Bulletin 492 Bureau oF Mines METHODS OF ANALYZING COAL AND COKE By A. C. Fieldner and W. A. Selvig UNITED STATES GOVERNMENT PRINTING OFFICE ¢ WASHINGTON : 1951UNITED STATES DEPARTMENT OF THE INTERIOR Oscar L. Chapman, Secretary BUREAU OF MINES James Boyd, Director For sale by the Superintendent of Documents, U. S. Government Printing Oifloe Washington 28, D. C. Price 50 contsForeword ‘This paper, first published in January 1912 as Technical Paper 8, was revised and enlarged in 1913, 1926, 1929, and 1938 to incorporate various improvements and modifications in laboratory procedures that had been made in the existing methods as weil as a number of new methods. ‘The 1938 edition of Technical Paper 8 described methods for determining sulfur forms and carbon dioxide in coal, the agglomerating index of coal, and iron forms in coul-ash slags and elinkers and for analyzing coal and coke ashes. ‘The present bulletin includes changes in the determination of carbon and hydrogen. by means of an electrically heated combustion furnace as given in the 1938 edition, methods for determining the agglutinating value of coal and the free-swelling index of coal, and tumbler test for coke. "The Bureau hopes that the paper in its present form will prove valuable to all persons interested in tho analysis of coal and coke,CONTENTS Foreword Introduetion, Preparation of iaboratory sample with preliminary “ary Method of caleuatog aid reporting anal reparation of laboratory sample’ without” pre= liminary air-drying. . eS - ca Relative merits of the two methods Total moisture deter Methods of analysis. ‘Moisture. Ash Volatile matter Modified method” for’ subbiiuminous coal, Tignite, peat, and eertain cokes, chars, at ‘Mraetfes; and semianthracites: 0 Use of platinum erueible with eapsule cover. Proper adjonzent of heating appara Siguifiance of results. Use of eleetrie furnace Fixed earbon Bultur. HEschika method > Bombawashing method." Garbon and hy droge Solutions requi Sulfur forms in coal. Sulfate sulfur Pyritie sulfur Organie sul Carbon dioxide in Determination of the calorific value of eoa Procedi Calculations ‘Determination of suifur—eorrection for acidity —— Constant-temperature tank for’ calorimeter ‘water. Standardization of the calorimeter Fusiility of oa! ash ‘Atmosphere. Furnace: Preparation of coal ash. a Preparation of eones Method of heating. Determination of agglomerating index of coal... Classification of coals by rank... Determination of true specific gravity of coal and ‘coke sb Detern “Method for large sample of eoke. Shatter test for coke. ‘Tumbler test for coke Samplin Tumbler machine. Sieves, Determination of free-swelling index of coal ‘Crucibles- Burner assembly Calibration of burner. Method of test Procedure for buttons that do not’ conform to standard profile Determination of aggtinating value of coal Porcelain eracibles Compression macs jutton compressor. Calibration of furnace Silicon earbide. Preparation of coal Method of test Analysis of coal and coke ash ‘Preparation of ash sample Ignition loss at 800° C. Phosphorus pentoxide Manganese oxide Caletm oxide. Magnesium oxide. Sulfur u Sodium and potassium oxides Solutions used for eoal- and coke-ash analysis. Method of reporting analysis Determination of ferric oxide, ferrous oxide, and ‘metallic iron in coal-ash slags and elinkers Preparation of sample. ‘Total iron Ferrous iron Metallic fron Solutions for iron forms- Details of drying oven. |; Apparatus for determining CO, . Furnace for determining tus A, Ce . Section of furnace arranged for fusion test 5; Brass cone mold for making ash eones 3; Residues from volatile-matter determination, shor . Apparatus for determining apparent specific gravity . + Sat tube - , Pull-seale standard profiles and corresponding free-welling index number : Relnionchi art haniarirodice aati aes |. Equipment for agglutinating-value test: A, Crucible stirrup; B, button compressor. CONTENTS: ILLUSTRATIONS Mixing wheel Moisture over ‘Ten-milliliter plat ining vol Electric furnace for determining volatile matter Percentage of volatile matter from different coals at various Electrically heated combustion furnace. Feng teesoope ad catheter ‘onstant-temperature apparatus termining fusiblity snes mounted on base: B, ts ig vatious degrees of waglomeration and cain Shatter-test machine. Details of burner assombly for determination of free-swelling index — Burner assembly, showing erucible in position. wf Index numbers.METHODS OF ANALYZING COAL AND COKE’ By A.C. Fieldner® and W. A. Selvig? Introduction HE Bureau of Mines has received many re- quests for information concerning” the methods its laboratories use for analyzing coal and coke and determining their heatin; value, ‘The fuel investigations now conducte by the Burean of Mines are an outgrowth of the work undertaken at the Government, coal- testing plant erected in 1904 at the Louisiana Purchase Exposition in St. Louis, and the methods for proximate and ultimate analysis of coal used by the Bureau of Mines are bised on those originally adopted by the chemists of the coal-testing plant and followed in the sub- sequent chemical work of the Government fuel- testing plants at St. Louis, Mo. Norfolk, Va., and Pittsburgh, Pa. The many valuable sug gestions and titiciems received. from N; W. 1nd, director of the School of Mines, and from E, E. Somermeier, professor of metallurgy, of ‘Ohio State University, at that time, are greatly appreciated; development of many of the methods for the proximate and ultimate analy- sis of conl now used by the Bureau of Mines is chiefly due to these men, PREPARATION OF LABORATORY SAMPLE WITH PRELIMINARY DRYING The coal samples are received at the labora- tory in zine-coated iron cans that contain when full about 3 pounds of coal. Each can has a close-fitting screw cap and is made virtually moisture-tight by a rubber gasket and a piece of electrician’s tape wrapped around the joint between the cap and the top of the can. “The methods of collecting and shipping samples are described in Technical Paper 1‘ and Handbook ‘on Coal Sampling® of the Bureau of Mines. ‘As soon as a sample is received at the labora- tory it is given a laboratory number and at ‘once placed in a weighed aluminum pan about 10 inches wide, 141 inches long, and 2 inches deep. The pan containing the sample is put ina large drying oven (fig. 1) in which a tem- perature of 30° to &5° C. ig, maintained. ‘hrough this oven a current of warm air is drawn by means of an ordinary desk fan mounted on top of the oven. A controlled tem- perature room is now used by the Bureau of ‘Mines because of the large number of air-drying determinations that must be made daily. The sample is dried under these conditions until the loss in weight between two successive weighings, made 6 to 12 hours apart, does not exceed 0.1 percent per hour. The purpose of drying is to reduce the moisture in the sample to'approximate equilibrium with the air of the laboratory and thus prevent changes in moisture ontent when the sample is handled during crushing and analysis. The loss of weight by air-drying is not re- garded as an accurate determination, and the ‘air-drying loss” reported in an analysis simply ‘means that the sample lost the stated percentage ‘of moisture before it reached the condition of approximate moisture equilibrium with the air of the room at a temperature of 30° to 35° C. “ork on manuscript completed June 1049. Silek, Bucks and’ Explosives Division, Bureau of Mines, WyRhnervising’chémtet, Coat Constitution and Migcelianeou ig Seqdog, Gentes ‘sperament’ Sation, ures of is Ay ‘Phe Sampling of Coal tn the Mine: Bureau | Baber, TL 2p, (ievised fn 1918517 10.) Snyder N. Hy Handbook on tial Sampling (Revision ot ‘Techical Paper 138)'" Bureau of Mines, 1800, 10 00. 12 METHODS OF ANALYZIN 2 COAL AND COKE 44" Fieve 1—DETAILS OF DRYING OVEN, Immediately after the last weighing the en- tire sample is pulverized to 20-mesh size quickly by being passed through a roll crusher. The 20-mesh ‘material is then reduced with a riffle sampler to about 200 grams. This 200-gram portion is at once put in the porcelain jar (1 gallon capacity) of a ball mill, sealed so that it is airtight, and ground to 60°mesh size. ‘The jar is slightly more than half full of well- Tounded flint pebbles, which are about % Ich in diameter, and it makes 45 to 50 revolutions per minute.’ ‘The wear of the pebbles and the interior of the mill does not increase the ash content appreciably. Bituminous coals require 10 to 20 minutes and anthracite about 1 hour of rotation to be pul- verized to 60-mesh size, Coke requires 4 to Lhour of grinding to be pulverized to 60-mesh size’ When pulverization is complete the con- tents of the jar are poured on a %4-inch, square- mesh screen, the sumple is separated from the pebbles by Shaking the screen over a rubber mixing cloth, and the pebbles and the inside of the jar are brushed earefully to remove any adhering pulverized coal. ‘The sample is then passed quickly through ‘a 60-mesh sieve. To minimize changes in moisture, the sieve is kept covered while the sample is being sieved. Usually some coarse particles, amounting to 0. to 0.4 percent of the sample, remain on the sieve. As these particles are likely to consist largely of * Selvig, W. A. Methgde of Taboratory Grinding of Coke for usieele MistaaGh Hes Rope of tavectiga ons So,PREPARATION [7] worm Gear OF LABORATORY SAMPLE WITH PRELIMINARY DRYING 3 ash-forming substances, they are reduced to 60-mesh on a bucking board and added to what passed through the sieve. Coarse particles of coke are reduced to 60-mesh by impact in a hard-steel, diamond mortar. The sample is mixed quickly and then reduced to about, 50 grams by passing it through a small riffle sampler.” ‘This 50-gram portion is put in a rubber-stoppered, wide-mouthed, 4-ounce bottle and constitutes ‘the laboratory sample. To minimize change in moisture, the mixing and reducing of the pulverized sample after re- moval from the ball mill are done with utmost Motor dispatch. ‘The total time elapsing from the opening of the porcelain jar to the stoppering of the laboratory sample bottle does not exceed 3 minutes, Before portions for the various analytical ° determinations are weighed the sample is mixed thoroughly. ‘The bottle containing the sample Ruy is placed in the mixing wheel shown in figure Proune 2—MIXING WHEEL. 2 and rotated for 20 minutes. The wheel is 30 inches in diameter and revolves 25 times in a minute, The sample bottles are held by spring clips on the periphery of the wheel at an angle of 45° with the axis. The bottles should not. be more than half-filled, and the speed of rota- tion should be slow enouigh to prevent cent gal action from hindering the thorough mi of the powdered coal. METHOD OF CALCULATING AND REPORTING ANALYSES Asall the analytical determinations are made the analysis of the air-dried coal in the follow- on the air-dried sample, the analysis of the eoal ing as received ner: lated from at the laboratory is calei Caleulation from analysis* of “air-dried” coal to coal “as received” (100—air-drying loss) Moisture x, ae +air-drying loss =moisture as received. ile matter x 100—air-drying 1088) — viatile matter as receive Volatile matter 100 ratile matte a red carbon x 200—air-drying Fixed carbon x va (100—airdrying loss) 100 (100—air-drying toss) 00 (100—air-drying loss), a (100—air-drving loss) 100 it 100 Osyguny E00 ai devine 108) 54, gi-dryingloss—ony gen as seceved. (100 —air-dryine loss) 10 fixed carbon as received. Ashx. 1h as received Sulfur, sulfur as received, Hydrogen x drying loss==hydrogen as received. Carbon x: carbon as received, (100: Nitrogen x TE} trogen as received, Calories x calories as received. TEDSEPE_E J. B, Davin supervising chemtxt, Coal Cas * AIL figures expressed in percent 912960 s1—24 METHODS OF ANALYZING COAL AND COKE, A statement of analysis is referred toa mois- or “as-received” analysis in the following ture-free basis by calculation from the “air-dry” manner: Calculation from “air-try” or as-received analysis to “dry coal” 100 Volatile matter x 359 sta’ folatile matter in dry coal. 100. Fixed carbon 75g 200g fized earbon in dry coal. 100 Ash x 799—moisture Sulturx gq as ash in dry coal. (Bydrogen—¥ moisture) X 99100 =hydrogen in dry coal. wolsture 100 Carbon x 755 Wg —earbon in dry coal, 100. ‘moisture Nitrogen x 799— nitrogen in dry coal. 100 (Oxygen—¥% moisture) x 59 —Nsistare oxygen in dry coal. fe 100 Calories x 799 moistare lories in dry coal. For making comparisons coal analyses are received, or moisture-free analyses in the fol- sometimes referred to a “moisture- and ash- lowing manner: free” basis by calculation from the air-dry, as- Calculation from air-dry, as-received, or moisture-free analysis to moisture and ash-free basis 100. Volatile matter iag—Gacintare aah) Volatile matter referred to moisture- and ash-free bass 00 i i Fixed carbon X 799—Gaaiemare Pash) —f8ed carbon referred to moisture- and asb-free basis. 00 Sulfur 769 Gaotstane Fash) TSMMUT Teferred to moisture- and ash-free basis, (Hydrogen —H moisture) x 5 10 drogen referred to moisture- and ash-free basis T00= (moisture Fash) 100 . Nitrogen 95a eras) MHOEEN refered to molsture-and asc bal (Oxygen Ho maar X95 Gragg OED err to one and alee Dsl 100 i Calories 55 —Gmaleeare Fas Tetloies refered to molature- and ash-free basis, Calorific values in calories are converted to (B. tu.) by multiplying by 1.8. calorific values in British thermal unitsMETHODS OF ANALYZING COAL AND COKE 5 PREPARATION OF LABORATORY SAMPLE WITHOUT PRELIMINARY AIR-DRYING The samples are received in zine-coated iron cans. If the coal is eoarser than Yj inch, it is first passed through a chipmunk jaw crusher, which reduces it to 14 inch and finer, and then through a roll crusher, which reduces it to 20- mesh fineness. Immediately after the mate- rial has passed through the rolls a 50-gram por- tion is taken for the determination of total moisture. It is taken with a spoon from var- ious parts of the material and is transferred to a rubber-stoppered, 4-ounce bottle. ‘The re- mainder of the sample is mixed thoroughly and then is passed repeatedly through a riffle sam- er until reduced to approximately 200 grams, ‘his 200-gram portion 1s ground to pass a 60- ‘mesh sieve by means of a planetary-disk pul- verizer or ball mill, again mixed, and reduced with a rifle sampler to about 50'grams, ‘This final sample is preserved in a rubber-stoppered, wide-mouthed, 4-ounce bottle. ‘Moisture is determined at 105° C. with 1-gram portions of the 60-mesh sample and with 5-gram portions of the 20-mesh sample, as deseribed on page 6, Analysis of the 60-mesh coal, which has become partly air-dried during sampling, is calculated to the dry basis by dividing each result by 1 minus its eontent of moisture. ‘The analysis of the coal as received is computed from the dry-coal analysis by multiplying by 1 minus the total moisture found in the 20- mesh sample, RELATIVE MERITS OF THE TWO METHODS ‘The first method is preferable for preparing laboratory samples that are intended for highly accurate analyses, The unavoidable loss of moisture during sampling is less than by the second method, especially in samples of wet. or freshly mined’coal. Such samples lose mois- ture rapidly on exposure to air. Air-drying should not be unnecessarily prolonged; other- wise an appreciable loss of heating value from oxidation takes place. ‘The second method, which was formerly used in the coal-analysis laboratory of the Bureau of Mines,’ permits a large number of samples to be handled in a short time. The moisture obtained by this method is usually somewhat less than that obtained by the first method. For coals that have lost part of their moisture content through exposure to the atmosphere, like the usual commercial shipments, this differ ence need not exceed 0.5 percent.” Wet sam- ples must be partly air-dried. ‘The disk pulverizer is not adapted to the fine grinding of coke and anthracite; the abrasive action of the coke on the iron surface of the disk pulverizer contaminates the sample se- riously, and anthracite is heated by the rub- bing surfaces to a degree that might change the composition of the sample. TOTAL MOISTURE DETERMINATION FOR LUMP COKE ‘The percentage of total moisture can be de- termined quickly within an accuracy of 0.8 by simply heating a large sample (10 pounds or more) of 1- to 2-inch lumps of coke th constant weight in any convenient oven or on a hot plate at a temperature of not less than 104° nor more than 200° C.* Because of its simplicity, this method may be used advantageously at points where coke shipments are sampled arge zinc-coated iron pans are used for the samples, Tf the pans are placed ‘on hot plates, the coke is spread in the pans to a depth of about 2 inches, and the pans are raised to give an air space of about 14 inch between the bottom of the pans and the hot plate to prevent overheating the coke; temperatures are determined by insert- ing thermometers in the bed of coke near the * See, The Fuel-Ineection Laborstors of the Buren ot Mina Pen ats Le aaa ae olature in Coke’ Hren’ of Mines ech: Paper 348, 184%, bottom of the pans. The loss in weight is caleulated and constitutes the percentage of total moisture in the coke. ‘The dried sample is crushed mechanically with a jaw crusher or by hand on a chilled iron or hard-steel plate by impact of a hard bar or sledge, avoiding all rubbing action; otherwise, the ash content will be increased materially by the addition of iron from the sampling appa- ratus, even though hardened iron or steel is used. Crushing is continued until all the sample passes through a 4-mesh sieve; the sam- ple is then mixed and reduced on a large riffle sampler to not less than 5 pounds. The 5-pound sample is erushed further by means of a hard- steel roll crusher to pass a 20-mesh sieve, mixed, and reduced on a small riffle sampler to 200 grams. The 200-gram portion is put in the por- celain jar of a ball mill and ground to 60-mesh as described on page 2.6 METHODS OF ANALYZING COAL AND COKE IE a ball mill is not available for fine grind- ing, the 200-gram sample of 20-mesh coke may be pulverized to pass @ 60-mesh sieve by means of a hard-steel roll crusher, in which the rolls, revolve at the same speed, or by impact in a hard-steel diamond mortar, ‘The use of rub- bing surfaces, such as a disk pulverizer or @ bucking board, is never permissible for grind- ing coke, Moisture is determined at 105° C. with 1-gram pottions of the mesh sumple as deseribed be- low. ‘The analysis of the 60-mesh coke is cal- culated to the dry basis by dividing each result by 1 minus its content of moisture. The analy- sis of the coke “as received” is computed from the “dry-coke” analysis by multiplying by 1 minus the total moisture found in the large sample of lump coke, METHODS OF ANALYSIS MOISTURE A L-gram sample of the 60-mesh coal is placed ina weighed 7%-1nch by 13¢-inch porcelain cap- sule (fig. 3, a) and heated for 1 hour at 105° C. in a constant-temperature oven (figs. 8 and 4). The capsule is then removed from the oven, covered with a special flat aluminum cover (fig. 3, 6), and cooled in a desiceator over concen- trated sulfurie acid. The loss in weight multi- plied by 100 is recorded as the percentage of moisture, ‘To determine the moisture content of the 20- mesh coal, a 5-gram sample is heated for 1% hours; the procedure is otherwise the same as for the 60-mesh sample, The oven used is a double-walled copper cylinder closed at one end and having a double- walled door at the other (fig. 3). The space be- tween the outer and the inner walls is filled with Frovrr 8—MOISTURE OVEN.‘METHODS OF ANALYSIS 7 FRONT ELEVATION SLIDING SHELF FOR CAPSULES ‘Piovne 4 DETAILS OF MOISTURE OVEN. 21:1 solution of Prestone (ethylene glycol) in rater the proportions being soadjusted thatthe boiling solution maintains a temperature of 105° C. in the interior of the oven. There is free circulation space for the Prestone solution on the sides and at one end of the oven. A re- turn condenser (fig. 3, ¢) keeps the concentra- tion constant. A solution of glycerin in water (specific gravity, 1.19 at 15° C.) was former! used, ‘The ethilene glycol solution has the ad. vantage of being Jess corrosive.” A current of air, dried by passing through sulfuric acid (fig. 3, d) and preheated by passing through a copper tube eneireling the oven between the walls, is forced through the inner chamber of the oven and over the samples; it takes w moisture from the samples and eseapes throng usinall opening in the door of the oven. The air is forced through the oven at a high enough rate to change the total volume of air in the oven two to four times in 1 minute. Maintai ing a constant temperature of 105° C. in this oven is not difficult. ‘The oven described is available on the market, It was formerly used by the Bureau but, because of the large number ‘of moisture determinations that now are made daily, ovens of the same principle but designed to handle large numbers of samples are used. For coke the circulation of air dried by sul- furic acid through the oven atmosphere, as specified for coal, is unnecessary ; there was no measurable difference in results between cireu- lating perfectly dry air through the oven and using in the oven the natural circulation of air from the room. An ordinary drying oven with openings for natural circulation of air may be used. ASH Ash is determined in the residue of dried coal from the moisture determination. Th porcelain capsule containing this residue is placed in a cold mule furnace and heated grad- ually so that the temperature reaches 500° C. in Lhour and 700° to 750° C. in 2 hours. ‘The ignition in the muffle is continued at 700° to750° C. until all particles of carbon have disap- peared; the ash is stirred occasionally during the heating period. The capsule with its con- tents is then taken from the mule, cooled in a desiccator, and weighed. After ignition it is replaced in the mutlle, heated for one-half hour, cooled in a desiccator, and weighed again. If the change in weight is less than 0.0005 gram, the weight is considered constant, and the weight of the capsule is deducted from the last weighing. If the change is greater than this, the ash is again ignited for 80 minutes and the process repeated until the variation in weight between two successive ignitions is 0.0005 gram8 METHODS OF ANALYZING COAL AND COKE or less, Coals of high iron content may present some difficulty in ignition to constant weight because of the oxidation and reduction of iron. oxides. Ash as determined by this method rep- resents ignited mineral matter remaining after the combustible substances of the coal are burned. Slow heating of the coal as described above prevents difficulty in obtaining satisfactory check results for coals that may contain con- siderable calcite and pyrite owing to varying amounts of sulfur retained as calcium sulfate in the ash. Slow heating also avoids coking the sample, which makes burning difficult; fur- thermore, if a coal high in volatile matter is heated rapidly, the gas generated tends to ex- plode within the capsule and thus mechanically carry off portions of the sample, An air aspi ator installed at the back of the muflle main- tains an ample supply of air in the mufile at all times to insure complete oxidation of the pyritie sulfur and provides circulation of air through the muffle to remove the SO, formed. VOLATILE MATTER A 1-gram sample of the fine (60-mesh) coal is weighed into a 10-ml, platinum erucible (fig. 5, a) with a close-fitting eapsule cover (fig. 5, 3). The crucible and contents are heated at Let Lf Pe il Le, ae al Pioue 5. — yan Litem PuatiNOa Giucient: Fon De TuMisind "vor if Ware 950° C. (220° C.) in a specially designed electric furnace of the vertical type, (See fig, 6.) After heating exactly 7 minutes the crucible is removed from the furnace, cooled rapidly, and weighed. Rapid cooling of the 1s Regs, 0. W., and Seleig, W. A, Determiation of Ash in aly High in Calelte and Pyrite: Ind eng, Chem, vols 14, 1042, pp. 200" 212, ee crucible to room temperature is accomplished by placing it on an air-cooled steel plate. The loss in weight minus the weight of moisture determined at 105° C. times 100 equals the per- centage of volatile matter. MODIFIED METHOD FOR SUBBITUMINOUS COAL, LIGNITE, EAT, AND CERTAIN COKES, CHARS, ANTHRACITES, ‘AND SEMIANTHRACITES Mechanical losses are incurred when peat, lignite, and subbituminons coal are heated sud denly;’ such losses also oceur with some low- temperature cokes, green cokes, chars, anthra- cites, and semianthracites. ‘This mechanical loss is usually designated as “sparking” and is caused by particles of the fuel being ejected from the erueible by the too rapid escape of steam or volatile maitter.® ‘These particles be- come incandescent in the flame of the burnin; volatile constituents and may be seen aroun the edge of the crucible cover, sometimes only one-fourth ineh above the crucible and at other times shooting several inches to the top of the furnace. In severe sparking ash deposits and occasionally unburned material are found on the crucible cover. Small deposits of ash are sometimes found on the crucible cover with moderately sparking fuels, All fuels that do not cake when volatile matter is determined are watched closely for sparking during the heating period; moreover, at the end of the test. the crucible cover is inspected for ash deposits, and any found are considered as evidence of sparking. ‘All fuels that spark appreciably when vol- atile matter is determined by the prescribed method are given a preliminary gradual heat- ing of not less than 5 minutes or more than 10 minutes at increasing temperatures. The crucible is gradually lowered into the furnace at, such a rate that sparking does not occur; after the preliminary heating, it is lowered into its regular position in the furnace and heated for exactly 6 minutes at 950° C, If sparking is then ‘observed, the determination is rejected and the test repeated until no spark- ing occurs either during the preliminary heat ing or during the 6-minute period at 950° C. ‘With some strongly eaking low- and medium- volatile bituminous coals the coke button may be broken with explosive violence because of liberation of volatile matter within the button, ‘This is designated as popping. It may blow the lid off the crucible and cause mechanical W., Experimental Work Conducted in the Chem: ony of {ng United, States Buel Testing bane at Mot Geol Survey ul. 333 (reprinted ea Bureat oF alines ‘ig, 28) 1041, pp ot Cooper, HAC, Gstood, BPD ind Solomon, R. B., The De termitation of Wolatie atstteg. ‘inom Temperatuce Cokes, nd Noncating (Coals Ibureau ot Mines Hept. of i! Seatigations S108, 1982, 11 vp.