inbye of the outburst face. Consequently,
the disruption to the coal seam caused by the fault contributed to the outburst. As in all
previously recorded cases at Collinsville, the outburst material was extremely sheared
with a sugary texture.
Analyses of coal samples taken from the outburst location ŽTable 6. reveal that the in
situ sheared coal and outburst coal have: Ž1. high crucible swelling numbers and Ž2.
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 49

Table 6
Coal analyses from outburst locality, 51 Level West Panel, No. 2 Mine, Collinsville
Sample details Inherent Ash Volatile Fixed Crucible
moisture Ž%, matter carbon swell
Ž%. db. Ž%, Ž%, number
db. db.
Outburst coal ejected onto miner head 1.9 16.2 23.0 60.8 8
Coal from outburst cavity 1.1 16.5 22.0 61.5 8
Dull coal from right hand side of the face above slip plane 0.6 16.8 19.4 63.8 2
Sheared coal from right hand side of the face below slip plane 1.6 17.9 19.7 62.4 5

relatively high volatile matter content. Both these parameters are consistent with
vitrinite-rich coal, which matches the seam profile assessment.
The area being worked in the No. 2 Mine had been undergoing gas drainage to
reduce the gas levels, and a borehole was located in the dull, upper part of the coal seam
ŽFigs. 15 and 16.. As a result of the different desorption rates of the dull and bright coal
lithotypes, the upper part of the coal seam would have drained to a safe level, however,
the bright, lower part of the seam would not have. This is reflected in the proximate
analyses ŽTable 6.. The dull coal had a very low moisture content, which would be
expected after 5 months of draining from a borehole. The vitrinite-rich coal also displays

Fig. 15. Face cross-sections A–A and B–B from Fig. 14.
50 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

Fig. 16. Face cross–sections C–C and D–D from Fig. 14.

a high ash content, reflecting the effects of mineral matter infilling cleats. The presence
of this material would also contribute to the slow desorption rate of the bright coal, due
to clogging of the macropore system. A crucible swelling number of 5 from the right
hand side of the face below the slip plane, would tend to indicate the presence of banded
coal. Therefore, the low EV of 0.50 ccrg recorded at the time of face development from
this part of the coal seam ŽFig. 13. is consistent with fast desorption in situ from this
type of coal, and a low retained gas content.

7.3. Instantaneous outbursts and gas content gradient

Generally, diagrams of instantaneous outburst mechanisms are presented which place

emphasis on a gas pressure gradient existing in the coal face environment ŽFig. 1..
Williams and Weissmann Ž1995. used this same diagram when discussing gas content
thresholds for outbursts. However, they believed the most important parameter is gas
desorption rate, in conjunction with the gas pressure gradient ahead of the face. The
laboratory results of desorption rates shown in Fig. 5 support this idea. Effectively, in
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 51

Fig. 17. Face conditions for outburst No. 22, No. 2 Mine, Collinsville, indicating high gas content gradient in
lower part of the working section.

the face environment as pressure is taken off the coal it is the different desorption rates
of the coal lithotypes which create a large gas content gradient, namely: Ž1. vitrinite-rich
or inertodetrinite-rich coal bands do not desorb rapidly, retaining their gas and thus
producing a steep gas content gradient and Ž2. fusinite-rich or semifusinite-rich coal
bands lose their gas more rapidly and thus have a shallow gas content gradient.
Consequently, as observed at No. 2 Mine, Collinsville and Leichhardt Colliery,
Blackwater the major thicknesses of vitrinite-rich and inertodetrinite-rich coal plies were
sufficient to create a major gas content gradient in the mining face which resulted in
outbursts occurring. Therefore, the outburst face example illustrated in Fig. 16 can be
explained using a gas content gradient model ŽFig. 17.. The key features of this model
are the different gas content profiles of the vitrinite- and inertinite-rich parts of the
working coal section. Naturally, the gas content gradient effect is exacerbated if the coal
seam gas is carbon dioxide due to its greater sorption capacity at a given pressure and
the strong retention effect of coal as noted by Greaves et al. Ž1993..

8. Conclusions

Instantaneous outbursts in underground coal mines continue to pose a hazard to safe,

productive extraction of coal. The problem results from a combination of the effects of
stress, gas content, and physico-mechanical properties of the coal. Research and
operational experiences have provided the opportunity to test theories on the mecha-
nisms of an instantaneous outburst as well as to set guidelines for assessing the potential
outburst-proneness of coal. For example, in the Sydney Basin, gas content thresholds
have been implemented, which take into account changes in gas composition from pure
methane to pure carbon dioxide. These thresholds are often modified for individual mine
conditions.
52 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

Many of the indices used for outburst-proneness assessment of face coal have shown
conflicting results for Australian Permian coals. Differences in the gas desorption
behaviour of banded coal lithotypes, which make up these Australian coals, help explain
these results. Dull coal bands rich in fusinite and semifusinite have fast desorption rates
from solid coal. Conversely, bright coal bands rich in vitrinite have slow desorption
rates from solid coal. Dull coal bands rich in inertodetrinite, behave in a similar manner
to bright coal bands.
The significance of this differential desorption behaviour of banded coals, is observ-
able in gas emission readings from a mine face. Vitrinite-rich coal sections of a seam
retain their gas content due to the slow desorption rate of the solid coal. Once the
particle size of the coal is reduced, either during coal cutting operations or retrieving a
sample for an emission value reading, the gas is rapidly released. Similarly, the resulting
differential desorption in the working face can produce a significant gas content gradient
ahead of the face, sufficient to initiate an outburst. This effect has been documented at
two mines in the Bowen Basin. At Leichhardt Colliery, Blackwater, a combination of
sheared vitrinite-rich coal surrounding an inertodetrinite-rich coal layer was responsible
for a major methanercoal outburst. At No. 2 Mine, Collinsville, a sheared vitrinite-rich
lower working section was responsible for a minor carbon dioxidercoal outburst.
Recognition of vitrinite-rich andror inertodetrinite-rich coal seam sections is war-
ranted in future delineation of outburst-prone areas of underground operations in the
Bowen Basin. Assessments based on rapid desorption must be treated with caution in
view of particle size and coal lithotype effects.

References

Airey, E.M., 1971. A theory of gas emissions in coal mining operations. MRDE Report, No. 21. Mining,
Research and Development Establishment, National Coal Board, Bretby, 25 pp.
Anon, 1991. Guide to the determination of desorbable gas content of coal seams: Direct Method. Australian
Standard AS 3980-1991. Standards Australia, North Sydney, NSW, 16 pp.
Ashton, B.H., 1990. Determination of gas content of coal. Undergraduate Project Report. Dept. of Min. Eng.,
Univ. Auckland, 41 pp.
Bartosiewicz, H., Hargraves, A.J., 1985. Gas properties of Australian coal. Proc. Aust. Inst. Min. Metall. 290,
71–77.
Beamish, B.B., 1981. CCP No. 2 Mine, 51 level west panel, report on continuous face monitoring using a gas
analysing system. Tech. Rept. R 101. Collinsville Coal Co. Pty, Collinsville, 49 pp.
Beamish, B.B., 1984. Overseas visit to West Germany. Tech. Rept. CCP-009-84. Collinsville Coal Co. Pty,
Collinsville, 38 pp.
Beamish, B.B., 1990. Continuous gas monitoring for the assessment of outburst-proneness at a mining face.
Int. J. Min. Geol. Eng. 8, 131–147.
Beamish, B.B., 1996. Sorption of gas by coal and associated phenomena in underground mining. Ph.D. thesis,
Univ. Auckland, 258 pp.
Beamish, B.B., Vance, W.E., 1990. Gas research at Huntly West mine, New Zealand. Tech. Rept., Dept. of
Min. Eng., Univ. Auckland, 32 pp.
Beamish, B.B., O’Donnell, G., 1992. Microbalance applications to sorption testing of coal. Proc. Coalbed
Methane Research and Development in Australia. Coalseam Gas Research Inst., Dept. of Earth Sciences,
James Cook Univ. of North Queensland, Townsville, 4, pp. 31–41.
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 53

Beamish, B.B., Crosdale, P.J., 1995. The influence of maceral content on the sorption of gases in coal and the
association with outbursting. Proc. Int. Symp.-cum-Workshop on Management and Control of High Gas
Emissions and Outbursts in Underground Coal Mines, NSW Joint Coal Board Health and Safety Trust,
Sydney, pp. 353–361.
Beamish, B.B., Gamson, P.D., Johnson, D.P., 1991. Investigations of parameters influencing gas storage and
release in Bowen Basin coals. Tech. Rept. CGRI TR91r4, Coalseam Gas Res. Inst., Dept. of Earth
Sciences, James Cook Univ. of North Queensland, Townsville, 120 pp.
Biggam, F.B., Robinson, B., Ham, B., 1980. Outbursts at Collinsville - A case study. Proc. The Occurrence,
Prediction and Control of Outbursts in Coal Mines, The Aust. Inst. Min. Metall., Melbourne, pp. 41–61.
Creedy, D.P., 1986. Methods for the evaluation of seam gas content from measurements on coal samples. Min.
Sci. Tech. 3, 141–160.
Crosdale, P.J., Beamish, B.B., 1995. Applications of coal petrography and microstructure to gas emissions,
outburst-proneness indications and alleviation in underground coal mines. Joint Coal Board Health and
Safety Trust Final Report, Coalseam Gas Res. Inst., Dept. of Earth Sciences, James Cook Univ. of North
Queensland, Townsville, 121 pp.
Curl, S.J., 1978. Methane prediction in coal mines. Tech. Rept. No. ICTISrTR04, Inter. Energy Agency Coal
Res., London, 78 pp.
Cybulski, W., Myszor, H., 1974. An attempt to define the dependence of methane emission on the coal output.
Arch. Gorn. 19, 145–155, ŽIn Polish..
Dunmore, R., 1981. Prediction of gas emission from longwall faces. Min. Eng. 140, 565–572.
Ettinger, I.L., Zhupakhina, E.S., Schterenberg, L.E., 1958. Methods of Allowing Forecasting in the Seams of
Coal Zoned Subject to Instantaneous Outbursts. Extract from Academy of Sciences of the USSR, Institute
of Mines Central Committee of Measures Against Instantaneous Outbursts, Moscow, Cherchar Translation
No. 111-50, Paris.
Fischer, F., Peters, K., Warnecke, A., 1932. Gases confined in coals. Brennst.-Chem. 13, 209–216.
Gamson, P.D., Beamish, B.B., 1991. Characterisation of coal micro-structure using scanning electron
microscopy. Proc. Queensland Coal Symp. The Aust. Inst. Min. Metall., Melbourne, pp. 9–21.
Gray, I., 1980. The mechanism of, and energy release associated with outbursts. Proc. The Occurrence,
Prediction and Control of Outbursts in Coal Mines. The Aust. Inst. Min. Metall., Melbourne, pp. 111–125.
Greaves, K.H., Owen, L.B., McLennan, J.D., Olszewski, A., 1993. Multi-component gas adsorption–desorp-
tion behavior of coal. Proc. 1993 Int. Coalbed Methane Symp. The Univ. of Alabama, Tuscaloosa, AL, pp.
197–205.
Hargraves, A.J., 1958. Instantaneous outbursts of coal and gas. Proc. Aust. Inst. Min. Metall. 186, 21–72.
Hargraves, A.J., 1962. Gas in face coal. Proc. Aust. Inst. Min. Metall. 203, 7–44.
Hargraves, A.J., 1969. Recent developments in instantaneous outburst control in Australia. Proc. Symp. on
Instantaneous Outbursts, Hungary. Economic Commission for Europe, Geneva.
Hargraves, A.J., 1980. A review of instantaneous outburst data. Proc. The Occurrence, Prediction and Control
of Outbursts in Coal Mines. The Aust. Inst. Min. Metall., Melbourne, pp. 1–18.
Hargraves, A.J., 1983. Instantaneous outbursts of coal and gas: A review. Proc. Aust. Inst. Min. Metall. 285,
1–37.
Hargraves, A.J., 1993. Update on instantaneous outbursts of coal and gas. Proc. Aust. Inst. Min. Metall. 298,
3–17.
Hargraves, A.J., Hindmarsh, J.W., McCoy, A.E.R., 1964. The control of instantaneous outbursts at Metropoli-
tan Colliery, NSW. Proc. Aust. Inst. Min. Metall. 209, 133–166.
Harvey, C.R., 1995. The development of an outburst data base for the Bulli coal seam. Proc. Int.
Symp.-cum-Workshop on Management and Control of High Gas Emissions and Outbursts in Underground
Coal Mines, NSW Joint Coal Board Health and Safety Trust, Sydney, pp. 405–412.
Hunt, J.W., Botz, R.W., 1986. Outbursts in Australian coal mines: A petrographic factor. Proc. Aust. Inst.
Min. Metall. 291, 59–63.
Jackson, L.J., 1984. Outbursts in coal mines. Tech. Rept. No. ICTISrTR25. Int. Energy Agency Coal Res.,
London, 55 pp.
´ M., Nodzenski,
Jagiełło, J., Lason, ´ A., 1992. Thermodynamic description of the process of gas liberation from
a coal bed. Fuel 71, 431–435.
54 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55

¨
Janas, J., Winter, K., 1977. Fruherkennung der Gasausbruchsgefahr auf der Grundlage der k-wert Bestim-
¨
mung, In: Frukerkennen ¨
und Verhindern von Gas-ausbruchen im Deutschen Steinkohlen-bergbau. Verlag
Gluckauf, Essen, pp. 21–28.
Juntgen, H., Karweil, J., 1966a. Formation and storage of gases in bituminous coal seams. Part 1. Gas
formation. Erdoel Kohle Erdgas Petrochim. 19, 251–258, ŽEnglish translation..
Juntgen, H., Karweil, J., 1966b. Formation and storage of gases in bituminous coal seams. Part 2. Gas storage.
Erdoel Kohle Erdgas Petrochim. 19, 339–344, ŽEnglish translation..
Kidybinski, A., 1980. Significance of in situ strength measurements for prediction of outburst hazard in coal
mines of Lower Silesia. Proc. The Occurrence, Prediction and Control of Outbursts in Coal Mines. The
Aust. Inst. Min. Metall., Melbourne, pp. 193–201.
Klebanov, F.S., 1971. External dynamics of gas emission from a worked seam. Sov. Min. Sci. 7, 469–472.
Lama, R.D., 1980. Results of some index tests on coal at Cook Colliery, Queensland Coal Mining Company,
Blackwater, QLD. GCM Tech. Rept. No. 11, CSIRO, Division of Applied Geomechanics, Mt. Waverley,
19 pp.
Lama, R.D., 1986. Improving the efficiency of gas drainage systems. National Energy Research Development
and Demonstration Program End of Grant Report 86r701. Dept. of Primary Industries and Energy,
Canberra, 88 pp.
Lama, R.D., 1995. Preface to Proc. of Int. Symp.-cum-Workshop on Management and Control of High Gas
Emissions and Outbursts in Underground Coal Mines. NSW Joint Coal Board Health and Safety Trust,
Sydney, pp. iii–iv.
Lamberson, M.N., Bustin, R.M., 1993. Coalbed methane characteristics of Gates Formation coals, northeastern
British Columbia: Effect of maceral composition. Am. Assoc. Petrol. Geol. Bull. 77 Ž12., 2062–2076.
Levine, J.R., 1992. Oversimplifications can lead to faulty coalbed gas reservoir analysis. Oil Gas J. 90 Ž47.,
63–69.
Lidin, D.L., Ettinger, I.L., Zhupakhina, E.S., Sazonov, L.V., 1954. Determination of the rate of gas emission
as a method of detecting zones liable to sudden outbursts. Ugol 29, 21–24.
Litwiniszyn, J., 1985. A model for the initiation of coal–gas outbursts. Int. J. Rock Mech. Min. Sci. Geomech.
Abstr. 22, 39–46.
Myszor, H., 1974. Emission of methane from high-output longwall faces. Przegl. Gorn. 30, 273–276, ŽIn
Polish..
Noack, K., 1976. Effect of face length, working speed and type of stowing on methane emission. Glueckauf
112, 1156–1159, ŽIn German..
Noack, K., Koppe, U., Kunz, E., 1995. Control of high gas emissions in underground coal mines. Proc. Int.
Symp.-cum-Workshop on Management and Control of High Gas Emissions and Outbursts in Underground
Coal Mines. NSW Joint Coal Board Health and Safety Trust, Sydney, pp. 15–24.
Paterson, L., 1986. A model for outburst in coal. Int. J. Rock Mech. Min. Sci. Geomech. Abstr. 23, 327–332.
Patterson, J.K., Elliott, W., Eyeington, S.R., McGhie, T., Prendiville, A.V., 1967. Report of commission to
inquire into disaster at Strongman Mine. Appendix to the Journals of the House of Representatives of New
Zealand C-4, Wellington.
Peng, S.S., 1978. Coal Mine Ground Control. Wiley Interscience, New York, 450 pp.
Ruckenstien, E., Vaidyanathan, A.S., Youngquist, G.R., 1971. Sorption by solids with bidisperse pore
structures. Chem. Eng. Sci. 26, 1305–1318.
Roberts, R., 1995. The gas menace. Australia’s Mining Monthly, December 1994rJanuary 1995, pp. 85–88.
Shepherd, J., Rixon, L.K., Griffiths, L., 1981. Outbursts and geological structures in coal mines: A review. Int.
J. Rock Mech. Min. Sci. Geomech. Abstr. 18, 267–283.
Smith, D.M., Williams, F.L., 1984. Diffusion models for gas production from coals: Applications to methane
content determination. Fuel 63, 251–255.
Smith, J.W., Gould, K.W., 1980. An isotopic study of the role of carbon dioxide in outbursts in coal seams. J.
Geochem. 14, 27–32.
Stutzer, O., 1936. Carbon dioxide eruptions from coal seams in Lower Silesia. Econ. Geol. 31, 441–452.
Thomas, J.W., 1876. On the gases enclosed in channel coals and jet. J. Chem. Soc. 30, 144.
Truong, D., Williams, R.J., 1989. Gas content and composition of Australian coals. Proc. 23rd Newcastle
Symp. on Advances in the Study of the Sydney Basin. Univ. of Newcastle, Newcastle, Australia, pp.
247–254.
 B.B. Beamish, P.J. Crosdaler International Journal of Coal Geology 35 (1998) 27–55 55

¨
von Meyer, E., 1872. Uber die in einigen englischen Steinkohlen eingeschlossenen Gase. J. Prakt. Chem. 5,
407.
¨ die Beschaffenheit des in Inselbad bei Paderborn zur Inhalation gebrauchten Gases.
von Meyer, E., 1873. Uber
J. Prakt. Chem. 6, 360.
Williams, R.J., Rogis, J., 1980. An analysis of the geologic factors leading to outburst-prone conditions at
Collinsville, QLD. Proc. The Occurrence, Prediction and Control of Outbursts in Coal Mines. The Aust.
Inst. Min. Metall., Melbourne, pp. 99–109.
Williams, R.J., Slater, M.I., 1995. Application of hole flow measurements and return gas monitoring to
validation of low gas mining conditions. Proc. Int. Symp.-cum-Workshop on Management and Control of
High Gas Emissions and Outbursts in Underground Coal Mines. NSW Joint Coal Board Health and Safety
Trust, Sydney, pp. 553–556.
Williams, R.J., Weissmann, J.J., 1995. Gas emission and outburst assessment in mixed CO 2 and CH 4
environments. Proc. ACIRL Underground Mining Sem. Australian Coal Industry Res. Lab., North Ryde,
12 pp.