Information Circular 9352

Fires In Abandoned Coal Minesand Waste Banks

By Ann G. Kim and Robert F. Chaiken

UNITED STATES DEPARTMENT OF THE INTERIORBruce Babbitt, Secretary

BUREAU OF MINES

Library of Congress Cataloging in Publication Data:

Kim, Ann G.Fires in abandoned coal mines and waste banks / by Ann G. Kim and Robert F.

Chaiken.

p. em. - (Information circular; 9352)

Includes bibliographical references (p. 57).

Supt. of Docs. no.: I 28.27:9352.

1. Abandoned coal mines-Fires and fire prevention. 2. Spoil banks-Fires andfire prevention. I. Chaiken, Robert F. II. Title. III. Series: Information circular(United States. Bureau of Mines); 9352.

TN295.U4 [TN315] 622 s-dc20 [622'.82] 92-20449 CIP

CONTENTSPage

Abstract. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2Nature of abandoned mined land fires. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Initiation and propagation of abandoned mined land fires 6Hazards of abandoned mined land fires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Extent of abandoned mined land fire problems 13Incidence of abandoned mined land fires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Control of abandoned mined land fires . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Abandoned mined land fire extinguishment-control 23Temperature and gas monitoring 24Conventional methods 30

Excavation 30Inundation-flooding 35Flushed barriers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36Excavated barriers 37Surface seals 38

Recent developments 40Mine fire diagnostics 41

Temperature monitoring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41Aerial infrared. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41Soil hydrocarbons. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41Magnetic anomalies 42Desorbed hydrocarbons-communication testing 42

Mine fire extinguishment 43Fuel removal methods 43

Burnout control 43Water jetting 48

Heat removal methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48Water 48Foam " . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . .. 51Cryogenic liquids 51

Oxygen exclusion (fire barriers) 54Research areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

Diagnostics 54Extinguishment 54Control-containment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Suggested approach to AML fire problems 55Conclusions 56References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

ILLUSTRATIONS

1. Coalfields of United States. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32. Geographic distribution of AML fires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33. Fire(s) in abandoned mine. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44. Bituminous waste bank, Albright, WV 55. Anthracite waste bank, Shamokin, PA 56. Fire triangle 67. Heat of combustion versus rank of coal 78. Rate of heat loss and rate of heat gain versus temperature 7

1l

ILLUSTRATIONS-ContinuedPage

9. Transmission of heat by movement of hot gases 810. Distribution of AML fire control projects by rank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 911. Smoke plume from AML fire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1112. Small vent at AML fire 1113. Fracture line above AML fire 1214. Sinkhole at AML fire. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1215. Percentage distribution of occurrence and cost of AML fire control projects by coal rank . . . . . . . . . . . . 2016. Average cost of AML fire control project versus coal rank 2117. Distribution of AML fire control projects by method 2318. Fire triangle and control methods: fuel removal, oxygen exclusion, and heat removal . . . . . . . . . . . . . . . 2419. Discontinuous fire propagation in abandoned mines 2520. Abnormal snow melt indicating subsurface heating 2521. Rate of subsurface cooling versus initial temperature 2922. Change in subsurface combustion zone temperature over 30-year period 2923. Variation in carbon dioxide and oxygen concentrations at Large mine fire . . . . . . . . . . . . . . . . . 3124. Carbon monoxide concentration versus temperature 3125. Ratio of CO2 to CO versus temperature. 3226. Cost and effectiveness of AML fire control projects by method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3227. Schematic of excavation of fire zone allowing propagation of fire into interior of mine . . . . . . . . . . . . . . 3328. Schematic of excavation to mine fire to inhibit propagation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3329. Schematic of propagation of waste bank fire during excavation 3430. Increased fire intensity during excavation 3431. Hazard of AML fire excavation 3532. Inundation by water infiltration from surface impoundment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3633. Noncombustible barrier injected from surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3634. Flushed boreholes during and after injection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3635. Plan view of injection boreholes for hydraulic flushed barrier 3736. Injection of foaming grout or cellular concrete to encapsulate burning coal 3737. Plan view of trench barrier for AML fire control 3738. Schematic of trench barrier for AML fire control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3839. Plan view of plug and tunnel barriers for AML fire control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3840. Surface seal to control AML fires in Eastern United States . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3941. Extent of surface seal, depending on condition of mine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3942. Surface seal used in western AML fires 4043. Changes in mine fire diagnostic ratio with temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4244. Values of mine fire diagnostic ratio during heating and cooling of selected coals 4245. Schematic of communications testing for mine fire diagnostics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4346. Mine fire diagnostic map of results of one communication test 4447. Mine fire diagnostic map produced from repeated communication tests. . . . . . . . . . . . . . . . . . . . . . . . . 4448. Changes in mine fire diagnostic ratio at three boreholes during extinguishment project . . . . . . . . . . . . . . 4549. Schematic diagram of Burnout Control with energy utilization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4550. Schematic of Burnout Control system - Calamity Hollow mine fire project . . . . . . . . . . . . . . . . . . . . . . . 4651. Artist's version of Burnout Control system at Calamity Hollow 4652. Results of Burnout Control test at Calamity Hollow 4753. Schematic of Burnout Control at Albright waste bank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4854. Water distribution system for quenching fire zone at Calamity Hollow site. . . . . . . . . . . . . . . . . . . . . . . 4955. Exhaust temperature during quenching at Calamity Hollow site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4956. Cumulative heat flow during quenching at Calamity Hollow 5057. Cumulative waterflow during water injection at Calamity Hollow site. . . . . . . . . . . . . . . . . . . . . . . . . . . 5058. Foam injection-fume exhaust system for extinguishing AML fires 5259. High and low temperatures during injection of liquid nitrogen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

III

ILLUSTRATIONS-ContinuedPage

60. Schematic of equipment to produce cryogenic slurry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5361. High and low temperatures during injection of cryogenic slurry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

TABLES

1. Fire control projects in eastern bituminous region, 1949-1972 142. Fire control projects in Pennsylvania Anthracite region, 1950-1987 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153. Fire control projects in Western United States, 1949-1977 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164. Abandoned mined land fires, 1977 195. Abandoned coal waste bank fires 196. Abandoned mine fires, AML inventory, October 1988 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207. Federal AML fire control projects by State, 1989 208. Mine fire control costs and effectiveness by rank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229. Mine fire control costs and effectiveness by method 22

10. Mean borehole temperatures, Large mine fire 2711. Mean borehole temperatures, Carbondale mine fire 2712. Variation in cooling rate at western mine fires 2813. AML research projects, 1987-1990 41

UNIT OF MEASURE ABBREVIATIONS USED IN THIS REPORT

Btu British thermal unit h hour

Btu. h. ft-2• 0p-l. in" British thermal unit per hp horsepowerhour per square foot perdegree farenheit per inch Hz hertz

Btu/lb British thermal unit per III inchpound

in-H2O inch of water°C degree Celsius

J. sec. m-2• °C-1• em" joule per second percal/g calorie per gram square meter per degree

celsius per centimetercal. min. m-2• °C-1• ern" calorie per minute per

square meter per degree kcal kilocaloriecelsius per centimeter

kcal/g kilocalorie per gramcm centimeter

kcal/rnole kilocalorie per molecm2 square centimeter

kW kilowattcm2/s square centimeter per

second lb pound

OF degree Farenheit Ib/ft3 pound per cubic foot

ft foot MMBtu million British thermalunit

ft2 square footMMgal million gallon

ft3 cubic footMW megawatt

g gramppm part per million

gal gallonpsi pound per square inch

gal/d gallon per days second

gal /h gallon per hourst short ton

gpm gallon per minuteV volt

g/s gram per second

FIRES IN ABANDONED COAL MINES AND WASTE BANKS

By Ann G. Kim 1 and Robert F. Chaiken2

ABSTRACT

Fires that occur in abandoned coal mines, waste banks, and in coal outcrops constitute a serioushealth, safety, and environmental hazard. Toxic fumes, the deterioration of air quality, and subsidenceconstitute the greatest hazards from these fires. Although fires on abandoned mined land (AML) occurin every coal-producing state, the severity of the problem varies. Methods to extinguish or control AMLfires, including excavation, fire barriers, and sealing, are generally expensive and have a relatively lowprobability of success.

This U.S. Bureau of Mines report includes information from a variety of sources, i.e., agencies of theFederal Government, State agencies, research reports, conference proceedings, product information, andtechnical literature. This information has been collated into a comprehensive discussion of AML fireproblems. Data on past fire control projects and on the estimated extent of the current problem havebeen compiled. Factors affecting the occurrence, propagation, and extinguishment of AML fires arediscussed. Conventional fire control methods are described, and their probable effectiveness isevaluated. Information on the hazards of AML fires and safety considerations is included. The statusof current technology, recent improvements in fire control methods, and areas of current research arediscussed.

ISupervisory physical scientist.2Research chemist.Pittsburgh Research Center, U.S. Bureau of Mines, Pittsburgh, PA.

2

INTRODUCTION

The emission of toxic fumes, the deterioration of airquality, as well as subsidence from fires in abandoned coalmines and wastebanks cause problems that range from im-minent danger to a decline in the quality of life. In ad-dition to the health and safety hazard, such fires usuallydepress property values for affected land and for adjacentareas. AML fires occur in every coal-producing state, butthe severity and extent of the problem varies. Dependingon the type of fire and its location, the cost of controllingan AML fire may be between $100 thousand and $100mil-lion. Even if funds are not a constraint, current methodsof extinguishing or controlling AML fires are not routinelysuccessful. The problems of AML fires are related togeology, past mining practice, and the limits of currentlyavailable technology.

Coal is defined as "a readily combustible rock. ....of car-bonaceous material" (1).3 It is not a homogeneous sub-stance. Most coals consist of varying amounts of visiblydistinct macerals (2). The three maceral groups, vitrinite,exinite, and inertinite, exhibit different chemical composi-tions and are petrographically divided into macerals andsubmacerals. The origin of the various macerals is relatedto variations in the original material from which the coalwas formed. Although the petrographic constituents varywith the rank of the coal, the rank as it relates to com-bustibility is generally defined by the percentage of fixedcarbon and the heating value (3).

Because of its combustibility, coal can be readily con-verted to other forms of energy. This property has beenessential to the economic and technical development of allWestern cultures, particularly in the United States. Coalfired the industrial revolution and was essential to the de-velopment of the nationwide railroad system. At present,when other forms of energy such as oil, natural gas, andnuclear power are available, coal still supplies 24% of theU.S. total energy market (4); more than 765 million st ofcoal is burned annually to generate electricity (5).

Burning coal produces usable energy, but the uncon-trolled burning of coal in place creates potentially lethalhazards and degrades the environment. Although fires inactive mines (6) can be catastrophic, they are usually con-trolled by the mine operator in a relatively short period.In contrast, fires in abandoned mines and waste banksoften affect people who had no connection with the orig-inal mining. They occur under different physical condi-tions and must be controlled or extinguished under adifferent set of institutional constraints.

This U.S. Bureau of Mines report focuses on the prob-lems associated with fires in abandoned coal mines, out-crops, and waste banks. It includes a discussion of sourcesof ignition, factors influencing the propagation of suchfires, currently available technology to control these fires,and current research in this area.

ACKNOWLEDGMENTS

This report includes information from extensive datathat has been accumulated on AML fires over the past 40years. U.S. Bureau of Mines' projects and methods weredocumented by M. O. Magnuson, supervisory mining en-gineer, in the Eastern United States and by F. H.Shellenberger, mining engineer, and D. L. Donner, miningengineer, in the Western United States. The factorsaffecting AML fires and the status of such fires inColorado were summarized by P. Rushworth, reclamation

specialist, Colorado Geological Survey. L. Roberts, chief,AML Reclamation, OSMRE's Washington Office and B.Maynard, hydrologist, OSMRE's Eastern Technical Centerprovided information of the AML Inventory and theFederal Reclamation Program. The authors also wish toacknowledge the many State and Federal officials whoresponded to our request for information, and thosemembers of the Bureau's AML Advisory Panel whooffered suggestions and support for this work.

NATURE OF ABANDONED MINED LAND FIRES

AML fires, also called wasted coal fires, generally occuras smoldering fires in the low oxygen environment of anabandoned mine or waste bank. Outcrop fires, eitherrelated to a mine fire or in unmined coal, are also

3Italic numbers in parentheses refer to items in the list of referencesat the end of this report.

considered within this category. They occur in most coal-bearing areas of the United States (fig. 1), and werecontemporaneous with mining in the east and predatedmining in the west. In 1765, soldiers from Fort Pitt, whowere digging coal from the outcrop, lit a fire near the baseof the Pittsburgh seam. The fire propagated into the coalseam and could not be extinguished. Visitors to the area

LEGEND~ Pennsylvania age coalfields~ Cretaceous coalfields

Black WorrioBasin

Western InteriorRegiono 500

, , I

Scale, miles

SouthwesternRegion

Figure 1.-Coalfields of United States.

gave an accurate description of the effects of a fire undershallow cover (10-11). The fire was active until at least1846.

In the Western United States, coal outcrop fires werea natural feature of the landscape. In 1805, Lewis andClark, in their exploration of the Missouri River, reportedthat coal seams were plainly visible in the bluffs along theriver and that some of the veins were burning, ignited byspontaneous combustion or by grass fires (12-13).

AML fires or wasted coal fires occur in abandonedunderground mines, abandoned surface mines, wastebanks, and in coal outcrops. Many of these fires can becategorized according to the area in which they occur(fig. 2). For example, most underground mine fires are inthe eastern coal-producing States. The characteristics ofeastern fires also vary depending upon whether they are inbituminous or anthracite seams. Waste bank fires occurin the eastern and central States where the majority ofcoal preparation plants were located. Outcrop fires aremore prevalent in the Western United States. These firesmayor may not be related to mining. A mine fire mayspread into the barrier outcrop or it may ignite coalsstratigraphically above the mine fire. Lightning and brushfires can ignite the outcrop of unmined coal seams. Undercurrent regulations, only fires related to past mining areconsidered AML fires.

16D

KEY14D ~ Surface fires

f20~ Underground fires

(/)w~ 100u,

u,0 80

~Wm so~::JZ

40

20

o

3

Figure 2.-Geographic distribution of AML fires. (Eastern geo-graphic region includes Appalachian region and anthracite fields.)

EASTERN EASTERN INTERIOR WESTERN WESTERN INTERIOR

GEOGRAPHIC REGION

4

In underground mines that used a room-and-pillarmining system, a relatively large proportion (30% to 50%)of the coal is left in place. Roof coals and carbonaceousshales also may have been left in place. The tonnage ofcombustible material left in the mine, therefore, mayexceed that extracted during mining. Older mines hadseveral entries at the outcrop for drainage, ventilation, andaccess. Fires usually started at the outcrop and propa-gated along the outcrop or through the interconnectedworkings. Heat moved by convection through the mine orby conduction into the overburden. The overburden servesas an insulator, preventing the transfer of heat away fromthe combustible material. As the overburden becamewarmer or as the coal pillars failed, the overburdensubsided, creating a system of cracks and fractures throughwhich smoke and fumes left the mine and fresh air enteredthe mine (fig. 3). Under these conditions, most aban-doned mine fires exhibit smoldering combustion, involvingrelatively small amounts of coal at any given time, withlittle visible flame and capable of burning with as littleas 2% oxygen (14). Such fires can continue to burn forextended periods (10 to 80 years) and are difficult toextinguish.

In abandoned surface mines, the coal outcrop may beleft exposed when stripping operations are terminated, orcoal refuse may be left in contact with the outcrop. Ineither case, fires are not unusual. If the stripping op-eration involved the barrier pillar of an abandoned mine,it is possible for a fire to propagate into the mine.

Surface disposal of coal waste, from mines and frompreparation plants, is an AML problem. Approximately25% of the coal removed from the mine in the UnitedStates is rejected and disposed of on the surface (15).

Abandonedmine opening._-----

~.>

Over the past 200 years, more than 3 billion st of refusehas accumulated in 3,000 to 5,000 active and abandonedwaste piles (fig. 4) and impoundments in the eastern coal-fields alone. It has been estimated that a billion cubicyard of anthracite waste (fig. 5) has been disposed of insurface piles in the anthracite region. The refuse consistsof waste coal, slate, carbonaceous shales, pyritic shales,and clay associated with the coal seam. The combustiblecontent of this material averages between 2,000 to 6,000Btu/lb. Material with a combustible content above 1,500Btu/lb will support combustion (16-17).

Construction standards for active piles are intended toprevent combustion and to reduce the infiltration of sur-face water that produces acid drainage (18). Older wastepiles, gob piles, or slate dumps were built wherever therewas sufficient land and where transportation costs could beminimized. Usually, no attempt was made to stabilize theslopes, prevent combustion, or reduce the production ofacid drainage. Many of these piles burned, producingacrid smoke and toxic fumes. The burnt material, knownas "red dog," is considered a good construction aggregate,and was not considered a waste problem. Because of pastindiscriminate dumping, there is no accurate estimate ofthe number of abandoned coal waste piles. Those pilesthat appear on inventories are generally those that pro-duce enough acid drainage to negatively affect the waterquality of nearby streams, and those that are on fire andbecause of suburban spreading are now causing safety andenvironmental problems in populated areas.

Another fire problem indigenous to western Statesis the outcrop fire, which can be either the surface ex-pression of an abandoned mine fire or in unmined for-mations. Depending upon their location, these fires can

Surface

Abandonedmine entries

Figure 3.-Fire(s) in abandoned mine.

5

Figure 4.-Bituminous waste bank, Albright, WV.

Figure 5.-Anthracite waste bank, Shamokin, PA.

6

threaten coal reserves, degrade environmental quality, andpresent a hazard to wildlife, grazing animals, and unwaryhumans. These fires frequently occur in lower rank coals,the high-volatile bituminous coals and lignites that areprone to spontaneous combustion. In general, an outcropfire spreads initially along the outcrop, under thin cover,where oxygen is relatively plentiful. It may slowly spreadinto the solid coal under deeper cover if the overburdenis naturally fractured or if subsidence promotes the de-velopment of cracks and fractures. The rate and directionof fIre propagation are determined by the availability ofoxygen.Any given AML fire may include more than one type of

fire, For example, waste banks may be in contact with anoutcrop and involve both waste bank and subsurface fires.Fires that begin in the outcrop can spread to undergroundworkings. Fires that appear to be related to past miningmay actually be in virgin seams above the seam that wasmined. Generally, fires are classified as surface (wastebank and outcrop) and underground; however, the classifi-cations are somewhat arbitrary and not always accurate.

INITIATION AND PROPAGATION OF ABANDONEDMINED LAND FIRES

As with any fire, wasted coal fires require three ele-ments: fuel, oxygen, and an ignition source (fig. 6). Incoal combustion, the fuel is the carbon in the coal. Ifcombustion is considered the exothermic reaction of car-bon and oxygen to form carbon dioxide, written as:

C + O2 ---- > CO2 + heat,

the amount of heat liberated is 93.7 kcal/mole, Theamount of heat liberated by the reaction of 12 g of carbonwith 16 g of oxygen is enough energy to raise the temper-ature of a liter of water 100° C. However, coal is notcomposed of elemental carbon. On a dry, mineral matterfree basis, coal contains between 60% and 90% carbon.The rest of the coal molecule is composed of hydrogen,oxygen, nitrogen, and sulfur. For example, the stoichio-metric combustion of coal can be written as (19):

Combustion reactions are exothermic, producing moreenergy than consumed, from 5 to 7 kcal/ g of coal. De-pending on the rank of the coal, combustion of coalproduces between 6,000 and 16,000 Btu/lb on a dry,mineral matter free basis (fig. 7).Oxidation of coal occurs constantly. The temperature

of the coal is a function of the rate of heat generation

versus the rate of heat loss. When these processes occurat the same rate, the temperature of the coal remainsconstant. When the rate of heat generation is greater thanthe rate of heat loss, the temperature of the reactingsystem increases. Since the rate of heat generation isan exponential function of temperature and the rate ofheat loss is a linear function of temperature (fig, 8), asthe temperature increases, the reaction rate increasesfaster than the heat loss (20). Ignition is a function ofthe amount of energy released by a reaction and the rateat which it is released, as well as the rate at which energyis transferred from the reacting mass to the surroundings.The reaction rate is a function of the concentration ofreactants, carbon and oxygen, the surface area, particlesize, temperature, and activation energy.There are two types of ignition, forced and spontane-

ous. In forced ignition, energy is added to the system toincrease the rate of reaction to the self-sustaining point.In spontaneous ignition, there is no external heat source;natural reactions supply sufficient energy to sustain com-bustion. For AML fires, forced ignition sources includelightning, brush and forest fires, improperly controlledcamp fires and spontaneous combustion in adjacent mate-rials. There are no statistics on the number of AML firesstarted by forced versus spontaneous ignition. It is gen-erally considered that lightning and other surface fires areprobably prevalent sources of ignition in the western out-crop fires. In the Eastern United States, trash fires inareas where the coal outcrop has been stripped are con-sidered the most common source of ignition.Spontaneous combustion in the coal or coal refuse is

related to the oxidation of the coal to form carbon dioxide,

Figure 6.-Flre triangle.

7

~ 100•..--EE 80Ez0(J)0:: 60<tu0w~u, 169 \0 11

HEATING VALUE (rnrnf ),KEY

subA SubbituminousA Ivb Low volatile bituminoussubB SubbituminousB sa SemianthracitesubC Subbituminous C an AnthracitehvAb High-volatile A bituminous ma MetaanthracitehvBb High-volatile B bituminous mmf Mineral matter freehvCb High-volatile Cbituminous mmmf Moisture, mineral matter freemvb Medium volatile bituminous

Figure 7.-Heat of combustion versus rank of coal.

carbon monoxide, and water. The oxidation of pyrite andthe adsorption of water on the coal surface also are exo-thermic or heat generating reactions that increase theprobability of spontaneous combustion (21). Thermophilicbacteria also may contribute to raising the temperature ofthe coal. In waste banks, most of the oxygen diffusingfrom the surface is consumed by bacterial activitywithin afew feet. However, enough oxygen is available at depth tosupport combustion.

The normal ignition temperature for coal is between4000 and 5000 C. In laboratory experiments, the minimumtemperature at which a coal will self-heat under adiabaticconditions (all heat generated is retained in the sample)was 35" to 1400 C (22). Normal underground tempera-tures are 11° C or less, and ambient air temperatures areusually below 35° C. To reach temperatures at whichcombustion is self-sustaining, heat generation must begreater than heat loss. In most abandoned mines andwaste piles, conditions favor the retention of heat. Aslisted above, there are several exothermic reactions thatrelease energy. Heat is lost by convection or conduction.Since abandoned mines and waste piles have an essentiallystagnant atmosphere, convection accounts for very littleheat loss. Most heat transfer is probably by conduction tosurrounding strata, but rocks tend to be good insulators,keeping heat within the mine or waste bank. Temperatureand the factors that increase temperature are one elementin starting and sustaining a wasted coal fire.

In addition to a physical environment that favors theaccumulation of heat, other factors influence the propaga-tion of wasted coal fires, i.e., geologic setting, previousmining, the rank of the coal and environmental conditions.

tt-<ll.IJ:

TEMPERATURE ••

Figure a.-Rate of heat loss and rate of heat gain versustemperature.

The geologic factors that affect the propagation of minefires vary depending upon the geologic setting. In bitumi-nous coalfields, the depth of overburden, the degree offracturing and the nature of the overlying strata are theprimary geologic factors. Mines under shallow cover areusually under more permeable strata and above the watertable; shallow strata tend to be more highly fractured.Fires propagate toward the source of oxygen, and in shal-low beds and in areas near the outcrop, the concentrationof oxygen tends to be higher. However, wasted coal firescan smolder in atmospheres with less than 3% oxygen(23-25). Where the overburden is fractured, barometricpressure changes cause the mine to breathe, exhausting

8

combustion products and bringing in fresh air. Underthese conditions, fires spread very slowly, but continue toburn for very long periods.

In anthracite mines and in some western mines, the dipor pitch of the beds also influences the propagation offires. In anthracite areas, the intense folding and faultinghave contributed to subsidence fractures extending fromthe coalbed to the surface. On steep pitches, differencesin temperature and elevation are sufficient to control thecirculation of oxygen and fumes. A fire near the outcropof a steeply dipping bed can draw air from within themine, propagating the fire downdip. The movement of hotgases can transfer heat to other areas (fig. 9). The dis-tance between coalbeds determines the transfer of heatbetween beds and the possibility of propagation of a firefrom the source bed to adjacent beds.

In abandoned mines, the extent of previous mining is afactor in the spread of fires. The amount and condition ofcarbonaceous material left underground determines the ex-tent of the fuel supply. Roof coals, rider coals, and/orcarbonaceous shales, which eventually collapse into themine, are capable of initiating and sustaining combustion(17). Coal that spalls from ribs and pillars as well ascarbonaceous material in gob areas add to the fuel vol-ume. Because the broken coal has a larger surface areathan solid coal, it is more combustible. Main entriesappear to be high oxygen areas along which a fire maypropagate. A fire can establish a natural ventilationpattern in which combustion gases are exhausted at one

Noncombustiblebarrier

Figure g.-Transmission of heat by movement of hot gases.

point and fresh air drawn in at another. The number ofopenings in the outcrop, the number of ventilation shafts,the competency of the overlying strata, and the prevalenceof subsidence-induced fractures are other factors thatcontribute to the prolonged propagation of abandonedmine fires. The condition of the mine determines theamount and surface area of fuel and the availability ofoxygen, which determine the direction and rate ofpropagation of the fire.

The rank of a coal is apparently not a primary factor inthe incidence of AML fires (fig. 10). Although the lowerrank coals (lignite and subbituminous) tend to be moresusceptible to spontaneous combustion, the incidence ofAML fires in these coals is no greater than in higher rankcoals (bituminous and anthracite). Moisture in the coal-bed or waste bank has both positive and negative effectson fire propagation. If water is present on the surface ofthe coal, heat generated by oxidation is dissipated by theevaporation of the surface water. Prolonged drying of thecoal allows for increased sorption of oxygen. The sorptionof water vapor on dried coal is also an exothermic proc-esses that raises the temperature of the coal, increasingthe probability of ignition.

If wasted coal fires were propagated by a flame spreadmechanism, coal adjacent to the burning mass would beheated by conduction or radiation to its ignition temper-ature (26). Propagation of the fire would follow a con-tinuous pathway. However, in many AML fires, fire zonesare discontinuous, with no apparent propagation pathway(8, 27-29). In these fires, the movement of hot gases isbelieved to be a factor in propagation. The fire inducescirculating air currents, which carry fumes and heat intononadjacent areas. The effective ambient temperature ofa large area of the mine is slowly increased. As thetemperature of the coal increases, normal bed moisture islost and the rate of oxidation reactions increases. If theheat is not dissipated, the temperature of the coal con-tinues to rise until spontaneous ignition occurs.

In a discussion of factors influencing the initiation andpropagation of wasted coal fires, it is apparent that allfactors are related to the three elements, fuel, oxygen, andenergy. The amount of combustible material, its particlesize, surface area, and tendency to spontaneous combus-tion are fuel related factors. The presence of fracturesthrough which air can be drawn into the fire zone, circula-tion caused by the fire, and changes in barometric pressurecontrol the amount of available oxygen. The rate of heatgeneration versus the rate of heat loss, the heat-generatingreactions (oxidation of coal, oxidation of pyrite, surfaceadsorption of water vapor, bacterial activity), and theinsulation provided by adjacent strata control the amountof energy within the system.

Natural barriers to subsurface fire propagation basicallyaffect the availability of fuel and the generation and/or the

9

70

60

C/)l- SOoWJ0a:: 40o,LL0rr 30WCO~:::) 20Z

10

0-1----ANTHRACITE E BITUMINOUS W BITUMINOUS

COAL RANKLIGNITE SUBBITUMINOUS

Figure 10.-Distribution of AML fire control projects by rank. (E Bituminous· eastern bituminous; WBlturnlnous > western bituminous.)

retention of heat. Faults with vertical displacement maydisrupt the continuity of the coalbed and limit the amountof fuel. Interior boundary pillars between mines are con-sidered natural barriers to fire propagation because solidcoal seams do not burn. However, the surface of the pillarand any fractured or faulted areas can be combustionzones, and in practice, many boundary pillars are breachedand therefore, do not constitute a fire barrier. Boundarypillars at the outcrop, because of weathering and theavailability of oxygen, are not considered fire barriers. Infact, many AML fires propagate along the outcrop. Thewater table serves as a barrier by limiting the amount ofoxygen and by absorbing energy released by the fire. Inthe absence of these natural barriers, a fire in anabandoned mine can, in an extended period, burn fromoutcrop to outcrop.

HAZARDS OF ABANDONED MINED LAND FIRES

The primary hazards of AML fires are the emission oftoxic fumes and subsidence. However, these fires also canaffect the conservation of coal resources, ignite surfacefires, and affect the value of adjacent property. Thetype and degree of hazard can influence the selection ofan extinguishment technique and the extent of the extin-guishment project.

The potentially most serious problem associated withAML fires is the migration of toxic fumes from the firethrough overlying strata into homes or other enclosedsurface structures. A fire produces carbon monoxide, car-bon dioxide and water, and consumes oxygen. Carbonmonoxide is the most serious hazard. This colorless, odor-less gas readily combines with the hemoglobin of theblood, which normally transports oxygen; it replacesoxygen and forms carboxyhemoglobin. At blood concen-trations of 10% to 40% carboxyhemoglobin, headaches,dizziness, faintness, impaired motor coordination, nausea,and vomiting are symptoms of carbon monoxide poisoning.At levels of 40% to 70%, symptoms include increased res-piration and pulse rate, collapse, coma, and convulsions.Respiratory failure and death occur when 70% to 80% ofthe hemoglobin has been converted to carboxyhemoglobin(30-32).The threshold limit value (TL V)4 for carbon monoxide

is considered 50 ppm, a level that will produce 8% to 10%carboxyhemoglobin. The recommended occupational ex-posure is 35 ppm for an 8-h workday (33). The effect ofcarbon monoxide exposure increases with duration of

'The threshold limit value (TLV) is a time-weighted averageconcentration of a substance in air to which workers may be exposedduring a normal 8-h day or 40-h week for an indefinite period withoutadverse effect.

10

exposure, higher humidity, and lower barometric pressure.The rate of effect also increases with increased physicalexertion. Other factors in individual response to carbonmonoxide exposure are age (very old and very young),pregnancy, heart disease, poor circulation, anemia, asthma,lung impairment, or the presence of drugs-alcohol in theblood.

Carbon dioxide, which is also produced by fires, isnormally present in air at a concentration of 0.03%. TheTLV for an S-h daily exposure to carbon dioxide is 0.5%,provided the percentage of oxygen is normal (34). Therecommended occupational exposure as determined byNational Institute of Occupational Safety and Health(NIOSH) is 1% carbon dioxide by volume for a 10-h shiftin a 40-h week. During prolonged exposure to elevatedcarbon dioxide concentrations (1% to 3%), excess hydro-gen and bicarbonate ions are produced. The body re-moves this excess acid through an increased breathing rateor through excretion. Prolonged exposure to elevated con-centrations of carbon dioxide may cause a loss of efficiencyin performing physical exercise, but has had no observedeffect on problem-solving or eye-hand coordination.

Prolonged exposure to slightly decreased levels ofoxygen has much the same effect as exposure to increasedcarbon dioxide. If the concentration of oxygen decreasesfrom the normal 20.95% to less than 16%, breathing andpulse rates increase. At less than 10% O2, nausea, vomit-ing, and loss of consciousness will occur. At less than 6%O2, convulsions and respiratory failure occur. Personswith cardiac, pulmonary, or hyperthyroid problemsexperience severe effects of oxygen deficiency at lesserreductions in oxygen concentration.

In most cases, the possibility of toxic fumes from anAML fire affecting people on the surface is very small.However, these fumes can migrate for considerable dis-tance through cracks and fractures in the overlying strata.They can enter houses through sewers or foundationcracks and can accumulate in closed, unventilated areaslike basements and closets. Usually, ventilation can beused to dissipate fumes in surface structures.

Fumes and smoke from AML fires create a seriousatmospheric pollution problem. The plume (fig. 11) seenat many AML fires is frequently a steam condensate.Smoke indicates the presence of particulates with thevapor emitted from the fire zone. In addition to toxicgases, fumes from a fire zone frequently contain coal dis-tillates, mercaptans, or hydrogen sulfide, which createnoxious and unpleasant odors. Fumes may be responsiblefor killing some species of surface vegetation in ventareas, although mosses appear to tolerate the fumes andthrive in the higher temperature areas. Small animals,seeking the warmth near vent areas, may be victims of thelethal fumes.

Subsidence occurs when a fire consumes a portion ofthe coal, removing support from the overlying strata. Thestrata over the mine then fall into the mine void. Thesurface expression of subsidence depends upon several

factors, such as the depth to the mine and the competenceof intervening rock units. The surface expression of thesubsurface fire may be a small vent (fig. 12), a fractureline (fig. 13), a slight depression, or a relatively largesinkhole (fig. 14). The width of a subsidence feature canvary from a few inches to several feet. The depth is alsovariable. If the coalbed is relatively shallow and theoverburden is primarily unconsolidated material, it ispossible for the subsidence feature to extend from thesurface to the mine void.

The hazards associated with subsidence due to minefires depend upon several factors. The first is locationwith respect to population density. A subsidence featurein a populated area is inherently more hazardous than onein an inaccessible or remote area. However, a subsidenceevent in an urban or suburban area is more likely to beabated quickly. In remote areas, subsidence may consti-tute a long-term threat to hikers and hunters. Subsidence,whether due to mine fires or simply to mining, affects thestability of surface structures, causing minor to major dam-age. In addition to the normal subsidence problems, ifsubsidence is related to a mine fire, cracks created infoundations can act as conduits for toxic fumes. Roads,surface streams, sewers, and waterlines also can be af-fected by any subsidence event. If the subsidence is re-lated to a mine fire, cracks in sewerlines may providepathways for fume migration into buildings. Fracturezones in the overburden serve as chimneys for combustionproducts and can supply fresh air to the undergroundcombustion zone.

Another hazard related to subsidence features is thepossibility that people or livestock will fall into the largersinkholes. Fumes from a fire compound this danger. Acorollary hazard is related to the tendency of people to usesuch sinkholes as trash dumps. Hot fumes from the minefire can accelerate the tendency of trash or garbage tospontaneously ignite, producing a fire that can spread tosurface vegetation and structures. Western outcrop fireshave been credited with starting brush, grass, and forestfires.

The hazards of mine fires are insidious. They are notlike hurricanes, tornados, earthquakes, or floods, in whicha single catastrophic event affects many people. Fires inabandoned mines and waste banks are protracted events;they can have a moderate effect on people for 20 years ormore. The most widespread effect is the environmentaldegradation caused by noxious odors and fumes. A moreserious, but less prevalent, effect is subsidence and/orfume migration into surface structures. The most exten-sive disruption, social and economic, caused by an aban-doned mine fire is the Centralia mine fire in the anthraciteregion of Pennsylvania (35). At Centralia, the inability tocontrol the mine fire forced the relocation of approxi-mately 1,000people at a cost of $42 million. Although theeffects of most mine fires are less extensive, they are noless severe to the people involved.

11

Figure 11.-Smoke plume from AML fire.

Figure 12.-Small vent at AML fire.

12

Figure 13.~racture line above AML fire.

Figure 14.-5inkhole at AML fire.

13

EXTENT OF ABANDONED MINED LAND FIRE PROBLEMS

An evaluation of the extent of the AML fire problem isbased on estimates of the current number of fires, the areaand population they affect, and probability of successfulextinguishment. The responsibility for AML fire controlefforts and the available funds have been and are factorsin the extent of the AML fire problem.

INCIDENCE OF ABANDONED MINED LAND FIRES

Although fires in outcrops and abandoned mines havebeen occurring for more than 200 years, prior to 1949 noFederal or State agency collected information on the prev-alence of AML fires. After 1949, the Bureau had theauthority to control fires in abandoned underground minesand in outcrops. In conjunction with this work, reportslisting fire control projects, extinguishment methods, andcosts were published (7, 36-37). Efforts were also made toestimate the extent of the AML fires and the total cost ofcontrolling such fires (38-39). With the passage of theSurface Mining Control and Reclamation Act (SMCRA)in 1977, AML fire control came under the authority of theOffice of Surface Mining, Reclamation and Enforcement(OSMRE) or the States that had an approved AMLprogram. Under this program, a national inventory ofAML problems, including fires, was compiled from datasubmitted by the States and Indian tribes. Based on thesesources, information on past fires is reviewed and anestimate made of the current extent of the problem.

For the work done by the Bureau, the country wasdivided into three regions: the Eastern bituminous region,the anthracite region, and the Western region includingAlaska. Between 1949 and 1972, 70 fire control projectswere executed in the Eastern bituminous region (7).Three of these projects were located in West Virginia, onein Kentucky, one in Maryland, and the remaining 65 werein western Pennsylvania (table 1). Surface sealing, aloneor in combination with other methods was used on 45 ofthe fire control projects. Ten of the fires were excavated,and fire barriers were used at 27. At 15 fire control proj-ects, either dry fly ash or a fly ash slurry was injected tocontrol the fire.

In the anthracite region, between 1949 and 1980, theBureau attempted to control 18 fires (27, 40). These in-volved 29 fire control projects; 18 included some formof excavation, 14 involved flushing, 3 surface seal proj-ects, and 1 control by natural inundation (table 2). Inseven of the projects, the planned fire control includedmore than one method. During the period between 1930and 1970, an additional 13 fires were controlled by mining

companies. Between 1980 and 1986, the OSMRE exca-vated six anthracite mine fires.

In the Western region, 590 coal fires were reported tothe Bureau between 1949 and 1979. Of 158 fire controlprojects during this period (table 3),511% used excavation,2% were isolation projects, and 87% utilized some form ofsurface sealing (9, 36-37).

In 1977, the Bureau listed 261 fires in abandoned minesand inactive outcrops (38). Twelve of these fires were inthe anthracite region, and 197 were in the western States(table 4). Ten coal-producing states reported no fires inabandoned mines or outcrops.

A survey by the Bureau in 1968 located 292 burningcoal refuse banks containing 270 million st of coal refuse(39). Of these, 132 were in West Virginia, 184 were inother eastern and midwestern States, and the remaining 24were in the western States (table 5). Forty-five percent ofthe banks were located within 1 mile of a community andsix of the burning banks affected communities of morethan 100,000 people. Sixty deaths were attributed to ac-cidents at burning coal refuse piles.

Under the SMCRA, the OSMRE authorized the Na-tional Inventory of Abandoned Mined Land Problems tolocate and identify AML problems and to estimate the costof reclamation (41). The inventory has also been used asa basis for allocating funds from the discretionary portionof the AML fund. The original inventory was updated in1986 and 1987 (42) and has been under review and reeval-uation since 1989 (43).

Under the inventory, AML problems were classified un-der six major categories and 16 keywords. Fire problemswere described as "surface burning" or "gases from under-ground burning." Problems were further classified as: pri-ority 1 - presenting extreme danger to the health, safety,and general welfare; priority 2 - protection of the health,safety, and general welfare; and priority 3 - restoration ofland and water resources degraded by past mining prac-tices. Evaluation of the seriousness of a problem is alsobased on the evidence of impact, the potential for propa-gation to populated areas and expression of concern byaffected people, as well as by the cost of reclamation ascalculated according to OSMRE guidelines (44).6

5There are more than 158 entries in table 3 since phases of a projectcompleted in different years are listed separately.

6The content, interpretation, and use of the AML inventory has beenthe subject of extended debate (49). In this report, it is used only toindicate the probable extent and distributions of the AML fire problems.

14

Project name

Table 1.-Fire control projects in eastern bituminous region, 1949-1972

Control methodLocationCounty, State

Date

Agnew Rd .Ardmore Blvd. . .Arlington Hts. . .Baldwin Borough .Becks Run Rd .Bedford Dwellings .Blairsville .Boyds Hollow Rd .Bradenville .Brinkerton .Brisbin Borough .Bullskin-U. Tyrone .Calamity Hollow .Carpentertown .Carroll Township .

Do .Catfish Run .Churchview Ave .Clairton .Coal Hollow Rd .Collier .Commonwealth Ave. ..Connemaugh .Cook Plan .Division St. .Division & 3rd .Fairmont .Fallowfield .Garden City .Grant St. .Green Valley .Harrison County .Hempfield Township ..Highland Terrace .Jefferson Borough .Johnsons Hollow .Ken Ridge Dr. .Kennedy .Klondike .Larimer .Liberty .Lick Run .Lloydsvllle .

Do .Longview Land .Lookout Ave. . .Masontown .Meyersdale .Monongahela City .Monroeville (I) .Monroeville (II) .Moon Township .Newell .Petermans Corners .Peters Creek .Pikeville .Plum .Pricedale .Robinson .Ross Farm .Rostraver .

Allegheny, PA .· .do .· .do .· .do .· .do .· .do .Indiana, PA .Allegheny, PA .Westmoreland, PA .· .do .Clearfield, PA .Fayette, PA .Allegheny, PA .Westmoreland, PA .Washington, PA .· .do .Allegheny, PA .· .do .· .do .· .do .· .do .· .do .Indiana, PA .Westmoreland, PA .· .do .· .do .Marion, WV .Washington, PA .Allegheny, PA .Westmoreland, PA .Allegheny, PA .Harrison, WV .Westmoreland, PA .· .do .Allegheny, PA .Fayette, PA .Allegheny, PA .· .do .Allegany, MD .Westmoreland, PA .Allegheny, PA .· .do. . .Westmoreland, PA .· .do .Allegheny, PA .Westmoreland, PA .Preston, WV . . . . . . . .Somerset, PA .Washington, PA .Allegheny, PA .· .do .· .do .Fayette, PA .Allegheny, PA .· .do .Pike, KY .Allegheny, PA .Westmoreland, PA .Washington, PA .Westmoreland, PA .· .do .

Surface seal .Plug-seal .Trench-seal .Excavation .· .do .· .do .Trench-plug-seal ..Plug-seal .· .do .Trench-seal .Trench-seal-flush ..· .do .Trench .Surface seal .· .do .Seal-flush .· .do .Surface sealTrench-seal .Su rface-seal . . .· .do .Plug-seal .· .do .Trench barrier .Seal-flush .· .do .Excavation .Trench .Seal-flush .Flush .Surface seal .Excavation .Trench barrier .Trench .Surface seal .· .do .Flush .Surface seal .Trench .Plug-seal .Excavation .Surface seal .· .do .Flush .Surface seal .· .do .Excavation .Trench .Seal-flush .Excavation .Trench-seal .Surface seal .Trench .Flush .Plug-seal .Surface seal .· .do .Barrier-seal .Plug-seal .Excavation .Seal-flush .

1954196219621953197219531965196719661964195819651963196919541970195819561962196419671964196419501958195919501961196519671957195919541961195519671969196719591971196019491949196819631958196119601969196019601971196019701968195919611952196819521956

15

Project name

Table 1.-Fire control projects in eastern bituminous region, 1949-1972-Continued

LocationCounty, State

Control method Date

Rostraver #2 .Santiago .Smith Township (I) .Smith Township (II) .Turnpike .SW Monessen .U. Tyrone .U. Wheyl .Young Township .

Westmoreland, PA .Allegheny, PA .Washington, PA .. .do .Allegheny, PA .Westmoreland, PA .Fayette, PA .Westmoreland, PA '"Indiana, PA .

Plug-seal .Trench .Excavation .Surface seal .Flush .Surface seal .Seal-flush .Plug-seal-flush ... do .

Project name

Table 2.-Fire control projects in Pennsylvania Anthracite region, 1950-1987

196619571961196219691953196919711972

Archbald .Carbondale .

Cedar Ave .

Centralia .

Coal Run .Eddy Creek .Enyon Street .Forestville .Hazelton .Hughestown .Kehley Run .Kulpmomt .

Larksville .Laurel Run .Maffett . . . . . . . . . . . . .Mt. Carmel .

North Scranton .Peach Mountain .Shamokin .Shenandoah .Sugar Notch .Swoyersville .Throop .Tower City .Warrior Run .

Anthracite field

Northern .· .do .· .do .· .do .· .do .· .do .W. Middle .· .do .· .do .· .do .Northern .· .do .NAl .

E. Middle .Northern .W. Middle .· .do .· .do .· .do .Northern .· .do .· .do .W. Middle .· .do .· .do .Northern .Southern .W. Middle .· .do .Northern .· .do .· .do .Southern .Northern .

Control method

Excavate .Flush .Excavation .Excavation-flush .Flush .Excavation .Excavation-flush .Barrier .· .do .Trench-flush .Flush .· .do .Excavation .· .do .· .do .· .do .Trench .Excavation .Excavation-trench .Trench-excavation ..Flush-excavation .Excavation .Seal .Seal-flush .Excavation .Flush .Excavation .Seal .Innundation .Excavation .· .do .· .do .Flush-excavation .· .do .

Date

1985195019741953196519731966197419781963197419651986196919841969195019581960198519711987195019521967196019491951196019841973196819541971

lNA Not available.

16

Table 3.-Fire control projects in Western United States, 1949-1977

Datecompleted

Project name! Location Control method

County, State

Alkali Butte. .. .. Fremont, WY Surface seal .Area "D" (I) . . . . . . . . . . . Kane, UT . . . . . .do. . .Area "D" (II) .do. . .do. . ..Area "D" (III) . .do. .. . .do. . .Area "D" (IV) . .do. .. . .do. . .Area "D" M . .do. . .do. . .Area "D" (VI) . . . . . . . . . . . .do. . . . .do. . .Arizona Black Mesa Navajo, AZ . . . .. . .do. . .Axial (I) Moffat, CO . . . . . . Excavation-seal .Axial (II) . .do. Surface-seal .Baker's Garden Richland, MT . .do. . .Barker Dome (I) San Juan, NM . .do. .. . .Barker Dome (II) . . . . . . . . .do. .do. .. . .Belfield Billings, ND Excavation-seal .Birch Creek (I) Emery, UT .. . . . . Surface seal .Birch Creek (II) . . . . . . .do. .do. . .Birch Creek (III) . .do. .do.. .Big Buck . .. Navajo, AZ . . . . . . .do.. .Big Smokey (I) Kane, UT .do.. .Big Smokey (II) .do. .do.. .Black Mesa No.2 (I) . . . . Navajo, AZ . . . . . . . Excavation-seal .Black Mesa No.2 (II) . .do. .do. .. . .Black Mesa No.2 (III) . . . . .do. Surface seal .Black Raven . . . . . . . . . . Garfield, CO . . . .do.. .Boyd No.1. . . . . . . . Navajo, AZ . . . . . . .do. .. . .Burnham No.2. . . . . . San Juan, NM .do.. .Burnham No.3. . . . . . . .do. .do.. .Burning Coal Mine . . . . . Converse, WY .do. .. . .Burning Hills No.2 (I) . . . Kane, UT .do.. .Burning Hills No.2 (II) .. .do. .do.. .Canaan Creek Garfield, UT . . . . . . . .do. . .Canfield. . . . . . . . .. . . . Campbell, WY . .do. .. . .Canfield No.2. .. .do. . .do. .. . .Castle Garden . . Freemont, WY . .do. .. . .Coal Bank Sheridan, WY . . . . . . . .do. ..Coal Draw Campbell, WY . . . . . .do. . .Coal Gulch " Mesa, CO . . . . . .do.. . . .Coalmont. . . . . . . . . . Jackson, CO . .do. .. . .Coalmont Mine No.1 .. do. . . . .do. . .Cottontail Butte Golden, ND .. . . . ExcavationCoyote Creek.. Powder, MT . . . Surface seal ..Crosby. . . . . .. Hot Springs, WY . . .do.. ..Crownpoint .. . . . . McKinley, NM . .. ... .do.. .Curry . . . . . . . .. Johnson, WY . . .do.. .Davis . . . . . . . . . Slope, ND ExcavationD & H (I) Garfield, CO . . . . . . Surface sealD & H (II) .. .. . . . .do. ... . . . .do.. .Debebekid Lake No.1 Navajo, AZ . . . . .do. . .Debebekid Lake No.2. . . .do, .. . . . .do.. .. ..Dead Horse No.1. . . . . Campbell, WY . . . .do. .. . .Dead Horse No.2. . . . . . .do. .... . . . .do. .. . .De Mores . . . . .. Golden, ND . . . . .do. . .Deer Creek. . . .. Dawson, MT . . . .do. . .Dinnebito NO.1. . . Navajo, AZ . . .do.. .Dry Creek .. .. . . . Campbell, WY .do. . .Duck Creek . . . . Converse, WY .. .do.. ..Dugger Rollins Delta, CO . . . . . .do. . .Dutch Creek. . . . . . . Sheridan, WY . . . .do.. . .. ..

Do. . . . .do. . . . . .do.. .East Canoncito.. Bernalello, NM . . .do.. .lRoman numerals in parentheses refer to separate phases of one project.

195619691969197019711973197419551964197219661966196619621964197019791964196719681962196319641972196419611961195019761977197119501964195419671969195319661974196619641967197219591977197319751964196419691969196719561964197619641952196119611963

Table 3.-Flre control projects In Western United States, 1949-1977-Contlnued

17

Datecompleted

Project name! Location Control methodCounty, State

East Dot Klish No.1. . . . . . Navajo, AZ . . . . . . . . . Surface seal .East Dot Klish No.2. . . . . . . .do. Excavation-seal .East Dot Klish No.3. . . . . . . .do. Surface seal .East Quitchupah Emery, UT . .do. . .Elk Creek. . . . . . . . . . . . . . Campbell, WY . .do. . .Farmer's Mutual . . . . . . . . . Mesa, CO . .do. . .Fish Canyon . . . . . . . . . . . . Routt, CO . .do. . .Ford Butte San Juan, NM Excavation .Fuller . . . . . . . . . . . . . . . . . Dawson, MT . . . . . . . . Surface seal .Garden Navajo, AZ . . . . . . . . . . .do. . .Gebo . . . . . . . . . . . . . . . . . Hot Springs, WY Excavation .George Harvey . . . . . . . . . . San Juan, NM Surface seal .Glendive Creek Dawson, MT . . . . . . . . . .do. . .Haas Garfield, CO . . . . . . . . Excavation .Haileyville Pittsburgh, OK . . . . . . Surface seal .Hart . . . . . . . . . . . . . . . . . . Richland, MT . .do. . .Hoffman Creek. . . . . . . . . . Carbon, UT . .do .Hoffman Creek (II) . .do. . .do, . .Hoffman Creek (III) . . . . . . . . .do. . .do. . .Hogback (I) San Juan, NM . .do. . .Hogback (II) . . . . . . . . . . . . . .do. . .do. . .Homer . . . . . . . . . . . . . . . . 3rd Judicial, AK Barrier .Horse Camp Campbell, WY Surface seal .Hot Point . . . . . . . . . . . . . . Garfield, CO . . . . . . . . . .do. . .Hunt Custer, MT . . . . . . . . . . .do. . .I.H.I. Mine Garfield, CO . . . . . . . . Barrier .I.H.1. #2 Mine. . . . . . . . . . . . .do. Surface seal .Indian Coulee. . . . . . . . . . . Rosebud, MT . . . . . . . Excavation-seal .Iron Springs . . . . . . . . . . . . Big Horn, MT . . . . . . . Surface seal .Jennison Richland, MT . .do. . .Killsnight Creek Big Horn, MT . . . . . . . . .do. . .Lame Deer . . . . . . . . . . . . . . .do. Excavation-seal .La Plata . . . . . . . . . . . . . . . San Juan, NM Surface seal .Laur . . . . . . . . . . . . . . . . . . Campbell, WY . .do. . .Unwood (Utah) Sweetwater, WY . . . . . . .do. . .Uttle Missouri 1 McKenzie, ND Excavation .Uttle Missouri 2 . .do. . .do. . .Uttle Missouri 3 . .do. Surface seal .Uttle Thunder. . . . . . . . . . . Campbell, WY . .do. . .Logging Creek . . . . . . . . . . Rosebud, MT . . . . . . . . .do. . .M.A. . . . . . . . . . . . . . . . . . . Sweetwater, WY . . . . . . .do .McKenzie County . . . . . . . . McKenzie, ND Excavation .Mesa Verde (I) San Juan, NM Surface seal .Mesa Verde (II) . . . . . . . . . . . .do. . .do. . .Mesa Verde (III) . .do. . .do. . .Mesa Verde (IV) . .do. . .do. . .Mexican Springs McKinley, NM . . . . . . . Excavation .Middle Prong Wild Horse .. Campbell, WY Surface seal .Minnesota Creek Delta, CO . . . . . . . . . . . .do. . .Moose Creek 3rd Judicial, AK . .do. . .Moyer Gulch Campbell, WY Barrier-seal .Mt. Garfield Mesa, CO Surface seal .Navajo 1 San Juan, UT . . . . . . . . .do .Nenamo . . . . . . . . . . . . . . . AI< . . . . . . . . . . . . . . . . .do. . .Newcomb San Juan, NM . .do. . .Newton Murphy 1 . . . . . . . . Dawson, MT . . . . . . . . . .do. . .Newton Murphy 2 . . . . . . . . . .do. ... do. . .Nine Mile 1 Sweetwater, WY . . . . . . .do. . .Nine Mile 2 . .do. . .do. . .Nine Mile 3 . .do. . .do. . .Ninilchik. . . . . . . . . . . . . . . AK . . . . . . . . . . . . . . . Excavation

lRoman numerals in parentheses refer to separate phases of one project.

1964196219641966195219691972196619551964197119601956196119681966195819591976196819721954197219651961194919531962196819611964196219541950195419601960196119501964197319591967196819711978196719701961195419501969196719721956195919591971197119711971

18

Table 3.-Fire control projects In Western United States, 1949-1977-Continued

Datecompleted

Project name! Location Control methodCounty, State

North Park. . . . . . . . . . . Jackson, CO Barrier .Nuxoll. . . . . . . . . . . . . . . Custer, MT . . . . . . . . . Surface seal .Onion Lake Montrose, CO . .do. . .Owens (I) . . . . . . . . . . . . Rosebud, MT . . . . . . . . .do. . .Owens (II) . .do. . .do, . .•........Owens (III) . .do. . .do. . .Owl Creek Hot Springs, WY . .do. . .Padlock. . . . . . . . . . . . . Campbell, WY . .do .Park. . . . . . . . . . . . . . . . Richland, MT . .do. . .Poposia . . . . . . . . . . . . . Fremont, WY . .do. . .Powder River Sheridan, WY . . . . . . . . .do. . .Pyle Dam . . . . . . . . . . . . San Juan, NM . .do. . .Recci . . . . . . . . . . . . Sevire, UT . .do. . .Reservation Creek Rosebud, MT . . . . . . . . .do. . .Rio Puerco . . . . . . . . . . . Sandoval, NM . .do. . .Robertson Converse, WY . .do. . .Rosebud No.1. . . . . . . . Jackson, CO . .do. . .Rosebud No.3. . . . . . . . . .do, . .do. . .San Juan . . . . . . . . . Sandoval, NM . .do. . .Skull Creek Rio Blanco, CO Barrier .Slagle Ouray, CO . . . . . . . . . Surface seal .Soda Lake Carbon, WY . . . . . . . . . .do. . .Soldier Gulch . . . . . . . . . Rosebud, MT . . . . . . . Excavation .Smokey Mountain (I) Mesa, CO Surface seal .Smokey Mountain (II) .. . . .do. Excavation-seal .Smokey Mountain (III) .. . .do. Surface seal .Smokey Mountain (IV) .. . .do. . .do. . .Smouse . . . . . . . . . . . . . San Juan, NM . .do. . .Snake River. . . . . . . . . . Carbon, WY . . . . . . . . . .do .Southeast Dot Klish Navajo, AZ. . . . . . . . . . . .do. . .Spotted Horse Campbell, WY . .do .Standing Rock . . . . . . . . McKinley, NM . . . . . . . . .do. . .Steamboat Springs Navajo, AZ. . . . . . . . . . Excavation .Stony Butte San Juan, NM Surface seal .Stove Canyon Garfield, CO . . . . . . . . . .do. . .Ten Mile Draw Sweetwater, WY . . . . . . .do. . .Terrett Custer, MT : .. do. . .Terry Prairie, MT . . . . . . . . . . .do. . .Three Forks Powder River, MT . . . . . .do. . .Toadlena San Juan, NM . .do. . .Traub. . . . . . . . . . . . . . . Powder River, MT . . . . Excavation-seal .Tsaya . . . . . . . . . . . . . . . San Juan, NM Excavation .Tsaya No.2. . . . . . . . . . . .do. Excavation-seal .Two Trees Powder River, MT . . . . Surface seal .Undem 2 Prairie, MT . . . . . . . . . . .do. . .Ute Mountain 1 (I) San Juan, NM . .do. . .Ute Mountain 1 (II) . .do. . .do. . .Ute Mountain 2 (I) . .do. . .do. . .Ute Pasture (I) Montezuma, CO . . . . . . .do. . .Ute Pasture (II) . . . . . . . . . .do. . .do .Ute Pasture (III) . .do. . .do. . .Virgil Widner Dawson, MT . . . . . . . . . .do, . .Watson Navajo, AZ. . . . . . . . . . . .do. . .West Canoncito Bernalello, NM . . . . . . . .do. . .West Quitchupah Emery, UT .. . . . . . . . . .do. . .White River No.1. . . . . . Rio Blanco; CO . .do. . .White River No.2. . . . . . . .do. . .do. . .White Rock San Juan, NM Excavation .Wild Horse Creek. . . . . . Campbell, WY . .do .Wilson Navajo, AZ. . . . . . . . . . Surface seal .Windmill . .do. . .do. . .Wise Hill No.3. . . . . . . . Moffat, CO . . . . . . . . . . .do. . .Yellow Jacket Pass. . . . . Rio Blanco, CO . .do, . .Zion (I) . . . . . . . . . . . . . . Kane, UT . .do. . .Zion (II) . .do. . .do. . .

19491964195519741975197619671951196119571972195019571958196019641963196919511951195419541967196119621963196419601953196419701963196719601966197219591956195819601966196619691957196519651966196519681971197719581964196319651959195919671970196419641976195419651974

!Roman numerals in parentheses refer to separate phases of one project.

Table 4.-Abandoned mined land fires, 1977

StateNo. uncon-trolled fires

Estimatedreclamation

cost, $K

Alaska .Arizona .ColoradoKentucky .....Maryland ..•..Montana ..•..New Mexico ...North Dakota ..Ohio ...•....Pennsylvania:

Anthracite ...Bituminous ..

South Dakota ..Texas .... " ..Utah .•.......Washington ...West Virginia ..Wyoming .

Total .

3104752659157

40309

1,641463263853233185920

123021172826

58,6538,165

2214

63659

1,0362,060

261 75,552

Table 5.-Abandoned coal waste bank fires

No. uncon- Estimated EstimatedState trolled fires area, acres reclamation

cost, $K

Alabama ..... 6 100 25,300Colorado .•... 15 130 27,500Illinois ....... 4 140 12,500Kentucky 27 160 37,000Maryland ..... 2 3 100Montana ..... 3 6 500Ohio ........ 6 20 3,200Oklahoma .... 1 1 100Pennsylvania:

Anthracite ... 26 680 96,100Bituminous .. 48 580 96,100

Utah ........ 4 30 4,900Virginia ..... . 17 80 6,300Washington ... 1 100 5,300West Virginia .. 132 1,190 153,100

Total .... 292 3,200 467,900

The AMLinventory is not comprehensive. It is a listof priority one, priority two, and some priority threeabandoned minesites requiring reclamation according toguidelines and restrictions established by OSMRE. Theaffected area is usually an estimate of surface area thatwould be included in the reclamation project. The esti-mated cost is based on the volume of the fire area (lengthand width of the inferred combustion zone times theaverage overburden depth) and an average unit value forexcavation.

Because of the way in which the data are collected, (i.e.,fires requiring immediate abatement or those posing a

19

potentially serious threat are included), the prevalence offires in the Eastern United States (table 6) may be relatedto the extent of past mining, the proximity to populatedareas, and to geological conditions, especially in theanthracite fields, However, priority one and two AMLfires are essentially those that have a serious impact onlocal populations. Fires that are located in remote orinaccessible areas are not included in the AML inventory.This may account for some of the difference between thedata previously reported and that in the AML inventory.

From the data in the inventory (table 6), fires inPennsylvania account for approximately 25% of the esti-mated cost of controlling surface (wastebank and outcrop)fires and 97% of the cost of controlling subsurface (aban-doned mine) fires. The subsurface fires in Pennsylvaniaare divided evenly between bituminous and anthracite re-gions. In Pennsylvania, bituminous mine fires account foronly 5% of the estimated cost, with the cost per projectranging from $150 thousand to $14 million with the aver-age cost at $2.8 million. Fires in the anthracite region thataccount for 95% of the estimated cost, range from $280thousand to $183million with the average at $42.6 million.If the data for Pennsylvania are excluded, the averageestimated cost of controlling a surface fire is $158 thou-sand, and the average estimated cost of controlling a sub-surface fire is $306 thousand. On an historical basis, lessthan 10% of AML fires have been in the anthracite coal-fields (fig. 15), but anthracite fire control projects havebeen at least 10 times as expensive as projects to controlfires in lower rank coals (fig. 16).

As of 1989, OSMRE had obligated over $45 million toan additional 278 priority 1 (emergency) fire control proj-ects in 19 States (table 7). Nine completed fire controlprojects were listed in the Abandoned Mined Land Inven-tory System (AMLIS) reports in 1989. Based on the dataavailable, there currently are over 600 fires associated withpast mining that require some form of remediation .

Although there is some degree of uncertainty in thisdata, particularly regarding size and cost of abatement,it can be used to indicate the magnitude of the AML fireproblem. Twenty-five percent of the priority two surfacefires and 38% of the underground fires are located in thewestern coalfields, Sixty-eight percent of the surfacefires and 60% of the underground fires are located inthe Appalachian coalfields. Of these, Pennsylvania andWest Virginia account for over 50% of the undergroundfires. One hundred thirty-four of the 154 surface fires inthe Appalachian region are in Kentucky, Pennsylvania, andWest Virginia. Less than 5% of all fires are found in theinterior coalfields, Almost 75% of the emergency fire con-trol projects are located in Kentucky, Pennsylvania, andWest Virginia.

20

Table 6.-Abandoned mine fires, AML Inventory, October 1988

State-Tribe Surface Est. area, Cost, $K Underground Est. area, Cost, $Kfires acres fires acres

Alabama ............ 14 57 601 1 NA 107Colorado ............ 9 20 173 11 12 885Hopi ............... 0 0 0 1 NA NAIowa ................ 1 10 97 0 0 0Illinois .............. 2 606 260 0 0 0Indiana ............. 3 68 833 0 0 0Kansas .............. 2 40 400 0 0 0Kentucky ............ 43 176 5,698 20 188 9,953Missouri ............. 6 9 474 1 1 NAMontana ............ 23 148 2,589 6 94 1,353Navajo .............. 3 31 320 0 0 0New Mexico .......... 4 31 686 1 30 414North Dakota .... , .... 1 NA NA 0 0 0Ohio ............... 4 5 280 0 0 0Pennsylvania ......... 33 143 11,963 35 4,933 721,552Tennessee ........... 2 1 5 0 0 0Utah ............... 9 20 731 6 32 5,187Virginia ............. 10 23 838 1 1 240Washington .......... 1 2 10 0 0 0West Virginia ......... 48 243 16,351 3 27 588Wyoming ............ 7 17 138 13 40 887

Total ............. 225 1,650 42,447 99 5,358 741,116NA Not available.

State

Table 7.-Federal AML fire control projects by State, 1989

Number of fires Obligated funds, $KAlabama .Arkansas .Arizona .Colorado .District of Columbia ..Iowa .Illinois .Indiana .Kansas .Kentucky .Missouri .Montana .North Dakota .New Mexico .Ohio .Pennsylvania .Utah .Virginia .West Virginia .Wyoming .

Total .

1 1451 112 85210 7851 321 14 252 322 578 6,3591 1,04311 4094 59 2556 4388 6,3583 315 92742 7,2957 593

278 45,312

100

90 KEY

~ Number of projects80 I2Q1l Project costs

....J 70~I-0 60I-u,0 50I-Z

40I.LJ()a:I.LJ 30Q..

20

10

0Anthracite Bituminous Lignite Sub bituminous

COAL RANKFigure 15.-Percentage distribution of occurrence and cost of

AML fire control projects by coal rank.

21

2.5.--------------------------------------------,

:2 2:2-t4

I-(I)0 1.5UI-UWJ00:::0...W0<{o:w><{ 0.5

O+ (A~ANlliRACITE E BITUMINOUS W BITUMINOUS LIGNITE SUBEITUMINOUS

RANKFigure 16.-Average cost of AML fire control project versus coal rank. (E Bituminous - eastern bituminous; W Bltumlnous - western

bituminous.)

CONTROL OF ABANDONED MINED LAND FIRES

In 1948, the Appropriations Bill of the Departmentof Interior, Public Law 841, 80th Congress, 2nd session,authorized the Secretary of the Interior to expend appro-priated funds to control or extinguish fires in inactive coaldeposits, paying not less than one half of the expenditure.The States, their subdivisions or the owners of privateproperty were responsible for the remaining cost. Prior tothis act, the Federal Government, through the Bureau, hadbeen limited to investigating reported fires and providingconsultation and advice. In 1954, Public Law 738, 83rdCongress, provided funds for surveys, investigations, andresearch into the causes and extent of outcrop and under-ground fires in coal formations and into methods for con-trol or extinguishment of such fires. It also allowed theBureau to plan and execute projects for the control andextinguishment of such fires under certain conditions: (1)in coal owned or controlled by the Federal Government,(2) in coal formations owned by the Federal Government

under privately owned land, (3) in privately owned coalformations that endangered Federally owned coal, and (4)in abandoned mines on privately owned land, providingthat the State, local government, corporation, or individualpaid half of the total cost of the work.In 1965, the role of the Federal Government in con-

trolling abandoned mine fires was expanded by Public Law89-4 of the 89th Congress, the Appalachian Regional De-velopment Act. This act authorized the Secretary of theInterior to plan and execute projects to extinguish under-ground and outcrop mine fires in the Appalachian Regionin accordance with the provisions of the 1954 law, withoutregard to appropriations ceiling. It also authorized theFederal Government to expend up to 75% of the cost offire control projects on non-Federal land.In 1977, the SMRCA established an abandoned mine

reclamation program, including a fund to reclaim andrestore areas affected by past mining (45). Money for thefund is collected from coal operators based on short tonsof coal produced. Collection of these fees was originally

22

mandated from the 1977 passage of the act until 1992, andthen extended until 1995. Monies from the fund financeFederal, State, Indian, and rural reclamation programs.Fifty percent of the money collected annually in any Stateor Tribe is reserved for that State or Tribe. As much as20% of the funds collected annually can be allocated tothe Department of Agriculture's Rural Abandoned MineProgram (RAMP). Ten percent, to a maximum of $10million, can be reserved for the Small Operators' As-sistance Program (SOAP). The remaining amount, atleast 20%, may be expended in any State or Tribe at thediscretion of the Secretary of Interior to meet the pur-poses of SMCRA. Money from the fund is available onlywhen appropriated by Congress and is generally con-sidered inadequate to solve all AML problems.

The 1977 act established priorities for expending fundscollected under the act. The first priority was to protectthe public health, safety, general welfare, and propertyfrom extreme danger of adverse effects of coal miningpractices. The second priority was to protect the publichealth, safety, and general welfare from the adverse effectsof coal mining practices. The fund was then to be used torestore land, water, and the environment previouslydegraded by mining practices, provide for research anddemonstration projects relating to reclamation and waterquality control, to protect, repair, replace, construct, orenhance public facilities, and to develop publicly ownedland adversely affected by coal mining practices.

States or Tribes identify projects according to the pri-orities and request annual grants to fund administration ofthe program and the projects, i.e., those listed in table 6(44). To be eligible for grants from the fund, the Statesmust have an approved regulatory program for active coalmining and an approved AML program. The requirementfor an approved regulatory program does not apply to theIndian tribes. Fire control projects in nonprogram States,those that do not have active mining operations, but dohave eligible abandoned minesites, are reclaimed withfunds from the Secretary's discretionary fund. RAMPfunds are used to reclaim rural sites by the department ofAgriculture's Soil Conservation Service (46).

Based on published data (8-9, 27, 40) and from privatecommunications on past fire projects, estimates were madeon the cost and effectiveness of AML fire control. Thecost of controlling abandoned mine fires has been highlyvariable. It depends upon the depth, location, and extentof the fire, on the extinguishment method, and the avail-ability of equipment and materials. Between 1950 and1988, the cost of fire control projects ranged from $3,000to $11.6 million.

The estimates of effectiveness are even less exact thanthe cost data. They were based on published evaluationsor inferred from repeated projects at a single site over anextended time frame. Because of the way in which thedata were collected and evaluated, the estimates in tables8 and 9 are considered only order of magnitude indicators.

Table S.-Mine fire control costs and effectiveness by rank

Number Average EstimatedRank of fires cost, $K effective-

ness, %1

Anthracite . . . . . . 22 2,350 40Bituminous ..... 102 111 60

East ......... 62 132 NAWest ........ 40 53 NA

Lignite ......... 13 4 NASubbituminous ... 46 18 44NA Not available.

Table g.-Mine fire control costs and effectiveness by method

Number of Average EstimatedMethod projects cost, $K effective-

ness, %

Excavation ......... 57 685 70Flushing ........... 16 533 27Seal .............. 86 28 42Trench ............ 4 119 <10

For the evaluations, projects listed in tables 2 and 3 atthe same location were considered related to a single fire.Projects listed separately for one location were consideredseparate projects; projects listed as phases were consideredone project. For example, there are six fire control proj-ects for Centralia, PA over a period of 20 years. This wascounted as one fire and six projects. The "Area D" fire,actually an outcrop fire affecting 20 to 25 coalbeds nearGlen Canyon City, UT was controlled in six phases be-tween 1969 and 1975. This was counted as one fire andone fire control project. Since complete information wasnot available on all projects, the number of projects givenin tables 8 and 9 does not agree with the number of proj-ects listed in other tables.

The effectiveness of past fire control efforts is a veryrough estimate. It was a common practice to consider afire controlled or extinguished at the end of a project.Generally, there was neither time, money, or personnelavailable to check on actual effectiveness. For this report,a fire control project was considered ineffective if the sameproject name and location reappeared after a few years inthe project lists. In this type of analysis, the degree ofuncertainty is relatively high. The tables are intended onlyto indicate costs and the relative probability of effectivecontrol.

In comparing fire control efforts versus the rank of thecoal, fires in anthracite are the most expensive to controland have the lowest probability of success (table 8). Bi-tuminous fires are two to three times more expensive tocontrol if they are in the Eastern United States than if inthe west. When comparing fire control methods, excava-tion and surface seals are most frequently used (fig. 17).Excavation is the most expensive, and surface seals are the

Flushing10%

Excavation

35%

• Trench2%

Seal53%

NUMBER OF PROJECTS

23

Trench1% Flushing

Seal5%

Excavation77%

TOTAL COSTSFigure 17.-Distribution of AML fire control projects by method.

least expensive control technique (table 9). Based on his-torical data, excavation has the highest probability ofsuccess. If because of location, cost, or other factors,excavation is not a control option, other currently availablefire control options have a lower probability of success.

There is actually no relationship between effective-ness and expense; over all fire control efforts, effective-ness could be considered a random parameter. In most

projects to control fires in abandoned mines, the extent ofthe fire was unknown. The selection of a method tocontrol the fire was often based on funds available, siteconstraints, and experience. There was often no flexibilityto alter the plan and no postproject evaluation. The con-ventional approach to fire control is not oriented towarda systematic evaluation of available options or their costversus probable effectiveness.

ABANDONED MINED LAND FIRE EXTINGUISHMENT-CONTROL

A fire requires three elements, fuel, oxygen, and energy.To extinguish a fire at least one of these elements must beremoved (fig. 18). Fuel is removed when it is consumedor when it is physically separated from the burning mass.Excavation and most barriers are fuel removal methods offire control.

Oxygen removal depends on either the introduction ofan inert atmosphere or on the isolation of the fire zonefrom sources of fresh air. The injection of inert gases isintended to suppress combustion by decreasing the avail-able oxygen supply. Flushed barriers are intended to iso-late the combustion area and also to interrupt the con-tinuity of the fuel supply. Surface sealing is the mostfrequently used oxygen exclusion method. It is based onthe premise that if atmospheric air can be excluded, thefire will eventually be extinguished.

Heat removal, the cooling of all fuel below the reigni-tion point, can be a method in itself or it can be used inconjunction with fuel removal or oxygen exclusion. With-out some form of heat removal, the chance of successfullyextinguishing an AML fire becomes smaller.

Heat removal can be accomplished by moving a heatabsorbing agent (usually an inert gas or water) through themine. It is more common, however, to allow heat to dissi-pate naturally while suppressing combustion. Convectionand conduction through the overburden account for mostof the heat loss in a mine fire. The normal lack of airmovement in a mine makes convection an extremely slowprocess. The time to cool a fire zone under 100 ft(3,000 cm) of cover by thermal conduction through theoverburden can be estimated by:

t = ~ /k = [3,000 cm]2/0.01 cm2/s = 9X 108s = 28.5years,

where t is the time, x is the overburden thickness, and k isthe thermal diffusivityof the overburden (sandstone in thisexample) (47).

To dissipate heat by natural convection would requirethat the combustion zone(s) be completely isolated whilea rapid influx of cooler air from the outcrop and theexhaustion of heated fumes through fractures in the over-burden removes stored heat. However, a typical fire

24

Figure lB.-Fire triangle and control methods: fuel removal,oxygen exclusion, and heat removal.

generates a higher pressure area that forces heated fumesdeeper into the mine while drawing cooler air from theoutcrop or from other areas of the mine. This type ofinternal convection cell may account for the discontinuouspropagation of mine fires. Even when the supply of oxy-gen is relatively low, continuous heating at low tempera-tures may drive off inherent water from the coal and allowfor increased adsorption of oxygen. Experimental data(14) has shown that accelerated oxidation begins at tem-peratures between 800 and 1200 C for anthracites, depend-ing to some extent upon the amount of prior oxidation.Accelerated reactions begin for bituminous coals at tem-peratures as low as 500 to 700 C. Convective heat transferin a mine occurs when heated vapors from the combustionzone migrate to other parts of the mine. If the coal hasbeen preheated or conditioned, spontaneous ignition mayoccur, even though these areas are remote from the orig-inal source of heat. In contrast, a flame spread mech-anism requires that coal adjacent to a burning area beheated by conduction and radiation to its ignitiontemperature of 4000 to 5000 C.Low temperature conditioning and spontaneous ignition

may be factors in the discontinuous propagation of minefires. In a hypothetical situation, a fire occurs near aportal or outcrop (fig. 19). The fire induces circulating aircurrents, which carry fumes and heat into the interior ofthe mine. The effective ambient temperature of a largearea of the mine is slowly increased. Drying and accel-erated oxidation continue to raise the temperature of thecoal. If in these localized regions, the heat is not readilydissipated and sufficient oxygen is present, spontaneousheating will eventually cause active combustion to occur.

After a period, several fires may occur in the interior ofthe mine without a continuous combustion pathway fromthe original firesite. Noncontinuous propagation of a minefire imposes two constraints on an extinguishment method.First, it must affect all the burning material and second,it must be effective until all the material within thecombustible zone is cooled below 1000 C. This applies toadjacent noncombustible rock that has been heated by thefire.Normal heat transfer to the overburden occurs by

conduction or radiation, which because of its low heatconductivity generally acts as an insulator. Roof coals andcarbonaceous shales even with carbonaceous contents aslow as 25% may exhibit spontaneous heating behavior.Heat conduction through the roof coals and shales mayserve as a pathway for the spread of the fire. Even if nocombustion occurs in the roof, heat transferred to theoverburden creates a very large heat reservoir. For ex-ample, the combustion of 1 st of medium volatile bitu-minous coal releases 30 million Btu. If this energy issimply adsorbed by the roof rock, it would raise the tem-perature of 75 st of rock, approximately 900 ft3,to 5000 C.Depending on the rock, the extent of combustion, and thelength of time the fire has been burning, the amount ofheat stored in the coal and adjacent strata can be morethan 1 billion Btu's. If all combustion ceases, it wouldtake 10 to 30 years for this amount of heat to dissipate byconduction through the overburden.To prevent reignition of the fire, all coal and heated

rock must be cooled below the reignition temperature.Even very small, isolated fires or areas where the coal isoxidizing at a high rate can serve as reignition points if thecontrol measure fails and oxygen becomes available. It isgenerally assumed that if the temperature is below 1000 C,the chance of reignition is small.In many cases, if extinguishment is technically im-

probable or economically impractical, controlling the fireis a desirable alternative. By limiting the propagation ofthe fire, by isolating it, or by slowing the combustion rate,safety considerations can be met until natural processesconsume all the fuel and/or dissipate the heat. Controlmethods require a commitment to maintain the appropri-ate conditions for an extended time and should include amethod to monitor the condition of the fire.

TEMPERATURE AND GAS MONITORING

Most fires in abandoned mines and wastebanks are notvisible on the surface. Occasionally, glowing coal can beseen in an abandoned portal, but this is the exception.Abnormal snow melt, changes in vegetation, and odors arethe usual indicators of a subsurface fire. Abnormal snowmelt (fig. 20), most pronounced in light to moderatesnowfalls, indicates areas where the heat transmission

Fresh air

Heat and fumes-.I

Subsidencearea

CombustionZone 1

25

Surface

LEGEND

~Air- ~ Fumes

DryingOxidationSpontaneous ignition

CombustionZone 2

Figure 19.-Dlscontlnuous fire propagation In abandoned mines.

Figure 20.-Abnormal snow melt Indicating subsurface heating.

26

through the underlying strata has raised the groundtemperature to above 0° C. The effect is usually transientand may be related to causes other than a ftre, such as thedistribution of heat absorbing minerals in a waste pile.Dead or dying vegetation (grass, shrubs, trees) may befound near some vents where the heat is sufficient toscorch roots (8). Conversely, some mosses seem to thrivein the high temperature area around vents. Odors areassociated with fumes from the fire that also may containhydrogen sulftde. Strong and persistent noxious odors, dueto the presence of coal distillates characteristic of burningcoal, are among the most serious AML fire problems, butare not a reliable indicator of a ftre's location.Smoke and vapor at vents, fractures, and sinkholes are

the usual indications of an AML fire. Smoky plumes con-tain particulates, indicative of relatively low temperatureand inefftcient burning (8). Steam condensate plumes in-dicate relatively high temperature, efficient fires, Theymay be visible only as heat refraction patterns or as vaporcondensing in cooler air. The emission of smoke and va-por is controlled by the location of vents and fracturezones formed by normal geologic processes. Therefore,smoke is not necessarily observed immediately above thecombustion zone in which it originates.Temperature and gases have been the conventional

indicators of unseen fires, Temperature measurementshave been used to delineate fires and to indicate long-termchanges in combustion activity. Temperature measure-ments alone are considered a poor indicator of fire loca-tion. Abandoned coal fires usually involve small volumesof smoldering coal; very hot spots are localized with arapid drop-off to cooler temperatures. These fires tend tobe discontinuous, and surrounding rock can provide addi-tional insulation. Borehole temperatures may be near nor-mal underground temperatures within several feet of acombustion zone. Under appropriate conditions, boreholetemperatures are a relatively quick and inexpensive meth-od to monitor changes in combustion activity.Subsurface temperatures are measured by thermo-

couples suspended in boreholes to the mine void. Tem-peratures measured by probes placed within the casing areaffected by the thermal conductivity of the surroundingrock and the casing. Updrafts and downdrafts also mayaffect the temperature measured in the casing as opposedto that in the mine void.

A 5-ft length of stainless steel sheath, type K (Chromel-Alumel) thermocouple temperature probe is recom-mended. This conftguration normally places plastic con-nectors and insulation above the casing bottom wheretemperatures above 100° C could melt the plastic. Aportable digital thermometer is used to read the downholetemperature. Permanent installation of the thermocoupleyields more accurate results, since lowering the thermo-couple for each reading is time-consuming, labor-intensive,and introduces sources of error. Generally, a thermo-couple measures the highest temperature within a volumeof 1 to 2 ft3immediately surrounding the thermocouple tip,i.e., a maximum radial distance of approximately 10 in. Tocompare temperatures over a period requires that eachmeasurement be made at the same point within the minevoid. Even careful insertion of the thermocouple is un-likely to meet this requirement. For example, a discrep-ancy of 1 in. in the placement of the thermocouple is a10% error in the radial distance. In a 50-ft borehole, thisrequires a placement accuracy of ±0.2%. Also, openinga borehole allows the mixing of subsurface and ambientair, depending on the relative pressures. This introductionof cooler or warmer air may distort the measured subsur-face temperature. If temperatures are high, these errorsmay be inconsequential. However, small changes to inferlong-term heating or cooling trends cannot be determinedunless the thermocouple installation is permanent.Even under stable conditions, subsurface temperatures

yield limited information about fire conditions. Boreholetemperatures were taken ata fire project in the Pittsburghcoalbed over a period of 2 years (48). Of 38 boreholes,three had temperatures greater than 100° C, 25 bore-holes had average temperatures between 30° and 100° C(table 10). Ten boreholes had average temperatures below30° C. The normal subsurface temperature in the minewould be expected to be in the 11° to 15° C range, andonly one borehole had a temperature in this range. Therelative standard deviation was between 1% and 29%; therelative variation in temperature for most of the holes wasless than 5% over the 2-year period. Based on tempera-tures alone, active combustion was occurring along theburied outcrop, and large areas of the mine were at higherthan normal temperatures. Heat capacity considerationsand gas composition data indicate that the active com-bustion zones extend to the interior of the mine.

27

At another mine fire project in the anthracite re- Long-term temperature monitoring can indicate heatinggion, borehole temperatures were taken over a period of or cooling trends. However, heating or cooling rates are3 months (24). Of 34 boreholes, two had temperatures very low and one annual temperature measurement mayover 100° C, 15 boreholes had average temperatures fall within the normal variation. For example, for abetween 30° and 100° C (table 11). Seventeen boreholes measured temperature of 30° C, the normal variation,had average temperatures below 30° C, but only one was assumed to be ± 5%, is IS C. Based on the data in ain the 11° to 15° C range. The relative variation in tem- Montana study (49), the average cooling-heating rate mayperature during the monitoring period was between be as low as 0.1° C per year (table 12). The rate of1% and 150%, with most boreholes in the 5% to 10% temperature change showed no correlation with the initialrange. Elevated temperatures were not well correlated temperature (fig. 21). Evaluations of control effective-with other combustion indicators. ness based on temperature monitoring should consider

that the magnitude of the temperature change should beTable 10.-Mean borehole temperatures, greater than the expected normal variation and should

Large mine fire show consistent trends over a minimum of 5 to 10 years(fig. 22).

Borehole n T, ·C Standard deviationnumber :!: ·C % Table 11.-Mean borehole temperatures,1 91 23 1 4.29 Carbondale mine fire2 96 36 1.86 5.116 93 47 4.21 8.99 Borehole n T, ·C Standard deviation7 98 32 4.94 15.39 number :!: ·C %8 98 59 4.25 7.23 1 11 73 2.4510 98 26 .59 2.26

3.372 6 38 1.50 3.96

11 98 44. 1.1 2.50 3 11 157 66.29 42.1112 63 18 .87 4.87 4 11 34 1.82 5.4113 67 54 3.78 7.00 5 11 25 1.47 5.8315 66 73 1.01 1.38 6 11 121 12.88 10.6116 64 62 1.6 2.58 7 10 40 6.47 16.2120 67 50 14.57 28.97 8 10 51 .99 1.9222 66 31 .49 1.58 9 8 69 3.10 4.4823 65 28 4.08 14.68 10 11 33 2.00 6.1528 20 30 .49 1.61 11 9 60 1.66 2.7529 20 21 .4 1.89 12 10 16 1.50 9.49

30 19 27 .44 1.62 13 11 18 1.43 7.9331 20 36 .34 .93 14 8 85 1.61 1.90

32 19 47 .04 .08 15 10 60 8.64 14.35

33 19 47 1.76 3.74 16 11 19 1.61 8.30

34 18 119 5.04 4.24 17 11 17 1.90 10.93

35 18 54 .89 1.64 18 10 18 6.49 36.18

36 20 37 .34 .93 19 9 19 1.38 7.12

42 19 60 .96 1.6120 8 16 .97 6.0521 8 14 1.66 11.70

43 15 47 .42 .89 22 8 32 47.45 150.3544 18 107 4.94 4.62 23 11 31 1.42 4.6445 17 353 19.18 5.44 24 10 22 1.34 6.0346 19 38 .36 .95 25 11 19 1.84 9.4847 19 29 .23 .80 26 11 23 2.46 10.7448 19 15 .7 4.62 27 11 17 1.269 7.3649 16 19 3.65 19.72 28 7 18 1.49 8.1650 19 38 .4 1.06 29 11 16 .82 5.2854 7 20 .91 4.45 30 11 16 .99 6.0855 7 35 .12 .34 31 11 21 1.60 7.57

56 7 34 .76 2.23 32 7 37 2.20 5.98

57 7 49 .43 .87 33 8 55 .80 1.44

58 6 34 .05 .15 34 8 47 .97 2.08

59 8 42 3.24 7.69

28

Table 12.-Varlatlon In cooling rate at western mine fires·

Borehole number Temperature Time, Cooling rate Borehole number Temperature Time, Cooling ratedecrease, ·C months ·C/year decrease, ·C months ·C/year

Frank Mine Fire: Muster Creek A-Continued1 ............ 2 67 0.31 7 ............ 6 33 2.183 ............ 23 67 4.06 10 ........... 8 33 2.974 ............ 9 67 1.60 13 ........... 1 33 .535 ............ 17 67 3.00 15 ........... 9 33 3.096 ............ 12 67 2.19 16 ........... 7 33 2.6311 ........... 9 67 1.65 17 ........... 1 33 .4412 ........... 17 67 3.12 18 ........... 2 33 .5513 ........... 7 67 1.33 19 ........... 2 33 .5515 ........... 24 67 4.34 22 ........... 10 33 3.4716 ........ ; .. 31 67 5.53 23 ........... 9 33 3.3518 ........... 10 32 3.73 25 ........... 9 33 3.1719 ........... 10 32 3.67 26 ........... 20 33 7.1120 ........... 10 32 3.69 27 ............. 6 33 2.0221 ........... 8 32 2.88 Muster Creek B:22 ........... 10 32 3.88 2 ............ 1 33 .3423 ........... 7 32 2.79 3 ............ 4 33 1.2924 ............ 8 32 3.08 4 ............ 9 33 3.1525 ........... 9 32 3.33 5 ............ 6 33 2.06

Shadwell Creek A: 6 ............ 11 33 4.082 ............ 14 33 5.19 7 ............ 11 33 3.824 ............ 9 34 3.29 10 ........... 18 33 6.715 ............ 5 22 2.76 13 ........... 1 33 .446 ............ 4 34 1.45 15 ........... 10 33 3.809 ............ 32 34 11.39 16 ........... 9 33 3.2710 ........... 5 34 1.90 17 ........... 2 33 .6511 ........... 40 34 13.94 18 ........... 2 33 .6512 ........... 16 34 5.71 19 ........... 2 30 .6213 ........... 9 34 3.02 22 ........... 24 30 9.4414 ........... 10 34 3.39 23 ........... 14 30 5.7615 ........... 20 4 59.17 24 ........... 35 30 14.1116 ........... 9 34 3.31 25 ........... 14 30 5.5817 ........... 8 34 2.98 26 ........... 22 30 8.6218 ........... 9 34 3.06 27 ........... 7 30 2.6919 ........... 10 34 3.65 Muster Creek C:

Shadwell Creek C: 2 ............ 2 33 .678 ............ 33 34 11.65 3 ............ 6 33 2.009 ............. 26 34 9.12 4 ............ 11 33 4.0410 ........... 7 34 2.39 5 ............ 9 33 3.1511 ... , ....... 34 34 11.90 6 ............ 23 33 8.4812 ........... 15 34 5.12 7 ............ 12 33 4.2813 ........... 10 34 3.57 10 ........... 18 33 6.6314 ........... 10 34 3.49 13 ........... 8 33 2.8915 ........... 33 34 11.53 15 ........... 16 33 5.6816 ........... 18 34 6.37 16 ........... 15 33 5.6217 ........... 14 34 4.80 17 ........... 6 33 2.2818 ........... 5 34 1.61 18 ........... 2 33 .6919 ........... 6 34 2.06 19 ........... 1 33 .3626 ........... 12 34 4.29 20 ........... 7 33 2.57

Muster Creek A: 22 ........... 39 30 15.492 ............ 1 33 .24 23 ........... 20 30 8.093 ............ 2 33 .85 24 ........... 26 30 10.244 ............ 6 33 2.24 25 ........... 13 30 5.005 ...........• 26 33 .89 26 ........... 20 30 8.006 ............ 4 33 1.56 27 ........... 5 30 1.93

lOata from Hanson (49).

29

20

18 KEY•F"RAflK MINE

16+

SHADWELL CREEK A0::: '*'<, 14 +:8:: SHADWELL CREEK C

U XMUSTER CREEK A

12 '*' '*' :8::* + MUSTER CREEK B

L&.l ••••• •••r- 10 MUSTER CREEK C« Z*~ .•••.:8::G 8 ••••• •••z >.&.....J

*0 6 + z ••••••0 zu + *..•..

*'4 t )Z.~ ~~• •l+~ z

••• • :.:2 ~ • ~><.&. * z .A

• + ~• ~ 4x010 20 30 40 50 60

INITIAL TEMPERATURE, DC

Figure 21.-Rate of subsurface cooling versus initial temperature. Based on data in reference 49.

110

90

KEY

~ 1950's

f22211987

-~

- I><

I>< I><

G1 G2 T11 T12 T2f T23 D1 D2 31 32 33 V1 V2 R2

BOREHOLE

100

80U0 70

W 60a::::=JI--« 500:::Wn,

40~Wf-

30

20

10

Q

Figure 22.-Change in subsurface combustion zone temperature over 30-year period. Based ondata in reference 49.

30

Gas composition as a combustion indicator has beenused occasionally to monitor abandoned mine fires, Gassamples must be obtained from the mine void or from anarea near the combustion horizon. The composition ofsamples taken from near the top of a borehole is depend-ent on the pressure differential between the atmosphereand the mine, the temperature gradient, and diffusion.Under any sampling conditions, subsurface fire indicatorsbased on the gaseous products of combustion, carbon diox-ide, carbon monoxide, and a decrease in oxygen are notsensitive or accurate. Ratios such as the Jones-Trickettratio", the airfree CO concentration and the ratio of CO toCOz8 (51), and the COjdOz (52) index? (53) are frequentlyused in active mines, particularly to monitor gob areas.Combustion product ratios often yield ambiguous resultswhen applied to abandoned mine fires. The dilutionfactor, gases from noncombustion processes, the ventila-tionpathway, high humidity, the adsorptive surface ofcoked coal and the accumulation of combustion productsover extended periods may be factors in the low accuracyof these indicators when used at abandoned mines.

In field and laboratory studies (17, 27, 48), variationsin the concentrations of carbon dioxide, carbon monoxide,and oxygenwere not consistent indicators of elevated tem-perature. It was found that the average carbon dioxideconcentration varied inversely with the oxygen concentra-tion irrespective of temperature (fig, 23). In controlledexperiments, it was found that carbon monoxide is notproduced from coal below a temperature of 1200 C(fig, 24). In the field study, only 9 of 38 boreholes hadaverage carbon monoxide concentrations greater than O.The absence of carbon monoxide can indicate that the coalis cold and no carbon monoxide is produced or it maymean that complete combustion is occurring in an oxygenrich environment and no carbon monoxide is produced.Of the boreholes in which carbon monoxide was detected,more than half had oxygen concentrations below 8%, sup-porting the theory that carbon monoxide is produced bycombustion reactions in a low oxygen environment. Theratio of carbon dioxide to carbon monoxide is variable,but asymptotically approaches a limiting value of three(fig. 25). Gas composition and the JTR can indicatewhether combustion is occurring in a relatively oxygen-richenvironment. Over an extended period, an oxygen concen-tration that remains constant or increases indicates asource of fresh air.

7Jones-Trickett Ratio (JTR) = ([COil + O.75[CO] - O.25[HiI)/(O.286[Nii - [Oil).

8(CO)Airfree= ([CO]/(lO0-4.76 [Oil))*100.90raham index = CO/dOz = [CO]/(.268[Nil - [Oil)

where [CO] = concentration of CO in ppmdOz = oxygen deficiency in pet.

The static differential pressure (the difference betweenthe ambient surface barometric pressure and the measuredpressure in the mine) can be determined. A differentialpressure of zero indicates that the mine is at the samepressure as the atmosphere. The differential pressure inthe mine may be influenced by changes in barometricpressure. However, if differences are due only to the rateat which the mine breathes, the differential pressureshould be uniform throughout the mine. Variations indifferential pressure indicate that the mine fire is caus-ing the formation of a convection cell within the mine.Elevated pressures may also be due to mass addition ofcombustion products to the airstream and to thermalexpansion of the hot gases. Areas that exhibit negativedifferential pressures may indicate areas in which air isbeing drawn into the mine. The static differential pressureis highly variable (relative standard deviations greater than100%), and has not been correlated to changes in otherfire related variables.

Temperature and product of combustion gas concentra-tion measurements are relatively easy to obtain. However,they are best suited to long-term monitoring of fire status.They are not effective indicators of a fire's areal extent orof short -term changes in combustion intensity.

CONVENTIONAL METHODS

Conventional methods of controlling andj or extinguish-ing AML fires include excavation i.e., dig out and quench,daylighting, and excavated barriers; flushing, either pneu-matic or hydraulic, and surface seals. Excavation and sur-face seals have been used most frequently (fig. 26). Thesecomprise a limited arsenal of dealing with the AML fireproblem. None of the methods are routinely successful.Most of the methods involve some degree of hazard, havevarying costs, and may disrupt more surface area than thefire threatens.

ExcavationExcavation (loading out, daylighting, dig and quench,

stripping) is a fuel removal method that is the mostsuccessful of the AML fire control techniques (54). Itinvolves physically removing the burning material and cool-ing it to extinguish the fire. The hot material is cooledeither by spraying it with water or by spreading it out onthe ground and allowing it to cool in air. Water, becauseof its relatively low cost and its general availability, ispreferred as the heat removal medium. It is also used toprotect equipment from high temperatures and to suppressdust. However, in some areas, particularly in the WesternStates, sufficient water is not available, and the hotmaterial is cooled by contact with the cooler air.

31

4-.-----------------------------

'*o-+---~.-_,__-_r-__=.,__---I~.f--____,--re~L.,--,__-.,__-_r-___1o 40 80 120 160 200 240

TEMPERATURE, OC

3

~~ 2oU

KEY

• Bailey refuse

+ Albright refuse

* Pittsburgh coal

>< F seam coal

D Zone C shale

*x

x•-

xX

0

* *X0

0 ••+

• !!!!!!+

Figure 24.-earbon monoxide concentration versus temperature.

32

4500

400 D

350 0

0 300I-«c:::0 250o0 200I- D

N

o 150u

100><

50

aXx

0

KEY

X Bituminous

o Anthracite

Do

~[D-X X A

100TEMPERA TURE,

Figure 25.-Ratio of CO2 to CO versus temperature.

100 -.--------------------------------,

90 KEY

~ Number of projects80

IZXI Average cost, $10K

rz:.l % successful70

50

50

40

.10

20

10

SEAL EXCAVATION FLUSHING

METHODTRENCH

Figure 26.-Cost and effectiveness of AML fire control projects by method.

300

The cooled material is disposed of as landfill; layeringwith incombustible material, such as clay, reduces theprobability of reignition. If another disposal site is notavailable, the cooled material is used to backfill theexcavated fire zone. .

To inhibit the propagation of the fire to the interiorof the mine (fig. 27), excavation should begin at the inbyside of the fire and proceed toward the outcrop (fig. 28).The excavation of a waste bank fire involves the problemof moving personnel and equipment on possibly unstableslopes. Also, the increased supply of oxygen may prop-agate the fire to the interior of the bank or to contiguous,buried outcrops. In waste bank fires, as in mine fires,excavation should proceed from the interior limit of thefire toward the surface if possible (fig. 29).

Excavation is inherently dangerous. It requires handlinghot material that emits toxic fumes. Moving combustiblematerial or even a shift in the wind can produce a locallyhigh concentration of carbon monoxide. Hot coal exposed

Surface

Fire zone

Surface

Planned limitr of excavation

Mine

Fire zone

Surface

Fire zone

Figure 27.-Schematic of excavation of fire zone allowingpropagation of fire into interior of mine. Top, Preexcavation;middle, first cut; bottom, final excavation.

33

to air can very quickly turn a large mass of coal into araging fire (fig. 30). Hot fines and dust, when disturbed,can generate an explosive cloud. In some cases, puttingwater in the wrong place in a fire zone can produce ex-plosive water-gas reactions. Although there have beenfew reports of injuries related to excavating AML fires,

A

I Planned limit",.. of excavationII

Mine

c~Oft excavation

Fire zone I~I--------~. Mine

Mine

Figure 28.-Schematic of excavation to mine fire to inhibitpropagation. A, Preexcavation;B, first cut; C, isolated fire zone;0, excavation completed and backfilled.

34

such projects are usually considered to involve hazardousworking conditions (fig. 31).

If properly applied, excavation is the surest method ofextinguishing AML fires. The recurrence of excavatedfires is usually due to one of two causes: the failure tocompletely excavate the fire or the failure to lower thetemperature of all the excavated material beyond the reig-nition point. In the first case, the extent of excavation isusually based on surface expression of the fire (vents andfractures) or on subsurface temperatures with a periphalmargin added for safety. The excavation may be insuffi-cient if either the fire exists beyond the original boundariesor if it propagates during excavation to previously coolzones. Since fires can be discontinuous, it is possible forthe excavator to be unaware that fire zones exist beyondthe excavated area. Such fires become apparent on thesmface usually within 2 to 10 years after completion of theoriginal project. Excavation also fails when incompletelyquenched material is used as backfill in the original exca-vation. This material can continue to smolder until suit-able conditions allow the fire to propagate.

The published cost of excavation for completed projectshas ranged from $5,000 to over $11million. Generally, thearea of the excavation, the depth of overburden, theavailability of equipment, the proximity to a disposal site,

Buried outcrop

-Buried outcropFire zone

Figure 29.-Schematic of propagation of waste bank fireduring excavation. Top, Burning waste bank adjacent to buriedoutcrop; bottom, fire zone propagated to burled outcrop duringexcavation.

Figure 30.-1ncreased fire Intensity during excavation (photo - courtesy of WVDNR).

35

Figure 31.-Hazard of AML fire excavation.

the access to large volumes of water, and the presenceof surface improvements will determine the cost of anexcavation project. These same factors are considered indetermining the suitability of a project for excavation.Particularly, the depth of overburden, the volume of exca-vated material, the lack of a disposal site, or the cost oftransporting hot material to an area where it can cool, andthe presence of utilities, homes, or roads may make exca-vation inappropriate as a fire extinguishment technique.

Inundation-Flooding

Inundation methods involve the underground use ofwater to lower the temperature of the burning material(heat removal). Covering the burning material with wateralso stops the combustion reaction by oxygen exclusion.To raise the water level, dams are constructed under-ground. The water level must cover not only the burningcoal, but also must reach the overlying heated rock. Thismethod is limited to use on fires that are small, are ator near the water table and have been burning for arelatively short time to minimize the amount of storedheat. It is expensive and often dangerous to constructdams in inaccessible areas or where strata are notcompetent to withstand the additional hydrostatic head.Confining water underground may be impossible near anoutcrop or where fractures extend below the coalbed. In

flooding, the consequences of the catastrophic failure ofremote dams or other confining elements must be con-sidered. Generally, flooding is not suitable for controllingunderground AML fires. For surface fires, the construc-tion of an impoundment more than 1 or 2 ft high is gener-ally not practical. The use of flooding presupposes thatsufficient water is available. It also assumes that waterlost by leakage or evaporation will be replaced until all theburning material has cooled below the reignition point.Water that has been in contact with coal or coal waste foran extended period may be acidic and will have to bedisposed of in accordance with regulations for acid minedrainage.

Another inundation method provides for the continuousflow of water through the hot material. This can be ac-complished by continuous .pumping or by gravity flow froma surface impoundment (fig. 32). The volume of waterrequired, the cost of high capacity pumps and time con-siderations have limited the utility of continuous pumping.The gravity flow method has apparently been tried onwaste banks by excavating a reservoir on the top of thebank and allowing water to flow naturally downward (55).The constraint on either of these methods is that the wateris not uniformly distributed. It may flow through channelsand may bypass the fire zone. If the water fails to cool asmall amount of material, the probability of reignition ishigh.

36

Area of elevated temperature

.......,-'r--- Hot spots

~.l.----Borehole

Typicalinfiltration

trench

Minedcoal

seam H=5' to 10'

W=3 x H

Figure 32.-1nundation by water infiltration from surface im-poundment (after (48».

Flushed Barriers

Flushing is designed to fill the voids in an undergroundfire zone with fine, noncombustible solids (fig, 33). Thenoncombustible material is intended to cover the burningmaterial and fill the interstices in adjacent rock, limitingthe amount of oxygen in the system and absorbing heat(fig. 34). The high percentage of incombustible material,if properly emplaced, is expected to form a barrier tofurther propagation of the fire, Flushing can be effectivewhere deposition of the noncombustible material can becontrolled, where the voids have a relatively simple geom-etry, and where the injected material will remain in place.

Sand, silt, red dog, crushed limestone, and fly ash arethe most commonly injected materials. Air or water areusually used to carry the material through a borehole intothe mine. With pneumatic (air) flushing, the noncombusti-ble material is deposited at the bottom of the borehole.It forms a conical pile that theoretically reaches to theroof of the mine. Material is pumped into the hole torejection, when it is assumed that the void is filled.Pneumatic injection has two constraints: the material doesnot penetrate fractured strata and the material tends toslump, which reduces the contact with overlying strata andallows air to flow near the roof.

- - -- ~ -.-.=...-=.',<--=-",:

Figure 33.-Noncombustible barrier injected from surface(after (48».

Duringinjection

Pneumatic flushing

Roof coa,

Afterinjection

Figure 34.-Flushed boreholes during and after inJection.

In hydraulic flushing, water is used to produce a slurryof the incombustible material. When the material is em-placed in the mine, the solids settle as the water drainsdown dip. This method is believed to carry materialfurther than dry flushing and to have some penetrationthrough porous material. To create an effective seal,injection boreholes are located on 10-ft centers (7). Threelines of boreholes, 10 ft apart, are generally used aroundthe perimeter of the fire area (fig, 35). The amount ofdrilling, availability of flushing material, transportationcosts, and the availability of water affect the cost offlushing projects. An average flushing project can con-sume 10 to 40 st of material per hole. Water consumptionmay range from 10,000 to 40,000 gall d.

Grout slurries have been pumped underground to formfire control barriers. Cement in the grout slurry solidifiesto form a competent seal, which also add support to col-lapsed strata. The addition of foaming agents and incom-bustible materials, like sand or soil, has been used to

Fire zone

••• ••••• Injection boreholes:::von 10 ft centers

• •••••••• •••••• ••

Figure 35.-Plan view of injection boreholes for hydraulicflushed barrier.

produce a lower density foam grout that hardens to a cel-lular concrete (fig. 36). By mixing various proportions offoam and port-land cement, grouts with dry densities be-tween 85 and 120 Ib/ft3 and compressive strengths between500 and 4,000 psi can be produced. The grout encapsu-lates the burning coal and limits combustion by limitingexposure to oxygen. The foamed grout has a low thermalconductivity, between 3 and 7 Btu. hr. ft-2• 0p-l. in" (80to 228 cal- min- m-2• 0C-1• ern"), and acts as an insulator,retaining heat within the coal. To remove heat, a thermalaggregate, small metal particles, can be added to the grout(56). To be effective, the remotely emplaced grout sealmust be complete, encapsulating all burning material andisolating it from other combustible materials, and thegrout barrier must be stable for extended periods while thematerial cools.

Excavated Barriers

An excavated barrier is intended to limit the spread ofa subsurface fire by removing the combustible materialaround the fire zone. When backfilled with noncombusti-ble material, the fire barrier is a dam between the fireand the contiguous coal. The barrier breaks the continuityof the coal and carbonaceous shales and must be wideenough to prevent heat transfer from the fire side to thecold side.

A trench barrier is constructed by excavating an opentrench between the fire and the threatened area and thenbackfilling it with incombustible material. Horizontally,the trench extends from outcrop to outcrop or to the waterlevel to isolate the combustion zone from the unburnedcoal. A trench extends vertically from the surface tothe bottom of the coalbed (fig. 37). A minimum width of12 ft at the bottom of the trench has been considered

37

Figure 36.-lnjection of foaming grout or cellular concrete toencapsulate burning coal.

outcrop

Limit offire zone area

Top of trench barrier

Bottom of trench barrier

Figure 37.-Plan view of trench barrier for AML fire control.

essential to prevent the spread of fire across the trench.The walls of the trench must be sloped at least 15°(fig. 38) to prevent their collapse during excavation (7). Atrench that is 12 ft wide at the bottom and 10 ft deep mustbe at least 16 ft wide at the top. If the trench is 100 ftdeep, it must be at least 66 ft wide at the top. Althougha trench may be benched to minimize the required width,the extent of surface disruption is a constraint on the useof a trench barrier as a fire control method. The limit ofthe fire area must be known prior to excavation of the

38

trench barrier and a margin of safety provided. Usually inplanning the excavation, the tendency is to restrict the sizeof the trench as much as possible to control costs. How-ever, if the trench is too close to the fire zone, the dis-ruption of the overburden can supply sufficient oxygen toaccelerate the fire, which can rapidly move beyond thepartially excavated trench. The trench must be backfilledas quickly as possible to prevent its collapse and to de-crease the amount of fresh air reaching the fire. Foreconomy, the excavated material is often used as backfill;noncarbonaceous material being separated from the coaland carbonaceous shale. The material is selectivelypushed back into the trench so that all noncarbonaceousmaterial is at the bottom of the trench and so that thecarbonaceous material is isolated and buried. Eventualsettlement in the fill material must be controlled.

Plug barriers and tunnel barriers (fig, 39) are variationson the excavated barrier. The plug barrier was usedwhere a complete outcrop to outcrop barrier could not beused because of excessive overburden. It begins at theoutcrop and terminates under more than 60 ft of cover, inflooded workings or in solid coal. The 60 ft of covercriterion was based on the belief that a fire would notpropagate into unmined coal under more than 60 ft ofcover if the surface is sealed (36). There is no factualbasis for the 60 ft rule and it was applied to any 60 ft ofcover. Depending on the overlying strata, it is possible fora fracture zone to extend more than 60 ft vertically. If thecoal is extensively mined, it is also possible for the fire toproduce a convection cell bringing in air from other partsof the mine. Heat transfer considerations and the failureto isolate the fire zone decrease the probability that plugbarriers are effective fire control methods.

A tunnel barrier is an underground tunnel backfilledwith noncombustible material. It is proposed as a meansof extending a trench barrier when the overburden depthis too great to excavate from the surface. Finely dividedmaterial is injected through boreholes to fill the tunnelvoid. In addition to the technical problems of excavatinga tunnel in an abandoned mine, unstable roof and toxicfumes make this inherently hazardous. Attempting to per-form such an operation in a safe manner escalates thecosts even further. No reference to the actual use of atunnel barrier to control an AML fire has been found.

Surface Seals

Surface sealing is a relatively inexpensive method ofcontrolling abandoned mine fires. It is intended to inhibitventilation of the fire zone (fig. 40). The exclusion of airand the accumulation of combustion products suppressesthe rate of fire propagation. If the seal can be maintained

Surface

Minimum surface width = 12+2(d x tan 15°)

Figure 38.-Schematic of trench barrier for AML fire control.

Top ofbarrier

Bottom ofbarrier

Fire

Coafbed

Flushing boreholes

Figure 39.-Plan view of plug and tunnel barriers for AML firecontrol.

while all the stored heat dissipates, the fire may eventuallybe extinguished. During this period, the seal must bemaintained. Surface seals frequently fail between 1 and 3years after construction. Failure may be related to set-tling' shrinkage, slope failure, drying, or increased fireactivity. Because surface seals can prevent the egress ofhot combustion products, the changed distribution of heatunderground can cause propagation of the fire to otherareas of the mine.

In eastern fires, a surface seal is generally constructedby pulverizing the surface to a depth of 4 to 8 ft (7). Thesurface seal begins at or near the outcrop (fig. 41),depending on whether the outcrop has been exposed,whether there are drift openings, and whether the area hasbeen stripped (fig. 42). The seal must extend beyond allsurface indications of the fire. All vegetation, includingbrush and trees should be removed from the area of theseal. The seal is constructed by plo-wingthe surface alongthe contour and then twice crossplowing at 45° angles tothe original contours. Drainage traps are constructed to

39

Before Sealing

Abandonedmine opening~-- -Airflow Airflow

After Sealing

Plowed surface seal

Fire area

Figure 40.-Surface seal to control AML fires in Eastern UnitedStates (after (7».

Figure 41.-Extent of surface seal, depending on condition of mine. A, Outcrop not exposed; B, drift open-ing into outcrop; C, outcrop contour strip mined.

40

A Open air passageways _

B

Sloped fire area surface)

I

c

/Open passageways

/

Figure 42.-Surface seal used in western AML fires. A, Normalsurface; B, borrowed and plowed surface seal; C, cut and coverseal.

allow water to flow from the coal mine while preventingair from entering it. Drains and ditches must be installedto prevent erosion from destroying the seal. The area isreseeded to minimize erosion and control runoff of surfacewater through the sealed area. Provision must be madefor regular inspection of the seal and for monitoring thecondition of the subsurface fire zones.In the Western United States, 85% of fire abatement

projects are surface seals. This is due to the relativelylow cost, the topography of the area, and the lack of waterneeded to implement other methods (9). Most of the firesin the western coalfields occur within either the MissouriRiver or the Colorado River drainage systems. Thesurface-sealing method as used in these regions differsslightly.In the Missouri River drainage region, the fire sites are

generally located on accessible, flat to rolling terrain with

a maximum slope of 25°. Generally, the surface strataover the fire area contains 2 to 10 ft of relatively thin,friable sandstones, shales, siltstones, and mudstones thatcan be easily worked by bulldozer. If necessary, the cover-ing materials are excavated from borrow pits, then trans-ported to the fire area and deposited in 6- to 8-in layers.With in situ surface sealing, the area's residual mantleand/or underlying strata are the source of the incombusti-ble material. With both sources of covering materials, thearea is sloped with a bulldozer to fill in large cracks andto grade uneven terrain. The bulldozer is used to coverthe area with incombustible materials and plow the af-fected area. The plowing, performed on the contour tominimize erosion, emplaces a competent 3- to 4-1/2-ft-thick seal. In the cut-and-cover technique, a bulldozerstarts at the lower perimeter of the fire area, and makesa cut along the contour. In succeeding upslope cuts, thematerial is cast onto the previous downslope cut, pro-ducing a seal between 4 and 8 ft thick. A diversion ditcharound the seal area controls erosion due to surfacerunoff.In the Colorado River basin, slopes are steep and sur-

face strata are massive sandstones, shales, siltstones, andmudstones. In addition to bulldozer cut-and-cover meth-ods, blasting was often used to produce material suitablefor sealing the fire area. It's use is now limited because ofcosts associated with bonding and insurance. Surface seal-ing under these conditions is more labor intensive, moretime consuming, and more costly. However, it is still themost practical and cost-effective method of controllingfires in this area.Surface sealing suppresses surface evidence of a fire.

If the seal is maintained for a sufficient length of time(10 to 20 years), the fire may be extinguished. A breachin the seal before the coal and associated rock has cooledbelow 100° C will probably reactivate the fire. Surfaceseals tend to contain the hot combustion products withinthe mine. Under appropriate conditions the hot productsmay spread the fire to other portions of the mine. How-ever, in sparsely populated areas where there are no sur-face structures threatened by the fire, surface sealsadequately control subsidence, inhibit unsightly venting,and limit the emission of noxious fumes. If extinguishmentis not critical, surface seals, with regular and periodicmaintenance, can provide an adequate control mechanism.

RECENT DEVELOPMENTSThe focus of the recent research on AML fires has

been to improve currently available techniques and todevelop remotely emplaced techniques that minimize

surface disruption, to improve cost effectiveness, and todependably protect people and property near AML fires.

The SMCRA authorized the use of the AML fund forresearch and development (priority four) (45). A researchprogram, established by OSMRE in 1985,was transferredto the Bureau in 1987. Of the 54 research projects funded,8 have dealt with various aspects of the AML fire problem(table 13). Funding for fire projects totaled $1.3 million;the average cost per project was $167 thousand. Prior tothe establishment of the AML research fund, studies onnovel or more effective, more efficient methods of control-ling AML fires were conducted primarily by the Bureau.The results of many of these projects, both under AMLand previous funding, are discussed below. Current orcontinuing work on topics such as diagnostics, extinguish-ment, and control-containment methods are discussed inthe "Research Areas" section.

MINE FIRE DIAGNOSTICS

One factor in the failure of some abandoned mine mecontrol projects was that the extent of the fire was un-known when the project was planned. A corollary factorwas that there was no adequate provision for monitoringthe propagation of the fire during and/or after the firecontrol project. To locate a remote fire zone, it is neces-sary that: (1) the fire have a measurable characteristic, (2)the characteristic be detectable through appropriate sam-pling methods, and (3) the data be interpreted correctly.

Temperature Monitoring

Borehole temperature monitoring provides point sourcedata that are of very limited utility. The area for whichthermocouple data is accurate is approximately 12 ft2around the borehole. To use subsurface temperatures asan indicator of subsurface combustion zones would requirethe use of boreholes on 10-ft centers. Even with this con-straint, elevated borehole temperatures may be related tothe movement of hot combustion products, rather than to

41

the proximity to burning material. Elevated temperaturesindicate that combustion is occurring, but give limitedinformation about the areal extent of the combustion zone.

Aerial Infrared

Attempts to use aerial infrared to locate subsurfacefires have not been particularly successful. The primaryconstraint on this method is that it measures temperatureswithin a few inches of the ground surface. This detectssurface fracture zones and vents, but cannot detect deeperheated areas. Triangulation methods to extrapolate sur-face temperature data to depth require that heated com-bustion products move along straight line paths from thesource to the surface, a condition that is not usually met.Also, heat absorbing features on the surface, like bodies ofwater, tend to appear on infrared as localized hot spots.In general, infrared methods have not been shown to lo-cate subsurface combustion.

Soil Hydrocarbons

Soil sampling for thermally produced hydrocarbongases, high-molecular weight alkanes, or the lower molecu-lar weight aromatics (benzene and toluene) has been pro-posed for locating subsurface fires. The simpler, lessexpensive sampling procedure would reduce the cost.However, the analytical equipment is the more complexgas chromatograph-mass spectrometer (GC-MS). Sam-pling at the surface assumes that the gases sampled havemoved by a relatively linear path from the source. Sincethe compounds of interest are gases, their migrationthrough a nonhomogeneous medium may be influenced bychanges in pressure, the location of natural fracture zones,areas of natural drainage, and differences in normal dif-fusion rates. At present, there is no data to support theuse of soil sampling of gases as an indicator of under-ground combustion.

Project

Table 13.-AML research projects, 1987-1990

Contractor Amount

Determining effectiveness of past mine firesurface seal abatement methods.

Self-heating characteristics of coal andrelated carbonaceous materials.

Use of foaming mud cement to terminateunderground coal fires and to controlsubsidence of burn cavities.

Location, extinguishing, and reignitioninhibition of refuse and underground firesthrough high pressure water-jet utilization.

Characterization of remote combustion .....Cryogenic extinguishment of waste bank firesImproved method for extinguishing coal

refuse fires.Development of fire diagnostic techniques for

burning coal waste piles.

L. C. Hanson Co., Helena, MT $99,604

Bureau of Mines, Pittsburgh Research Center .. 160,000

Colorado School of Mines Research lnst. andWyoming Dept., Environmental Quality.

251,456

University Missouri-Rolla and MontanaDepartment of State 'lands.

201,478

Bureau of Mines, Pittsburgh Research Center .... do .MSA Research Corp. . .

237,100243,000160,000

Bureau of Mines, Pittsburgh Research Center .. 237,500

42

Magnetic Anomalies

It has been suggested that the measurement of smallsubsurface magnetic anomalies can be used to delineatefire areas, particularly in coal refuse. The effect is as-sumed to be due to the change in remnant magnetismwhen magnetic minerals are exposed to high temperatures(57). This method has had limited use in attempts tolocate buried refuse and as a diagnostic technique forsubsurface fires, Depth and the presence of steel boreholecasings have an effect on the measured magnetic anomaly.At present, there is insufficient data (laboratory or field)to evaluate this technique as a mine fire diagnostic meth-od. A surface survey of electrical potential anomaliesrelated increased conductivity to the presence of cokedcoal (58).

Desorbed Hydrocarbons-Communication Testing

The Bureau has developed a mine fire diagnosticmethodology in which the measured characteristics are thetemperature, pressure, and hydrocarbon concentration atthe base of an array of boreholes (51, 59-60). The sam-pling method uses an exhaust fan to impose a pressuregradient to control the direction of airflow. Changes inpressure determine whether the fan has influenced theflow of gas at any borehole. The temperature is measured(or comparative purposes, but is not a variable in thedetermination of fire signatures. Changes in a hydrocar-bon concentration ratio are compared with an empiricalscale based on the laboratory determination of fire sig-natures. Interpretation is based on ability to control andtherefore know the direction of airflow, and on relativechanges in the fire signature.

The combustion signature used in the Bureau's minefire diagnostic methodology is a ratio based on hydro-carbon desorption from coal. The low molecular weighthydrocarbons, methane through pentane, are adsorbed onthe. internal surface of coal. At normal temperatures,methane and a small percentage of higher hydrocarbonsare desorbed. As the temperature of the coal increases,the rate of desorption increases and the concentration ofhigher hydrocarbons increases. In a laboratory study oftemperature dependent desorption (59-61) from coals andcoal wastes, a ratio Rl, relating the concentrations ofmethane and total hydrocarbons, was defined as:

1.01[THC) - [CH4]Rl = x 1000[THC) +c

where concentration of total hydrocarbons,ppm,

[THC)

concentration of methane, ppm,

and c = constant, 0.01 ppm.

The ratio was found to increase with increasing temper-ature and decrease with decreasing temperature (fig, 43).The concentration of desorbed gas and the ratio were re-lated to the rank of the coal. For bituminous samples, thevalue of Rl is interpreted according to the empirical scale:

R1 Relative Coal Temperature

o to 50 Normal «300 C)

50 to 100 Possible heated coal (300 to 1000 C)

>100 Heated coal (>1000 C).

The absolute value of Rl is not related to a particulartemperature. For a given coal, Rl values during heatingmay not be the same as those during cooling at the sametemperature (fig, 44). The value is influenced by rank,by internal surface structure, and by the amount of gasadsorbed on the coal surface. However, for bituminous

1,000

Ratio type800 --I

_ •••••••• IT--m

600

Rl400

200

oTEMPERATURE, ·C

Figure 43.-ehanges In mine fire diagnostic ratio withtemperature.

KEY.--- Pocahontas e>-- - Illinois0--- Black Creek _--- D seam_.- Pittsburgh

1.000800 __.>.............•--------..•..------- ..•-----

- - -----.------- ..-------R 1 600 .=.==-=-=-..:::....-:-_-=--:--.. <, ......

~. ----""""-,,"'-"_-- '- - <,,- ---t:.. <,____..• r --__ ~~ "-

150 200 250 200TEMPERATURE, DEG C

400

o100

Figure 44.-Values of mine fire diagnostic ratio during heatingand cooling of selected coals.

samples, the ratio increases during heating and decreasesduring cooling within the ranges listed on the previouspage. The hydrocarbon ratio is sensitive, unaffected bydilution, and requires relatively simple analytical methods.The emission of higher molecular weight hydrocarbonsfrom anthracite is relatively low. In a field study, changesin the absolute concentration of methane was found tocorrelate with hot and cold subsurface zones.

The mine fire diagnostic methodology, in addition tothe hydrocarbon ratio, uses communication testing to de-fine hot and cold subsurface zones (24). This assumes thata sufficiently large negative pressure (vacuum) applied tounderground regions will cause the gases in the mine at-mosphere to flow from some distance toward the point ofsuction. Repeated sampling at all points in a boreholepattern" provides data to determine the presence orabsence of fire along pathways between sampling points(fig. 45). A measured change in pressure of at least0.01 in H20 defines the area in which the fan produceda pressure change and therefore a flow of gas. Foreach test, a quadrant indicating hot (Rl > 100), cool(Rl < 50), or insufficient hydrocarbons, is placed on astraight line drawn through the borehole and the suctionpoint (fig. 46). Reiteration of the tests using variousboreholes as suction points produces a composite map ofheated and cold zones (fig. 47). The tests are repeated,sometimes with the drilling of additional boreholes, untila cold boundary can be defined. To date, three fieldstudies have supported the use of the Bureau's mine firediagnostic methodology to define remote combustion zones(23-24, 51). Sampling the mine atmosphere during extin-guishment can indicate the current status of the fire. Atone project, the mine fire diagnostic method defined threenoncontiguous areas of combustion. During attempted ex-tinguishment, determination of Rl indicated that one areawas cooling, one area showed no change, and one areawas becoming hotter (fig. 48).

The Bureau's mine fire diagnostic technology is asignificant improvement in locating abandoned mine fires,in the essentially two-dimensional plane of a mine. TheBureau is using a physical model of a waste bank to verifythe application of the diagnostic methodology in the three-dimensional volume of a waste bank. The test pile, 4 by4 by 8 ft, is instrumented with pressure probes; two bore-holes, 4 and 2 in, were placed to a depth of 30 in from thetop and can be used alternately as suction points. Airflowthrough the waste bank material was determined to bewithin the limits of Darcy flow. Therefore, flow in a waste

lOGenerally, the borehole pattern is dictated by the size and thetopography of the site. As a first approximation, a pattern of 8-inboreholes on 100-ft centers is adequate.

43

bank under vacuum can be characterized as laminar anda model based on conventional fluid flow equations canpredict the combustion source within a waste bank (62).

It cannot be emphasized more strongly that the firststep in a successful fire control-extinguishment project isknowing where the fire is, and of equal importance, know-ing where it is not. It is also important to determineduring the course of the project that the fire is not prop-agating to previously cold areas. The last step in a suc-cessful fire control extinguishment project is continuedmonitoring to determine that the fire is out and remainsout.

MINE FIRE EXTINGUISHMENT

The extinguishment of any fire requires that at least oneelement of the fire triangle be removed, fuel, oxygen, orheat. Recently, completely new methods of accomplishingthis or adaptations of conventional methods have beentried.

FUEL REMOVAL METHODS

Burnout Control

The Bureau has developed a technique called BurnoutControl that can control fires in abandoned coal mines andwaste banks, and can also extract the thermal energyrepresented by such wasted coals (fig. 49) (63-69). Thetechnique involves complete combustion of the coal inplace while maintaining total control of the resulting heatand fumes. The thermal energy produced and brought tothe surface as high temperature flue gas (up to 1,0000 C)can be 20 times the equivalent thermal energy required tooperate the Burnout Control system.

Induced mine gas flow

KEY8 Suction hole

o Communicating hole

(+) Combustion gas present

(-) No combustiongas present

~ Possible fire location

(&-~-~--~-~--A-Case: A

(+) ~\t (+) ,,'II (-)0--®--o--~--o- - ---

Case: B

~-~--b-----6--- --Cose-C

Figure 45.-Schematic of communications testing for mine firediagnostics.

44

o32

o31

23 180

APARTMENTBUILDING

030 0

29

APARTMEBUILDING

L.INE

o22

o34

o20

o16

LEGENDo Borehole

k;';~:dlHeated zone

§ Cold zone

DNo data

o 100I I I

Scale, It

Figure 46.-Mine fire diagnostic map of results of one communication test.

--- ----

'08S__<,

~1

APARTIlENTeUILDIN8

o29

o 100I I !

Scare,lI

flt°,'"/// o22

__ -1090----o19

o16

LEGENDo Borehole

l'i,:,··!;v~J Hea'ed zone

§ Cold zone

DNa data

-1090 Elevation

Figure 47.-Mine fire diagnostic map produced from repeated communication tests.

Figure 4S.-ehanges in mine fire diagnostic ratio at threeboreholes during extinguishment project.

Burnout exhaust

Boiler

Hot combustiongases

Electricgenerator

Steamcondensatereceiver

Figure 49.-Schematic diagram of Burnout Control with energyutilization.

During Burnout Control (fig. 50), the burning wastebank or coal mine is placed under negative pressure rela-tive to the atmosphere. Air flows into the undergroundcombustion zones through natural fractures, crevices, andpores in the ground and/or through specially drilled airinlet boreholes. An exhaust ventilation system, consistingin part of a large borehole, which acts as a combustionmanifold, and a fan that draws hot gases from the firezone, pulling them out at a single point. Burnout Controlhas several advantages:

1. The affected mine or refuse bank will be at negativepressure, relative to ambient; hence, few or no fumes willbe emitted to the atmosphere except at the fan exhaustpoint.

45

2. Accumulation of all the fumes at the fan exhaustpoint will enable postburn incineration of the exhaust toinsure complete combustion of products to carbon dioxideand water. If needed, flue gas scrubber treatment also canbe applied to remove air pollutants such as sulfur dioxideand particulates.

3. The heat of combustion of the burning coal willappear as sensible heat in the exhaust - at temperatures ashigh as 1,0000 C. This heat is recoverable for producingsteam and/or electricity.

4. The complete burnout of carbonaceous material andpyrites in a mine or waste bank will permanently solve theenvironmental problems of an active fire. In contrast, firesextinguished by cooling and 'sealing leave wasted coal withits potential for reignition and acid water formation.

5. The product of complete burnout of a coal refusebank is "red dog," a gravel substitute with commercialvalue. The "red dog" material, if injected into mine voids,also has potential for mitigating subsidence and acid minedrainage.

Burnout Control has not yet been applied to completelyburn out an AML fire, but it has received two limited fieldtrials during its development. The first trial was a 4-month controlled burn of a shallow, abandoned coal minefire in a confined area of the Pittsburgh seam (65-69).The second trial was a 1-year test of the process as appliedto a 0.9-acre section of a 6.5-acre abandoned coal wastepile fire (64).

At Calamity Hollow (fig. 51), exhaust control (i.e., vac-uum >0.1 in H20) was maintained over an undergroundarea of about 2 acres that encompassed the fire zones.Over 102 days of fan operation, an estimated 1,100 st ofcoal were burned, producing exhaust gases at an averagetemperature of 6000 C and thermal power level of 3.2 MW(fig. 52). At the maximum design level output of 9000 Cand 5 MW, it would have taken 1.6 years to completelyburn the 10,000 st of coal at the site. The total thermalenergy, if converted to electricity through a small mobilesteam turbine-generator system, could have produced 14to 18 million kilowatt-hours of electricity, about 20 timesmore energy than that required to operate the BurnoutControl system.

During the course of the mine fire field trial, significantslow subsidence and ground fracture did occur, but ap-parently not to an extent that affected the burnout processitself. Surface disturbances were handled by relatively sim-ple techniques involving construction of additional footingsfor equipment support, and sealing the ground surface forfissures. In terms of air pollution, test data indicated thatall the fuel-sulfur appeared as sax in the exhaust, but only5% of the fuel-nitrogen appeared as NOx' From an en-vironmental viewpoint, it is probable that removal of saxfrom the exhaust will be required in the actual practice of

46

Diesel electricgenerator

40'-------112' 6"-ll------120'. I

I ~46'--i

Drop-out I I'

tower

i ';-rJ~I1

!, ,

Vertical manifoldwith fluid jacketand refractory liner<

~~

'UOJ.0

ooo

Top of coalbed,\

Figure SQ.-Schematic of Burnout Control system - Calamity Hollow mine fire project.

~~~=~'_.~ .,

Manifold (refractorylined to here)

'-Expansioncollar

Subsidence./""'" , 'zone

bustibleIn corn trench

fire 'rbarne

Figure S1.-Artist's version of Burnout Control system at Calamity Hollow.

47

Temperature

u0

W 6000:::::>~0::~ 400::Ew~

200

09,000

Mass flow

~Cl 6,000

?i9l.L.(f)

~ 3,000::E /~\, ,, ,

/ \,/ \

OL...-_-L-_--L_---l__ .L.-_--l-_----L__ L-_...L-_---.L_---l__ L.-_-L:-_--L_---J

7,OOOr---,---r-----,-----,-----,-------,----.-----r---,---------r---.---------r--,--~

Heat flow6,000

~ 5,000~?i 4,0009l.L.I-: 3,000<:(

~ 2,000

1,000

O/--~---.:.----=2'=-0--::I30=---:l40=--~50=--~--::I=--~-...,9~0::------:1~0-=0--::1,+10=-----:~---=.....,..:!,=-.....,....,..J

Figure 52.-Results of Burnout Control test at Calamity Hollow.

Burnout Control. A royalty-bearing license to use BurnoutControl for power generation from underground mine firesis currently held by Coal Dynamics, Inc., a wholly ownedsubsidiary of Environmental Power Corp.

The field trial of Burnout Control at the Albright wastepile was also designed to operate at a 5-MW (thermal),9000 C exhaust output; however, the engineering of thecombustion manifold was considerably different from thatused at Calamity Hollow (70). At Albright, a 3-ft diame-ter, 140-ft-Iong stainless steel combustion manifold was sethorizontally at the bottom of the pile along its base

(fig, 53). A 50-ft-Iong perforated section of the manifoldserved to draw heat and fumes from a volume of waste es-timated at over 10,000st, which at the design output wouldhave taken 0.75 year to burn producing 113 billion Britishthermal units of thermal energy. Engineering designproblems and the need to control air pollution during thetrial prevented long-term continuous burnout operations atthe site. However, during 1,600 h of fan operations, anestimated 700 st of waste were burned producing 8.12billion British thermal units of exhaust thermal energy.The designed output was exceeded over several days

48

, ,

, '_ I I, Waste bank. '

_';/~--:/f;2~p,/,~BoreOOle

from surface

•

Figure 53.-5chematic of Burnout Control at Albright waste bank.

before burnout operations were terminated when excessivetemperatures and vacuum combined to collapse the com-bustion manifold.A key finding at the Albright field trial was that as

the fire was spread under Burnout Control, the pile actedas a large gasifier; hence, there was a vital need toafterburn the exhaust gases in the combustion manifold.Only 30% of the fuel-sulfur appeared in the exhaust, butwithout full stoichiometric burning, sufficient amounts of.reduced sulfur gases (e.g., HzS, COS, and CSz) werepresent to cause odor problems for nearby residents.Considerable subsidence occurred on top of the waste pileover the fire zones, but as in the case of the CalamityHollow trial, the subsidence holes could be readily filled.With improved engineering designs and developments,particularly with regard to maintaining the structuralintegrity of the combustion manifold and improving itsoperation as an afterburner, it is believed that BurnoutControl can be successfully applied to controlling AMLcoal waste pile fires.

Water Jetting

Another fuel removal method tested was the excavationof burning coal by high pressure water jet (71). The firewas located in an outcrop above the Yellowstone River inMontana. Holes were drilled horizontally and verticallythrough 25 to 30 ft of overburden. Although the drillingcould be accomplished, the direction of the water couldnot be controlled. In most circumstances, it would forma channel out of the coal seam. When the water did reachburning coal, it produced an explosive reaction and highconcentrations of carbon dioxide. The drilled holes werefound to serve as sources of additional oxygen and theoverlying rock tended to collapse. These conditions made

it difficult to remove all the burning coal. In general, thistechnique was not considered a potentially useful methodof excavating AML fires.

HEAT REMOVAL METHODS

Water

Removing heat, one of the three essential elements,from an underground fire requires the introduction of aheat -absorbing medium, its controlled movement throughthe mine and the removal of the heated substance (72).Theoretically, moving very large volumes of cold airthrough a mine will remove heat and lower the tempera-ture. However, the air introduced usually supplies oxygen,which increases the oxidation rate and increases tempera-ture. Water, where it is plentiful and inexpensive, is adesirable heat exchange medium because of its high heatcapacity and its latent heat of vaporization.In two projects, the Bureau has used injected water

with the suction induced removal of heated gases. Intheory, the water is injected into heated subterraneanareas. The water is converted to steam. Under the influ-ence of an exhaust fan, the steam is moved through themine, adsorbing more heat from the heated coal and roofstrata. A fan can be used to exhaust the heated steam tothe atmosphere or to a heat exchanger where its thermalenergy can be used.The Bureau's water injection-fume exhaustion method

had mixed results. During the cool down phase of theCalamity Hollow Mine fire project (66), water was injectedby gravity flow through an array of 1-in hose and 2-inpipes. Simple pinch clamps were used to regulate the flowof water (fig. 54). Prior to quenching, the average

2-in water hose supply line4-in plastic tee with screwedconnectionand 4-in plug for filling

4-in plastic pipe

4-in plastic tee

Water storagepool,18-ft diameter by 4-ft high,

approx 18,000 gal

\To similardrain system

49

1~ -In 4-way(typical)

2-in hose \

Figure 54.-Water distribution system for quenching fire zone at Calamity Hollow site.

-"-:'1 ,

temperature of the exhaust gas was 6000 C. The systemwas operated at an injection rate of 2 to 3 gpm per bore-hole for 8 h each day for 21 days. For an additional10 days, the water was injected continuously. Over the30 days, 2.8 million gal of water was injected. Thetemperature of the exhaust was lowered to 1620 C (fig. 55),and 300 million Btu of heat was removed from the mine(fig. 56). Although the method did lower the temperatureof the fire zone, the estimated heat removal efficiency wasnot high, 7 caljg compared with the heat of vaporizationof 540 calj g. It was known during the operation of thewater injection-fume exhaustion system that much of thewater was flowing out of the mine and was not reachingthe burning material. When the water injection rate wasincreased, the seepage rate also increased (fig. 57).However, the system achieved its objective of cooling thecombustion zone prior to excavation in a relatively shortperiod.

Based on the Calamity Hollow project, an attempt wasmade to extinguish a fire in an abandoned mine at Renton,PA (23, 51). Three noncontiguous combustion sites,totaling about 10 acres, were located on the perimeter ofthe 60-acre site. Water was supplied by gravity to spraynozzles located at the casing bottom in each borehole.The fan was operated for approximately 6 h per day, andthe water injection rate was approximately 1 gpm per

1,000

800

uo 600W0::=>!;:0::Wa,:2:wI-

400

200

o119 126 133 140 147

JULIAN TIME, days

154 161

Figure 55.-Exhaust temperature during quenching at CalamityHollow site.

50

300280260240

:::J 220OJ~ 200~« 1800--l 160LL.

~ 140w:I: 120

100806040

124

3.0 r------.---,--,------,--.------,----,--,.--,----,--,-------,--....-----,----,-----,

2.82.62.42.22.0

.6

.4

.2o ~~l9=O=:B:8~~~~~~::::i::::::::::L:~----l.----L---l.-~124 126 128 130 132 134 136 138 140 142 144 146 148 150 152 154 156

JULIAN TIME, days

Figure 57.-Cumulative waterflow during water injection at Calamity Hollow site.

c~ 1.8~ 1.6.;r-~z«:Ja

1.41.21.0.8

126 128 130 132 134 136 138 140 142 144 146 148 150 152 154 156 158 160

JULIAN TIME, days

Figure 56.-Cumulative heat flow during quenching at Calamity Hollow.

KEY0- H20 injected

/:; - H20 seepage

borehole. The water injection-fume exhaustion system wasoperated in this configuration for about 6 months. Thenthe injection tubes were reinstalled on the outside of thecasing about 1 to 2 ft below the surface. In this manner,the water would flow down the casing and through thestrata above the mine void. However, this modificationproduced no general improvement in cooling efficiency.

Analysis of the data led to the conclusion that combustionwas occurring in the roof coal and carbonaceous shales.It was unlikely that the water reached these areas and nosignificant conversion to steam occurred. Not enough heatwas removed to slow or stop the combustion process. Abetter delivery system for placing the water in contact withthe heated material was needed.

Foam

In the Renton fire project, it was determined thatunless the injected water is rapidly converted to steam, themovement of the water through heated areas of the minecannot be controlled. The use of foam as an extinguishingagent is common for surface fires; a variety of foams areused for well servicing in the petroleum industry. If anyof these foams can be used to retain moisture in the firearea, foam injection might be used as a heat transfermethod. A stable gelled foam can be pumped into thestrata around the fire zone to act as a cold barrier. Thefoam injection-fume exhaust (FIFE) extinguishment meth-od is considered a potentially significant improvement inremote fire control (fig. 58). The use of foam as a watertransport agent has two distinct advantages: it would keepthe water in place long enough for it to be entrained inthe airstream; in voids, the foam could build upon itselfto reach the mine roof where combustion may be seated.

Foam is a dispersion of gas in water. It is made ofwater, a suitable gas, either air or nitrogen, and a surfac-tant. Depending on its purpose, it also may contain fireinhibiting chemicals. Two sources of foams, which are po-tentially useful in extinguishing subsurface fires, are firefighting foams used on surface fires and injectable foamsused in oil field operations.

High expansion fire-fighting foams used on surface firesthose with a high volume to water ratio, can quickly fillvoids, but are relatively fragile and break down when incontact with uneven surfaces. Low-expansion foam hasgreater stability, but does not build upon itself, i.e., it willnot fill voids. Medium-expansion foam is thought to incor-porate the desirable qualities of the other foams, beingstable and capable of filling voids. The standard fire-fighting foams tend to break down when pumped throughporous media. Thickeners or gels added to the foam arebelieved to improve the lateral distribution of foamspumped through material the size of mine refuse (73).

A wide variety of foams are used for well servicing inthe petroleum industry. These are designed for pumpingunder pressure in relatively tight formations. Additives areused to adjust the stability of the foam and its viscosity.Gelling agents can be used to increase the stability of thefoam. Because of the complexity of the subsurface fireproblem, probably a variety of foams will be needed forthis application. Gelled foam because of its stability canbe used to reinforce barriers. Medium quality well-fracturing foam can be pumped through fractured stratafor greater lateral dispersion (74). Medium expansionfire-fighting foam can be used to fill voids. The foams areused to distribute water underground, to keep water inplace until it can be entrained in an airstream, to protectnonfire areas from the migration of hot gases, and to

51

reach areas of the mine that are not within reach of asimple gravity feed system.

Cryogenic Liquids

The use of cryogenic liquids as a heat removal mediumin wasted coal fires has the potential advantages of uni-form distribution of the liquid and isotropic expansion ofa cold gas. If water is injected into a waste bank, gravitycauses it to flow downdip. As the water flows, erosioncauses the size of the drainage channel to increase. Thedistribution of the heat removal agent, water, affects arelatively small area, and cannot be controlled. If a cry-ogenic liquid is injected, moisture in the material freezes,displacing the injected liquid to another area. This in-creases the size of the area affected by the injected liquid.Also, as the temperature of the gas increases, the gas ex-pands, creating a cold pressure front that moves from thepoint of injection to the surface of the bank.

In small-scale tests of cryogenic injection using a1/2 in injection line to a central point in a 55-gal drumfilled with coal refuse, cryogenic CO2 was used as the heattransfer liquid. At atmospheric pressure, the injected CO2

forms a solid, like snow. In a relatively short period, thissolid forms around the injection point and blocks the flowof liquid.

In medium-scale tests conducted in a 320-ft3 box of coalwaste, liquid nitrogen was used as the heat transfer medi-um. These tests indicated that if the refuse was wet, theformation of ice during cryogenic injection could containand direct the flow of the liquid nitrogen. However, at anypoint where the refuse was dry, the nitrogen would act likea liquid and flow to the bottom of the box.

In a test conducted in the Bureau's Surface TrenchBurn Facility at the Pittsburgh Research Center, 11,OOOlbof liquid nitrogen was injected into the center of a 540-ft3

(20 st) bed of coal at the rate of 100 gal/h. Prior to injec-tion, the maximum temperature in the coal was approxi-mately 1000 C. Within a few minutes of the start of injec-tion, the temperature near the injection point was -680 C.In 1/2 h, the temperature of half of the trench was lessthan 00 C, and -1000 C in an hour. The lowest temper-ature recorded was -1700 C. Allowing the N2 to evaporatethrough the open top of the bed, it took 30 days for thetemperature of the coal to reach -200 C (fig. 59).

For the trench experiment, changes in temperatureshowed the distribution of liquid-gaseous nitrogen through-out the bed from a single injection point. When the cryo-genic liquid was contained, the vaporization of the nitrogenwas effective at lowering the temperature of the coalbetween 1000 and 2000 C at the rate of I Ib of nitrogen per4lb of coal.

52

Heat and waterremoval fan

150

100o0

50wf.'Z 0«a:::w0::::J -50I-«0:::wn, -100~wI-

-150

-200

Foam injection --r--r--....,-----------,.boreholes -..-F..rrt7S"O'"

Burning abandonedmine workings

Gelled foam orgrout for cold

barrier

Figure 58.-Foam injection-fume exhaust (FIFE) system for extinguishing AML fires.

111111111,/11111111 III II 1 III I 1 III I[ II

I

53 7

9 13 17 21 25 29 33 37 41 45 49 53 57 6111 15 19 23 27 31 35 39 43 47 51 55 59 63

THERMOCOUPLE NUMBERFigure 59.-High and low temperatures during injection of liquid nitrogen.

The use of cryogenic gas as the extinguishing agent inwaste bank fires assumes that freezing the water normallypresent in the bank will enhance the uniform distributionof the extinguishing agent and that as the liquid vaporizes,the expansion of the cold gas will distribute the extinguish-ing agent upward to the surface. Tests with liquid nitro-gen have shown that the injected liquid causes a relatively

quick cooling of surrounding material and that the expan-sion of the evaporating gas maintains the cool atmospherefor an extended period. To overcome the flow constraints,an apparatus" has been designed to produce a pumpableslurry ofliquid nitrogen and carbon dioxide (fig, 60). This

llpatent applied for.

53

slurry has the distribution properties of a liquid. As thenitrogen evaporates, the carbon dioxide remains in place.Using a batch injection system, 300 gal of the cryogenicslurry reduced temperatures in the trench from 4000 C tobelow the reignition point (fig. 61). Further development

of this technique involves scale-up of the equipment and afield trial (75). The potential applicability of this noveltechnique cannot be determined until research is complet-ed, but appears to be promising. Eventual use will dependupon the cost and the availability of cryogenic liquids.

From CO2supply

C021N2Mixing tank

Sight level guage

From N2-Cl~-~'"supply

CO2 IN2 Slurry lineto injection probe

Pump discharge

Gravity flow-+

N2 wash downtflow

Jet pumpNitrogen Centrifugal pump

Figure 60.-Schematic of equipment to produce cryogenic slurry.

600

.' .. , , ' I . " , I , I I, , I, ,II I III I IIIII I II I

o4000

wCJZ<{0:::W 2000::::JI-<{0:::W0...~ awI-

-2005

3 79 13 17 21 25 29 33 37 41 45 49 53 57 6111 15 19 23 27 31 35 39 43 47 51 55 59 63

THERMOCOUPLE NUMBERFigure 61.-High and low temperatures during injection of cryogenic slurry.

54

OXYGEN EXCLUSION (FIRE BARRIERS)

Recently, foaming grouts have been used as under-ground fire control agents. These have been normal ce-mentitous grouts mixed with a preformed foam and foam-ing mud cements that use local soil as an aggregate (56,76). The foamed grouts are light in weight and can bepumped through boreholes. The thermal conductivity islow, and decreases with the amount of air in the slurry.The thermal conductivity can be increased by adding anappropriate aggregate to the cement slurry. The foamedconcrete when pumped into the voids of an abandonedmine is believed to flow over burning material, encapsu-lating it, and preventing further combustion by oxygenexclusion. The foaming grout has the additional propertyof filling voids, providing subsidence control.

In many abandoned mines, the roof has collapsed andvoids are relatively small. The configuration of an aban-doned mine can more accurately be described as a non-homogeneous, porous bed. Whether foaming grouts canbe forced to flow through such a bed has not beendetermined. Also, although the grouts can reach the mineroof, there is no data on how well it will fill cracks or howfar it can be made to migrate along roof fractures.Presently, the suggested method of using foaming groutspresupposes that they are introduced at or near the com-bustion. If this assumption does not hold, their effec-tiveness is unknown.

RESEARCH AREAS

Diagnostics

The primary need in any AML fire effort is to knowwhere the fire is and what area it affects. The Bureau'smine fire diagnostic technology is a significant improve-ment; however, it requires extensive drilling and gas anal-ysis. The need for repeated tests and possibly additionalrounds of drilling are time consuming and add to the over-all cost. It does have the advantages of being more accu-rate than other methods, of determining cold boundaries,and of being useful for long-term monitoring.

Two methods are being considered to improve the minefire diagnostic technique. The first, ventilation analysis,is being tested. This bases the location of fire zones onpressure changes and anisotropic resistances betweenborehole pairs. It eliminates the need for hydrocarbon gasanalysis. The second improvement would be the use of anon-site sampling and analysis system to replace laboratorygas analysis, paired with a semi-automated computerizeddata evaluation.

A remote reconnaissance method to determine thelocation and extent of subsurface AML fires would im-prove the evaluation and control of these fires by severalorders of magnitude. Ideally, such a method would besimple, quick, inexpensive, and require no drilling.Although methods based on geochemical or geophysicalprospecting may be applicable, the use of such a methodwould require extensive research, including design anddevelopment, controlled testing, field testing, and compara-tive evaluation. Development of any remote reconnais-sance method must consider that the strata between thesurface and the fire are not homogeneous, that permeabil-ity is influenced by the presence of fracture zones and thatthe energy released by the fire and the gaseous productsof combustion are not limited to straight line migrationpaths. Remote reconnaissance is probably the area ofgreatest need in AML fire research; it is also, at present,the area least likely to produce immediate results.Another evaluation technique, which could be more ex-

tensively used in AML fire projects, is the borehole cam-era. The use of a camera in open entries is obvious.However, if the mine has collapsed, could the boreholecamera be used to evaluate the condition of the mine, in-cluding such factors as size and distribution of voids, theheight of the combustible zone, size distribution of rubble,condition of the overburden, directions of normal waterflow, and gross diffe~ences in permeability? Can the bore-hole camera be used to observe the emplacement of a bar-rier or extinguishing agent to estimate potential effective-ness? The borehole camera is an available instrument andhas potential for use in improving the effective applicationof available fire control techniques.

Extinguishment

With respect to extinguishment techniques, excavationis currently the most effective. The implementation of ex-cavation could be improved by the use of diagnostics, morecomprehensive project planning, and the inclusion of post-project monitoring. More attention to safety considera-tions would probably be of value.

For extinguishment techniques other than excavation,the delivery system is the primary problem. Water is aneffective extinguishing agent if it can be placed in contactwith the burning coal. There is no indication that extin-guishing agents other than water are more effective inquenching AML fires. The use of foam and of cryogenicliquids are methods to improve the delivery system, tokeep the extinguishing agent within the fire zone. Re-search on the use of fire retardant chemicals in subsurfacefires is unwarranted unless there is an effective method ofgetting the extinguishant to the fire.

Control-Containment

The use of barriers, either as independent fIre controldevices, or as a corollary to excavation is another fertilearea for research. Foaming grouts or cellular concretes(77-78) have been considered for use as fire barriers. Theuse of stable, nontoxic chemical agents, i.e., plasticizers orgelling agents, also could be considered for heat-absorbingbarriers or toxic fume barriers. Slurry walls, either for fireisolation (rectangular trench barriers) or as containmentdevices for inundation, are potentially applicable to fireproblems. These are currently used for seepage controland for isolating contaminated ground water. In somecircumstances, these may apply to AML fire problems.It should be recognized that under certain conditions,

it may be unnecessary to completely eliminate an AMLfire to solve the fire problem. If the problem is theemission of noxious odors, the possible migration of toxicfumes or subsidence caused by the fire, these can beaddressed by methods that treat the symptoms. As exam-ples, if the fire is in an isolated and inaccessible area, butvents are producing noxious odors that can be detected inpopulated areas, it may be possible to construct a surfaceseal and insert a tall chimney and wind turbine to dispersethe fumes. A seal suppresses surface evidence of a fire,but may not extinguish it unless the fire is relatively small,and has limited sources of oxygen. If the fire is in anabandoned mine with extensive workings, the seal canforce the movement of hot gases into another area of themine, actually spreading the fire, The use of a chimney orother pressure balancing method to control the movementof heat and fumes may significantly improve the effective-ness and longevity of a seal. Pressure balancing can beaccomplished by creating either a high or low pressurearea. Even creating an area of higher permeability (i.e.,a French drain) may control the migration of toxic fumes.The subsurface injection of polyurethane has been used

to control subsidence and water infiltration (79). It may

55

also be useful for forming an impermeable barrier to toxicfumes and as an improvement in surface sealing. Its ex-pansive characteristics, its strength, and its compatibilitywith water make it suitable for barriers. Its combustibilitycan be modified by combining the polyurethane with othermaterials, such as sodium silicate. The near surface in-jection of polyurethane grout to increase the shearstrength of unconsolidated materials (80) may be effectivein improving the longevity of surface seals.Blasting has been suggested as a means of collapsing

the overburden and limiting oxygen in subsurface fires. Inmost cases, the overburden is already collapsed; mine fIresburn in a low oxygen environment « 3%) and the explo-sive fragmentation of combustible material increases thesurface area. Because of these factors, it is unlikely thatblasting would extinguish fires and it could lead to en-hanced propagation rates. In the case of some outcropfires, however, properly placed explosives might be used toremove the burning material from the rest of the seam.The use of the foregoing methods assumes that there is avalid method for determining the degree of risk, the extentto which a subsurface fire may be hazardous to people onthe surface. If that hazard can be evaluated, it may beappropriate to control the symptoms and simply monitorthe fire. At present, there are no guidelines for makingthis assessment, and certainly, careful consideration wouldhave to be given to all factors in any project attempting toprotect surface features without controlling the fire.The above discussion of research needs is intended

to indicate that there are many areas and technical ap-proaches that can be considered. The cost of conductingresearch or of implementing new technical approaches hasnot been a factor in this discussion. However, given thecost of AML fire control projects, the historically poorsuccess rate and the few available options, it is apparentthat money spent on AML fire research would be a goodinvestment in more effective, more efficient reclamation ofAML lands.

SUGGESTED APPROACH TO AML FIRE PROBLEMS

The foregoing information on AML fires should indi-cate that there is no typical fire and certainly no standardcontrol or extinguishment method. The methods currentlyin use are generally expensive and not routinely successful.In an attempt to improve the efficacy of current methodsand to limit costs, the Bureau suggests an approach toAML fires involving an assessment of the hazard andselection of the appropriate response. The steps in thisapproach are as follows:

1. Determine the location and extent of the fire.2. Assess the degree of risk.

Nature of hazard.Toxic fumes.Subsidence.Noxious odors.

Size and distribution of affected population.3. Select appropriate strategy.

Extinguishment.Control-containment.Surface amelioration-cosmetics.Do nothing.

4. Select method most likely to succeed.5. Monitor completed project.

56

The first step in an AML fire project should be todetermine the extent and location of the fire. Determiningthat within a lO-acre site, the fire is limited to 1 acre canconsiderably reduce the cost of the project. Conversely,discovering that a supposed outcrop fire extends into anabandoned mine will affect how much money must be allo-cated to control a fire. And discovering in the middle ofa fire project that what was assumed to be the limit of thework area passes through a hot zone can have a significantimpact on the expenditure of time and money, particularlyif it is necessary to obtain additional rights of entry. Timeand effort put into finding where the fire is and how big itis will usually result in a less costly, more efficient firecontrol effort.The second step is to assess the degree of risk, in-

cluding the type of risk and the population affected. Is thehazard the migration of toxic fumes, subsidence under ornear inhabited structures, smoke and noxious odors in acommunity, subsidence in a remote, but accessible area,fumes in a remote area, etc.? Estimating the probabledirection of propagation is also a factor in the degree ofrisk. If the problem is currently minor, will it in timeaffect more people or be significantly more costly tocontrol? This type of evaluation is essential to cost-effective planning.The next step is to select an appropriate strategy.

Based on the location and extent of the fire and on thehazard it presents, a variety of options may be available.It may be that the degree of risk imposes a responsibilityto extinguish the fire. It's possible that some fires, ifproperly contained or controlled, will burn themselves outor at least cease to be hazardous. In some situations, cos-metic alterations to the surface, i.e., filling in sinkholes orsuppressing vapors, will resolve the problem. And in somesituations, a do nothing option is the correct choice.

Once the strategy has been selected, the method ofimplementation should be planned carefully. Since mostfires have unique characteristics, and since there is nostandard method, the project, including the technique, itsimplementation and the cost, must be tailored to the fire.Designing a fire control or extinguishment project basedon the amount of money available limits the availableoptions and decreases the probability of success. Forexample, the first three attempts to control the Centraliamine fire were stopped before the fire was controlled, be-cause funds had been expended. At least twice, work wasdelayed until additional funding could be approved, andtwice, the control method believed to be most effectivewas rejected in favor of a less costly alternative (27). Theavailability of funds was a factor in the failure to controlthe Centralia mine fire. As a corollary observation, suc-cessful projects are 2 to 10 times more expensive than theoriginal unsuccessful project.Some simple precautions during an extinguishment proj-

ect may limit the probability of reignition. For example,determining the temperature of quenched material beforeit's used as backfill may seem like additional work, but itmay prevent hot material being returned to the excavation.As another example, in many projects, it is probably worththe additional effort to work from an interior cold bound-ary toward the surface. This simple step limits the pos-sibility of propagating a fire beyond the original fire zone.The final step in improving the probability of a suc-

cessful fire project is postproject monitoring. Although itis natural to be optimistic, experience shows that manyAML fires are neither extinguished or controlled at theend of a fire project. Periodic monitoring at least permitsappropriate remedial action to be taken.

CONCLUSIONS

The problem of AML fires is serious. It may not be aspervasive as acid mine drainage, but on a local level AMLfires can involve a greater degree of hazard and can beless amenable to solution. AML fires are usually remoteand may involve outcrops, waste banks, and abandonedworkings. Because of the nature of these fires, it isunlikely that the the extent of the problem or the costof solutions will decrease in the near future. Althoughnew techniques for locating fires, and for controllingand extinguishing fires have been or are being developed,the majority of current AML fire control projects uti-lize conventional methods and techniques. Despite theseriousness of the problem, in some cases, costs are

prohibitive and current methods are inadequate. In manycases, however, evaluation, assessment, and planning canimprove the implementation and cost effectiveness of a firecontrol effort. Wasted coal fires have occurred in thiscountry for over 200 years. Considering the extent ofabandoned mined lands, they may continue to occur forthe next 200 years. Experience has shown that they aredifficult and expensive to control. Most methods currentlyused to control AML fires are less than 70% effective.Research in new technology and in the adaptation of tech-nology available in other fields may significantly improvethe effectiveness of fire control methods, and is essentialto reducing the cost of AML fire control.

57

REFERENCES1. Schopf, J. M. Definitions of Peat and Coal and of Graphite That

Terminates the Coal Series (Graphocite). J. of GeoI., v, 74, No.5, 1966,pp. 584-592.

2. Stach, E., M-Th. Mackowsky, M. Teichmuller, G. H. Taylor,D. Chandra, and R Reichmuller. Stach's Textbook of Coal Petrology.Gebruder Borntraeger, Berlin, 1975, 428 pp.

3. Hessley, R K, J. W. Reasoner, and J. T. Riley. Coal Science.Wiley, 1986, 269 pp.

4. Energy Information Administration. Coal Production 1988. U.S.Dept. of Energy, DOE/EIA-0118(88), 1989, p. 85.

5. Keystone Coal Industry Manual. Mining and Construction Group.Maclean Hunter, 1991, p. 692.

6. Mitchell, D. W. Mine Fires. Maclean Hunter, 1990, 200 pp.7. Magnuson, M. O. Control of Fires in Abandoned Mines in the

Eastern Bituminous Region of the United States. BuMines IC 8620,1974,53 pp.

8. Rushworth, P., B. D. Haefner, J. L. Hynes, and R K Streufert.Reconnaissance Study of Coal Fires in Inactive Colorado Coal Mines.CO Geol. Surv. IS26, 1989, 60 pp.

9. Shellenberger, F. H., and D. L. Donner. Controlling Fires inUnmined Coal Deposits and Abandoned Coal Mines in the WesternUnited States and Alaska. BuMines Prog. Rep. 10047, 1979, 150 pp.

10. Eavenson, H. H. The Pittsburgh Coal Bed - Its Early History andDevelopment. Trans. AIME, v. 130, 1938, pp. 9-10.

11. __ The First Century and a Quarter of American CoalIndustry. Privately printed, Pittsburgh, PA, 1942,701 pp.; available onloan from Carnegie Library of Pittsburgh, Sci. and Technol. Div,

12. Lavender, D. The Way to the Western Sea. Lewis and ClarkAcross the Continent. Harper and Row, 1988, p. 190.

13. Thwaites, R G. Original Journal of the Lewis and Clark Expedi-tion - 1804-1806. Arno Press, 1964.

14. Scott, G. S. Anthracite Mine Fires: Their Behavior and Control.BuMines B 455, 1944, 206 pp.

15. Doyle, W. S. Deep Coal Mining Waste Disposal Technology.Noyes Data Corp., Park Ridge, NJ, 1976, 392 pp.

16. Chaiken, R. F. Controlled Burnout of Wasted Coal on Aban-doned Coal Mine Lands. BuMines RI 8478, 1980, 23 pp.

17. Kim, A. G., and R F. Chaiken. Relative Self-Heating Tendenciesof Coal, Carbonaceous Shales, and Coal Refuse. Paper in Proceedingsof the 1990 Mining and Reclamation Conference, Charleston, WV,Apr. 23-25, 1990. WV Univ. Publ. Serv., 1990, pp. 535-542.

18. Cassidy, S. (ed.), Elements of Practical Coal Mining. Coal Div.,SME-AIME, NY, 1973, 614 pp.

19. Chaiken, R. F. Heat Balance in In Situ Combustion. BuMinesRI 8221, 1977, 11 pp.

20. Kanury, A. M. Introduction to Combustion Phenomena. Gordonand Breach (NY), 1975, 411 pp.

21. Kim, A. G. Laboratory Studies on Spontaneous Heating of Coal:A Summary of Information in the Literature. BuMines IC 8756, 1977,13 pp.

22. Smith, A. c., and C. P. Lazarra Spontaneous Combustion Studiesof U.S. Coals. BuMines RI 9079, 1987, 28 pp.

23. Dalverny, L. E., and R F. Chaiken. Technical Services to Extin-guish the Mine Fire in the Abandoned Renton No.1 Mine in Plum Bor-ough, Allegheny County, Pennsylvania. Rep. to OSMRE, InteragencyAgreement No. HQ-51-CT-6-01492, 1988, 62 pp.

24. Justin, T. R, and A. G. Kim. Mine Fire Diagnostics: Carbon-dale, PA. Rep. to OSMRE, Wilkes Barre, PA, Interagency AgreementNo. CT745011, 1987, 52 pp.

25. Leitch, R D. Some Information of Extinguishing Refuse-BankFires Near Mahonoy City, PA. BuMines IC 7104,1940,16 pp.

26. Chaiken, R F., J. M. Singer, and C. K Lee. Model Coal TunnelFires in Ventilation Flow. BuMines RI 8355, 1979, 32 pp.

27. Chaiken, R F., R Brennan, B. Heisey, A. G. Kim, W. Malenka,and J. Schimmel. Problems in the Control of Anthracite Mine Fires: ACase Study of the Centralia Mine Fire. BuMines RI 8799, 1980, 93 pp.

28. Dalverny, L. E., R F. Chaiken, and A. G. Kim. Mine Fire Diag-nostics in Abandoned Bituminous Coal Mines. Paper in Proceedings ofthe 1990 Mining and Reclamation Conference. (Charleston, WV,Apr. 23-25, 1990). WV Univ. Publ. Serv., 1990, pp. 527-534.

29. Michalski, S. R, L. J. Winschel, and R E. Gray. UndergroundMine Fires on Abandoned Mine Lands. Paper in Proeedings 10th An-nual AML Conference, PA/D ER, Wilkes Barre/Poconos, October 10-14,1988,29 pp.

30. American Conference of Governmental Industrial Hygienists.Documentation of the Threshold Limit Values for Substances in Work-room Air. 1971, pp. 39, 41-42.

31. Patty, F. A. (ed.). Industrial Hygiene and Toxicology. Inter-science, v. 2, 1963, pp. 913-915, 924-938.

32. Plunkett, E. R Handbook of Industrial Toxicology. ChemicalPubl., 1976, pp. 85, 88-89.

33. U.S. Dept. of Health, Education, and Welfare. Criteria for a Rec-ommended Standard: Occupational Exposure to Carbon Monoxide,1972, 122 pp.

34. __ . Criteria for a Recommended Standard: Occupational Ex-posure to Carbon Dioxide, 1976, 169 pp.

35. DeKok, D. Unseen Danger. Univ. PA Press, 1986, 299 pp.36. Griffith, F. E., M. O. Magnuson, and G. J. R Toothman. Con-

trol of Fires in Inactive Coal Formations in the United States. BuMinesB 590, 1960, 105 pp.

37. Jolley, T. R, and A. W. Russell. Control of Fires in Inactive LandDeposits in Western United States Including Alaska, 1948-58. BuMinesIC 7932, 1959, 22 pp.

38. Johnson, W., and G. C. Miller. Abandoned Coal-Mined Lands:Nature, Extent, and Cost of Reclamation. BuMines Spec. Publ. 6-79,1979,29 pp.

39. McNay, L. M. Coal Refuse Fires, An Environmental Hazard.BuMines IC 8513, 1971, 50 pp.

40. Philbin, D. R, and J. A. Holbrook, II. Progress Toward Abate-ment of Mine Fires in the Anthracite Coal Fields. BuMines IC 9184,1988, pp. 330-336.

41. Office of Surface Mining, Reclamation and Enforcement. A Na-tional Inventory of Abandoned Mine Land Problems: An Emphasis onHealth, Safety, and General Welfare Impacts. 1983, OSM/TR-4/83.Dep. of Interior.

42. __ AML Inventory Update Manual. Technical SupportBranch, OSMRE, 1987, 42 pp.

43. Maynard, B. Distribution and Characteristics of Outcrop Fires inHorizontal Strata in Eastern and Western Coal Provinces. Paper in Pro-ceedings 1990 Mining and Reclamation Conference, v. 2, WV Univ. Publ.Serv., 1990, p. 505.

44. National Research Council, Committee on Abandoned MineLands, Board on Mineral and Energy Resources. Abandoned MineLands, A Mid-Course Review of the National Reclamation Program forCoal. Natl. Acad. Press, Washington, DC, 1986, 221 pp.

45. U.S. Code of Federal Regulations. Title 30-Mineral Resources;Chapter VII-Office of Surface Mining, Reclamation and Enforcement,Department of the Interior; Subchapter R-Abandoned Mine Land Rec-lamation; July 1, 1989.

46. Maley, T. S. Handbook of Mineral Law. MMRC Publications, 2ded., 1979, pp. 377-387.

47. Corslaw, H. S., and J. C. Jaeger. Conduction of Heat in Solids.Univ. Press, Oxford, 2d ed., 1959, 510 pp.

48. Kim, A. G., and L. E. Dalverny. Large Mine Fire DiagnosticStudy, Report to PA/DER, Memorandum of Agreement 14-09-0050-3320, 1992, 108 pp.

58

49. Hanson, L. c., J. E. Hodges, and K. M. Redmond. Determiningthe Effectiveness of Past Mine Fire Surface Seal Abatement Methods.Final Report, (contract J0278004). BuMines OFR 12-91, 1990,235 pp.

50. Jones, J. H., and J. C. Trickett. Some Observations on the Exam-ination of Gases Resulting From Explosions in Colleries. Trans. Inst.Min. Eng., Y. 114, 1955, pp. 768-791.

51. Dalverny, L. E, and R. F. Chaiken. Mine Fire Diagnostics andImplementation of Water Injection With Fume Exhaustion at Renton,PA. BuMines RI 9363, 1991,42 pp.

52. Graham, J. I. The Normal Production of Carbon Monoxide inCoal Mines. Trans. Inst, Min. Eng., v. 60, 1920-21, pp. 222-231.

53. Miron, Y., A. C. Smith, and C. P. Lazarra. Sealed Flask Test forEvaluating Self-Heating Tendencies of Coals. BuMines RI 9330, 1990,18 pp.

54. Chaiken, R. F. Extinguishment of Wasted Coal Fires: A CriticalReview-New Directions. Paper in Proceedings National Symposiumand Workshops on Abandoned Mine Land Reclamation, Bismark, ND,May 21-24,1984, pp. 457-484.

55. Bruhn, R. W., and S. R. Michalski. Control of an UndergroundMine Fire in Pennsylvania's Anthracite Region. Proceedings of 11th An-nual Assoc. of Abandoned Mine Land Programs Conference (Williams-burg, VA), Oct. 16-19, 1989, pp. 56-62.

56. Terry, R. C. Terminating Persistent Underground Coal Fires.U.S. Pat. 4,641,711, Feb. 10, 1987.

57. Schueck, J. Using a Magnetometer for Investigating UndergroundCoal Mine Fires, Burning Coal, Refuse Banks, and for Locating AMDSource Areas on Surface Mines. Paper in Proceedings of the 1990 Min-ing and Reclamation Conference, Y. 2 (Charleston, WV, Apr. 23-25,1990). WV Univ. Publ. Serv., 1990, pp. 493-502.

58. Rodriguez, B. D. A Self-Potential Investigation of a Coal MineFire. M.S. Thesis, CO Sch. Mines, Golden, CO, 1983, 130 pp.

59. Kim, A. G. Signature Criteria for Monitoring Abandoned MineFires. Paper in Proceedings of the 8th Annual National AbandonedMined Lands Conference, 1986, Billings, MT, pp. 8-26.

60. __ Laboratory Determination of Signature Criteria forLocating and Monitoring Abandoned Mined Land Fires. BuMinesRI 9348, 1991, 19 pp.

61. __ Laboratory Determination of Signature Criteria forLocating and Monitoring Abandoned Mined Land Fires. Report toOSMRE, Interagency Agreement No. HQ-51-Cf-6-01535, 1988, 45 pp.

62. Jones, J. R., and R. F. Chaiken. Fire Diagnostic Simulation forBurning Coal Waste Banks. Paper in Proceedings of the 1990 Miningand Reclamation Conference, Y. 2 (Charleston, WV, Apr. 23-25, 1990).WV Univ. Publ. Serv., 1990, pp. 521-526.

63. Chaiken, R. F. Method for Controlled Burnout of AbandonedCoal Mines and Waste Banks. U.S. Pat. 4,387,655, June 14, 1983.

64. Chaiken, R. F., and L. G. Bayles. Burnout Control at the AlbrightCoal Waste Bank Fire. BuMines IC 9184, 1988, pp. 337-342.

65. Chaiken, R. F., L. E. Dalverny, and A. G. Kim. Calamity HollowMine Fire Project (In Five Parts) 2. Operation of the Burnout ControlSystem. BuMines RI 9421, 1989, 35 pp.

66. Chaiken, R. F., E. F. Divers, A. G. Kim, and K. E Soroka. Ca-lamity Hollow Mine Fire Project (In Five Parts) 4. Quenching the FireZone. BuMines RI 8863, 1984, 18 pp.

67. Dalverny, L. E., R. F. Chaiken, and K. E. Soroka. CalamityHollow Mine Fire Project (In Five Parts) 3. Instrumentation for Com-bustion Monitoring, Process Control, and Data Recording. BuMinesRI 8862, 1984, 23 pp.

68. Irani, M. c., R. F. Chaiken, L. E Dalverny, G. M. Molinda, andK. E. Soroka. Calamity Hollow Mine Fire Project (In Five Parts) 1.Development and Construction of the Burnout Control System.BuMines RI 8762, 1983, 29 pp.

69. Soroka, K. E, R. F. Chaiken, L. E. Dalverny, and E. F. Divers.Calamity Hollow Mine Fire Project (In Five Parts) 5. Excavation andEvaluation of the Fire Zone. BuMines RI 9001, 1986, 25 pp.

70. Garvin, J. P., and H. F. Moomau. Field Investigation, Design, andConstruction of a Coal Waste Bank Fire Control System. Final ReportFrom Green International, Inc. on U.S. Bureau of Mines contractS0348046 (May 1986),122 pp.; available from R. F. Chaiken, BuMines,Pittsburgh Research Center, PO Box 18070, Pittsburgh, PA.

71. Mazurkiewicz, M., and G. Galecki. High Pressure Water Jet FireExtinguishing Research. Paper in Proceedings of the 8th Annual Na-tional Abandoned Mined Lands Conference, Billings, MT, 1986,pp.28-46.

72. Dalverny, L. E. Heat Removal Methods for Control of Under-ground Abandoned Coal Mine Fires. BuMines IC 9184, 1988,pp. 343-347.

73. MSA Research Corp. Improved Method for Extinguishing CoalRefuse Fires. Final Rep. contract J0398007, 1991, 100 pp.

74. Haliburton Services (Duncan, OK). Foam Stimulation. 1985,62 pp.

75. Kim, A. G., and A. M. Kociban. Cryogenic Extinguishment ofWaste Bank Fires. Paper in Proceedings of the 14th Annual AbandonedMined Land Conference, Chicago, IL, Natl. Assoc. Abandoned MinedLand Programs, 1992, pp. 276-289.

76. Lucero, R F. Use of Foaming Mud Cement to Terminate Under-ground Coal Fires. Paper in Proceedings of the 8th Annual NationalAbandoned Mined Lands Conference, Billings, MT, 1986, 27 pp.

77. Mearl Corp., Foam and Chemicals Div. Fire Fighting and Sub-sidence Control. Technic. Bull. GM601, 1989, 1 p.

78. Mearl Corp., Foam and Chemical Div, Cellular Concrete Pro-duction Methods and Equipment. Technic. Bull. FC 703, 1986, 11 pp.

79. Ackman, T. E., and J. R Jones. Methods to Identify and ReducePotential Surface Stream Water Losses Into Abandoned UndergroundMines. Environmental Geol. Water Sci., v, 17, 1991, pp. 227-232.

80. Jones, J. R., and T. E. Ackman. Shear Strength Analysis ofGrouted Coal Waste Materials to Improve Mine Outslope Stability: ALaboratory Analysis. Int. J. of Min. and Geol. Eng., v. 8, 1990,pp. 227-240.

INT.BU.OF MINES,PGH.,PA 29761

*U.S.G.P.O. 1993-709-008/80020