Collapse of Colebrook Coal Mine

Coalbrook mine was laid in 1905 on lands 100 km south of Johannesburg , within the modern city of Sasolburg ^{[5] [6]} . Coal reserves in this field consist of three layers ^{[5] [6]} . The most valuable layer No. 2 , with a thickness of about 6 m , lies at a depth of about 140 m from the surface and has virtually no fall , with the exception of hard-to-reach areas on the outskirts of the field ^{[K. 2]} . Above and below it are beds No. 1 and No. 3 , with a thickness of up to 2 m and up to 3 m, respectively, separated from layer No. 2 by layers of slates and limestone ^{[5] [6]} . A solid dolerite slab with a thickness of about 40 m lies between formation # 1 and the surface ^{[5] [6]} .

In the first decades of the mine’s operation, coal mining was carried out exclusively at bed No. 2 using a chamber-pillar system ^{[c. 1]} without collapse of the worked out workings. The development plan was compiled conservatively: the distance between the axes of adjacent chambers was 24.4 m , and the width of each chamber did not exceed 7.3 m - thus, the columns remaining after cleaning (supporting interchamber pillars ) had dimensions of at least 14 × 14 m ^{[5 ]} . The width of the barriers (pillar pillars) between adjacent panels was also 24.4 m , and the bottom height was only 2.4 m ^[5] . The mining of the remaining upper part of the seam was not carried out due to the low quality of these coals, and from the used lower part more than half of the reserves remained in pillars ^{[5] [6]} . Since 1932, the leadership of the mine began to sequentially compact the mine ^[5] . By 1949, the distance between the centers of the chambers was reduced to 18.3 m , with a corresponding decrease in the size of interchamber pillars, and the thickness of the interpanel pillars was reduced to 12.2 m ^[5] . An additional excavation of “pockets” of coal from supporting pillars was carried out, further weakening their bearing capacity ^[5] . Thanks to these measures, by 1950 coal production increased to 1600 tons per day ^[5] .

The vast majority of mine workers were black people from neighboring Basutoland and the East African possessions of Portugal ^[9] . Places of foremen, instructors and explosive masters were legally assigned to whites ^{[10] [11]} ; in fact, dangerous blasting was also performed by blacks ^{[12] [11]} . Most black workers studied at the workplace. Only a few specially trained and specially tested workers received vocational training, who occupied the positions of boss ( English ) boys in underground work and controlled the masses of ordinary Africans ^{[13] [14]} .

In 1950, the mine owners won a tender for the supply of coal to a new power plant ^[5] , planned for construction in ^{[k. 3]} . To fulfill the contract, it was necessary to increase Coalbrook's coal production to 10 thousand tons per day within four years of construction, however experts invited by power engineers doubted the mine’s capabilities ^[5] . In their opinion, the weak bearing capacity of formation No. 2 did not allow to increase the height of the face above 2.9 m , while at least 60% of the coal from this layer should be left in pillars ^[5] . The owners did not agree with the experts, and began to forcefully increase production ^[5] . The dimensions of the mine field increased to 5.6 km along the west-east axis and up to 3.7 km along the north-south axis ^[3] . By 1957, the bottomhole height increased throughout the mine to 4.3 m , and in some of its sections to 5.5 m - which caused a series of local, not yet catastrophic, collapses ^[5] . After the intervention of mining inspectors, the company was forced to reduce the height of the face to 4.3 m ; 1958 annual production amounted to "only" 2.26 million tons - 17 times more than in 1954 ^[5] .

The owners of the mine, dissatisfied with the achieved volumes, decided to increase them due to the re-drilling of old workings - with an increase in the height of the face and with a partial selection of the pillars left earlier ^{[5] [6]} . To verify this decision, they carried out a pilot re-cleaning of site No. 10 worked out in 1952, located on the northern outskirts of the mine field ^{[5] [6]} . From the west, it was adjoined by a layer not yet touched by the penetration; from the east, the main active sections of the mine were located ^[5] . From the point of view of safety and the correctness of the experiment, it should have been carried out in older sections located in the west of the mine field, but for reasons of convenience, the management chose the tenth section ^[5] .

The experimental work — cleaning the upper part of the reservoir, reducing the size of the pillars, removing “pockets” from the reduced pillars — affected an area of 3 hectares and successfully completed by March 1959 ^[5] . During the experiment, there were no extraordinary incidents, and the leadership gave the green light to spread the new practice throughout the site ^{[5] [17]} . In the meantime, in the areas adjacent to it, production also continued, sequentially weakening the bearing capacity of the formation ^[5] . There were no means to measure, or at least evaluate it at the mine - the engineers relied only on their own hearing and sight ^{[5] [6]} .

Around 19:00 on December 28, 1959, the tenth site collapsed. The worked out “experimental training ground” was forever buried by the masses of the rock ^[5] . By a lucky coincidence, the incident occurred on a non-working shift when there was not a single miner on the site ^{[5] [6]} . Only one person, who was far from the place of collapse, was knocked down by an unexpected stream of air and was injured ^[5] . A survey of workers employed in the previous day shift at an adjacent site did not reveal any signs of impending collapse ^[5] . After the collapse in this area, noises of deformed arches and pillars were clearly audible, but within three days they completely died down ^{[5] [6]} . The company chose to keep the incident secret, and the inspector who visited the mine on January 11 did not find anything suspicious ^{[5] [18]} .

Around 16:00 on January 21, 1960, miners working underground to the west of Section No. 10 heard a loud noise - a kind of "shots" coming from the side of the workings that collapsed in December - and the characteristic crack of pillars settling from excessive load ^[5] . The shift supervisor stopped work and brought people out of the faces; along the way, they were knocked down by a strong gust of air from the side of section No. 10 ^[5] . Around 16:20, similar noises and claps were observed south of plot No. 10 ^[4] . At 16:45, seismographs recorded an earthquake of magnitude 0.5 on the Richter scale ^[4] . The head of the underground work and the acting director of the mine (the director himself was on vacation that day) immediately went underground to find out the situation. As it turned out, all the ventilation jumpers ^{[to. 4]} in the vicinity of plot No. 10 were destroyed ^[4] . The air increased methane concentration, but there were no traces of carbon monoxide - which excluded the version of the explosion of mine gases ^[4] . On the surface of the earth above the plot number 10 , a failure was formed ^[4] .

Having assessed the situation, the managers concluded that the failure removed the excess stress accumulated in the rocks above the plot No. 10 , and it no longer poses a danger ^[4] . Two sections directly adjacent to section No. 10 were closed and evacuated, the rest continued to work ^[4] . A small group of miners began to restore ventilation lintels, and most were still concentrated in two opposite, isolated areas in the northwest and on the eastern edge of the mine field ^[18] . No one at the mine knew that at the same time a crack was growing on the surface, which reached 1.2 km in length by 18:30 ^[14] . According to later reports by the left-wing newspaper, a group of five African miners attempted to escape to the surface, but were stopped by security ^[21] . Two rebels went under arrest, three agreed to return to the mine, but at the last moment managed to break free ^[21] . Probably, professor of Durban University Dennis Schauffer believes, the miners guessed about the impending disaster due to the unusual, panic behavior of rats ^[14] . Managers, according to Schauffer, went underground not to solve technical issues, but to force miners to work ^[14] .

At the beginning of the eighth hour of the evening, the workers, who were reconstructing the ventilation lintels, heard a growing rumble from the emergency tenth section; at the same time, the instruments recorded methane emissions ^[4] . The miners rushed to run, but along the way they were knocked down by a powerful stream of air and dust ^[4] . The hurricane whirlwind did not subside for ten minutes, preventing people from rising; at 19:26, seismographs recorded the beginning of a five-minute series of tremors of magnitude 1.0 on the Richter scale ^[4] . When the whirlwind gradually slept, people rushed to run ^[4] . As it turned out later, all the repairmen and all the workers of the north-western areas managed to safely surface; out of 438 eastern workers, not a single one got out ^[4] . The leaders of the mine again went down to the ground and found that all the passages to the eastern sections were littered with collapsed rock ^[4] . Only one of the 438 missing people, who were engaged in hauling coal to the central pile , was found alive and brought to the surface ^[4] .

Examination of the surface showed that the failure over section No. 10 reached a diameter of 150 m and a depth of 1.8 to 2.1 m ^[4] . A network of cracks spread to the east of it, on an area of ​​approximately 324 ha , and the surface level dropped by an average of 0.5 m ^[4] . So it became clear that the underground collapse affected the entire eastern part of the mine field. Support pillars, like domino bones , one after another collapsed under load. The chain reaction stopped only reaching the border of the mine, or to the old mine with wide pillars ^[4] .

The rescue operation, which lasted several weeks, attracted the attention of reporters from all the country's central newspapers and radio stations ^[22] , as well as the special department of the South African police ^[21] . Rescuers immediately realized that you can get to the eastern sections of the mine only from the surface. Drilling rigs urgently delivered from neighboring mines were deployed on the outskirts of the mine field ^[11] . The dolerite plate did not succumb to drill bits for a long time; only eleven days after the disaster did the first well reach voids in the place where coal was recently mined ^[11] . A microphone lowered into the well recorded only the sound of falling drops of water ^[11] . Sensors from other wells detected completely flooded voids, a high concentration of methane, but no signs of life ^[4] .

A few days later, the operation was terminated; 437 people were declared dead. Among them were six white miners ^[11] and 210 black people from the East African possessions of Portugal ^[21] ; the remaining victims belonged to the South African tribes of Basutoland ^[9] . The widows of white miners received lifelong pensions ranging from 30 to 75 South African pounds per month ^[23] . The widows of blacks received only one-time benefits, not exceeding 252 South African pounds , paid monthly at 3-4 pounds ^{[23] [24]} .

According to the local historian Horst Muller, the mine resumed coal mining on March 18, 1960. In March 1961, another collapse occurred, and the northern part of the mine, along with the main, northern shaft, was permanently closed. Extraction in the southern part of the mine field, which was served by a special southern shaft, continued until 1990 ^[25] .

According to the memoirs of the journalist , already in the first days after the catastrophe there were persistent rumors that its cause was the extraction of a larger share of coal than the bearing capacity of the rock allowed ^[11] . State authorities generally supported this opinion, but did not prosecute the owners and managers of the mine ^{[11] [26]} , limiting themselves to recommendations for the future ^[4] . The opinion about the excess mining of coal and the insufficient margin of safety of “optimized” pillars has become dominant among professional miners ^[4] . In the framework of this hypothesis, the immediate “culprit” of the incident was the fracture of the dolerite plate over section No. 10 ^[27] . Under normal conditions, the plate perceived and redistributed the load from the overlying rock layers, protecting the supporting pillars from excessive loads ^[27] . After the first fault, the once-monolithic plate cracked and turned into a pair of unconnected consoles ; after the second fault, the free ends of these consoles squeezed and destroyed the overloaded, small pillars of the eastern sections, causing a chain reaction of collapse ^[27] . Oversized pillars of the old western and southern sections resisted ^[4] . The small pillars of the northwestern sections also withstood the load, as they were protected by massive interpanel barriers ^[4] .

The miners urgently needed a reliable method of calculating the strength of the pillars; the related problem of the strength of mine arches for the time being remained in the background ^[27] . The long-known formulas for the resistance of materials were poorly suited for coal mines due to the large spread of coal characteristics and the lack of methods for extrapolating the characteristics of samples to large arrays of pillars and barriers ^[27] . Researchers focused, on the one hand, on testing the strength properties of coal, and on the other hand, on the statistical processing of data on already-occurring collapses ^[27] . In 1967, Miklos Salamon and A. Munro, who worked in South Africa, summarizing the characteristics of 27 collapsed and 92 serviceable pillars, proposed the empirical Salamon – Munro formula , which has become classical ^[27] . Since Salamon’s filing, the notion of “safety factor” ( safety factor ), a dimensionless characteristic of a pillar characterizing the probability of its collapse, has entered the everyday life of miners in the English-speaking world ^[27] .

The empirical Salamon method, based only on collapse statistics, did not need accurate measurements and time-consuming calculations, but had a fundamental drawback - the dependence on the sample of historical data underlying the model ^[27] . For this reason, in subsequent decades, the technique was refined several times taking into account the latest data ^[27] . Since the 1990s, researchers have been using separate statistics on mines with strong and weak coals (which include all the mines of the Waal basin , including Colebrook) ^[27] . According to the estimates of the beginning of the 21st century, the consistent application of safety standards dating back to the Salamon-Munro method has reduced the probability of collapse of South African mines by four times, at the cost of losing about 600 million tons of coal in unrecoverable pillars ^[28] .

• Gorodnichenko V.I., Dmitriev A.P. Fundamentals of mining . - M .: Mountain Book, 2016 .-- ISBN 9785986724348 .
• C. Didier. Retour d'expérience sur la catastrophe de Coalbrook // Rapport Scientifique INERIS, 2013, 2012-2013. - 2013 .-- P. 70–72.
• JN van der Merwe. Beyond Coalbrook: what did we really learn? // The Journal of The Southern African Institute of Mining and Metallurgy. - 2006. - Vol. 106, No. December. - P. 857-868. The author is an employee of the Department of Mining, Metallurgy and Geology, University of Pretoria .
• JN van der Merwe. Lessons From Coalbrook. Presentation to SACMA 22 August 2013 // South African Colliery Managers Association. - 2013. - P. 1-9. The author is an employee of the Department of Mining, Metallurgy and Geology, University of Pretoria .
• HJ Simons. Death in South African Mines // Africa Today. - 1961. - P. 41–55.
• Schauffer, Dennis. The Clydesdale Coalbrook Colliery Disaster // E-Leader Warsaw 2018.- 2018.- P. 1-6. The author is professor emeritus of the University of Kwazulu-Natal in Durban .