State Geologic nn IIM 111 i nun ILLINOIS STATE GEOLOGICAL SURVEY 3 3051 00000 1754 Digitized by the Internet Archive in 2012 with funding from University of Illinois Urbana-Champaign http://archive.org/details/cokefromillinois64thie STATE OF ILLINOIS HENRY HORNER, Governor DEPARTMENT OF REGISTRATION AND EDUCATION JOHN J. HALLIHAN, Director DIVISION OF THE STATE GEOLOGICAL SURVEY M. M. LEIGHTON, Chief URBANA BULLETIN No. 64 Coke From Illinois Coals GILBERT THIESSEN ASSOCIATE CHEMIST IN CHARGE OF FUEL CHEMISTRY WITH THE COLLABORATION OF WALTER H. VOSKUIL, MINERAL ECONOMIST, AND PAUL E. GROTTS, ASSISTANT IN GEOCHEMISTRY CONTRIBUTION OF THE SECTION OF GEOCHEMISTRY FRANK II . REED, Chief Chemist PRINTED BY AUTHORITY OF THE STATE OF ILLINOIS URBANA, ILLINOIS 1937 STATE OF ILLINOIS Hon. Henry Horner, Governor DEPARTMENT' OF REGISTRATION AND EDUCATION Hon. John J. Hallihan, Director Springfield BOARD OF NATURAL RESOURCES AND CONSERVATION Hon. John J. Hallihan, Chairman Edson S. Bastin, Ph.D., Geology William A. Noyes, Ph.D., LL.D., Chem.D., D.Sc, Chemistry John W. Alvord, C.E., Engineering William Trelease, D.Sc, LL.D., Biology Henry C. Cowles, Ph.D., D.Sc, Forestry Arthur Cutts Willard, D.Engr., LL.D., President of the University of Illinois STATE GEOLOGICAL SURVEY DIVISION Urban a M. M. Leighton, Ph.D., Chief Enid Townley, M.S., Assistant to the Chief GEOLOGICAL RESOURCES Coal G. H. Cady, Ph.D., Senior Geologist L. C. McCabe, Ph.D. James M. Schopf, Ph.D GEOCHEMISTRY Chief Chemist Earle F. Taylor, M.S. Charles C. Boley, B.S. Non-Fuels J. E. Lamar, B.S. H. B. Willman, Ph.D. Oil and Gas A. H. Bell, Ph.D. G. V. Cohee, Ph.D. Frederick Squires, B.S. James L. Carlton, B.S. Areal and Engineering Geology George E. Ekblaw, Ph.D. Victor N. Fischer, M.S. Subsurface Geology L. E. Workman, M.S. J. Norman Payne, M.A. E. A. Atherton, Ph.D. Donald G. Sutton, M.S. Stratigraphy and Paleontology J. Marvin W 7 eller, Ph.D. W. A. Newton, M.S. Metrography Ralph E. Grim, Ph.D. Richards A. Rowland, Geol. Physics R. J. Piersol, Ph.D. M. C. Watson, Ph.D. Dona i. d O. Holland, Frank H. Reed, Ph.D W. F. Bradley, Ph.D. G. C. Finger, M.S. Mary C. Neill, M.S. M.S. Fuels G. R. Yohe, Ph.D. P. E. Grotts, B.S. Non-Fuels J. S. Machin, Ph.D. F. V. Tooley, M.S. Analytical O. W. Rees, Ph.D. Norman H. Nachtrieb, B.S. George W. Land, B. Ed. P. W. Henline, B.S. MINERAL ECONOMICS W. H. Voskuil, Ph.D. Mineral Economist Grace N. Oliver, A.B. EDUCATIONAL EXTENSION Don L. Carroll, B.S. PUBLICATIONS AND RECORDS George E. Ekblaw, Ph.D. Dorothy Rose, B.S. Alma R. Sweeny, A.B. Meredith M. Calkins Consultants: Ceramics, Cullen Warner Parmelee, M.S., D.Sc, Pleistocene Invertebrate Paleontology, Frank Collins Baker, B.S Topographic Mapping in cooperation with the United States Geological Survey University of Illinois; Universitv of Illinois. (327'AS) Sep Contents PART I— INTRODUCTION Chapter I Page Purposes of the report 13 Scope of the report 15 Acknowledgments 16 PART II— THE COKE MARKET AND COMPETITIVE FUELS IN THE ILLINOIS COAL MARKET AREA Chapter II — The economic basis of a coking industry in Illinois 17 Geographic limits of the Illinois coal market area 17 Nature of the fuel market 18 Immediate outlook for a coke industry based upon Illinois coal IS The metallurgical market IS The domestic market 18 Fuel oil 19 Natural gas 20 Briquets 20 Factors favorable to the use of Illinois coal for making coke 21 Disposal of by-products 22 Development of new markets for coke and its by-products 22 Chapter III — The production, disposal, and uses of coke 23 Production and distribution by uses of coke in the United States 23 Production of coke in the Illinois coal market area 23 Chapter IV — The use of coke as a domestic fuel 37 Management of a coke fire , 39 PART III— TECHNOLOGY OF COKE MANUFACTURE Chapter V — Reasons for the manufacture of coke 41 Chapter VI — Methods of producing coke 43 High- versus low-temperature carbonization 46 Chapter VII — Historical survey of the production of coke from Illinois coals 49 Introduction 49 Use of beehive type ovens for coking Illinois coals 49 Use of by-product ovens for coking Illinois coals 52 Application of low- and medium-temperature carbonization processes to Illinois coals . . 54 Parr process * 54 Green-Laucks process 55 Other processes 55 Small scale carbonization tests on Illinois coals 55 Conclusions from the historical review of the art of producing coke from Illinois coals. . 57 Chapter VIII — The requirements for coking coals 59 Criteria for coking coals 59 Standard specifications for gas and coking coals 60 [3] Page Sampling and analysis 60 Chemical and physical properties - 60 A. Special requirements for gas coals 61 B. Special requirements for coking coals 61 Classification of Illinois coals with reference to coke-making properties on the basis of their proximate and ultimate analyses. ....":. 73 Rank of coal 73 Ultimate items 77 Specific volatile index 82 ^Conclusions from the application of various criteria for coking coals to Illinois coals. ... S3 PART IV— EXPERIMENTAL INVESTIGATIONS OF THE PRODUCTION OF COKE FROM ILLINOIS COALS Chapter IX — Introduction to the experimental work 85 Chapter X — Methods used in the experimental work 87 Selection of methods 87 Chemical analyses -•...'.." 87 Tests indicating the behavior of coal under the influence of heat 88 Gas, coke, and by-product yields 89 Softening and swelling characteristics 90 Agglutinating or caking power 92 Coke quality tests: retort for carbonizing two- to three-kilogram charges of coal. ... 92 Coke shatter tests 93 Chapter XI — Samples used in the investigation 95 Chapter XII — Coke, gas, and by-product yields 99 Variation in yields with geographic location 99 Comparison of results of tests by several laboratories on No. 6 coal from or near Orient Mine 101 Comparison of results on coals from other states 103 Chapter XIII — Behavior of Illinois coals under heat as shown by the results of experi- mental work 109 Softening temperatures and temperature range of plasticity. .-. 109 Significance of the plastic range temperature Ill Softening temperature and plastic temperature intervals obtained from Illinois coals. ... Ill Agglutinating or caking strength values found for Illinois coals Ill Characteristics of the coal which affect its agglutinating value 113 Mineral impurities 113 Freshness 113 Fusain content 113 Conclusions 114 Chapter XIV — Effect of the banded components and impurities of coals on coke structure . 115 Banded components 115 Carbonizing properties 116 Composition of Illinois coals in terms of banded components 116 Impurities 117 Moisture 117 Ash or mineral matter 118 Sulfur • ... 122 Occurrence in Illinois coals 123 Removal from Illinois coals 123 [4] Page . Behavior of coal sulfur during coal carbonization 124 Studies on face samples 126 Sulfur in coke and total sulfur in coal 127 Sulfur in coke (on basis of original coal) and total sulfur in coal 127 Sulfur in coke (on basis of original coal) and pyritic sulfur in coal 130 Coal components 133 Effect of weathering 133 Conclusions regarding behavior of sulfur during coal carbonization 135 Sulfur behavior in the United States Steel Corporation carbonization 135 Removal of sulfur from coke 137 General conclusions 137 Chapter XV — Coke from Illinois coals and blends; small scale tests 141 Tests in six-inch retort 141 Coals used in the tests . . 141 Results of tests in six-inch retort 143 Effect of coal used . 143 Effect of the size of coal as coked '..".- 143 Effect of the addition of non-caking carbonaceous material 150 Effect of low volatile coal : T f 150 Effect of the addition of tar or petroleum oil ..,?.... 150 Influence of the rate of heating and final temperature 151 Effect of added moisture 151 Effect of removal of mineral matter , 151 Box tests in Knowles sole-flue ovens 166 Results of box tests in Knowles ovens 166 General conclusions on the use of Illinois coals for making coke 170 Chapter XVI — An industrial coking plant utilizing Illinois coal 179 Operation of Knowles sole-flue ovens 181 Objectives of the investigation .:.'.". 183 Temperature measuring equipment 184 Procedure for temperature measurements , 185 Sampling and testing of coal, coke, and gas 186 Results of temperature measurements -.-..' 187 Temperatures in charge 187 Relation of time and temperature 191 Travel of the plastic zone 192 Temperature in flues, regenerators, and stack 195 Analyses of coal coked and of coke produced 196 Influence of oven conditions on coke character 197 Volatile matter gradient in coke made during 10j^ hours coking time 202 Gas analyses „ 202 PART V— CONCLUSIONS AND RECOMMENDATIONS Chapter XVII .....' 205 Scope of the report 205 Conclusions 206 Recommendations for additional work 207 [5] APPENDICES Page Appendix A — Description of the by-product analysis method as applied to Illinois coals. . . 209 Method of the coal carbonization assay 209 Character of the assay 209 Preparation of the sample 212 Apparatus and procedure 213 The furnace 213 The distillation tube and its preparation 214 The tar piece 214 The tar filter 214 The absorption train 215 The freezing tube 215 The gasometer 216 Manipulation 216 Heating the combustion train 216 Disconnection of the train at the end of the test 217 Analytical procedure 217 Separation and estimation of naphthalene 217 Determination of free ammonia 217 Determination of combined ammonia 218 Determination of hydrogen sulfide 219 Calculation of results 219 Standardization of the assay 222 Appendix B — Description of the Agde-Damm plastic range test as applied to Illinois coals. . 225 Method 225 Apparatus and procedure 225 Corrections and interpretation 225 Appendix C— The test for agglutinating value as applied to Illinois coals 227 Method 227 Apparatus 227 Procedure ^ 228 [«] Illustrations Figure Page 1. Bituminous coal consumed for beehive and by-product coke manufacture in the • United States, 1917 to 1936 15 2. Distribution, of by-product and beehive coke sold for furnace, foundry, other in- dustrial, and domestic use, 1918-1936 25 3. Number and location of by-product coke ovens in the Illinois coal market area in 1925 . 26 4. Number and location of by-product coke ovens in the Illinois coal market area in 1930 . 26 5. Number and location of by-product coke ovens in the Illinois coal market area in 1935 . 26 6. Coke produced, and domestic and total sales for Illinois, 1918-1936 ^ 7. Cut-away block model of a modern by-product coke oven 44 8. An installation of by-product coke ovens at South Chicago 45 9. Relative sizes of coal carbonizing equipment 46 10. Coke produced in Illinois, 1880-1936 50 1 1. General view of abandoned beehive coke ovens near Sparta, Randolph County, Illi- nois 51 12. Abandoned ovens at Sparta: View of oven door and arch with part of stone facing removed ^>2 13. County average carbon ratios for coals Nos. 1, 4, and 7, and Assumption coal in Christian County 65 14. County average carbon ratios for coal No. 2 : . 66 15. County average carbon ratios for coal No. 5 67 16. County average carbon ratios for coal No. 6 68 17. Coal bed numbers and county average moisture ("as received") of Illinois Coals 69 18. Coal bed numbers and county average ash (dry basis) of Illinois coals 70 19. Coal bed numbers and county average total sulfur (dry basis) of Illinois Coals. 71 20. Map showing location of low-sulfur coal in the Franklin-Williamson and Big Muddy districts 72 21 . Classification chart of typical coals of the United States 74 22. Classification chart of Illinois coals with the coal bed numbers indicated 75 23. Classification chart of Illinois coals with the location of coals indicated 7 6 24. Division of State into northern, central, and southern areas 77 25. Three-component diagram of ultimate analyses of United States coals 78 26. Three-component diagram of ultimate analyses of 150 coking coals of the United States : 78 27. Three-component diagram of Illinois coal ultimate analyses 79 28. Three-component diagram of Illinois coal ultimate analyses with the coal bed num- bers indicated 80 29. Three-component diagram of Illinois coal ultimate analyses with the location of coals indicated 80 30. Mining districts or coal areas in the United States which contain coking coal 81 31. Coal fields of the United States 81 32. Classification of Illinois coals on the basis of specific volatile index 82 [7] Figure Page 33. Classification of Illinois coals on the basis of specific volatile index with the coal bed numbers indicated 83 34. Agde-Damm apparatus 90 35. Apparatus for determining volatile matter loss 91 36. Location of the mines from which samples were taken, and the tests which were ap- plied 96 37. Location of mines from which samples were tested for coke, gas, and by-product yields 100 38. Gas, coke, and rank indices along the line A-A' (Fig. 37) 101 39. Gas, coke, and rank indices along the line B-B' (Fig. 37) 102 40. Gas, coke, and rank indices along the line C-C (Fig. 37) 103 41. Agde-Damm and volatile matter evolution curves for fresh No. 2 coal 110 42. Agde-Damm and volatile matter evolution curves for weathered No. 2 coal 110 43. Coke made from raw and from washed screenings 120 44. Relation of sulfur content of coke to sulfur content of coal 125 45. Sulfur content of coke, expressed on basis of original coal, as function of total sulfur content of coal 125 46. Sulfur content of coke, expressed on basis of original coal, as function of pyritic and organic sulfur contents of coal 126 47. Cokes from Franklin County, Illinois, No. 6 coal, made in six-inch retort 152 48. Cokes from Franklin County, Illinois, No. 6 coal, made in six-inch retort 154 49. Cokes from Franklin and Randolph counties, Illinois, No. 6 coal, made in six-inch retort 156 50. Cokes from Randolph County, Illinois, No. 6 coal, made in six-inch retort 158 51. Cokes from Randolph County, Illinois, No. 6 coal, made in six-inch retort 160 52. Cokes from Saline County, Illinois, No. 5 coal, made in six-inch retort 162 53. Cokes from various coals, made in six-inch retort 164 54. Metal basket used in coking tests 166 55. Basket test coke and plant-run coke 171 56. Cokes made in basket tests 172 57. Rose cross-sections, natural size, of cokes made in basket tests 174 58. Rose cross-sections at a magnification of 7 diameters 175 59. Rose cross-sections at a magnification of 7 diameters 176 60. Rose cross-sections at a magnification of 7 diameters 177 61. Knowles ovens at West Frankfort, Illinois, in 1936 . 180 62. Construction details of Knowles sole-flue oven 182 63. Coke side of Knowles sole-flue ovens 183 64. Coke side of Knowles ovens with the quenching car in position 184 65. Thermocouple well assembly 185 66. Temperature-time relations at various points in the sole-flue oven charge 186 67. Isothermal relations through the sole-flue oven charge 190 68. Isochronal curves of temperature distribution through the sole-flue oven charge. . . . 190 69. Coke taken from the end of the thermocouple assembly 191 70. Method of determining position and magnitude of maximum heat gradient from an isochronal curve 193 71. Travel of plastic zone and position of maximum heat gradient 194 72. Relation of heat gradient and volatile matter gradient at completion of coking 195 73. Rose longitudinal section, natural size, of sole-flue oven coke 198 74. Rose cross-sections, natural size, of sole-flue oven coke 199 75. Stalagmite carbon on surface of coke made in sole-flue oven 200 76. Hair carbon on surface of coke made in sole-flue oven 201 I 8] Appendix Figure Page I. Travel of heat and evolution of fixed gas in a carbonization assay test (Carbonization No. 44) 210 II. Carbonization assay equipment 211 III. Specimens of silica, coal, and coke 212 IV. Distillation tube, tar piece, and tar filter for carbonization assay 213 V. Freezing tube for the collection of light oil 215 VI. Apparatus for filtering pentane-naphthalene solution 218 [9] Tables Page 1. Quantities of bituminous coal consumed in the United States for beehive and by- product coke manufacture, by the public electric utilities, and total consumption, 1917-1936 , 1-1 2. By-product and beehive coke produced and sold, or used by the producer in the United States, 1918-1936 24 3. Percentage distribution of by-product and beehive coke sold or used by the producer in the United States, 1918-1936 27 4. Summary of ovens in operation in Illinois coal market area, 1925, 1930, and 1935. ... 28 5. Quantity and source of coal used in the manufacture of by-product coke in the Illi- nois coal market area, in 1925, 1930, and 1935 ^2. 6. Disposal of by-product coke produced by states in the Illinois coal market area in 1925, 1930, and 1935 34 7. Distribution by uses of by-product coke produced and sold or used by the producer in Illinois, in 1918-1936 35 8. Results of tests of various fuels in house-heating boilers S8 9. Relative total costs of heating typical Columbus, Ohio, house with various fuels 38 10. Average efficiencies of use of various fuels in house-heating furnaces, Columbus, Ohio. 39 11. County average analytical values for Illinois coals 62 12. Tabulation of tests applied to the Illinois coals investigated and location of the re- sults in the text 95 13. Results of carbonization assay tests on Illinois coals 104 14. Comparison of gas, coke, and by-product yields obtained in various tests on Orient Coal, No. 6 bed, West Frankfort, Franklin County, Illinois 108 15. Plastic range temperatures and agglutinating values for Illinois and other coals 112 16. Agglutinating values of various coals and blends of coal and coke or of coal and de- dusting plant dust 113 17. Segregation of larger impurities in smaller sizes of coke produced by shatter breakage. 119 18. Variations in ash content with size in naturally produced screenings from Illinois coal. 121 19. Sulfur data for face samples of Illinois coals 128 20. Comparison of determined and calculated coke sulfur contents 131 21. Sulfur data for component bands of Illinois coals 134 22. Sulfur data for samples of weathered Illinois coals 136 23. Relationship between distribution of sulfur varieties in coal and sulfur contents of cokes made in carbonization assay tests 138 24. Analyses of coals tested or used for blends 142 25. Summary of coke quality tests 144 26. Size distribution of cokes as recovered from the retort and results of shatter tests .... 146 27. Analyses of coals used in basket tests in Knowles ovens, with analyses of cokes, coke breeze, and results of shatter tests 167 28. Temperatures in charge of Illinois coal coked in oven 9, Knowles oven installation of Radiant Fuel Corporation, West Frankfort, Illinois 188 [11] Page 29. Location of points of maximum heat gradient at hourly intervals after charging, and heat gradient over distance one-fourth inch each side of that point 193 30. Analyses of coal coked and of coke produced during test (dry basis) 196 31. Analyses of coal coked and of coke produced at various times in the sole-flue ovens (dry basis) 196 32. Analyses of cross sections of coke specimens made during test 197 33. Analyses of cross sections of coke specimens taken from new ovens 202 34. Analyses of coke oven gas produced in sole-flue ovens 203 Appendix A I. Carbonization assay test No. 44 211 II. Data for comparative assays 222 [12] Coke from Illinois Goals Gilbert Thiessen with the collaboration of w. h. voskuil and paul e. grotts Part 1 — Introduction CHAPTER I PURPOSES OF THE REPORT THE manufacture of by-product and beehive coke has furnished a market for almost fifteen per cent of the bituminous coal mined in the last decade in the United States. This quantity approaches that used for locomotive fuel and is considerably larger than that used by public electric utilities 1 (Table 1 and fig. 1 ) . Illinois coals, however, have found a very small outlet in this major market provided by the coke manufacturing industry. The Illinois State Geo- logical Survey in its program of work designed to assist the mineral industries of Illinois undertook certain studies of Illinois coals in an effort to determine whether and in what way these coals could better participate in the coking coal market. This bulletin reports the results of certain of those investigations. The purposes for which this report has been prepared are : ( 1 ) To consider the possible market for coke in the Illinois coal market area and the competitive position of coke from Illinois coal in that market. (2) To review critically the present state of the art of coking Illinois coals, and to show the position occupied by Illinois coals in the various schemes which have been set up to evaluate coals for coke-making purposes. (3) To present the results of experimental investigations designed to furnish fundamental data concerning the coking properties of Illinois coals, the effects of impurities, the influence of test conditions, and the results of blending coals. (4) To review the operation of a commercial installation for coking Illinois coals, (5) To present a summary concerning the coke which may be made from Illinois coals, its qualities and possible markets. i U. S. Bureau of Mines, Minerals Yearbook 1936. [13] 1 1 INTRODUCTION MS D H a -r! In D X «*1 s w M c U H U J a o »; Ph vo >H m pq t-H c hN * T-H «u ON w > £ X O w H a. pq s r* D g fc O U H -f j H < C/3 ft Q W h h a ^ 2 »— ' < w CO a UJ H H fc hJ 1-1 H Q t-J w 2 CO H ^ U C w u < u c U 03 CO C3 Cm - SZ I S H D H >- h Total consumption thousands of tons © On tO On ■f "CO to On to CO rt< to -+ t-h on ©On 00 On On © t-h t-H t-H 00 W) © 00 © CN tO On IO 00 00 to On O t- OO CO ^ co NO t-H -HH tHH ON oo On t- © ^ cn OOrHOO CN co OO © VO iO ^ to t-h no 00 -r 1 On On cn t-h 00 On CO "* to -* t* CN On 00 On -t 1 co On On t-h io tHOtHNON W O IN t)( O TH CO co co co co t^ en a 3 _cj CJ CJ 3 Xi o u a, PQ Per cent of total consumption O On -+ r-~ CO NO IO tH tH^ ©t- t-H 00 ** NO t-h O 00 ON CO NO NO -H 00 ^ On © © t-h rs] (-M -fH -fH -HH CN © <"N i~^l CO to Thousands of tons NO 00 t— to O NO to o iO 00 00 CN CO CO NO t-H © t-H IO t^- O t-H t— © NO On t-h "+ -t NO IO CN NO CN t-H t-~ IO NO J^ t-H CO NO NO 00 00 O co © On t-h no to *?»< CO co co ^ XrHrfONt^ CN -HH io rJH to CO co © NO iO NO NO f~- f~- NO O © 00 -f On co -r co co -+ -^ o u _^ o u o _> IS pq Per cent of total consumption On t-h cn CO <^J t-h 00 co iO NO cn t-h on On tO CO IO tO - H t-» On On nO no CN co to CO co rOr-MH Thousands of tons NOO« -t O co OO CN t-h r^ On to NO ^f <* co J- M 00 tH ^l t+i rsi O On -t 1 IO 00 00 00 ^ CN © t-h cs oo CN CN © © CN r-. © t-h io On co nO co NO co O On r-~ © -* NO -it OO CN 00 On t-h tO tHH CN CO oo co © to r~~ t-H CO t-H T-H On t-h t-. o ^ T-H T-H T-H T-H T-H CN 5 V 1- OO On O On On On On t-h cn co ^+ io cn cn cn cn cn On On On On On no t-» oo On © r^ rv\ r^i CN ro On On On On On T-H CN CO Th tO NO <*0 co co CO CO CO On On On O^ s - On I* . o o o a! rt +-> SCOPE OF THE REPORT 15 90 80 70 60 50 40 30 20 A / \ -i \ / .*/- 1 // ^1i ° h- \ / ^ \\ \ \ 1 ~jl N ^ / > / O/ / ^/ // o/ ^ \ \ \ // o/ /// \ \ / ^ /// // 4/ // \ i \ 7 -X/ col / \ \ v// \ ^' \ \ / \ \ £- \ \ La / \ c 0/ / \ \ *r* \ \ \ y \ \ \ . s •' s \ \ \ > K 1 •^ -~" 1917 '18 '19 '20 '21 '22 '23 '24 '25 '26 '27 '28 '29 '30 '31 '32 '33 '34 '35 '36 Fig. 1. — Bituminous Coal Consumed for Beehive and By-product Coke Manufacture in the United States, 1917-1936. SCOPE OF THE REPORT This report presents a discussion of the economic position of the coke industry in Illinois, the place occupied hy Illinois coal in the various schemes for classifying coal for coking purposes, a hrief review of the art of coke manufacture, a history of coke manufacture in Illinois, and a statistical review of the industry, as well as the results of a variety of investigations undertaken in the laboratories of the Illinois Geological Survey. This report is not to be considered as a complete monograph on the manu- facture of coke nor as containing all that is desirable or necessary to know about the coking of Illinois coals. Some of the material has been previously published as individual articles. 2 2 Thiessen, G., Behavior of sulfur during coal carbonization: Incl. and Kng. Chem. vol-. 27, pp. 473-478, April, 1935. Thiessen, G., Coke from Illinois coal, temperature conditions in sole-flue ovens: Ind. and Eng\ Chem. vol. 29, pp. 506-513, May, l!t:;7. 16 INTRODUCTION ACKNOWLEDGMENTS The author is under great obligation to his associates on the staff of the Illinois State Geological Survey and to many others who have assisted in or made possible much of the work which this bulletin reports, and he wishes to take this opportunity to express his appreciation. To Dr. M. M. Leighton, Chief of the Illinois State Geological Survey, the author expresses his sincere appreciation for the active interest and support received during the entire progress of this undertaking. This work was one of the early projects for which he sought the establishment of the Survey's new laboratories. This bulletin is a report of work carried on in the Geochemical Section of the State Geological Survey, of which Dr. Frank H. Reed is Chief Chemist and to whom the author is especially indebted for guidance and suggestions during the course of the work and during the preparation of the manuscript. Much assistance and many helpful suggestions during the course of the work have come from other associates on the Survey staff, especially Dr. G. H. Cady, Dr. O. W. Rees, Dr. Walter Voskuil, Dr. L. C. McCabe, and during the earlier stages of the work from Dr. C. F. Fryling. The samples used in the tests were collected by members of the Coal Division of the Geological Resources Section of the Survey. Much credit is due my assistant, Mr. P. E. Grotts, for his part in carrying out much of the experimental work and in preparing many of the illustrations. Messrs. J. W. Teter, A. L. Ryan, B. Bilder, and J. E. Pentecost assisted in this work during three months in 1934 under a C.W.A. project. Most of the analytical work was performed under the direction of Dr. O. W. Rees by J. W. Robinson, Jr., Carl Westerberg, C. E. Imhoff, L. D. McVicker, G. C. Finger, and W. F. Bradley. Mr. L. D. Vaughan assisted with the photographic work and in many other ways. Mr. A. W. Gotstein was of much assistance in the design and construction of equipment. The author wishes to express his thanks to Mr. E. T. Johnston, Chief Chemist of the Central Laboratories of the American Steel and Wire Company at Joliet, Illinois, for aid furnished in standardizing our procedure in the use of the United States Steel Corporation dry distillation test ; to Dr. A. C. Fieldner, to Mr. J. D. Davis, and Mr. W. A. Selvig of the United States Bureau of Mines for their assistance in connection with the use of many of the test procedures; and to Mr. M. D. Curran, president of the Radiant Fuel Corporation for permission to make and publish the tests on the Knowles oven installation at West Frankfort, Illinois, and to Mr. George Curran, plant superintendent, Mr. F. E. Dodge, chemical engineer, and Mr. C. E. Case for their assistance during the tests. Many others have assisted with suggestions, criticisms, supplying samples and information for which the author expresses his thanks. Part II — The Coke Market and Competitive Fuels in the Illinois Coal Market Area CHAPTER II— THE ECONOMIC BASIS OF A COKING INDUSTRY IN ILLINOIS By Walter H. Voskuil , GEOGRAPHIC LIMITS OF THE ILLINOIS COAL MARKET AREA The Illinois coal market area is defined as the territory included in the states of Illinois, Missouri, Iowa, Minnesota, and Wisconsin, the eastern cities of Kansas and Nebraska, and a small section of the Dakotas. The boundaries of this so defined "Illinois coal market area" are determined by competition from other coal fields and other forms of fuel. Within the area so described 90 per cent of Illinois coal is marketed. In the southwest part of the area fuel oil and natural gas dominate the market almost to the exclusion of coal. The westward movement of Illinois coal in Kansas, Nebraska, and the Dakotas is met by an eastward flow of coals from Colorado, Wyoming, and Montana. In the lake shore counties of Minnesota and Wisconsin the market is dominated by Appalachian coals, cheaply carried over the Great Lakes and reaching the ports of Lake Michigan and the head of Lake Superior. Illinois coals, however, are marketed to a considerable extent in the southern and western sections of these two lake states. Only small quantities of Illinois coals are shipped east of the Illinois line. This territory is occupied almost entirely by the neighboring coal fields of Indiana and the Appalachian coal fields of Ohio, West Virginia, eastern Kentucky, and Pennsylvania. Within this market area itself there are also consumed Appalachian coals, and also fuel oil and natural gas, but bituminous coal is by far the most important source of fuel for industry, railroads, and domestic heating. For further in- formation on this point, the reader is referred to Bulletin No. 63 of this Survey. [17] J 8 COKE MARKET AND COMPETITIVE FUELS NATURE OF THE FUEL MARKET Fuels consumed in the Illinois coal market area find an outlet in the manu- facturing industries, smelting, railroad fuel, public utilities, mining and quarrying, domestic heating, and miscellaneous smaller uses. The miscellaneous uses include the heating of large non-factory buildings, such as hotels, apartments, stores, offices, theatres, garages, and also a number of non-domestic uses, such as water- works, threshing engines, construction work, power laundries, grist mills, and small factories. Altogether the consumption of fuel in this area in a year of high productivity is approximately 102,000,000 tons of coal and its equivalent in liquid and gaseous fuels. The consumption of coke in this area in 1929 was approximately 7,000,000 tons of which 5,000,000 tons were consumed in the manufacturing industries and 2,000,000 tons were used as domestic fuel. Public utilities, mining, and quarrying use practically no coke. In the manufacturing industries, the principal outlet for coke, possibly 80 per cent is used in the blast furnace industry and the remainder is distributed among industries such as commercial baking plants, cast iron pipe manufacture, chemical manufactures, foundry and machine shop products, gas manufacture, smelting and refining of nonferrous metals, and in such metal products industries as engine manufacture, stoves and ranges, steam fittings and plumbers supplies, and motor vehicles. Clay products, lime manu- facture, meat packing, and paint and varnish making also use coke to some extent as a fuel. IMMEDIATE OUTLOOK FOR A COKE INDUSTRY BASED UPON ILLINOIS COAL As has been pointed out, the two largest outlets for coke in the Illinois coal market area are for metallurgical purposes and for domestic heating. The metallurgical market. — Coke for metallurgical use is obtained mainly from eastern coke ovens and from coking coal imported from Kentucky, Pennsylvania, and West Virginia. In 1935, for example, a total of 2,500,000 tons were shipped into the state for coke-making purposes. There is, at present no Illinois coal used in the manufacture of metallurgical coke, although in times past it has been employed to some extent. The domestic market. — The principal hope for an outlet of coke made from Illinois coal lies in an expansion of the domestic market rather than the metallurgical. The annual domestic fuel requirements in the Illinois coal market area are approximately the equivalent of 35,000,000 tons of coal divided among the various fuels as follows: ECONOMIC BASIS OF COKING INDUSTRY 19 Tons of coal or its equivalent Pennsylvania anthracite (1936) 1 ,000 ,000 Fuel briquets (1935) 575 ,000 By-product coke (1929) a 2 ,078 ,000 Oil heating (1935) (19,006,000 bbls.) 4,751,200 Natural gas (1935) (67,770,000,000 cu. ft.) 2 ,591 ,000 Bituminous coal, estimated 24 ,000 ,000 Total 35 ,095 ,000 a The latest year for which detailed coke distribution data are given. The use of coke for domestic heating increased consistently until it reached a peak in 1933 and declined in 1934 and 1935. The Bureau of Mines offers the following explanations for the recent decline i 1 ( 1 ) Greater industrial activity and a greater use of coke by industrial users with the result that less coke reaches the domestic market (2) Increasing installations of domestic and commercial stokers with cheaper coal replacing some coke (3) Increasing competition of natural gas (4) Declining prices of anthracite (5) Increasing prices of raw coal resulting in higher prices of coke (6) Growing competition from cheap trucked coal, both anthracite and bituminous. In the Illinois coal market area, coke occupies a position between the higher priced fuels and the prepared sizes of bituminous coal. For those con- sumers who wish to obtain the advantages of cleanliness, smokelessness, and convenience, coke must compete to some extent with natural gas, anthracite, heating oil, and fuel briquets. For those consumers who are interested mainly in a low priced fuel, coke will find difficulty in competing with prepared sizes of bituminous coal. The market outlet probably must be sought among those who desire smokelessness but will not go to the extent of using gas, heating oil. or anthracite. Shipments of anthracite are confined principally to Chicago and the lake shore counties of Wisconsin, and in 1936 aggregated 1,000,000 tons. The importance of anthracite as a domestic fuel in this area appears to be declining. The increased use of natural gas, fuel oil, and coke are tending to displace anthracite. Fuel oil. — Since 1926, the year for which data are first available, the consumption of fuel oil in the Illinois coal market area has shown an almost constant upward trend. If the consumption of fuel oil is broken down into uses for domestic heating and for all other purposes, an interesting trend is observable. Fuel oil for manufacturing, bunkers, smelting and refining, and transportation has shown no increase since 1926. On the other hand, the use of fuel oil for heating purposes has shown a rapid upward trend and will probably continue i Tonne, W. H., Bennil, H. L., and Plein, L. N., Coke and by-products: U. S. Bureau of Mines Minerals Yearbook, pp. 581-624, 1936. 20 COKE MARKET AND COMPETITIVE FUELS to increase in the immediate future, as indicated by the continued increased rate of sales of oil burners. Consumption of heating oils in the Illinois coal market area in 1935 was approximately 19,000,000 barrels or an equivalent of about 4,750,000 tons of coal. There are, however, certain factors that must be considered in evaluating fuel oil as a domestic fuel in the future. The consumption of fuel oil in this area in 1935 was practically equal to the quantity of furnace oils of the heating grade obtained in the process of refining in the Illinois-Indiana refining area. If the rate of oil burner installation continues it is not unlikely that the demand upon the oil refining industry for heating oils will become so heavy that price rises will result which will tend to discourage further increases or the oil industry will be compelled to meet this demand by increasing runs to stills over the present high level. A substantial rise in fuel oil consumption will necessitate readjust- ments in the refining processes in which the percentage of gasoline recovered is decreased in winter months and the percentage of fuel oil recovered is decreased in the summer months to correspond with the seasonal demands for these two products of oil refining. Natural gas. — The supply of natural gas available to cities in the Illinois coal market area by pipe lines from Texas and Kansas to Chicago, to cities in central Illinois and to cities in Iowa, eastern Nebraska, and southern Missouri, appears to be sufficiently large to last for several decades. Consumption of natural gas in the Illinois coal market area is now equivalent to approximately 2,600,000 tons of coal and will probably increase somewhat in the future. The limiting factor in the use of natural gas is not a shortage of supply but price. Another factor limiting the use of natural gas is the fact that its distribution is limited practically to cities. The high cost of distribution will prevent its use in small communities except near newly discovered sources. Briquets. — At the present time the manufacture of briquets in the Illinois coal market area is governed largely by the amount of screenings which result from the transportation of coal over the Great Lakes with resultant degradation and breakage. In order to dispose of these fine sizes, briquetting plants are in operation in St. Paul, Minnesota; Omaha, Nebraska; Superior, Sheboygan, Ashland and Milwaukee, Wisconsin. There are also some briquetting plants in North Dakota using lignite in an attempt to improve this fuel. Consumption of briquets is at present less than a million tons in this area. The total consumption of anthracite, natural gas, fuel oil, and briquets is approximately equivalent to 8,000,000 tons of coal. An estimated 70 per cent of the domestic fuel consumption consists of bituminous coal, from various sources. Coal used in the domestic fuel market is obtained from Illinois and from several eastern fields, the principal contributing states being West Virginia and eastern Kentucky. There are certain factors which favor the importation of large quantities of eastern coal for use in the domestic fuel market, such as higher ECONOMIC BASIS OF COKING INDUSTRY 21 heating value per ton, cleanliness, and lower ash content, for which the domestic consumer is willing to pay a higher price. The possibility of coke from Illinois coal in replacing these fuels depends upon the ability of a coking industry to supply a fuel which gives the advantages of present solid fuels at a lower price than is now paid for eastern coals or to supply a fuel of improved characteristics over such bituminous coal, for which the householder is willing to pay an added price. FACTORS FAVORABLE TO THE USE OF ILLINOIS COAL FOR MAKING COKE Experience in the coking of Illinois coals indicates that is is economically and technically difficult to produce metallurgical coke, but that a coke which is suitable and desirable for domestic heating and for other nonmetallurgical uses is feasible. Experimental investigation indicates that coking-coal resources of Illinois are confined principally to central and southern Illinois. The two largest domestic fuel markets in the Illinois coal market area and near to these producing fields are St. Louis and Chicago. At present about 85 per cent of the coal shipped into the St. Louis district is obtained from the southern Illinois fields. This market area consumes probably about the equivalent of 1,800,000 to 2,000,000 tons of fuel for domestic heating. The Chicago market consumes approximately an equivalent of 5 to 6 million tons for domestic heating. This area appears to be less favorable for a market for coke from southern Illinois coal because of a higher freight rate than to the St. Louis market and because of greater competition from fuels in the higher price brackets. Natural gas for domestic use is available by pipe- line from the Panhandle of Texas, heating oils in large quantities are available from nearby refineries, and eastern coals are available by lake transportation. At present, coal is available in southern Illinois at a cost of $1.60 to $1.75 per ton at the mine and screenings can sometimes be obtained at a lower price. The freight rate on coal, or coke, from the producing fields to the two principal markets and to the smaller cities in the Illinois coal market area give to the southern Illinois coal district an advantage in delivered cost of fuel over coke or high rank coal from the more distant eastern coal fields. The most difficult problem of developing a coke industry is that of establishing in the minds of the domestic consumer the advantages of coke as a domestic fuel. Coke has burning characteristics that are different from either anthracite or bituminous coal. To introduce coke into a market area, it is necessary to carry on a campaign of educational advertising. Because coke possesses certain properties, it requires special handling as a domestic fuel. It is distinctly more bulky than bituminous coal or anthracite. Some of the difficulties encountered in marketing coke for domestic heating have resulted from attempts to use the larger sizes in furnaces that were too small. 22 COKE MARKET AND COMPETITIVE FUELS The chemical characteristics of coke usually do not concern the domestic uses unless the coke employed has unusually high ash or high sulfur content. Sulfur produces corrosive products of combustion which are highly objectionable. In producing coke care should be taken with regard both to the ash and sulfur content. DISPOSAL OF BY-PRODUCTS In the development of a coking industry, the problem of the disposal of coke-oven gas, tar, light oil, and ammonia must be considered. The principal outlet for gas other than that used as fuel by the plant is in public utility distribu- tion, and as fuel in industry. In southern Illinois, gas could be disposed of in public utility distribution systems for domestic cooking and for such local industries as commercial bakeries, laundries, restaurants, etc. A coke oven located in the St. Louis area would meet competition from natural gas. The marketing of gas from a coke oven plant presents a peculiar problem. It must be used within a few miles from the plant at places which can be reached by gas mains or pumping lines. Moreover, it is very bulky and, therefore, cannot be stored for more than a few hours. In these respects gas differs from other products of the coke oven which not only can be stored but also can be shipped considerable distances before they are used. The disposal of tar, light oils, and ammonia presents no particular problems in this area since they can be shipped over greater distances than the gas. DEVELOPMENT OF NEW MARKETS FOR COKE AND ITS BY-PRODUCTS Although coke for domestic heating offers the largest immediate outlet for coke it is conceivable that in the future other markets may be developed. Particular reference should be made to the production of motor fuel in the event of a declining crude oil supply. For the production of liquid fuels, two processes are now in progress of development, the Bergius process for the liquefaction of coal by hydrogenation of a mixture of pulverized coal and tar under high pressures and temperatures with the aid of a catalyst, and the Fischer-Tropsch process for the gasification of coke and the synthesis of liquid hydrocarbons from gas and hydrogen. The latter process seems to be gaining favor and may provide an outlet for coke to meet future motor fuel requirements. CHAPTER III— THE PRODUCTION, DISPOSAL, AND USES OF COKE By Gilbert Thiessen and Paul E. Grotts PRODUCTION AND DISTRIBUTION BY USES OF COKE IN THE UNITED STATES The production and the quantitative and relative distribution by uses of by-product and beehive coke in the United States since 1918 are shown in Tables 2 and 3 and graphically in figure 2. The continued rise of the domestic coke market and its importance as an outlet for coke during periods of depression are obvious. PRODUCTION OF COKE IN THE ILLINOIS COAL MARKET AREA Much coke has been and is being produced in the Illinois coal market area. 1 Most of this coke is or has been made for metallurgical purposes at plants located close to Great Lakes navigation, or as a by-product in the manufacture of city gas from coals originating outside of the Eastern Interior basin of Illinois, Indiana, and western Kentucky. At the lake steel plants a combination of price and quality factors practically excludes Illinois coals at present. The amount of gas-works coke produced is much smaller than the amount produced primarily for metallurgical purposes, but a greater proportion of it goes into the domestic market. Table 4 gives the number, type, and location of the coke ovens in operation in the Illinois coal market area in 1925, 1930, and 1935. The number and location of the ovens are diagrammatically shown in figures 3, 4, and 5. The origin of the coal coked in these ovens in 1935 as closely as can be estimated from Bureau of Mines statistics 2 is shown in Table 5. i Voskuil, W. H., The competitive position of Illinois coal in the Illinois coal market area: Illinois State Geol. Survey Bull. 63, pp. 10-11, 1936. 2 Young, W. H., Bennit, H. L., and Plein, L. N., Coke and by-products: U. S. Bureau of Mines Minerals Yearbook, pp. 581-624, 1936. [23] 24 COKE MARKET AND COMPETITIVE FUELS p w S H fc 1-1 W u N Q &H C -D OS PL, -a c W a a H (Si a a •§ rt co O ,_ 13 5---' U o ex rt 3.M « . xt >^ rv C . . . I-, rv ro co ^ 2 0\ On On ON P-l ,—l ,-H ,-H ,-H ^— I 00 On O O On O CM CN TH.^H^-I^HfNr^ MfOINtN fOMMiM CM cm CO l^ GO CM HM(N OOOnO --h cm co -f 'O HHfS M r>i M CM <^i OnOnOn OnOnOnOnOn NO I s - 00 On O »— I CM co -f >0 NO ci (N (nj f^i ro f) f5 ^ fO f) f) OnOnOnOnOn OnOnOOnOOn ^ m o <» +-> ^ as ST- to . SiS fl- ed ^ • OT3 OJ >»tj o ri 3 a ri"^ 3S S8 o-H ~c/2 ri ^ O m <^ «>» ."X furnace use. ted prior to 1 oducer" and es of the Un indicated. OOfths. ^ ft Sri CO >0 cp 0) :-, ,q co >, P-H (p $£«tf « 3 ^ ^ orations coke w ' and s reau of linerals ted corp dustrial roduced ates Bu Mines A .S3 .5 a M^ affil ther coke ited au o o o- s £ o ri f»+3 .. ke s oke betw fers tates O O CP W a a _ des esti ren R." itec P c ^ c PRODUCTION AND USES OF COKE 2S FURNACE * FOUNDRY OTHER INDUSTRIAL DOMESTIC cooorjcviro^ttocoi^-coooiirvjroTfinco — — (U<>J(\j(VI(\|(Vi(\i. (\j(\j(Vj~rr) (T >( f )r r )'rQ (*)(*) 100 FURNACE h- FOUNDRY ER DUSTRIAL DOMESTIC Fig. 2. — Distribution of By-product and Beehive Coke Sold for Furnace, Foundry, other Industrial, and Domestic Use, 1918-1936. *Includes coke used by producer in blast furnace. 26 COKE MARKET AND COMPETITIVE FUELS • | *' ^p^ vo o z •to P >' )o S^ (Ts'' O 1^ •2 (1 5 J 1 * ■f^i / «o oi) O * V. • < 2 2 (0 2 2 2 < m Q Q 1 2 to ! 2 — •in 1 jjxy *■ •rv^_ £ JjL W" Proc. A.G.A. 11: 1026 (1929 65 Koppers 196 3 65 Koppers 196 3 Wisconsin 80 Semet-Solvay 100 Koppers 15 Koppers 195 2 56 Koppers. 8 Piette 64 1 80 Semet-Solvay 100 Koppers 15 Koppers 195 2 Missouri 56 Koppers 8 Piette 64 1 Proc. A.G.A. 11: 1027 (1929); 12: 1096 (1930) 30 COKE MARKET AND COMPETITIVE FUELS TABLE 4- City Company Date in- stalled 1925 Battle Creek. . . fackson Michigan Battle Creek Gas Co. (M) Consumers Power Co. (M) Total 1924 1925-30 1 1 Koppers 11 Plants 1 a Based on statistics in U. S. Bureau of Mines Mineral Resources Reports and Minerals Year- books and additional data. b Listed under the name "By Products Coke Corporation" at S. Chicago in 1925. Included with S. Chicago ovens on 1925 map. c Listed under name "International Harvester Co." in 1925. d Several batteries of these ovens have again been put into use in 1936. e Only western half of State considered. (P) "Furnace plant" or plants associated with iron furnaces and steel works. (M) "Non Furnace Plants'" or other plants including merchant plants, plants associated with industries other than iron and steel and plants supplying gas under contract for city use. PRODUCTION AND USES OF COKE CONCLUDED 1930 1935 Source of additional data 18 Koppers 15 Koppers 33 2 Michigan 18 Koppers 26 Koppers 44 2 Proc. A.G.A. 12: 1098 (1930); 13: 943 (1931) 32 COKE MARKET AND COMPETITIVE FUELS Table 5. — Quantity and Source of Coal Used in the Manufacture of By-product Coke in the Illinois Coal Market Area in 1925, 1930, and 1935 a (In tons) Source of coal Illinois Indiana Kentucky. . . . Pennsylvania. Virginia West Virginia Illinois Kentucky. . . . Pennsylvania. West Virginia. Kentucky. . . . Pennsylvania. Virginia West Virginia 325,095 53 1,925,000 35 ,939 41,451 1,971,753 4,299,293 565 ,362 1,312,004 503,591 2,727,597 5,108,554 625,194 466,967 1,382,938 2,475,099 State in which coal was used Indiana Minnesota Wisconsin 13 Missouri 1925 1,923,486 156,834 4,786,851 204,810 246,708 340,474 6,867,171 1930 791,992 2,569,228 382,166 3,948,759 247,507 395,903 294,328 6,900,153 1935 2,181,970 150,637 226,484 2,677,234 5,236,325 937,738 189,029 256,144 184,577 629,750 156,500 954,300 941 ,600 257,000 313,500 309,100 a Data from U. S. Bureau of Mines Mineral Resources and Minerals Yearbook. b Estimated from concealed figures on basis of number of ovens. The distribution by uses of the coke produced in Illinois, Indiana, and Minnesota in the years 1925, 1930, and 1935 are presented in Table 6. The figures for coke produced in other states in the Illinois coal market area are not available. The distribution of coke produced in Illinois for each year since 1918 is presented in Table 7 and figure 6. In figure 6 the difference between total production and total sales is accounted for largely by the coke consumed by the producer and by changes in stocks. As is the case for the United States as a whole, the domestic market has been an important outlet for coke in Illinois, especially in the years 1932-35 when the amount of coke used by the producer fell to very low figures. The statistics presented here do not include the coal coked at the installation of 26 Knowles type sole-flue ovens at West Frankfort, Illinois, the only plant at which coke was made for commercial sale from Illinois coal in 1936. This plant and its operation is described later in this report for the reason that it is a new type of oven and is now operating with Illinois coal. PRODUCTION AND USES OF COKE 33 Up to the summer of 1935 two batteries of 40 ovens each of Roberts type ovens were operated by the St. Louis Coke and Chemical Co. at Granite City. These ovens at times operated on Illinois coal alone, at other times on mixtures Fig. 6. — Coke Produced, Domestic and Total Sales for Illinois, 1918-1936. A-B represents domestic and miscellaneous industrial coke. containing substantial proportions of Illinois coals, and at other times on out-of- state coal alone. The abandonment of this installation leaves the Radiant Fuel Corporation's plant as the only plant making coke entirely from Illinois coal. 34 COKE MARKET AND COMPETITIVE FUELS Q .5 2w 0-3 3 3 rt O •Si 6 s u Ttf L— -tH • vOOONCAroa "^ -CM -NO lO 00 ^ IT) >o O -H cm on th j>- thh no © J>- -00 • Tt C/3 4J c c c c c c O o H U K D U U io o cm ro O- On NHNro LO O O -CM -NO if; N O h h ro ro © On CM NO CO NO IO tHH rHfNOOf^NCN io • ro ■ J>- vOJ^CSNvOOs © cm ro t~- NO f— ro ro fO 00 ro t~» t~~ i— i no On -00 -CM CM O tH O t^ O t~~ ^* CO tH ro CM Ni^NroOrO lo On • f— tHh tHh t-( ro r— CM CM tWh rh fO no 00 CM tH t-H rH J-~ CM o • ON • -HO On O ro NO T3 C hH ONro^r^OOO NO -00 • tH • ro 00 ro NO no NO -00 -00 -00 ON t-H O ro CN !>■ LO ro CM Tf tHh co O O O O O O ■*o on t-i I-. CM NO - 1^ - io - ON CM >0 NO T-i NO T-I •HH fO 't 1 t— NO ON O IO t-I 00 ■ O ro oo t~~ O • t- NO -oo 00 T-H ro O t- o 00 • T-, O O OJ u CD CD O NO no O t~~ CM CM -t-I • -HH io >o r- O ro ro -IO -tHH - OO NO t-H 00 IO t-I CM CM LO O LO O CM O lo O O O 00 o CO tH CM t-H O t-H 00 t-H O t-h tHh ro LO NO tHH C/3 4-1 V) 4J in *-> C C C C C C o CJ 4-1 U 4J U rt ^ Ph(^ o o oj cp PRODUCTION AND USES OF COKE 35 c a ei n b£ a c n c D Q 3 O * o P< 0> w v OB n c U O D u e Q c C a* Ph t* pq CN -^ -H LO en CO CN d '^ O CO ^ t- i^CN^OhN o!.>f ^ H* co co hh i—i oi t"» • lo NO NO t» NO LO f^>T3 >, . C «- a d ex (£, & d< d< d< o* a cu « i— r •— r i— r M " " | r ^^ ^ . ^ NO NO NO \ZZ C/3 M M M M *"! ^ '"l *"! *"! *~> " . . . V 1) Cm Ph Ph Oh On Ph P-iPh P-t Ph %> cx ex a£ , V O fN TjH »o ON00OO iHMHlOOt^ H >H >H* >-I >H s s s s sss's's s"s.ssss ^— / ^-^. K^. ^^^ fNlOtN lo i— i "* N On NO O 00 iO i-i ^^(NMCnIO CO 00 00 ONOO't i— 1 rf CO -H/l 1>- O LO ^H HH ^H NO ~rt ^■^l ° NO 00 00 HH NO co i—i no -H^ r— co 00 -H/i t^ co co rHO\(N ON NO CM CO LO CN NO LO i-H 00 rHl^CANHCO O t^»Ht^ CO i-H CO NO -H/i CO CO OO i-H to i— 1 CO 1— 1 N LO LO H OH OO i^OO»0\H \OMM VOO On i-h i-h ^h ^h co rH ^ ^H ^H ^H ^H ^| ^H ^H 1-1 ^ ^ ^ ^ ,c MOM NO co 00 00 ^H HNOM'IO CJ M00O OOO-HN co ^ i-h CN 00 On HNH 00 Tf rfi -H/l CNI CN lo CO i-i O 00 en V i-H On On co H^i on ^h co co "H/i 00 lo lo On H^ O On CM rjn On 1OM00 LO LO CO 00 ^H CO CO NO 00 no t^ On 00 !>■ O On O On On o Q Hl^lO •<* tjH r-. lo LO ro co OOHN CM n O CO T3 X! CNtNMD lOiO-HfOM Ttl LO NO NO O r— i-i lo ^H t— 00 LO NO ^H O CN "* CN CO N 00 i-H NO On i-i O ^ i-H lo ^3, 'tiOO 00 H ON O i-i co lo co i* > oi co co GONCNOO NO g IONIO ^OlONOMTtt o_ t^ NO On O NCOO 00 no O co On o""^^^^^ -S^"S^t- "* o oo 3 HO^H N lo 00 nO O •H/l S y^ y^ y ^ , w.^nO On lo co o (NCNCS H^rHHM Ol i— I Ol ta GOO O fOna •OiHHNM 00 00 i-H 00 cd lo *» t~~ On 00 n r^ On co CJ LO ^ N N LO co 00 00 LO . 00 f- co 00 i* O ■H/i CO i-t LO OI NO _K CJ co oo io nO^OMM i-H LO 00 NO i-H NO —i On O) i-h Hh (A ~Xj _q rt cn r— oi NO O On co 00 t-~ NO 00 LO H ^l^MHOOO 13 2 5 i-H ,-H i-H i-H rt i-H HHHMM T _ H " ex <+i OfOro 00 NO 00 TfHf- MOHON iH HH O t-~ CO NO T3 iH O On r— O NO N On \Ot|*vOhN OOfOMOtNH cu 1) o OON i-h On i-i -^ Tfi On On On tH lo On co O On lo lo lo co nO MCNNlOrt NO CO O ^ NO 00 00 i— i On 00 CN o-d ooofo CO 00 00 «0 i— I CJHitON NtNCtNOOO sJ o CNt^rH fOCAHfOO CO O CXI oi LO •H/i rjn LO NO NO O (Nrt (N i— i i— i CO CN CO CO co co "Hh co f\)HHr-IHM cu CO ON O rHMfOlflO NO N 00 On O i— 1 Csl CO HfH U-) NO HHCS co co co co co rsj r-1 cv] ex) co CO co co co co co On On On On On On On On On On On On On On On On On On On 2 43 u o o o u - - o _2 r-j •-/. CP H I c B a o ■^ ^5 c O al m 4) W C o o o o — rt w ^3 ri S F "O o o Q ^ CHAPTER IV— THE USE OF COKE AS A DOMESTIC FUEL By Gilbert Thiessen The domestic fuel market for coke from Illinois coal is especially important for these coals because as coke they enter the market as a high quality, economical, smokeless fuel in a region quite far removed from sources of anthracite and smokeless coal. Although coke from Illinois coals may in time play an important part in the iron and steel industry, certain of its characteristics make its present competition difficult with coke from coals of the eastern fields. Coke is definitely a smokeless fuel. The seriousness of atmospheric pollution, especially by coal smoke, is recognized. It has been estimated that the cost of smoke pollution in the larger cities in the soft coal fuel districts amounts to from 10 to 20 dollars or even more per capita per year. 1 The general public is gradually being educated to the damaging effects of atmospheric pollution and already demands that something be done to minimize it in our larger cities in the soft coal districts. Any such abatment action will place a premium on smokeless fuels. Smoke abatement movements and ordinances will be more popular if a smokeless fuel is available which permits the average householder to heat his home as economically and conveniently as with bituminous coal. Gas, oil, and stoker fired coal also provide domestic heat without smoke but results of various tests have indicated that none of these provide domestic heat as cheaply as does coke. United States Bureau of Mines tests have shown that coke is more efficiently utilized in domestic heating furnaces than is bituminous coal. 2 These tests in six different steam and hot water boilers showed that coke was burned as efficiently as anthracite and that the efficiencies obtained in burning Pittsburgh coal were approximately 10 per cent lower and Illinois coal 20 per cent lower than those obtained in burning coke or anthracite. Fuel tests with domestic heating boilers by the University of Illinois Engineering Experiment Station have given similar results with respect to efficiencies of use. 3 Averaged values for results of similar tests have been selected and presented as examples by Kreisinger and Fieldner. 4 These values are presented again in Table 8 for easy reference and comparison. 1 Kreisinger, H., and Fieldner, A. C, Why and how coke should be used for domestic heating: U. S. Bureau of Mines Tech. Paper No. 242, p. 6, 1919. 2 Kreisinger, H., Blizard, J., Jarrett, H. W., and McKitterick, J. J., Comparative tests of by-product coke and other fuels for house-heating boilers: U. S. Bureau of Mines Tech. Paper No. 315, 1923. 3 Snodgrass, J. M., Fuel tests with house-heating boilers: Univ. of Illinois Fng Exp Sta. Bull. 31, 1909. 4 Kreisinger, H., and Fieldner, A. C, Why and how coke should be used for domestic heating: U. S. Bureau of Mines Tech. Paper No. 242, p. 17, 1919. [37] coke market and competitive fuels Table 8. — Results of Tests of Various Fuels in House-Heating Boilers Tests Made at the Engineering Experiment Station of the University of Illinois.' 1 Fuel Efficiency of boiler and furnace per cent Fuel fired at each firing pounds Average interval between firing hours Rated capacity developed per cent By-product coke. .' Gas house coke Anthracite Illinois coal, Williamson Co Pocahontas coal 61.63 56.22 51:93 48.00 46.51 75 75 75 75 75 3.71 3.21 2.47 2.92 3.30 64.46 65.48 66.00 64.04 63.88 a Data from Snodgrass, J. M., Fuel tests with house-heating boilers: Univ. of Illinois Eng. Exp. Sta. Bull. 31, p. 82-83, Table 12, 1919. (Referred to erroneously by Kreisinger and Fieldner as U. of I. Eng. Exp. Station Bull. 19.) Recent tests in Columbus, Ohio, showed that the over-all cost of heating homes under conditions prevailing there at the time the tests were carried out was lowest with coke. 5 The results are summarized in Table 9. Table 9. — Relative Total Costs of Heating Typical Columbus, Ohio, House with Various Fuels Fuel Unit fuel price Relative cost per cent Bituminous coal, hand fired $6.75/T 17.00/T $8.35/T 15.00/T $0.55/M.C.F. $0.065/Gal. 100 Semi-bituminous coal, hand fired 97 Coke, hand fired 95 Bituminous coal, stoker fired Natural gas Oil 119 151 177 The average efficiencies with which these fuels were used in the furnaces tested in actual use in house heating are given in Table 10. In addition to being less smoky and more economical, coke has the advantage of being cleaner than coal. The walls in houses heated with coke remain cleaner than do those in which bituminous coal is burned. It is also possible to maintain a more even temperature if the coke fire is correctly regulated. Less frequent firing is required and in general operation of the heating plant is less troublesome. Another important but more intangible benefit is the resulting more efficient use of coal, effecting thereby some conservation of the nation's fuel resources. ■• Sherman, R. A., and Cross, R. C, Efficiencies and costs of various fuels in heating: Bituminous Coal Research, inc. Technical Report No. :>, 1936. COKE AS A DOMESTIC FUEL 39 Table 10. — Average Efficiencies of Use of Various Fuels in House-Heating Furnaces, Columbus, Ohio Fuel Furnace or boiler Test efficiency per cent Average efficiency per cent Assumed seasonal efficiency per cent Warm air Steam Warm air Warm air Warm air Warm air Warm air Steam Hot water Warm air Warm air a Steam a Warm air b Warm air Steam 54 46 41 54 52 76 75 69 56 55 69 76 70 65 51 47 53 75.5 60 72 58 Hand fired semi-bituminous coal Hand fired coke 45 50 Stoker fired bituminous coal Natural gas Oil . 70 55 70 75 70 60 a Designed for gas. b Converted to gas. MANAGEMENT OF A COKE FIRE Combustion conditions in a fuel bed of coke differ from those in a bed of high-volatile-content coal and a different firing procedure must be adopted. Many of the complaints against coke as a domestic fuel and much of the difficulty in attempts to introduce it as such arise from improper firing conditions which are usually aggravated by equipment in poor condition. The use of coke under improper conditions usually results in a greater amount of trouble than is experi- enced when coal is incorrectly fired. Coke is more sensitive to the conditions under which it is fired than is coal. Directions for burning coke in the domestic plant have been presented in two publications by the United States Bureau of Mines. 6 The directions are similar, and those given by Nicholls and Landry are quoted here for convenient reference. o Nicholls, P., and Landry, B. A., Coke as a domestic fuel: TT. S. Bureau of Mines. Report of Inv. 2980, 1929. Kreisinger, H., and Pieldner, A. C, Why and how coke should b< used for domestic heating: U. S. Bureau of Mines Tech. Paper 242, 1919.. 40 COKE MARKET AND COMPETITIVE FUELS "1. Before starting to use coke, inspect the furnace and see that it is in good repair and that the dampers will close; see that the ash pit is tight and has no leaks which will admit air in spite of its damper being closed. "2. Clean the flues and smoke pipe. "3. Use the size of coke recommended for the heater. There is no standard scale for coke sizes, and producers differ in their methods of grading, both on size and on the number of sizes carried ; consequently, recommendations have varied. "4. Carry a deep fuel bed, and fire as large a charge as the furnace will hold ; do not carry a shallow bed but fill the fire pot to its capacity, even in mild weather. "5. Do not poke the fire ; be content with levelling it and remove the ashes by shaking the grates, leaving a layer of ash to protect the grates and to increase the resistance of the fuel bed. "6. Open the dampers the minimum distance necessary to obtain the desired heat. "Such instructions can only be very general and they leave a lot for the user to learn from experience if he is to get the most satisfactory results; his problem with coke is, how- ever, no worse than with other fuels, except that the adjustment of the dampers may have to be finer. Success with coke does, however, call for more self-restraint than with other fuels; it is easier to obtain a fierce fire, and trouble is less likely to result if one exercises patience and gives the house time to warm up without making use of the full possibilities of the extra hot fire obtainable with coke." 7 It is very desirable that the heating system be provided with automatic control devices. A room thermostat operating the dampers will greatly improve operating conditions. Further improvement is obtained by the addition of a high temperature limit control in the stack pipe to prevent the development of excessive temperature in the fuel bed, loss of heat in hot stack gases, overheating of the house, and the formation of clinkers. Reliable devices for such control are on the market. i Nicholls, P., and Landry, B, A., Op. cit. p. Part III — Technology of Coke Manufacture By Gilbert Thiessen CHAPTER V— REASONS FOR THE MANUFACTURE OF COKE Coal is coked to produce a fuel having certain desirable characteristics for use in metallurgical and industrial processes, to provide a smokeless domestic and industrial solid fuel, to produce gas and other distillation products, and to utilize smaller sizes of coal. The production of metallurgical coke was the main incentive for the development of the coke industry. In more recent years the demand for a smokeless domestic fuel, that is cheaper than oil, gas, and anthracite has furnished a steadily growing market for coke which has approached in mag- nitude the industrial and metallurgical market. Bituminous coal is not very suitable for blast furnace and many other metallurgical and industrial uses because of the volatile matter it contains or because it becomes soft and semi-fluid when heated. Anthracite is not a competitor either because of price or because of its low reactivity, high ash content, or other characteristics. Although other methods for the reduction of iron ore have been proposed and some have been adopted, the modern iron and steel industry relies on coke and the by-products made in coke manufacture for the operation of its fundamental units, the blast furnace, the open hearth, and the cupola. The rising price of anthracite has caused a considerable shift to the use of coke in the northeastern part of the United States where the use of smokeless fuel is compulsory in many localities. This change was not made without con- siderable effort on the part of coke producers who undertook elaborate edu- cational programs. In other parts of the United States coke has come into use as smokeless fuel which is cheaper, if less convenient, than oil and gas. The manufacture of illuminating gas was the reason for the first establish- ment of the coal carbonization industry. The carbonization of coal has continued to be the main source of manufactured gas. Formerly, practice was directed toward complete gasification of coal, usually in steps, with little attention paid to the quality of coke produced during the process, since this coke was not marketed. The growing domestic market for coke has caused a change in the manufactured gas industry. Recent practice is directed toward the production of gas and high quality coke in coke ovens, the ovens being heated either by combustion of a part of the coal-gas produced in the ovens or by combustion of [41] 42 TECHNOLOGY OF COKE MANUFACTURE water-gas made from a part of the coke breeze. The use of by-product coke ovens in the manufactured gas industry is especially noticeable on the eastern seaboard where the coke is sold in competition with anthracite to customers who have learned to appreciate it, or who are compelled by ordinances, to use smokeless fuel. Because the coking process utilizes very small sized coal it offers a means of converting certain sizes of coal into a more valuable and more easily marketable fuel. However, because of the usual tendency of the mineral matter to con- centrate in the naturally formed smaller sizes of coal, it is usually necessary that such coal be cleaned before being coked, if the ash content is to be kept within reasonable limits . The importance of small coal in providing a source of cheap fuel for coking depends upon conditions determining the local marketability of such coal. CHAPTER VI— METHODS OF PRODUCING COKE All processes for producing coke from coal are similar in that they provide means for heating coal in the absence of air to a temperature at least high enough to pass the coal through its plastic state and to set the plastic mass to cellular porous coke. The first method of coking coal was to build specially arranged piles in which flues were built out of blocks of the coal itself. Heat for the coking process was furnished by combustion of part of the coal by air brought in through the Hues. These piles were covered with earth or clay to conserve heat and coke. This very primitive method of coke manufacture developed into the beehive process in which the coal is coked in a layer on the floor of a beehive-shaped oven by the heat provided from combustion of the volatile matter and part of the coke. The coking process is initiated by heat stored in the oven walls from the previous charge. Very desirable metallurgical coke has been and is being made in beehive ovens from suitable coals. At one time there was a strong prejudice among blast furnace superintendents in favor of beehive oven coke over by- product coke. Coke for metallurgical purposes has been produced by the beehive oven or similar methods since the middle of the eighteenth century. For many years likewise, independently of the metallurgical industry, coal has been carbonized to obtain volatile products, such as illuminating gas, tar, and ammonia, with domestic coke often produced as a by-product. The manufactured gas industry carbonized coal in special types of retorts or ovens. The coke so obtained was frequently of poor quality. Development work on large chamber ovens in which good metallurgical coke could be produced, at the same time permitting recovery of by-products, dates from about 1850. The change to by-product coke ovens in the United States reached its full stride in the period 1915-1920, under the influence of the war-time demand for benzene and toluene. Today by-product coke ovens supply the basic demand for coke, while beehive ovens, because of their greater flexibility, supply the requirements in excess of pro- duction by operating by-product ovens. The modern by-product coke oven in the slot form which has become almost universally adopted consists of a rectangular refractory brick chamber, about 40 feet long, 10 feet high and 18 inches wide. Heat is supplied by the com- bustion of gas in flues built into the side walls of the ovens. The ends are closed [43] 44 TECHNOLOGY OF COKE MANUFACTURE by doors which are removable to permit the coke to be pushed out. Coal is fed into the ovens through charging holes built into the tops of the ovens. The gas offtakes are located at the ends of the ovens and also are built into the top. These ovens are constructed in groups, or batteries, side by side, separated by sets of heating flues common to the adjoining ovens. Recuperators or regenerators for conserving heat by preheating the incoming air and fuel gas by means of the dftJUJ Fig. 7. — Cut-Away Block Model of Modern By-product Coke Oven. Courtesy of Koppers Co. heat contained in the combustion gases coming from the heating flues are usually located beneath the ovens. A cut-away block model of one of the widely adopted constructions is shown in figure 7, and a view of an installation of such ovens is shown in figure 8. Rectangular coke ovens have been constructed in a wide, low form with the heating flues beneath the floor or sole of the oven on which the coal is coked. Such ovens, termed "sole-flue" ovens have been built in a variety of designs one of which is shown in cross section in figure 62 and an installation of such ovens in figure 61 (pp. 180, 182). METHODS OF PRODUCING COKE +5 Figure 9 shows the relative sizes and shapes of the more common types of chambers in which coal is carbonized at high temperatures to produce a coke substantially free from volatile matter. Commercial coke is now produced mainly in slot-type by-product ovens, beehive ovens, and (in Illinois) sole-flue Fig. 8. — An Installation of By-product Coke Ovens at South Chicago. Courtesy of Koppers Co. ovens. Heat is supplied to the coal in the Knowles ovens mainly through the floor, in the beehive ovens largely from above, in the slot-type ovens from the sides, and in the gas retorts from the entire periphery. Gas offtakes are from the tops of the ovens or charges in all cases. The temperatures attained in the charges are in the range of 750°-1200°C. (1382°-2192°F.) in almost all cases. The temperatures attained at the completion of coking in beehive and slot-type by-product coke ovens in which most of the coke produced today is made is about 1000°C. (1832°F.) 46 TECHNOLOGY OF COKE MANUFACTURE HIGH- VERSUS LOW-TEMPERATURE CARBONIZATION The high temperatures (about 1000°C. or 1832°F.) employed in by-product ovens subject the distillation products which first arise at temperatures up to about 550° C. (1022°F.) to rather severe decomposition or "cracking" and also decrease the ease with which the resulting coke may be ignited and kept lit at low rates of burning. The cracking which the tarry products first liberated from the coal undergo as they pass into zones of higher temperature on their way out of the oven results in a lower tar yield, a higher gas yield, and the formation of the compounds such as benzene, naphthalene, and phenols characteristic of r b Hi I =u -fT. oven! knowles type by-product BEEHIVE SLOT TYPE BY- PROD HORIZONTAL VERTICAL INCLINED stop-end|through GAS RETORTS coal 1 5 jons charge! 6 TONS 12 TONS 450 LBS. 1700 LBS. 1700 LBS. 1 TON Fig. 9. — Relative Sizes of Coal Carbonizing Equipment. high-temperature coke oven tars. This thermal alteration of the so-called primary decomposition products of coal has been considered objectionable for various reasons and much attention has been given to the problem of coking coal at temperatures only slightly higher than the temperature required to form solid coke from plastic coal. Reasons for tins effort have been the desire to produce ( 1 ) more easily ignitable and combustible cokes which could be used as smoke- less domestic fuel with greater convenience than could high-temperature coke, and (2) larger quantities of liquid products from which motor fuels could be METHODS OF PRODUCING COKE 47 prepared. The second reason has been given its greatest emphasis in those countries which have an insufficient domestic supply of petroleum. This has been especially true at times when a world petroleum shortage has seemed imminent and previous to the development of other processes for producing liquid fuel from coal. The advantages for domestic use of a coke which is more easily combustible are well known. One of the main difficulties encountered in low-temperature carbonization systems is the low capacity due to the low temperature gradient permissible if the coke and tar are not to be overheated. Considerable attention is now being given to so-called medium-temperature carbonization processes which produce coke more easily combustible than high-temperature coke, yet which permit higher heat gradients to be used than can be used in low-temperature carbonization processes. Failure of true low-temperature coal tars to live up to their estimated commercial value and usefulness has been another point in favor of the medium-temperature processes which can produce a tar more like high- temperature coal tar. In spite of the large amount of money spent on the develop- ment of so-called "low-temperature carbonization" systems (estimated at 40 million dollars in the United States alone) very few plants are operating success- fully today from either a technical or financial standpoint. Most of these which are operating have some form of subsidy, either direct or indirect, to assist their financial position. CHAPTER VII— HISTORICAL SURVEY OF THE PRODUCTION OF COKE FROM ILLINOIS COALS INTRODUCTION Efforts at the production of coke from Illinois coals followed the general progress of the coke industry in the United States. Before 1900 attention was directed almost entirely towards the use of the beehive or similar ovens. With the introduction of the by-product coke oven in this country on a large scale after about 1907 and especially during the war time boom from 1915 to 1920, attention was directed towards the use of these ovens or to the development of such ovens for making metallurgical coke from Illinois coals. More recently effort has been directed towards the production of domestic fuel with recognition that the high ash and high sulfur content of many Illinois coals makes the pro- duction of metallurgical coke from them an uncertain proposition. The development of ways of making domestic coke have followed several lines. At one time it was thought that low-temperature carbonization had great possibilities. The production of coke in relatively small, low capital cost units utilizing local screenings and supplying gas to nearby users is now receiving attention. A brief summary of the early work directed towards the production of coke from Illinois coals was prepared by Ovitz in 191 7 1 under a cooperative agreement between the United States Bureau of Mines, the Illinois State Geological Survey, and the Department of Mining Engineering of the University of Illinois. This bulletin was published before any appreciable work had been done on the carbonization of Illinois coal in by-product coke ovens. Much of the early history of the manufacture of coke from Illinois coal is obscure. USE OF BEEHIVE TYPE OVENS FOR COKING ILLINOIS COALS Illinois coal is not well suited for coking in beehive coke ovens. Nevertheless a considerable quantity of beehive coke has been produced from Illinois coal, mainly in the period 1880-1900 (fig. 10). By 1910 beehive coke oven operation in Illinois had practically ceased. The largest operations were : ( 1 ) at Mount Carbon, near Murphysboro, Jackson County, where, according to Ovitz, from three to four hundred thousand tons of coke were produced; (2) at Equality, Gallatin County, where there was an installation of 24 ovens, 20 feet long, l Ovitz, F. K., Coking of Illinois Coals: U. S. Bureau of Mines Bull. 138, 1917. [49] 50 TECHNOLOGY OF COKE MANUFACTURE 5 feet high, and 20 inches wide, perhaps not beehive ovens; and (3) at South Chicago where there was an installation of 24 beehive ovens fitted with flues in the floor and an arrangement for introducing hot or cold air over the charge to increase the oven temperature and improve control with a view to coking Illinois coals. 2 A A r \ A g / \ , Q Z \ A t\ 1 / V 3 / / H 5 / — 16 80 '84 '88 '92 '96 1900 '04 \l i A v / \ V J \ 1 v / z o V l~ Z o _i _i 5 r i i Fig. 10. — Coke Produced in Illinois, 1880 to 1936. The most exhaustive experimental attempts to produce beehive coke from Illinois coals were made at the fuel testing plant of the United States Geological Survey, located at the Louisiana Purchase Exposition from 1904 to 1907, and at Denver, Colorado, from 1907 to the summer of 1909. : Two standard sized beehive ovens were used in these tests. Briefly, these tests showed that a fair metallurgical coke could be made from some Illinois coals, that washing the coal in many cases improved the quality of the coke or made possible its successful 2 Moss, R. S., Improved Heminway Process: Mines and Minerals, vol. 21, No. 6, pp. 112-14, April 1901. 3 Parker, E. A., Holmes, J. A., and Campbell, M. A., Report on the operations of the coal-testing plant of the United States Geological Survey at the Louisiana Purchase Exposition, St. Louis, Mo., 1904: Part III Producer-Gas, Coking, Briquetting and Washing Tests. TJ. S. Geol. Survey, Prof. Paper 48, 1906. Holmes, J. A., Preliminary report on the operations of the fuel testing plant of the United States Geological Survey at St. Louis, Mo., 1905: U. S. Geol. Survey Pull. 290, 1906. Holmes, .1. A., Report of the United States fuel-testing plant at St. Louis, Mo., January 1. I'.mm; to June 30, 1907: CJ. S. Geol. Survey Bull. 332, 1908. Belden, A. W., Delamater, G. LI., and Groves, .). VV., Washing and coking tests of coal at the fuel-testing plant, Denver, Colorado. July 1, 1907 to June 30, LUIS: V. S. Geol. Survey Bull. 368, 1909. Belden, A. W., Delamater, G. R., Groves, J. W., and Way, K. M., Washing and coking tests of coal at the fuel-testing plant, Denver, Colorado, July 1, 1908 to June 30, 1909: U. S. Bureau of Mines Bull. 5, 1910. HISTORY OF COKE PRODUCTION 51 production, and that the sulfur contents of the cokes were in many cases higher than was was desirable for foundry or blast furnace cokes. At that time there was little or no demand for coke as a domestic fuel. The coke was, therefore, judged as to quality by its suitability for use in a cast-iron foundry cupola. Much of the coke classed as undesirable for metallurgical use by these tests because of its sulfur content or flngery structure would probably be acceptable as domestic coke today. Inasmuch as Illinois coals in most cases rapidly lose much of their coking power on weathering and since the length of time elapsed between the Fig. 11. — General View of Abandoned Beehive Coke Ovens near Sparta, Randolph County, Illinois. mining and the testing of the coals is not given, it is possible that many of the coals which did not yield good cokes in the test would have done so had they been fresh and coked under more favorable conditions. In addition, there have been several semi-commercial attempts by private organizations to coke Illinois coal in beehive type ovens. In most of these cases only a few ovens were built and they were soon abandoned. An installation of 50 beehive ovens was located on the Mobile and Ohio Railroad, southwest of Sparta in Randolph County. Only a few were ever put into operation and these were later abandoned. Their present state is shown in figures 11 and 12. It is said that satisfactory coke could not be made in them from coal obtained at an adjoining mine and that the battery of ovens was built before adequate tests had been made. 52 TECHNOLOGY OF COKE MANUFACTURE USE OF BY-PRODUCT OVENS FOR COKING ILLINOIS COAL The close proximity of the Illinois coal fields to several of the large markets for metallurgical and industrial coke has naturally resulted in attempts to use Illinois coals for by-product coke manufacture. This was especially true during the World War when supplies of eastern coals and transportation facilities westward were curtailed. The following paragraphs contain a brief review of these attempts and summarize the conclusions drawn from them. Fig. 12. — Abandoned Ovens at Sparta: view of Oven Door and Arch with Part of Stone Facing Removed. The United States Bureau of Standards made a study of the possibility of coking Illinois coals in an installation of 24 Roberts type recuperative ovens at Canal Dover, Ohio, at various times during 191 7-191 8. 4 Trouble was encountered with the operation of the recuperators during the tests but a satis- factory metallurgical coke was produced from No. 6 Illinois coal and used in a blast furnace. The conclusions to which the Bureau of Standards came were as follows : "At least the majority of the coals from Franklin County, Illinois, and some from Pike County, Indiana, located in the midcontinent field can be coked in by-product coke ovens, and a fairly satisfactory grade of metallurgical coke can be produced. Likewise by-products of reasonable quantity and good quality can be obtained from these coals. 4 U. S. Bureau of Standards Miscellaneous Publications 46 standards, i>i». 73-82, <'<>ke-oven investigations, 1921. War Work of tli<> Bureau of HISTORY OF COKE PRODUCTION 53 "The Bureau believes that at least some of the coal from the midcontinent field should be used in the coke-oven plants in the central section of the country. Even though the coke, on the whole, proved not to be uniformly as satisfactory as that from the so-called coking coals, the advantage due to the elimination from the coke-oven plants in the eastern section will go far in overcoming any lack of quality of the coke. There can be no doubt that where domestic coke is wanted these coals can fulfill all requirements. Such uses will materially aid in the proper development of the fuel resources of the country." The Bureau of Standards in cooperation with the Bureau of Mines also made a large scale test using about 7600 tons of Illinois coal from Orient No. 1 mine in Franklin County in an installation of Koppers ovens at St. Paul, Minn."' The conclusions from these tests were that it was "clearly demonstrated that some of the Illinois coals can be coked in the 'chamber type' oven without radical change in operating methods for the production of coke which can be successfully used in a blast furnace", and that it appeared that "the temperature at which Illinois coal should be handled for the production of the best coke is somewhat lower than the best operating temperatures for eastern coals, and moreover the speed of coking of the Illinois coal is somewhat less . . . The coke produced was very irregular in size, had a longitudinal fracture, was fingery, brittle, and shattered easily. The cell structure was very small and irregular. The coke was lighter than the average by-product coke, weighing only 23 pounds per cubic foot." The coke was tested by use in a blast furnace and found to be satisfactory. At the time the tests were carried out there was a shortage of coke, for which reason the suitability of the coke as a domestic fuel was not considered. It is said that at this time also, Illinois coals were being coked experimentally on a large scale at Gary, Indiana, by the United States Steel Corporation and at other places. As an outgrowth of the development work on the Roberts type ovens at Canal Dover, Ohio, 80 ovens of an improved Roberts form in two batteries of 40 ovens each were placed in operation at Granite City, Illinois, in January 1921 for making metallurgical coke from southern Illinois coals, ° and were operated until June 15, 1935 when they were abandoned. These ovens were operated for several months with Illinois coal alone from Franklin and Williamson counties. The more usual operation was with mixtures of about 85 per cent Illinois coal and 15 per cent West Virginia Pocahontas coal. The Illinois coal came from Franklin and Williamson counties in the following proportions: Year 1928-1930 inclusive 1931 (per cent) (per cent) Franklin Co 65 80 Williamson Co 35 20 r> McBride, R. S., and Selvig, W. A., Coking- of Illinois coal in Koppers type oven: U. S. Bureau of Standards Tech. Paper 137, 1919. <; Ditto, M. W., Design and operation of Roberts coke oven: Trans. Amer. Inst. Min. Met. Eng. vol. 69, pp. 483-512, 1923. 54 TECHNOLOGY OF COKE MANUFACTURE The Illinois coal had to be reasonably fresh to give good results. If the coal were stored more than two weeks difficulties arose due to the charge sticking in the ovens. Considerable quantities of this coke were sold for domestic fuel purposes in the St. Louis area during the last few years of the plant's operation. On the basis of the results obtained during the operation of two experimental ovens at St. Louis, a battery of 10 Knowles sole-flue type ovens was erected at West Frankfort, Franklin County, in 1934. 7 This installation was increased to 26 ovens by the addition of two batteries of 8 ovens each in 1936. This plant is in regular operation, coking screenings from Illinois No. 6 coal produced at nearby mines. The coke is sold for domestic and industrial purposes. Because this plant successfully cokes Illinois coal and because its operation has not been described elsewhere in detail, a more detailed discussion of the operating conditions of these ovens is presented in this report (pp. 179-203). APPLICATION OF LOW- AND MEDIUM-TEMPERATURE CARBONIZATION PROCESSES TO ILLINOIS COALS The Parr process. — The pioneer in work on low-temperature carboniza- tion of Illinois coals was the late Professor S. W. Parr, who was interested in the production of smokeless fuel from them. He has written as follows : s "First and historically this work was begun at the time of the great anthracite strike in 1902, but our idea from the start was to work on fundamental factors involving especially the role of oxygen, both inherent in the chemical composition of coal and that absorbed from the time of breaking out at the mine. We worked almost entirely on laboratory studies and with laboratory apparatus, but we moved from tests involving the use of a few ounces of coal up to a few pounds — twenty pounds, then thirty-three pounds, then six hundred, and finally, in ton lots. At the end of about ten years of experimentation I became thoroughly convinced that the low temperature range, say from 700° to 900°F. (360°- 480°C.) was not at all suited to our economic conditions and environment. From that time on we centered our efforts upon a mid-temperature range of from 1,400° to 1,500° F. (760°-815°C). Our final apparatus might well be considered a semi-industrial size of plant. . . "I may say, moreover, that this outfit operated continuously for three hundred and sixty-five days, twenty-four hours per day, without a single shutdown or delay other than for making minor repairs or adjustments, and that while we operated upon car-lots from as far west as Iowa and as far east as Pocahontas and Pittsburgh fields in Pennsylvania, and as far south as Birmingham, Alabama, including Kentucky, Indiana and substantially all the fields of Illinois, we did not fail in any case to make a high grade coke " t Anon., Knowles oven widens market for No. 5 Illinois screenings by turning- fines into domestic coke: Coal Age, vol. 39, pp. 421-423, Nov. iy;>4. McBride, it. S., Processing coal in Knowles coke ovens: Chem. and Met. vol. 42, No. (>, pp. 300-3, .June 192F,. 8 Proceedings of the Symposium on Fuel and Coal. McGill University, pp. 186-187, 1931. HISTORY OF COKE PRODUCTION 55 The Parr process is a two-stage process in which the coal is heated to within 50°C. of its plastic temperature in the first stage in an externally heated, rotating horizontal retort and then charged into a coke oven and coked. During the preheating stage the rise in temperature of the coal is slow until 100°C. has been passed and then rapid to 300°C. Coking is rapid in the second stage due to the fact that moisture and carbon dioxide have been driven off and the coal is almost at the fusion point. Coke oven operators claim that great difficulties arise in charging dry, preheated coal into hot coke ovens because the rapidly evolved gases in the oven carry much fine coal dust with them. This dust deposits in collector mains and tar-handling equipment and may plug them. 9 Even relatively small amounts of dust are undesirable, since they increase the so-called "free-carbon" content of the tar. It is said that in some cases practically the entire charge of dry, hot coal has been carried out of the oven. This process is not in commercial or experimental operation today. The experimental plant operated by Professor Parr has been dismantled. Greene-Laucks process. — In the Greene-Laucks process the coal is pro- pelled upwards on a worm conveyor shaft in a vertical retort. Heat is applied on the outside of the retort and inside the hollow conveyor. This method leaves 15 to 17 per cent volatile matter in the coke. A pilot plant having a rotary conveyor 3 feet in diameter and 18 feet long and a capacity of 15 tons of coal per day was operated in Waukegan, Illinois, for two years. No full size plant was built and the pilot plant has been dismantled. 10 Other processes. — Illinois coals have been experimentally coked in the course of development or promotion of a large number of low-temperature carbonization processes. The reports of these tests are rarely given in sufficient detail to have any value. In most cases the process in question "successfully processes" the Illinois coal. Since most of these processes did not go beyond the experimental stage and since none are at present operating or appear likely to operate in the future with Illinois coal, they are not discussed further here. Small scale carbonization tests on Illinois coals. — Because of the expense involved or the danger of damage to large scale equipment in conducting large scale coking tests, there is a natural tendency to try to determine the coking properties of coals on much smaller scales. The use of small scale equipment introduces the difficulty of so conducting the test that the results may be directly applied to full scale practice. However, even though the results may not be comparable to full scale work, small scale tests in many instances are comparable o Discussion by Ulmer, C. D., Rarasburg, C. J., and Good, J. B., of paper by Warren, W. B., The relation of work of the laboratory to practical carbonization: Proceedings of Technical Meeting held under the auspices of the Coal Research Laboratory, Carnegie Institute of Technology, Pittsburgh, Pa., Dec. 3, 1936, mimeographed, pp. 17-19. :o Greene, F. C, The Greene-Laucks process: Proc. First Int. Conf. on Bit. Coal, pp. 712-728, 1926. 56 TECHNOLOGY OF COKE MANUFACTURE among themselves. Small scale tests are especially valuable in studying coals about whose properties very little is known and for which there would be a very great risk of failure if they were tested on a large scale. It is impractical, if not impossible, to refer to and review all of the small scale carbonization tests which have been made on Illinois coals. Only those which appear to the author to be of most significance are referred to here. The late S. W. Parr probably conducted more tests on the possibilities of producing a smokeless fuel from Illinois coal than any other man. The scale on which he worked was, as quoted above, gradually increased in the course of some twenty years work until his last tests were made in a commercial scale experi- mental oven. Professor Parr's efforts were largely directed towards the pro- duction of a low-temperature coke to supply a demand for a smokeless domestic fuel. 11 As a result of Parr's investigation on the carbonizing characteristics of Illinois coals he came to the conclusion that these coals produced a much better coke if they were given a preliminary conditioning heat treatment in an inert atmosphere at a temperature just below the softening temperature, that is, at about 350°C. This conditioning treatment in an inert atmosphere is one of the important features in the Parr low-temperature coking process. The conditioned coal is charged into an oven similar in construction to a standard by-product oven and carbonization is finished at a temperature of from 750°C. to 800°C. The United States Bureau of Mines in cooperation with the American Gas Association undertook a survey of the gas, coke, and by-product making properties of American coals during the course of which they examined coal from No. 6 bed at West Frankfort, Franklin County, Illinois. 12 This was the same coal used in the full sized tests at the plant of the Minnesota By-Product Coke Company, carried out cooperatively by the Bureau of Mines and the Bureau of Standards. 13 The results of their cooperative tests on the Orient (Illinois) coal are of special interest because this was one of a group of thirty or more coking coals from various parts of the United States to which the systematized tests 11 Parr, S. W., Anthracizing bituminous coals: Illinois State Geol. Survey Bull. 4, 1907. Parr, S. W., and Francis, C. K., Artificial modification of the composition of coal: Illinois State Geol. Survey Bull. 8, 1908. Parr, S. W., and Francis, C. K., Univ. of Illinois Eng. Expt. Sta. Bull. 24, 1909. Parr, S. W., and Olin, H. L., The coking of coal at low temperature with a preliminary study of the by-products: Univ. of Illinois, Eng. Expt. Sta. Bull. 60, 1912. Parr, S. W., and Olin, H. L., The coking of coal at low temperatures, with special reference to the properties and composition of the products: Univ. of Illinois Eng. Expt. Sta. Bull. 79, 1915. Parr, S. W., bow temperature carbonization of coal: Proceedings of the Second Inter- national Conference on Bituminous Coal, vol. 1, pp. 54-70, 1928. 12 Fieldner, A. C. et al., Carbonizing properties and constitution of No. C> lied from West Frankfort, Franklin Co., Illinois: U. S. Bureau of Mines Tech. Paper 524, 19:52. i ■". McBride, It. S., and Selvig, W. A., Coking of Illinois coal in Koppers type oven: U. S. Bureau of Standards Tech. Paper 137, 1919. HISTORY OF COKE PRODUCTION 57 were applied, providing truly comparable results. In comparing the carbon- izing characteristics of the Orient coal from Franklin County, Illinois, with those of the eastern gas coals, Fieldner and his coworkers arrived at the following conclusions quoted from Technical Paper 524. "Correlation of the chemical composition and physical properties of the Orient coal is of particular interest in view of its relatively low rank in the scale of classification of bituminous coal. The high oxygen content of Orient coal is reflected in its carbonizing properties as compared to eastern gas coals by: "1. A higher yield of liquor. "2. A lower yield of highly oxygenated tar, containing nearly 30 per cent of tar acids in the low-temperature tar. "3. A moderate yield of gas rich in oxides of carbon and therefore of somewhat low calorific value, 2,650 B.t.u. in stripped gas per pound of coal at a carbonizing temper- ature of 900 °C. "4. A lighter and more porous coke. "5. A lower agglutinating index. "6. Less development of plasticity and a lower swelling coefficient." They also conclude that: "On blending with 25 per cent of low-volatile coal, however, it makes excellent metallurgical or domestic coke" and that: "The physical properties of the coke obtained in the test retort at a carbonizing temperature of 900 °C. were similar to those of coke obtained in tests of coal from the same mine in by-products ovens, and the oven yields of gas, coke, and tar were between those obtained in the retort tests at carbonization temperatures of 900° and 1000° C." The United States Bureau of Standards also tested this coal on a small scale at about the time it carried on the large scale tests in the Koppers type ovens at St. Paul, Minn. Four pounds of coal were carbonized in a cast iron box placed in a special muffle furnace regulated so that the temperature of the vapor over the coal being coked was 840°, 775°, 700°, 600° and 605°C. in the five tests. By-product and coke yields and general observations on the quality of the coke are reported. 14 The investigators warn against making any conclusions from the tests. CONCLUSIONS FROM THE HISTORICAL REVIEW OF THE ART OF PRODUCING COKE FROM ILLINOIS COALS The production of coke from Illinois coals has occupied the attention of the coal industry and of investigators for more than sixty years. During this period the art of coking Illinois coal followed closely the art of coking the more easily coked coals. From about 1880 to 1900 some coke was produced from Illinois coals in beehive or similar ovens, the coke being sold for light industrial purposes. Illinois coals apparently are not nearly as well adapted to being coked in beehive ovens as are some of the Appalachian coals. In an effort to improve the pro- 14 McBride, R. S., and Brumbaugh, I. V., TCxperimental retort tests of Orient coal: IT. S. Bureau of Standards Tech. Paper 134, 1919. 58 TECHNOLOGY OF COKE MANUFACTURE duction of coke from Illinois coals various modifications of the beehive oven were introduced. In comparison with the total United States production of beehive oven coke, the production of beehive coke from Illinois coal was un- important. The rapid loss of coking power on exposure to air and the high moisture and volatile matter content probably were the controlling factors in the manufacture of coke from Illinois coals by the beehive oven process. The introduction of the by-product recovery coke oven into the Illinois area led to trials of Illinois coals in them. Again, the rapid loss of coking power and the much shrunken, fingery coke formed because of the high volatile matter content placed the coals from Illinois in an unfavorable position when compared with the eastern coking coals. As before, efforts were made to modify the by-product coke oven to make it more suitable for the coking of these coals. Until recent years coke was made primarily for metallurgical purposes and the quality and values of the coke produced were judged mainly from the stand- point of the blast-furnace operator. A growing realization of the importance of the domestic fuel market for coke has led to a reconsideration of the coke which may be made from Illinois coals. Cokes which would be inferior for metallurgical uses may be well adapted for use as a domestic fuel. The mechanical problems of removing the fingered and shrunken coke from the oven may be and have been solved by modifications in oven construction. The new possibilities in the disposal of coke from Illinois coals makes desirable a survey of the coke-making properties of the coals found in Illinois, their improvement for coke-making purposes, and an investigation of the conditions under which they produce the best coke. Present production of coke from Illinois coals is in an installation of sole- flue ovens, having characteristics quite different from those of the more common slot-type ovens. Because these ovens seem to be well suited to the carbonization of Illinois coals, certain features of their operation, particularly the temperature conditions throughout a charge during a coking cycle, have been investigated and are reported in detail together with analyses of the coke as produced (see pp. 179-203). In these ovens a coke resembling slot-type, by-product oven coke or a coke grading from high-temperature coke at the bottom to a higher volatile, lower temperature coke at the top can be produced at will and easily removed from the oven. The higher volatile content cokes may have great advantage over low volatile cokes as domestic fuels. CHAPTER VIII— THE REQUIREMENTS OF COKING COALS The difficulties encountered in endeavors to coke Illinois coals indicate the desirability of a study of the requirements of coal for such use. Such a study should include consideration of the commonly accepted criteria for determining whether or not a coal is a coking coal, and also experimental studies with respect to the yield and characteristics of the products of carbonization of a variety of Illinois coals, even though such investigations may be possible only on a labora- tory scale. Such studies and investigations will provide some answer to the questions concerning the nature of coking coals and the extent to which Illinois coals are coking coals. CRITERIA FOR COKING COALS Gluud's Handbook 1 defines coking coals as follows: "The commercial value of coke depends upon its purity, and upon such factors as size, strength and structure. These physical characteristics are determined not only by the kind of coal used, but also by the degree of pulverization and moisture content of the coal, the amount, nature and size of impurities such as slate, the temperature of the oven walls, dimensions of the oven, and other variables, all of which affect the coke at some stage of its formation. Therefore, it is not surprising to find that the only thoroughly reliable method to determine the quality of coke that can be made from a coal under particular conditions, is to subject the coal to a full-scale oven test under the desired conditions. "For practical purposes, a coal may be classed as a coking coal if it will yield a merchantable coke when carbonized by a commercial method in an existing type of coke oven. In order to be merchantable, a coke must, in general, be of adequate purity, size, strength and structure, for the uses to which it is to be put. Furthermore, the coke from any coal or coal mixture carbonized in by-product ovens must be strong enough and have sufficient shrinkage from the walls to permit its discharge from the ovens without difficulty. Coal mixtures are commonly used at by-product coke-oven plants, and this practice has resulted not only in the production of superior coke, but in the extensive utilization of coals which would not be very suitable for coke manufacture, if used separately." It is to be noted that this definition indicates that a coke should be judged according to its suitability for the use to which it is to be put. In the past cokes from Illinois coals have almost always been judged as to their suitability for metallurgical purposes. These cokes should now be judged as to their suit- ability for use as a smokeless and convenient domestic fuel. Inasmuch as the requirements which must be met by the coke for domestic use are different and i Gluud, W., International handbook of the by-product coke industry: American Edition by D. L. Jacobson. New York 1932, p. 156. [59] 60 TECHNOLOGY OF COKE MANUFACTURE not so rigorous as for metallurgical purposes, it seems quite possible that a different valuation should be placed upon many Illinois coals as coking coals. The American Society for Testing Materials has adopted standard speci- fications for gas and coking coals which may be used as a guide in the selection of coking coals. It should be emphasized that these specifications were set up with the production of metallurgical coke in slot-type by-product ovens primarily in mind and are rather rigorous; however they do furnish a means for evaluating coals comparatively. Coals which fall outside the limits for coking coals set up in this specification may still yield coke of good quality entirely satisfactory for many purposes. These specifications are reproduced here: 2 STANDARD SPECIFICATIONS FOR GAS AND COKING COALS A. S. T. M. Designation: D 166-24 "These specifications are issued under the fixed designation D 166; the final number indicates the year of original adoption as standard, or, in the case of revision, the year of last revision. Issued as Tentative, 1923 ; Adopted, 1924 "1. Gas and coking coals must yield both merchantable gas and coke when distilled in a retort or oven by commercial methods. The type of coals may vary within rather wide limits according to the treatment in the retort and the market for the products. These specifications, therefore, merely give the limits within which gas and coking coals will usually fall, and indicate the circumstances under which further restrictive conditions should be imposed. Sampling and Analysis "2. The coal shall be sampled in accordance with the Standard Method of Sampling Coal (A.S.T.M. Designation: D 21) of the American Society for Testing Materials. "3. Analyses of the coal, when required, shall be made in accordance with the Standard Methods of Laboratory Sampling and Analysis of Coal and Coke (A.S.T.M. Designation: D 271) of the American Society for Testing Materials. Chemical and Physical Properties "The carbon ratio, that is, the ratio of fixed carbon to volatile matter, while not entirely reliable, is the best simple index to the behavior of the coal when carbonized. The carbon ratio in the case of gas coals will vary from 1.4 to 2.0 and for coking coals from 1.4 to 5.0. The latter includes a wide range of coals varying from high volatile gas coal to low volatile or 'smokeless' coal. "4. (a) The percentage of moisture in the coal as mined shall be subject to agree- ment by the purchaser and the seller. "(b) In the absence of a definite agreement between the purchaser and the seller, the mine moisture in the coal as mined shall not exceed 4.0 per cent. The moisture shall be determined by the general average composition of coal from the mine in question and an analysis of each shipment shall not be required. 2 Amor. Soc. for Testing Materials: A. S. T. M. Designation P 166-24, A. S. T. M. Standards: pp. 379-381, 1936. Reprinted by permission of the Society from its copyrighted publications. REQUIREMENTS OF COKING COALS 61 "5. The fusion temperature of ash of coal, the coke from which is intended for domestic and industrial use, shall not be below 2200° F. In the case of coke for use in the manufacture of water gas, the fusion temperature of the ash of the coal shall preferably be higher than 2300°F. The fusion temperature of the ash shall be determined in accord- ance with the method for determination of fusibility of coal ash appearing in the Standard Methods of Laboratory Sampling and Analysis of Coal and Coke (A.S.T.M. Designation: D 271) of the American Society for Testing Materials. "NOTE. — The fusion point of ash is not usually important for metallurgical work. It is important, however, in the case of coke for domestic and industrial furnace use and for the manufacture of water gas. A. Special Requirements for Gas Coals "6. Gas coal shall contain not less than 35.0 per cent of volatile matter when deter- mined on the moisture and ash-free basis. "NOTE. — This is equivalent to 30.8 per cent volatile matter for a coal containing 12.0 per cent of combined ash and moisture. "7. In the case of gas coals, the ash in the dry coal shall not be over 9 per cent. "8. The composition of gas coal shall be such that the dry coke produced therefrom will contain not over 1.5 per cent of sulfur and the resultant gas will contain not more than 30 grains of sulfur, in the form of compounds other than hydrogen sulfide, per 100 en. ft. of gas. "9. Gas coal shall be such that the coke produced therefrom will be of sufficient size and strength to be suitable for domestic use or for the manufacture of water gas. "NOTE.— These physical characteristics of coke are not amenable to simple explicit definition and must necessarily be left to the judgment of experienced operators." B. Special Requirements for Coking Coals "10. In the case of coking coals, the ash in the dry coal shall be not over 9 per cent. "11. (a) If metallurgical coke is to be produced, the composition of the coking coal shall be such that the dry coke produced therefrom will not contain more than 1.0 per cent of sulfur in the case of foundry coke, and 1.3 per cent of sulfur in the case of blast-furnace coke. "(b) If gas is to be sold for domestic use, the composition of the coking coal shall be such that the resultant gas will contain not more than 30 grains of sulfur, in the form of compounds other than hydrogen sulfide, per 100 cu. ft. of gas. "12. A limitation as to phosphorus, which may be required when coke is used for metallurgical purposes, shall be subject to agreement between the purchaser and the seller. "13. The composition of coking coal which is to be charged into a by-product oven without admixture shall be such that the coke produced therefrom will shrink sufficiently to permit of its being discharged from the oven without difficulty. "NOTE. — The mixing of coals for by-product coke-oven use is widely practiced, and such mixtures usually contain, as a very important component, low volatile or 'smokeless' coking coals, which when carbonized alone would not give the requisite shrinkage. "14. The composition of the coking coal shall be such that the coke produced there- from will meet such requirements as to size, strength, and structure as are necessary for good practice in the industries using the coke. "NOTE. — These physical characteristics of coke are not amenable to simple explicit definition and must necessarily be left to the judgment of experienced operators." In the discussions which follow various county-average analytical values and values derived from them will be used to determine where Illinois coals stand when judged by the foregoing A.S.T.M. specifications. The values used are (,2 J9)1)BUI 3!pB[0A o; xapui 8IpB|0A DlJI09dg c)U90 J9d 80U9J9Jfip A"q ((boo iiun) U93XXQ JU90 J9d QBOO^iun) roSojpA'ij ^U90 J9d (p3oa ;iun) ubqjuQ xgpui ^we-ft !)U90 J9d ([■boo ^tim) uoqjBO p9xi^ JU9D J9d (99JJ-qS'B 'A\ip) J9^BOI 911^10^ c)U90 J8d (Aip) J9^BUI 9]I^|0A c)U90 J9d (ifcp) uoqjBO p9xij -)U99 J9d (yfjp) mjpg ;u90 J9d (A\.p) qsy ^)U90 J9d P9AI309J sb 9jn;siop^ B UOl?B0O| oiqdBjaogr) "°N TECHNOLOGY OF COKE MANUFACTURE OCiC0»0^ OOCMOSCMUSOS©TtOCOO ©rH^OliO tOOOHC* N © CO CM •— i -cH -** 1 OS It- OS OONlOffi (DMIOON CM' — NtO«5 CO ^P ^ OS N OS CO CM b- OS © CM i-i OS CO Tt< CD © -* -h CO OS CM © CO t© 00 i-i CO i-i CM CO CM CO CO CM CM CO CM CO CM CM -^f CO CO CO CO CO CM CM CM CO CO CO CO CO CM CO CO CO CO CM CO CO CO CD US "S • ■ CO to CO CO • OO US ■ • . CM to US O CO CO • t- •>*< -IOCS • N OirH . . . _;CM~* • ■ CMCMOO Ort ■ rt rn N 0 OO CM OS OS CO Cm-h-hCMCM CMCMCMCOCO CM « -9> ^* ^ cM-hCM^-* ^h CO CO CO CM CM^CMCMCM CM CM -h CM CM 0)tO!D-iO ■^CM-Jtiio'O -fOOaoO 00 CO -* US CM to CO CO O i— I © © CO CO US NtOOrnOO lO 0-*CMCM CO -*cK CO OS CM CO N ^ OS OS COCMCOCMCM SocOOli no us >o us us us us us us in»0!Du5io us >o io us us us to us us us inmio^" loiomio^f rt*< us CO OS © CMoOOOOCM CM -* N N- CO OS CO CO OO CM inONOOOl NNNihX N CO OS CM CM NOOOOOO© tONrtN OS US OS ■* ^ CO 00 US CM i— i Tt< us -en tp -* ^-^^ti^p^f tj< ■* cc Th Tf ^fTjiTt'Tti^f 'Si^^^ji-^^ti ■■# -tF -f "5 "^ ■*'*'*ioifl OS OO CO 00 O CO >0 r)H CO CO ©cet-ifN CM ^h CO to CO IO io ■— ' OO OO N US OS "5 tO OO to O CO 00 OS US OS CO >0 CM -H CM N CO fhOOUINN CM US CM CO CO N |>! 00 N CM -* OS ^ch 00 'O OS CM i— i N CO co ■* co •«*< "* -*-*'*tOTf Tf fs co co co -* ^ -* 1 ■* -f co co co co ^ti Tfi co Tt< "* "* p° "* -* 1 ■"* ~f 0>-*CCMO O tO ^p CM CM CM US i-i OS cm OOONCOih f MONO CM US Tt< CM >0 i— i i-l O IO OO NUSit CON CO CO CO CO US US' © US © ,— i CON US US "O NcOrtINN CDMOC* «5lOC5tMrt ^< •* ■* Tt< ■* **^U5^( it "S US "J US -^f -c*h ^T ^ch -^f tUJUJlO'* ^f ^ch M< "* "^ -*"#-tfl^f-* 00 US OO CO CD US M io CO CO CO CO CO CO Q COCO^NO 00-^-^rficO CO CO O O N CMcDlO-fHCO CO CO CO CM CO -cf> ^ CO i— IO CO CO CO CO CO itCO'^'lO'* -^rnMiHio -^ CO CO "■*< "* "^ CO CO CO -f CM00-S«OO NOcNuSlO CqNcqNrH OO^^OO'-O i— ijoCCJO'OcO i— lONM^ i— iMOO-f CM 00 tO O 00 ~ ^h rt ci *-, COOO-^^H rtMNOrt US CO O CC O cicOCOS'-H 05 >C CM O OS -f CCHNMO lr^ 00 IO CM CM COOrnOJcO ^hOoiO i— i CO "* US CO O 05 US OO CM OS US to CO CO rn' CO N rt CO CM-HCMOSrt US>0-*cH-c«^f -^t^i-^US-* OS — J OS 00 -ccfi USl^-»ti"*PC3 i-HCOtO^cO to to « CO — .2.5 M c l— ' cu cu fe cu o to 5^to to- ■g b t a b s«2 g-o^ n n rt f3 c d d a; C 1 ^ -- kKBKK ^h^>pc;o ffiSKWtf ^WhJWo: CU M ^ ~ M. ».5J3a5.£ >- a ° cj E. O to O hJ Uu O^^OO OOOmz o^cc^oi OJz^Zlz; !?Of»OZ !z;^;^;^;^; ^.OO^D a. a. a C ■ ss :Ph c c c - _ -a >, >» > o o o o n a cs njCC*-'- ^-S!5 , ^=;_t;~ — = j; a a c e G"ri"r, 033J2-C J32T)u3 3 3 « « * fc t! b ii3 u2 J^Scil MPQP300 OOWPhPm ^PhCCO OOOKW K >->>-> X ' o _ n pi « t «s co n ce oc o co r- oo os © REQUIREMENTS OF COKING COALS 63 OSHlflffl OMONOC NMON-t O "*T 00 OO to rnOrtfflrt °-29S2 cc **• ^ S 2S HOOH91 OHClrtO i-h ■* pq ^ ^-l i-HO-^OO r-H .-c CM ~h ^ NOOOM CO -'J 1 -* 1 CM t— CO C3 t~- CM NNO>00 CO CM -*P >0 -"t 1 ■"*< t- CM O -* 1 tfOOlNO MNhO* SNOB O-f ©ON O OO i-H ■* CO M«D-*iOt»i 00 00 ■* 00 CO QOTfSOlO i— i lO t- 1 t— r- 1 NClO^ c^j cm cm — i co cm ■ co os -*t< o oo • cm oo cm • o cm CM CO CM O • OO CM • OHff) -CM 'HOHN ! • t- 1 O O <-H © -(NO CM -OO tooooo • os oo -«o • oo co co -co ■ no to r- to . ost~i>- co oo .ost- o -toto mwjioio • «5«5 • »o • iflioio • >o -iflioxsio • ws «o uo >« "O . m m co • uo uo OOC* • CO^fn -co • j-^oq,— i - (NOONNtD OOONHrt OJ^CONo CM lO uO t^ CO COM-* C-l N O) (M C-l -h CM CM CM CO CMCMoqcNIoq HNMNN jsi CM CM ^ CM NNNNCC CM CO CO CM !Dt^Tioiioio usinujiom mmoiom uo ^c >o «o 10 io «o >o "5 "5 >o »o co uo N!DO>«5m C35 O t-i CO — I *— I ,— I uo O t~- OhmOio ^ CM CM b- M< UO CM -* t- 1 OS HUJNOO NCOOitDo OS tr~ i— I t^ 00 NrHiotOtO t~- OS O OO 00 NCOK5W5N KJCSO® N CO >— I O CO ^"*^t<-*flO -*l^tlU0-*l^l -^^t 1 -*!-*-*! -f -tf< -**< ^ Tf -# "f "f ^f -* if^U}'*'* -^-ttl-^l^t 1 P H i H .»0 COONno HOOHStD NONrt» cdON'-IH *!OTj(tDK W5K)®N CM CO -5 © to Hrt«5rnN H CO O © ©' CMCOCOCMCM CO © OS © CM CO CO kO CM 00 © CO c© O -^-rf^'^^i Tr^f^f-rji^t 1 **cri co -*f -^f -*t< **t< ^F co •**< -*f -^t 1 "■cf co "^ ***• -*f -*f --^ ■**< co "ct< co co "*f i— I CO CO OS ifj t^ CM t— 10 -* HNHNw HtOcctOm HiO(N!ON (OH|>NlC «5 ifl Tf M OW-^Ouj -r-it^COlDlO tONCCNtD t^ -* 1 -«f tf5 IC 00 t^ 00 »^ CO 00 K5 ■* ■* i-i CO — I CO ti -* tji -^ ^ ^1^5^^,^, ^1^,^-,^,^, ^,^^,^1^, ^i ^ ^^ ^t, u«, minifjin OtONM M t-lOOCOOOCO COOOOHtH CCNNCOX CMi—iO'OOO NINcO ^ OS 00 ifi CM -F "* -* -# ;r5 •'fcomiji-* CO i— I CO »C ^F >o ■* CM ■* -<*i in CO tF ■* CO CM CO CO ■* CO *-H CO cm cm OS CO to CO >j0 t^ OS CO 00 CO XincOCOr-l N ■* O M ffl ® » rt M N OO CO OS CM 00 © © © *-t rf rt H M oo CO O © i-l 0 i-l 00 CO CM CO lO -* 1 O O »JO CO CO CO O no "3 IC tO CO tO tF O O O O CO»-ltD-*cO CM »C i— I t— lO ^f CO CO O 00 U5 t> 64 TECHNOLOGY OF COKE MANUFACTURE presented in Table 11, the original data for which come from or are derived from Table 4 of Bulletin 62 of the Illinois State Geological Survey. 3 The ratios of fixed carbon to volatile matter for Illinois coals as shown by county-average analyses are shown in Table 11 and figures 13, 14, 15, and 16. County average ratios of fixed carbon to volatile matter of 1.4 or greater are found only in the southern part of the Illinois coal field in Franklin, Gallatin, Jackson, Jefferson, Perry, Saline, and Williamson counties. On the basis of the ratio limits given in this specification, Illinois coals are at the lower limit, some just managing to be included. Illinois coals as they exist in the bed all have higher moisture contents than 4.0 per cent. (See fig. 17.) The ash require- ment of 9.0 per cent maximum can be easily met by careful preparation and cleaning. The county average ash values are given in figure 18. Illinois coals are less fortunate as regards sulfur. County average sulfur values are shown in figure 19. A detailed map of the low sulfur area 4 is shown in figure 20. If we assume that the sulfur content of the coke will be 0.8 of the sulfur content of the coal (see p. 127), coals having a sulfur content of less than 1.3 per cent and 1.6 per cent will be required to produce cokes having respectively sulfur contents of 1.0 per cent or less and 1.3 per cent or less. Coals having sulfur contents of 1.3 per cent or less are found in Illinois only in parts of Franklin, Jackson, Jefferson, and Perry counties. These counties also contain the only coals having sulfur contents of 1.6 per cent or less. For most non-metallurgical uses, the sulfur limit need not be so low or so rigorously adhered to as set forth in the A.S.T.M. specifications. 3 Cady, Gr. H., Classification and selection of Illinois coals: Illinois State Geol. Survey Bull. 62, 1935. i Cady, G. M., Ijow sulfur coal in Illinois: Illinois State Geol. Survey Bull. :»8, fig'. 58, p. 432, 19*22. REQUIREMENTS OF COKING COALS 6S Fig. 13. — County Average Carbon Ratios for Coal Nos. 1, 4, and 7 and Assumption Coal in Christian County. 66 TECHNOLOGY OF COKE MANUFACTURE Fig. 14. — County Average Carbon Ratios for Coal No. 2. REQUIREMENTS OF COKING COALS 67 Fig. 15. — County Average Carbon Ratios for Coal No. 5. 68 TECHNOLOGY OF COKE MANUFACTURE Fig. 16. — County Average Carbon Ratios for Coal No. 6. REQUIREMENTS OF COKING COALS 69 -j .DAVIESS STEPHENSON I WINNEBAGO | § , MeHENR A LAKE / - L r-yM J_\ (^ | DEKAl - B | " \|oupage| o \ /HITESIQE^f" lee' I y|_ O \ / vP^ ; ^ J t7^"7nvJ J jfcl , 1 UCENDALLl J I ^_^V^|-I5.9T 2 -I6.I J2-I4.54.JJ-/ 1 I r ~Z^r*. . x ?_ J HENRY BUREAU / T ,/ w I l l ' J*tBj| 2-14.5 L.,7.7 J^^jLigfKi^Aj mercer [6-I9JJ 1 ^( N am|6-|3 2 6 " l4 -°r^v J V-|5.A i ^^f7TARK|2--S-L: '-i— J K.W^| (J [li ox 6-1751 ffi s "f"-, I L I 1 // WARREN W^Oy^A.S/ 1 — ^H —TON H f j / v """" PEORI a/ WOODFORD I c_ | | Q k~^ / I-I3.2I6H79 6-15.9} — |2-i4.5 r L ^-U* y I .«oouoT1 ' ' -/ Vt„.3 L r J L i TAZEWELL McLEAN F0RD ' 1 M , AI J vermilJn 7-iaij DOUGLAS r J ^-1 EDGAR I MASON CHAMPAIC jersey 6-13.3 6-13.2 r-- MAOISON -13. 6-1 - L T' I c L *^_Y r,chland 6-11.3 ST. CLAIR , MARION ■ CLINTON .^ I J6-I0.3] WAYNE LAWRENCE WASHINGTON 6-10 ~ 10. .-I0.7FI IDOLPH PEWY -I03fe: 1-9. Scale iO 10 20 30 40 ".Of Fig. 17.— Coal Bed Numbers and County Average Moisture ("As Received") of Illinois Coal. 70 TECHNOLOGY OF COKE MANUFACTURE Fig. 18.— Coal Bed Numbers and County Average Ash (Dry Basis) of Illinois Coals. REQUIREMENTS OF COKING COALS 71 Fig. 19. — Coal Bed Numbers and County Average Total Sulfur (Dry Basis) of Illinois Coals. 72 TECHNOLOGY OF COKE MANUFACTURE R1E Scale in miles Coal containing Coal containing less than 1 per less than 1.25 per cent sulfur. cent sulfur. Fig. 20. — Map Showing Location of Low-Sui.fur Coal in the Franklin-Willia and Big Muddy Districts. The Big Muddy District in Jackson County has been worked out. {("ady, C. //., Op. MSON cit.) REQUIREMENTS OF COKING COALS 73 CLASSIFICATION OF ILLINOIS COALS WITH REFERENCE TO COKE- MAKING PROPERTIES ON THE BASIS OF THEIR PROXIMATE AND ULTIMATE ANALYSES Systems for classifying coals in groups having similar properties on the basis of their analyses or the results of various tests applied have been proposed as aids in the selection of coals for given purposes. Such classification systems have been found to be of considerable practical value and are being used to an increasing degree. Some of the proposed classification systems were applied to Illinois coals in an effort to determine the relative positions of Illinois coals with respect to each other and to other coals. The various classification systems which have been proposed are not of equal value in evaluating coals for the production of coke, and in some cases it is not certain whether the system will evaluate coals for coke making. An attempt is made to indicate the value of the classification systems to be discussed for purposes of estimating the coking characteristics of coals. Rank of coal. — The position a coal occupies in the series from peat to anthracite is known as its rank. Increase in rank in coal is due to various geological factors which are not discussed here. Progressive change in the rank of coal causes a progressive change in properties, other things, such as type, being equal. Coals of the same type and the same rank can, in general, be assumed to have the same properties. The lower ranks of coal do not fuse and form a coherent cellular coke. This property appears with the bituminous rank and is one of the characteristics which distinguish bituminous from sub-bituminous coals. The coking characteristics of coals increase with rank through a maximum in the low volatile coals and disappears again in the anthracite rank. With few exceptions, Illinois coals fall in the high-volatile B (moist mineral-matter-free calorific value of 13,000 or more, and less than 14,000 B.t.u. per pound) or high-volatile C (moist mineral-matter-free calorific value of 11,000 or more, and less than 13,000 B.t.u. per pound) bituminous groups of the A.S.T.M. tentative specifications for classification of coals by rank, D 388-36 T. 5 To be placed in group C, rather than in the lower sub-bituminous group, coals must be either agglutinating or non-weathering. To be agglutinating they must form a coherent coke button during the standard volatile matter test. All Illinois coals so far tested are agglutinating coals. They are placed in the bituminous rather than in the sub-bituminous rank by virtue of their estab- lished coking quality. Coals of the high-volatile B group are found only in the southern part of Illinois. Some coals belonging to the high-volatile A bituminous group (having • r > Amer. Soc. for Testing- Materials Tentative Specification for Classification of Coals by Rank, A.S.T.M. Designation D 388-36 T., A.S.T.M. Tentative Standards 1936, pp. 520-26, 1936. 74 TECHNOLOGY OF COKE MANUFACTURE a dry mineral-matter-free fixed carbon content of less than 69 per cent and a moist mineral-matter-free calorific value of 14,000 or more B.t.u. per pound) 2 . . V • Chart of typical coals of the United States Symbols • Caking or agglutinating • Noncaking a No information concerning caking properties •< Qd. »• % % * *( E£ 90 • • 1 86 • Low_ /ol. bit rcent 00 •' I ■„• ri 78 A , V E~ a =j x> o 74 ■o _: -o «g i_ a, 70 r|— £ £ 66 < E I I 62 * ;.» • • • .»*« •• /" .5? •= X c t 58 •*• > ■ • ' •• • •, *' ■■ • 54 • ' ••• .»< 1 •; ■ « • . m ^. • • ■ 1 .* ■ 1 • 'V • • •• •m • n ■ ■ • • 1 50 • < t • • • 46 • 42 i -i — 16, ooc ] 4,C )00 12, OOC 10, DOC 8,0 00 6,0 00 Moist, mineral-matter-free B ibbit. JSu E bii bbil •* " B ***" C Hi.-vol.lHi.-vol. C bit.| Subbit. ISubbit, High-vol. A bit. -hi-- L!1 -4^r .tu.l '"4-u Lignite Fig. 21. — Classification Chart of Typical Coals of the United States. (From Fieldner, Sclvig, and Frederick, Op. cit.) are found in a small area in southern Gallatin and southeastern Saline county, known as Eagle Valley, where there are no shipping mines. There may be small bodies of such coal at other places along the southern margin of the coal basin. ,; 6 Cady, C. II., Classification and selection of Illinois coals: Illinois State Geol. Survey Bull. 62, p. 33, 1935. REQUIREMENTS OF COKING COALS 75 Rank may be indicated graphically on a chart upon which fixed carbon on the dry mineral-matter-free basis is plotted against the moist mineral-matter-free calorific values of a coal, as has been done by Fieldner, Selvig, and Frederic 7 (fig. 21). The plotting of a large number of values representing a wide variety 70 66 64 62 60 > ^ X 9 •X X :: m 56 54 52 50 48 46 44 _x ■ • A •* ■# ? • i D x P {(A * ■>. O P LEGEND O NO. 7 COAL # NO. 6 COAL X NO. 5 COAL ▲ NO. A COAL ■ NO. a COAL &L NO. 1 COAL X i ft 42 40 16000 12000 15000 14000 13000 I I MOIST MINERAL-MATTER-FREE B.t.u. 1 1000 HIGH-VOLATILE A BITUMINOUS HIGH- VOLATILE B BITUMINOUS HIGH-VOLATILE C BITUMINOUS Fig. 22. — Classification Chart of Illinois Coals According to A. S. T. M. Tentative Specifications for the Classification of Coals by Rank, D 388-34T. of coals results in a chart upon which the positions of the coals lie within a fairly narrow band in the diagram. In this diagram the position of the high volatile bituminous coals should be observed, as this is the position occupied by Illinois coals. An enlarged portion is used to plot the position of the county 7 Fieldner, A. C, Selvig, W. A., and Frederic, W. H., Classification chart of typical coals of the United States: U. S. Bureau of Mines Rpt. of Inv. 3296, 1935. 76 TECHNOLOGY OF COKE MANUFACTURE averages of Illinois coals, first in such a way as to differentiate the coal beds (fig. 22), and then to differentiate the position in the state, northern, central or southern (figs. 23 and 24). 70 68 z 66 Ul O a. UJ 64 Q- Z o 62 dJ (XL < 60 O Q LJ X 58 U. UJ u 56 - 46 CJL a 44 42 40 i } • ¥ •* • *3 • ^y 3 r o ©x ^ xcy X c! r x JO 0^ 9 LEGEND £ SOUTHERN ILLINOIS O NORTHERN ILLINOIS X CENTRAL ILLINOIS X ( S 16000 15000 12000 14000 13000 I I MOIST MINERAL-MATTER-FREE B.L.u. 1 1 000 HIGH-VOLATILE A BITUMINOUS HIGH - VOLATILE B BITUMINOUS HfGH-VOLATILE C BITUMINOUS Fig. 23. — Classification Chart of Illinois Coals, by Geographic Location, According to A. S. T. M. Tentative Specifications for the Classification of Coals by Rank, D 388-34T. Coals from the same parts of the state show a greater tendency to have the same rank than do coals from the same bed irrespective of their location. The Illinois coals which have received the most attention as possible coking coals have been from the southern counties and as a group are the coals of highest rank in the state. REQUIREMENTS OF COKING COALS 77 Ultimate items. — When ultimate coal analyses are plotted on a triangular diagram, the coals arrange themselves across a part of the diagram in a narrow 1 '~v~' OXYGEN 5 JO IB 20 .<, ^ Bituminous "coalV ! s^^TUM^Uey; :^.ign it%S IOO 95 90 85 SO «H> CARBON 75 70 Fig. 25. — Three-component Diagram of Ultimate Analyses of United States Coals. (From Rose, Op. cit.) % OXYGtEN A 3 6 7 IO HIGH VOLATILE * (f COKING COALS. GAS COALS O z 111 h x 0^ Jl+" MEDIUM VOLATILE COKINGc COALS .1 /j - V J LOW VOLATILE • COKING COALS "*"• V + . + * / J: I30 TYPICAL COKING COALS OF U.S> CAl.COI.ATED FREC OP MOISTURE, ASH, SuL" FUR. AND NITR.OC.EN , KEY PENNSYLVANIA WEST VIRGIN \A KENTUCKY VIRGINIA ALABAMA ARKANSAS ILLINOIS,' INDIANA OHIO / UTAH/ COLORADO WASHINGTON S3 55 h 9 ir 1 — w -W 87 56 % CARBON as 63 82 Fig. 26. — Three-component Diagram of Ultimate Analyses of 150 Coking Coals of the United States. (From Rose, Op. cit.) REQUIREMENTS OF COKING COALS 79 useful if the analyses have first been calculated to a dry mineral-matter and sulfur- and nitrogen-free basis. H. J. Rose has published such a diagram show- ing 150 coking coals of the United States, with the states of origin of the coal indicated. 8 This diagram is reproduced in figure 26. Figure 27 is a similar plot showing the positions of the county average ultimate analyses of the various coals of Illinois for which analyses are available. 9 These analyses are also shown on an enlarged scale and differentiated according to coal bed in figure 28 and according to their geographic location in figure 29. Rose says: "All coking coals appear to fall within fairly definite chemical limits, so far as their ultimate o , 2 3 PER A CENT OXYGEN 5 6 UNIT COAL BAS 7 8 IS 9 10 „ 12 13 / / / / 1 / 1 / / / / / / / / / / 1 TZ 1 7Z / / / / / / / / 1 / 1 / / / / • /. 1 / If) / ..•*/ • / • / CD / 1 • / / .,-/ < O / / • \ T\ • / < • / t / I • / D / 1 / / O a. a i z 7_ a: / t / / / / 1 1 1 / / / / / / / ■ / / / / / 1 / / / / / / / / / / / 1 / / / / / / 89 88 87 86 85 PER CENT CARBON UNIT COAL BASIS Fig. 27.— Three-component Diagram of Illinois Coal Ultimate Analyses. analyses are concerned, and it is the writer's experience that from the location of a coal within these limits, the coking properties and by-product yields may be predicted with reasonable accuracy, if all essential data on the chemical and physical characteristics of the coal as it will be charged, and the operating con- ditions which will exist, are known." The location of the coals in the United States which Rose, on the basis of his experience, classes as coking coals are shown in figure 30 taken from his paper s and may be compared with the coal fields of the United States shown in figure 31. It will be noted that Rose included only the coals of southern Illinois and Vermilion County with the coking coals. He may have discriminated on the basis of sulfur and ash content, since his area in Illinois includes only the low sulfur and lower ash coals. His area excludes a large area of Illinois coal from which satisfactory domestic and industrial fuel coke mav be made. 8 Rose, H. J., The selection of coals for the manufacture of coke: Trans Amer. Inst. Min. Met. Eng. vol. 74, pp. 600-636, 1926. 9 Cady, G. H., op. cit. Table 4, pp. 314-26. 80 TECHNOLOGY OF COKE MANUFACTURE PER CENT OXYGEN UNIT COAL BASIS 10 O 6.0 84 83 82 PER CENT CARBON UNIT COAL BASIS Fig. 28. — Three-component Diagram of Illinois Coal Ultimate Analyses with the Coal Bed Numbers Indicated. Numbers beside points refer to lines in Table 11. PER CENT OXYGEN UNIT COAL BASIS 10 II 84 83 82 PER CENT CARBON UNIT COAL BASIS 60 Fig. 29. — Three-component Diagram of Illinois Coal Ultimate Analyses with Location of Coals Indicated. Numbers beside points refer to lines in Table 11. REQUIREMENTS OF COKING COALS 31 Fig. 30. — Mining Districts or Coal Areas of the United States Which Contain Coking Coal. (From Rose, Op. cit.) - x Fig. 31. — Coal Fields of the United States. From "The Design of Steam Generating Plants" Part II, "The Three Natural Fuels" by F. H. Rosenerants, published in "Combustion," May 1932, p. 37, probably redrawn from original in colors by M. R. Campbell, U. S. Geological Survey. 82 TECHNOLOGY OF COKE MANUFACTURE Specific volatile index. — For classifying coals, using the items of the proximate analyses, Burrough, Swartzman, and Strong have devised a new value which they term the "specific volatile index", 10 which is the calorific value in B.t.u. per pound of one per cent of the volatile matter in the coal. It is arrived at by subtracting the calorific value of the fixed carbon in the coal from the calorific value of the coal and dividing the result by the percentage of volatile matter in the coal, according to the following formula: S.V.I.= Determined B.t.u. in coal — ( 14,500 X weight of F.C.) Per cent of volatile matter tf L IG N m c 5U B-B ITU .4INI 3US A B ■ s C i V.'v, >i • ** , D . E 30 n F z C H NC IN- CO UN j < or. CO Kir IG 1 20 u 2 1 IN BY U -_l PR DDUiCT CO KE ov INS J o < AN K THRACI tc 1 100 110 120 130 150 160 170 180 190 200 210 220 230 240 250 SPECIFIC VOLATILE INDEX Fig. 32. — Classification of Illinois Coals on the Basis of Specific Volatile Index. When the specific volatile index is plotted against the volatile matter content of the coal, as expressed on the dry ash-free basis, a continuous band of varying width which includes all coals is obtained. The authors of the system have divided the plot so obtained into various areas which include coals with similar coking characteristics. Their chart, on which the Illinois coals have been plotted using values obtained from county averages published by Cady, 11 is reproduced here in figures 32 and 33 in which the area occupied by Illinois coals has been enlarged. In this system, as in the others, the points representing the various coals fall in a narrow band across the chart with the Illinois coals grouped closely together on the lower edge of the coking coals. This system does not differentiate coking coals from non-coking coals, some of the non-coking sub-bituminous coals from the United States and Canada lying in the same area with Illinois coals. 10 Burrough, E. J., Swartzman, E., and Strong-, R. A., Classification of coals using specific volatile index: Canada Dept. of Mines, Mines Branch No. 725, Fuels and Fuel Testihr 1930-31, pp. 36-50, 1933. 11 Cady, G. H., op. cit; REQUIREMENTS OF COKING COALS 83 60 55 50 45 40 > 35 s u B - B B 1 T U M 1 N o u s c • ■ ■ • f • • » x LEGEND O NO. 7 COAL # NO -6 COAL X NO. 5 COAL A NO .4 COAL ■ NO. 2 COAL 53 NO . 1 COAL □ ? COAL • ■ •• X • 121 X 1 1 1 20 125 130 135 140 SPECIFIC VOLATILE 145 150 NDEX 155 Fig. 33. — Classification of Illinois Coals on the Basis of Specific Volatile Index with the Coal Bed Numbers Indicated. CONCLUSIONS FROM THE APPLICATION OF VARIOUS CRITERIA FOR COKING COALS TO ILLINOIS COALS Definitions of coking coals and classifications which claim to provide definite criteria for separating coking from non-coking coals are based in general upon the average characteristics of coking coals. Since most coke has been prepared for metallurgical purposes, it is the characteristics of coals which produce that kind of coke that have provided the criteria for determining coking coals. Apply- ing the criterion that coking coals are coals from which a saleable commercial coke can be made, much of the coal in Illinois has definitely been shown to be coking coal. This is especially so in view of the growing market for non- metallurgical coke for which the requirements are different than for blast-furnace coke. The coals from southern Illinois usually occupy a more favorable position with respect to the established coking coals when plotted in the various systems of classification than do the coals from central and northern Illinois. The differences between the various Illinois coals are, however, small in comparison to the range of variations possible among the coking coals. Part IV — Experimental Investigation of the Production of Coke from Illinois Coals By Gilbert Thiessen and Paul E. Grotts CHAPTER IX— INTRODUCTION TO THE EXPERIMENTAL WORK In spite of the work which has been done toward the production of coke from Illinois coal, very little definite information is available, particularly con- cerning the comparative coking characteristics of coals from different parts of the state. Many of the tests from which results are available were made in apparatus which would no longer be used, for example, in beehive or special types of ovens. Many tests were made in connection with the development or pro- motion of new types of carbonizing equipment, the conditions and results of which are not available or only in a general way. Many economic and technical changes have also occurred since the problem of the production of coke from Illinois coals was last reviewed by the Illinois State Geological Survey. 1 It appeared desirable, therefore, to assemble a systematic body of data in regard to the coking characteristics of a variety of Illinois coals, at the same time having the advantage of newer experimental methods, for the sake of providing funda- mental information concerning the coking characteristics of these coals. Although these data provide general indications of the coking characteristics, much yet remains to be done before we can say that we thoroughly understand the mechanism of coke formation from any coal, not only Illinois coal. In determining the coking quality of a coal, small-scale experimental tests provide essential information not supplied by the standard types of commercial analysis. Proximate and ultimate analyses provide values on which probabilities can be based, as was shown in the previous chapter; but for border-line coals, particularly such as those found in Illinois, additional factors are important. That is to say, differences among coals cannot be completely revealed by the standard types of analysis. This is particularly true with respect to thos-e differences arising from variations in the physical constitution of the coal, known as type differences, since variations in chemical constitution have not been sufficiently well correlated with such differences. Experimental tests are, i Ovitz, P. K., Coking of Illinois coals: U. S. Bureau of Mines Bull. 138, 1917. [85] 86 EXPERIMENTAL INVESTIGATIONS therefore, necessary to determine the exact behavior of a coal, a condition which will continue until the relationships between type differences and the behavior of coals under the influence of heat are more definitely understood. Because of the expense involved, full-scale tests are usually out of the question even if ovens were available. There is, however, some compensation in the use of small-scale tests in the fact that such tests permit much greater freedom in controlling conditions of operation and, therefore, in determining the extent to which the coking characteristics of a coal may be modified, at least in a general way, under a variety of conditions. On the olher hand, it is well to keep in mind the statement by H. J. Rose 2 that ". . . the only thoroughly reliable method to determine the quality of coke that can be made from a coal under particular conditions is to subject the coal to a full-scale oven test under the desired conditions." 2 Rose, H. J., the selection of coals for the manufacture of coke: Trans. Amer. Inst. Min. Met. Eng-. vol. 74, pp. 600-636, 1926. (Especially 604.) CHAPTER X— METHODS USED IN THE EXPERIMENTAL WORK SELECTION OF METHODS Some methods for testing coal, particularly those used in the proximate and ultimate analyses, have been highly standardized and widely used. A high degree of standardization of methods is essential in coal testing if the results from different laboratories and from tests at different times are to be comparable, since most of the methods are highly empirical. Most of the methods used in studying the coking properties of coals have not been so standardized, partly because they are relatively new and untried as only a few laboratories have use for them, and in many cases the essential factors in the methods have not yet been discovered. The investigator must, therefore, select his methods cautiously. One of the factors considered in the selection of the methods used in this work was the extent to which they had been adopted and used in the United States. Those adopted include methods set forth in the standard and tentative standard methods of the American Society for Testing Materials, in the "Methods of the Chemists of the United States Steel Corporation for the Testing of Coal, Coke and By-Products, " and in the reports of the United States Bureau of Mines on the work in cooperation with the American Gas Association on the coke, gas and by-product making properties of American coals. CHEMICAL ANALYSES The methods for the chemical analysis of coal have been well standardized in the United States. Methods used in this investigation were the standard methods as set forth in the A.S.T.M. Standard Methods of Laboratory Sampling and Analysis of Coal and Coke, A.S.T.M. Designation: D 271-30. These methods were carried out in strict conformity with the methods as set forth, both as to procedure and equipment. As a check on the accuracy of the work, samples were periodically exchanged with the Coal Analysis Laboratory of the United States Bureau of Mines at Pittsburgh. The proximate analyses, calorific values and sulfur contents of all the coals used in these tests were determined in order to evaluate them as to quality and rank and to make easier their comparison with other coals. [87] 82 EXPERIMENTAL INVESTIGATIONS TESTS INDICATING THE BEHAVIOR OF COAL UNDER THE INFLUENCE OF HEAT Coal when heated loses first moisture, then a portion of its volatile matter, and then it softens and becomes plastic, at which stage decomposition increases greatly and volatile matter is liberated. In the plastic stage the individual particles may completely fuse together, may merely sinter (stick together), or may remain completely free. The behavior of the coal while plastic is dependent upon whether or not the particles fuse together. When complete fusion occurs at the same time that a considerable proportion of the volatile matter is given off a much swollen coke is produced which, if permitted to expand freely, may be several times the volume of the original sample. If fusion does not take place the charred product will be smaller in volume than the original coal, due to loss of volatile matter. As heating is continued the sample loses more volatile matter and finally reaches a temperature at which the plastic coal, regardless of whether fusion has occurred, sets to a solid. Further heating of the coke results in its devolatilization and in a slight shrinkage. The temperatures at which the coai particles are most plastic and the temperatures at which the coal decomposes most rapidly may not be the same. Their relationship to each other is an important factor in production of coke from a given coal. The temperature range during which a coal is plastic is also important when blends of coal are coked, because the various coals in the blend should be plastic at similar temperature ranges. From the standpoint of operating practice, the yields and quality of the coke, gas, tar, light oil, and other by-products of those coals which will produce coke are important. From the standpoint of the present investigation the ultimate interest concerns whether or not coke of quality suitable for use as a domestic fuel may be produced from a given coal. Certain conventional tests are used in judging the quality of coke. On the basis of the above considerations, tests were selected to yield the following information : ( 1 ) The yields of coke, gas, tar, and other by-products obtainable from a given coal, together with some information concerning the analysis of the coke and gas (2) The temperature at which the coals tested became plastic (3) The temperatures at which the plastic coals set to form coke (4) The coking or agglutinating power of the coals, that is, the ability of the individual coal particles to form a coherent mass (5) The quality of the coke obtainable under experimental conditions from a variety of Illinois coals and blends of coals (6) The influence and behavior of various impurities in coal on the quality of the resulting coke METHODS USED 89 Gas, coke, and by-product yields. — Coke and by-product yields were determined using a procedure and equipment adopted by the laboratories of the United States Steel Corporation' 1 with a few modifications developed in the course of this work to facilitate duplication of results. A detailed description of the apparatus and method as used and the pro- cedure adopted is given in Appendix A (pp. 209-223). Transparent fused quartz combustion tubes were used in place of the hard glass combustion tubes, as it was found difficult to obtain glass which would retain its shape at the required temperature of 900°C. The quartz tubes were exactly the same dimensions as specified for the glass tubes. One alteration of procedure consisted of the introduction of a snugly fitting piece of transparent fused quartz tubing 128 mm. long into the open end of the combustion tube. Removal of this tube with the tar deposited in it took the place of cutting off this portion of the glass tube in the ordinary procedure. The tar collecting between the two snugly fitting tubes was recovered with negligible loss. The inner tubes projected into the tar filter. The directions call for heating the portion of the charged combustion tube containing the crushed silica brick to the constant temperature of 720°C. It was found difficult to maintain this temperature uniformly with gas heat. A small electric heater was built to take the place of the first four burners. The use of this electric heater greatly simplified the operation of the equipment. Four thermo-couples were used to measure and automatically record on a 4-point recording potentiometer pyrometer the temperatures in the electric furnace and at the two ends and center of the coal charge. Light oil was recovered by the freezing tube method using a mush of acetone and dry ice to produce a low temperature. When these tests were first started the laboratories were supplied with artificial gas which produced a hot concentrated flame. Shortly after the tests started, however, the gas supplied was changed to natural gas which gave a slow flame that made it difficult to localize the heat. To overcome this difficulty arrangements were made for mixing air and gas before the mixture was passed into the burner manifold. Both the air and gas lines were equipped with orifice flow meters. The proportion of air and gas which would give the required rise of temperature to 900°C. in ten minutes was found by trial and this proportion was maintained during subsequent tests. The mixture of air and gas had burning characteristics similar to the artificial gas previously supplied. The greatest difficulty encountered in carrying out these carbonization tests was in obtaining tar yields comparable to those obtained in by-product oven practice. In the results presented in another section it will be noticed that the tar yields are low bv about one-half. 3 Methods of the chemists of the U. S. Steel corporation for the sampling- and analyses of coal, coke, and by-prodncts, 3rd Edition, 1929. Dry distillation test for coal, pp. 130-143. 90 EXPERIMENTAL INVESTIGATIONS Softening and swelling characteristics. — The apparatus and pro- cedure used for determining softening and swelling properties of the coals which include a determination of the initial softening temperature, the initial active decomposition temperature, the temperature of maximum swelling, the temperature of volatilization, and the temperature range during which the coal Fig. 34. — Agde-Damm Apparatus. was plastic was that designed by the United States Bureau of Mines after Agde and Damm, 4 illustrated in figure 34. A detailed description of the apparatus and procedure as will be found in Appendix B (pp. 225-226). The relation- ship between the plastic temperature range and the liberation of volatile matter from the coal was determined as follows: 4 Fieldner, A. C., Davis, .1. l>. ( Thiessen, R., Kester, 1<:. B., and Selvig, W. A., Methods and apparatus used in determining (he gas, coke, and by-product making properties of American coals: (J. S. Bureau of Mines Bull. 344, pp. 14-16, L931. METHODS USED ( >i A support was constructed so that a Westphal specific gravity balance could be supported with its hook directly above the opening of a standard coal volatile matter furnace. The standard volatile-matter platinum crucible was suspended centrally in the furnace from a long stirrup attached to the balance hook. Con- vection currents were minimized by covering the furnace opening with a series of perforated mica plates. One gram of coal was placed in the crucible, the capsule Fig. 35. — Apparatus for Determining Volatile Matter Loss. lid firmly put into place and the crucible suspended by the stirrup in the cold furnace. By means of weights the beam was put into balance. The temperature of the furnace was gradually raised at the uniform rate of 6°C. per minute. At intervals of one minute the beam was put into balance and the loss of weight determined. The temperature in the furnace was measured at the same time by means of calibrated chromel-alumel thermocouples. From these readings a graph was plotted which showed the evolution of volatile matter with tempera- ture change. The apparatus is illustrated in figure 35. 92 EXPERIMENTAL INVESTIGATIONS Agglutinating* or caking power. — The caking or agglutinating power of a coal is its ability to cement itself and inert material into a coherent mass upon carbonization. To measure this property the finely ground coal must be diluted with a sufficient amount of suitable inert matter until only the cementing power of the coking coal is effective. Insufficient dilution permits the formation of coke cells w T hich cause anomalous and untrustworthy results. A number of such test procedures have been described in the literature. 5 This type of test has been extensively studied at the U. S. Bureau of Mines fi with the result that a modification of the Marshall-Bird test was devised and published by the A.S.T.M. as a proposed method. 7 Wider use of this test brought to light the fact that the values obtained by use of the test are very de- pendent upon the surface condition of the sand grains used as inert material and that it is practically impossible to supply a sand which would be satisfactory except at a prohibitive price. Accordingly, results by various laboratories are comparable among themselves as long as the same batch of sand is used. Values varying two-fold have been obtained using different sands. The Marshall-Bird test 8 was first used in the study of Illinois coals until the Bureau of Mines modification was developed and adopted. This method as used is described in detail in Appendix C. Coke quality tests: retort for carbonizing two- to three-kilogram charges of coal. — As only small amounts of coke were obtainable from the United States Steel Corporation by-product apparatus, it was necessary to con- struct a special retort to yield a sufficient quantity of coke from which an estimate of its quality could be obtained. A cylindrical steel retort was con- structed from a piece of 6-inch steel pipe 9 inches long. A sheet of steel of the same thickness as the walls of the pipe was welded to one end to form a bottom, the other end of the pipe was threaded to take a 6-inch pipe cap. Two holes were drilled in the cap cover, one centrally located to take a 1-inch pipe fitting and the other 2 inches from the center to take a >j/J-inch fitting. A one-inch nipple carrying a pipe cross at the other end was screwed into the centrally located hole. 5 Anonymous, Methods of analysis of coal: Dept. of Sci. and Ind. Research, Great Britain, Fuel Research, Physical and Chemical Survey of the National Coal Resources, No. 7, pp. 10-12, 1927 (reprinted 1929). Gray, T., The determination of the caking power of coal: Fuel, vol. 2, No. 3, pp. 42-45, 192:',. Kattwinkel, R., The examination of coking coals and estimation of their value: Fuel vol. 5, No. 8, pp. 347-355, 1926. Marshall, S. M., and Bird, B. M., Test for measuring- the agglutinating power of coal: Trans. Amer. Inst. Min. Met. Eng. Coal Division 1930, vol. 88, pp. 340-388, 1930. Marshall, S. M., and Bird, B. M., Agglutinating, coking and by-product tests of coals from Pierce County, Washington: IT. S. Bureau of Mines Bull. 336, 1931. <> Selvig, W. A., Beattie, B. B., and Clelland, J. B., Agglutinating value test for Coal: I 'roc. Amer. Soc. Testing Materials, vol. 33, pt. II, pp. 741-760, 1933. t Report of Committee D-5 on Coal and Coke. Proc. Amer. Soc. Testing Materials, vol. 34, pt. I, pi). 457-, 1934. 8 Marshall, S. M., and Bird, B. M., op. cit. METHODS USED 93 Pieces of pipe screwed into the horizontal openings of the pipe cross served to support the retort in the gas-heated pot furnace and to carry off the carbonization gases. Stuffing boxes were fitted into the top of the cross and into the ^4-inch opening to hold fused quartz thermocouple tubes in place. Temperatures in the retort were continuously recorded during the course of the test by means of a recording potentiometer pyrometer and chromel-alumel thermocouples in the thermocouple tubes so that the couple was about half way down the charge. No effort was made to recover the gases or other by-products. Several of these retorts were kept on hand so that one could be charged while another was being heated. Coke shatter tests. — The resistance of the coke to shattering was deter- mined by means of equipment similar to the A.S.T.M. shatter test for coke but made smaller in order to be comparable to the size of the sample. The box holding the coke was 12 inches square and 6 inches deep fitted with two hinged doors which open from the center upon the release of a catch. The coke was dropped 6 feet onto a steel plate. Before the test the coke samples were sized on three screens having square openings of 2 inches, 1 inch, and 1/2 inch. The shatter test was made on the coke retained by the 1-inch screen, and after each of the four drops the coke was sized and the fractions weighed. From these weights curves were drawn which showed the degradation of the coke. CHAPTER XI— SAMPLES USED IN THE INVESTIGATION In the course of the work carried on in the present study of the coke making properties of Illinois coals, samples of coals from 18 counties in Illinois were used. The various tests to which the samples of coal from the various counties were subjected are indicated in Table 12 which also indicates where the results of the various tests may be found in the report. In almost all cases samples from several places in each county were used. The tests indicated as having been Table 12. — Tabulation of Tests Applied to the Illinois Coals Investigated and Location of the Results in the Text (X indixates that test was applied) County a Coal No. Dry distilla- tion Caking value Plastic range 6-inch retort b Basket test in Knowles oven Bureau 7 6 6 7-Streator 6 6 6 5 6 6 6 6 5 6 6 7 6 6 2 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Christian Franklin Grundy Henry X Knox Macoupin Menard Montgomery. . . . Perry Randolph St. Clair Saline — Stark Vermilion Vermilion Washington Williamson Woodford — a More than one sample from each county may be represented. b These coals were also used in blends as well as separately. Test Results discussed on pages — Results presented in Table No. Figure No. Dry distillation Caking value Plastic range 6-inch retort Basket in Knowles oven 99 111 109-111 141-165 166 13 15 15 24, 25, 26 27 38 - 40 47 - 53 55 - 60 [95] 96 EXPERIMENTAL INVESTIGATIONS made on samples of coal from any one county were not necessarily, and probably were not, made on the same sample. The reason for this is that in many cases the various tests were in use at different times during the time the work was ZV^ i -JEWELL .. , c . u ' F0 |aCHUVLEH ^Z^ ~_H L OGAN ° EWITT / CHAMPA.GN \™»r2 brownS c * s s ^J- 1 -^ J P,ATT ' I MORGAN ( SAN <3*"ON ^ | ^ [ C DOUGLAS [ OULT/llEI H EDGA " ) L 1 ^__| c^,a„ , ^jCOLES, ; rl GREENE ' ' r jSHELBV rrrJcuARK^ }§\ -H MACOUPIN | MONTGOMERY | |DUM="' I MACOUPIN I MONTGOMERY pUMBERLANDI f /^V : EFFINGHAM . N 1 f ^1—1 FAYETTE JASPER i L BOND I 1 10N I I I 1 | CLA ' 1 ' I MARION I | cuntonI j r"tc3Hi-—i/k OE^l \W I JEFFERSON I Hams-Is I Fig. 36. — Location of the Mines from which Samples were Taken, and the Tests which were Applied. 1. Dry distillation (coal carbonization assay) 2. Caking power (agglutinating value test) 3. Plastic range (Agde-Damm test) 4. Small scale coking test (six-inch retort) under way. Since these Eastern Interior basin coals are rapidly influenced by exposure to air and since most of the tests must be made on fresh samples of the coal if the results are to be used in evaluating the coals for coke making, samples SAMPLES USED 97 obtained for one series of tests were usually too old or changed by exposure to be suitable for later tests. The uniformity of Illinois coals over geographical areas of considerable magnitude permits a considerable latitude in the selection of samples which may be considered equivalent for many purposes. The locations of the mines from which the samples used in this investigation were taken and the tests to which these samples were subjected are indicated in figure 36. CHAPTER XII— COKE, GAS, AND BY-PRODUCT YIELDS The coke, gas, and by-product yields were determined on samples of No. 6 bed coal from 15 mines and for samples of No. 5 coal from two mines. The proximate analyses, calorific value, rank index, 1 and results of the carbonization tests for these samples are given in Table 13 and the locations of the mines in figure 37. VARIATION IN YIELDS WITH GEOGRAPHIC LOCATION Three lines can be drawn so that most of the samples studied lie along or close to them. One of these lines A- A' (fig. 37) is drawn in a northwest-south- east direction through Belleville in St. Clair County and Harrisburg in the center of Saline County, another, B-B', in an almost north-south direction from Springfield in Sangamon County to Marion in the center of Williamson County, and the third, C-C, from Taylorville in Christian County to Belleville in St. Clair County. Distances along these lines have been used as abscissae against which the rank indices, yields of ash-free coke in per cent of ash-free coal, and yields of gas per ton of ash-free coal expressed in therms have been plotted as ordinates in figures 38, 39 and 40. Certain trends are evident from these graphs. Along the line A-A' the rank indices and coke yields increase and gas yields decrease in going toward Franklin County. The coke yield increases and the gas yield decreases greatly as the DuQuoin anticline is crossed in the eastern part of Perry County. Samples 13 and 14 are from the No. 5 coal and are not strictly comparable to the others. Along B-B' (fig. 37) the rank index rises slowly but regularly as does coke yield, while the gas yield drops sharply after the DuQuoin anticline is passed at the western edge of Marion County. From north to south along C-C' (fig. 37) the rank index rises very gradually carrying with it a very gradual increase in coke yield and a decrease in gas yield. The well known improvement in rank of Illinois coals in passing from west to east across the position of the DuQuoin anticline and as one continues southeast reveals itself likewise in the results of the determinations of the yields of coke, gas, and by-products. Inasmuch as the more valuable and probably highest grade portion of these coals is rapidly nearing exhaustion, supplies of coal for coke-making purposes will before many years have to be obtained from l Cady, G. H., Classification and selection of Illinois coal: Illinois State Geol. Survey Bull. 62, pp. 29-30, 1935. [99] 100 EXPERIMENTAL INVESTIGATIONS lower grade or lower rank coals or both. It is important to note, however, that higher gas yields in terms of therms per ton characterize the lower rank No. 6 coal found west of the DuQuoin anticline, although coke yields are lower. Fig. 37. — Locations of Mines from which Samples were Tested for Coke, Gas. and By-product Yields. Graphical comparison of coke yields with rank indices and with fixed carbon contents shows that there is only a general trend towards increasing coke yields with increase in fixed carbon content or rank index. COKE, GAS, A\D BY-PRODUCT YIELDS 101 72 71 rt 70 < m 69 Ul u. 68 to .-6 < 67 66 4 * 58 57 Jt 1 / 1 — ^ G*>- ^ 1 / 54 53 52 r^ """ ySC \ if / / \ / _ COKE S I 1 65 51 1-40 130 < 120 no 100 J I 140 " \ 120 1 10 100 \ MILES 10 20 30 4-0 50 60 70 f 1 80 90 100 1 CARBONIZATION SAMPLE 7 8 NUWRFR 9 II 13 15 • 14 A CROSSES B-B' A Fig. 38. — Gas, Coke and Raxk Indices aloxg the Line A-A' (Fig. 37). COMPARISON OF RESULTS OF TESTS BY SEVERAL LABORATORIES OX XO. 6 COAL FROM OR XEAR ORIEXT MIXE The value of empirical tests, such as the present, lies largely in the comparison of results obtained from a considerable number of coals subjected to the same set of conditions. That exactly similar results will be obtained in other labora- tories is not probable, even though the specifications for making the tests are scrupulously followed, because of local and uncontrollable differences in pro- cedure and equipment. However, where it is possible to compare the results of empirical tests by two or more laboratories such comparison is of interest since it furnishes a means of bridging the gap between different series of tests by different organizations. The best available values for comparison are those obtained on coal from the No. 6 bed from the Orient No. 1 mine in Franklin County or from a mine very close to it which because the geological conditions in the region are very uniform should be practically the same. Determinations of coke, gas, and by- product yields using the United States Steel Corporation dry distillation test were made by the Illinois State Geological Survey on coal from the Orient mine and from an adjacent mine. A modification of the dry distillation test was used as one of the small-scale tests by the United States Bureau of Alines in it- cooperative study with the American Gas Association of the gas, coke, and by-product making properties of American coals, 2 coal from Orient No. 1 mine being one of those tested. The coal was also carbonized in the larger sized retort 2 Fielclner. A. C Davis. J. D., Thiessen, R., Kester. E. B., and Selvig. W. A., Methods and apparatus used in determining- the gas. coke and by-product making properties of American coals: U. S. Bureau of Mines Bull. 344, 1931. 3 Fieldner. A. C. and others, Carbonizing properties and constitution of Xo. 6 bed coal from West Frankfort. Franklin County. 111.: U. S. Bureau of Mines Tech. Paper 524, 1932. 102 EXPERIMENTAL INVESTIGATIONS used in that cooperative survey. Coal from this mine had previously been used in the large-scale tests at St. Paul on the practicability of producing coke from RANK INDEX PER CENT OF COKE ON DRf ASH -FREE BASIS - f\> U) Jk IP -J NET THERMS OF GAS PER TON RANK INDEX OF DRY ASH-FREE COAL Fig. 39. — Gas, Coke, and Rank Indices along the Line B-B' (Fig. 37). Illinois coals in Koppers type ovens. 4 The quantitative yields of coke, gas, tar, and other products obtained in the various tests have been compiled for compari- son in Table 14. I McRride, It. S., and Selvig, W. A., Coking- of Illinois coals in Koppers type oven: II. S. Bureau of Standards Tech. Paper L37, 1919. COKE, GAS, AND BY-PRODUCT YIELDS 103 71 C 70 69 68 67 66 65 64 i _ GAS 6 J i \ 1 ~~~~ „ ^\ ^^ \ ^~4 C£KE__ 1 >_ 1 » 130 R ANK INDE) 120* 1 10 100 * <^ MILES 10 20 30 40 50 60 70 80 90 100 4 CARBONIZATION SAMPLE 5 3 j NUMBER 1 2 1 C CROSSES B-B' C,' 50_ 130 0- ui O , L. I 0$ Fig. 40. — Gas, Coke, and Rank Indices along the Line C-C (Fig. 37). The agreement between the results given by the United States Steel Corporation carbonization test in the two laboratories is very good, especially when consideration is given to the fact that the samples were taken years apart and probably from places quite far apart in the mine. They point to the uni- formity of Illinois coals. COMPARISON OF RESULTS ON COALS FROM OTHER STATES No tests were made on coals from other states during the course of the present investigation. More than 30 coals have been tested by this method by the Bureau of Mines in its cooperative study with the American Gas Association. In view of the good agreement between the tests at the Illinois State Geological Survey and at the United States Bureau of Mines on the Orient coal there is good reason to assume that the tests reported by the Bureau of Mines 5 may be directly compared with results on Illinois coals reported here. In making such com- parisons it is well to keep in mind the difference in moisture and ash content of the various coals. Attention has already been called (page 57) to the con- clusions arrived at by the Bureau of Mines from a comparison of the Bureau of Mines, American Gas Association carbonization test results for the Illinois coals with the results for eastern coals. The Bureau of Mines finds a close correlation between the results of the Bureau of Mines, American Gas Associ- ation test at 900°C. and the results of the United States Steel Corporation carbonization test. 5 Fieldner, A. C, and Davis, J. D., Gas-, coke-, and by-product making properties of American coals and their determination: U. S. Bureau of Mines Mon. 5, p. 119, 1934. 104 EXPERIMENTAL INVESTIGATIONS h ^ X t^OO^(N(N^lO0(N^(N t^O\VO(N O O CN CSrH(N(N(N(N(NfCfOt^fOfOrOfOfOt^y)tOCN rt.S y S^ te *"i 5 oooooooooooo ©■© o o o ■<* o "CPQ § OOOOiOfOOOiOfOiOONOfONO\^MNi< >Or- iroCNfOOt^-ONOOONONOO v OrOfNOsON'— i On J2 n, &< iHNNMMNrHiHNMCSCNf^^CONNINfN > tile ter ent O^ONvO'*\OOMO'-i>^i0^ 1 Oa)rorOiOiOO « -m y . 22 c/5 JO ixed rbon rcent ioOMDoot^ooioafO^<*(Nvo^oofo^rHfo )^ [tj 03 . < •u "rt _C « lOf^iOOOfSXMOONNlNrH'OrtrooONO'O U o^oaoHNNcoaiOiHO fOHfT|NOOO'0(NfO'H^OtONOOCNH\ON co u a^cN^fooowoNooooovoioooov^^^ J2 a; S & >> >^ _, v- C c 3 O I. c c c o G C c c c d o u r? o o '5 5?j= -§ ^.3 33 rt 5 £ £ CO CO fO fO t~» f) O0 f) Q\ r^ fO 1^ Os ^ On On OnOOON''— iOnOncniOnOnOnOnOn©On < i i i i i i i i i i i i i i i i T I i t/3 g UUUUUUUUUUUUUUUUUUU rt ^o cs r<- ^ iO VC t> X o c CN PC ^t "~ nC nC nC mm COKE, GAS, AND BY-PRODUCT YIELDS 105 Gas dry cu. ft. per ton, 30" Hg. 60°F ooooooooooooooooooo OnOOOOOnOnOnOnOnO\OnOnOnOnOnOOOn OS o fDM^Of s )'*fO(N^*OOvXMt^t^'^MfOOOO CNrOfDf^f^^r^MNfS'HMf / 5fOrO(NfCifO he .2 N 'S ID " rOfOfOMMCNro^^rO't't^fOMrOi^mfO o u T3 M £2 OOO^hO-hOOOOOOOOOOOOO "o bO ONooa^iooNOiOM^fOfTi^MOOOH ccifO'>-iPCi , «*'-tiorNooiO'^t l '*ONCNOiO'* l '^-iON CNtNCNM»H(NINrH<*fOCNfOrOl0^t<)r^^(N o U COOOMN^COONVO'^IO'ONMD^^OO^OO N\OOONNNNNNNNNNNNNNN 06 EXPERIMENTAL INVESTIGATIONS J** rt O u, © o .t: £ ^ cc ^rHChrOCMN-HOOONHvOOONa'OMOW ■^•^^•^'O^^i^iO^'^'tUT* OOrHNNOOOlO^lOvOO^OONO^NOOlO cot^t^t^-oooo<~ s >^iO'— i fo 10 n n o o a 3 . . o PQ ^^(N^HOOOOOOOONaMrorOlNfNi/) (NfNrHlOt^'*^fO'^N>OON'-llON^a^ON u HNNOt v 10\(XI'HHr-llOO^r-lT-(t>lOf^fO O NCSfCMfOMOKNO'^MOOiO'HO'HrHH MroO'OM'HioO^OOiflOOO'-HCMNroONN COKE, GAS, AND BY-PRODUCT YIELDS 107 e -Jd 1) , o -C ° «i (D**- ^ (/) CO rt rt o (Nioa^a^oo^^ fOt^rHOO CO ^3 rt * J rt rt c o V u Ot^foO'H»HfO OOMOOONO'OOOOOOOOO'fOOOCNOO roOoOfOO^OOOOOOMDr^roiOO^HioarO Oio-^^ooioior-OOONi^ CO o i>- -^ *0 O so ro On CO lO 'tt^fO^tOTTfO'HHTHTHCSN £ (Nf0i H >')\0NC00\O'HMf^"fi0O l 0ON 108 EXPERIMENTAL INVESTIGATIONS 55 rt < O O 55 CO co O co C ^ 0\ OiOCO t a . C iO Ov t— vo r- 1 00 co<£ P VO CM 1>IOOO^ rt ^ *-• HtN OOfONrH ° 3 & CO i-l OM^CO ° a "" ^" s-, o a 1 vo t^ vo ON Ov • t^ r— 00 t^ rH • J Ih i-i co ^^cnt^ c cu o co co >OiO^^ cu Oh CO 13 CO IOCN (NOOOO £ o j>- r~- r^ © cni co H rH HCNCNCN CO Ov H 00 -* O o 0\0 l>- r}n -^ iO vo vO vo vo vO vo U • ^ g • 2 .S *^ U o ° J-> rt rt +j 4-1 +J^5 S-i J-i j, ° S S o . ~. .^,oo^ £< &m C 3 3 2 g }h 'g rt cti cu .x O >, £ W o a <-> rt **3 > r u 0, C c C JJ ■fif s^ .— fO fO ^O ro CO x* • NO t^ ■ © co ■ ^ O CO HI '© ©t^ONCOt^coONNOCNNOOT^coONt^-cot^- -^©ONrNNOiOTfio^T^ioioeNCNNONOeN lONOiOiO^OiOiOiONOiOiOiOiOiOiOiOOO T-irOfONHrHCMMfOHHHMCN|NrJ*lO S £ as bfl X>-nOnO -, ; C u S : 3 c o £ l_ i_ u £ . d . dS-d co O as O «-, S-, «i ^ H -H W Oi ■£ +J +J U'-a-'-i U fofaQj:^Sp^c^oococo>> : >&H ; : I'd :::::::::: : : : : ^ £? .^.d CO CO CO C/3 CO C/3 co co c/3 co co co en c/3 cn y p> o o o o o C C C C C O O O O c d d c 3 3 3 3 3 *-> -m c c d c c c g — 1 _d i — i . — i . — I .— h . 1 , 1 1 1 ( 1 ^h-h^^k^^^ 02 >> ^ t^rOl^rOfO^ONCsiro vO^iO^n^OOfOOO oo • «r> f^ -co PC ■ o ro -On ^ • T*H OnOnOOOO'-hcXJOnoO'O 00"*(NIOTj*lOfON (N^rHtNNOaONO OOionOOOnOOOncn M cu (U O TjH co cu 2 it* j I. Si ^ SP| 6x r- 1 J- o O^ • C «J £ 2^j d o-d o 0U"u OJ cu C a! X MQ jy ^.2hd ai be Jjja O — cd^T! u, O as — i ,_« -p tr- ON no 'O r^rO^OOOiOOOOOOO'H-frOCOfNr^H iOON"-H-troro-H-+-f-t H -+-t l ONO"-HO'-t UUUUUUUUUUUUUUUUU o O > d BEHAVIOR OF ILLINOIS COALS UNDER HEAT 113 CHARACTERISTICS OF THE COAL WHICH AFFECT ITS AGGLUTINATING VALUE Mineral impurities. — The removal of mineral matter from coals high in ash should improve their agglutinating powers. A Washington County coal having an ash content of 13.7 per cent had an agglutinating value of 2.9 kg. After cleaning at a specific gravity of 1.35 by the float-and-sink procedure the coal had an ash content of 6.4 per cent and an agglutinating value of 3.2 kg. The 28.8 per cent ash refuse had an agglutinating value of less than 0.2 kg. With an increase in agglutinating value proportional to the ash reduction a value of 3.1 kg. would have been obtained. This is within experimental error of the value 3.2 kg. found experimentally. Freshness. — The agglutinating values of Illinois coals are variously decreased by oxidation. Exposure of the mine face for some weeks permits oxidation sufficient to affect the agglutinating values of channel samples. For example, a sample of Franklin County coal taken from a fresh face had an agglutinating value of 3.64 kg., while a sample from a face that had been exposed for several weeks had a value of 2.75 kg. Fusain content. — In attempts to reduce the formation of shrinkage cracks and to produce blockier coke, dedusting plant dust or coke breeze was added as low volatile carbonaceous matter. The dedusting plant dust contained about 50 per cent fusain and had no agglutinating value. The addition of dedusting plant dust or fusain to coke oven charges to improve coke quality has been patented by Mott and Wheeler 2 in England on the claims that fusain is "wet" by and becomes better incorporated into the coal than coke. Some agglutinating values of blends of coal coked in experiments reported elsewhere in the report are of interest in that connection and are presented in Table 16. Table 16. — Agglutinating Values of Various Coals and Blends of Coal and Coke or of Coal and Dedusting Plant Dust Composition of sample 100% Franklin Co. coal 80% Franklin Co. coal 20% dust 80% Franklin Co. coal 20% coke 75% Franklin Co. coal 20% coke, 5% dust. 100% Randolph Co. coal 80% Randolph Co. coal 20% coke 100%, Woodford Co. coal 80% Woodford Co. coal 20% dust Agglutinatim value, kg. 3.2 4.0 0.4 0.4 4.3 0.0 3.3 3.Z 2 Mott, R. A., and Wheeler, R. V., and The National Federation of Iron and Steel Manufacturers (Corporate Organization) Limited. "Improvements In and Relating to the Manufacture of Coke." British Patent 351,546, accepted June 29, 1931. 114 EXPERIMENTAL INVESTIGATIONS Conclusions. — Illinois coals have relatively little caking or agglutinating power. The agglutinating power they have is easily destroyed by oxidation. Oxidation also destroys the ability of the plastic coal particles to coalesce to form the viscous mass which ultimately would form the coherent cellular coke. For these reasons Illinois coals should be coked as soon as possible after they are mined, and should be taken from freshly worked places in the mine. Inert material such as mineral matter adversely affects the agglutinating power of these coals and should be present, therefore, in amounts as small as possible. CHAPTER XIV— EFFECT OF THE BANDED COMPONENTS AND IMPURITIES OF COALS ON COKE STRUCTURE Though the usual analyses and tests treat coal as a homogeneous substance, it is far from being so. Even casual examination of a piece of coal with the naked eye reveals the fact that the common varieties of coal are made up of bands of materials, some having a vitreous, black appearance, others a finely striated, silky appearance. Fracture surfaces parallel to the bedding plane frequently are covered with a charcoal-like, friable, cellular material often having the appear- ance of small chips of carbonized wood. These bands are not merely the same substance in different forms but are actually different materials which had their origin from different parts or different mixtures of parts of plants. The impurities, likewise, are not uniformly distributed throughout the mass of coal. During the mining and preparation of the coal the banded components and impurities, because of their different properties, may be present in the various sizes of prepared coal in different proportions than they existed in the coal as it occurred in the bed. Since the various components and ingredients behave differently during carbonization and since the situation is more or less common to all coals, they may be treated in a general way. BANDED COMPONENTS In Illinois coals the banded components are mainly of three kinds : ( 1 ) a bright, vitreous material known to be derived from larger pieces of woody material, variously called "anthraxylon" 1 or "vitrain" 2 ; (2) non-vitreous appearing, striated materials, sometimes having a silky lustre, which forms the bulk of the coal and which separates the bright bands from one another, derived from plant debris of all kinds, but mainly of small size and called "attritus" or "clarain" ; and (3) a charcoal-like material practically always present and in minor amount, derived in some unknown manner from woody material and called "fusain". The banded structure of Illinois coals is largely, if not entirely, due to the presence of the bright bands which range in thickness from fine slivers to infrequently as much as one inch. A less common variety of banded coal is composed of finely banded material having a dull gray or slaty appearance. i Thiessen, R., and Francis, W. R., Terminology in coal research: IT. S. Bureau of Mines Tech. Paper 446, 1929. 2 Stopes, M. C, The studies on the composition of coal: The Four Visible Ingredients in Banded Bituminous Coal, Proc. Royal Soc. Ron. 90R, p. 470, 1919. [115] 116 EXPERIMENTAL INVESTIGATIONS These coals are characterized by toughness and a blocky fracture and are known as "splint coals", dull coal "durain", or in England also as "hards". Cannel coal, also a less frequent variety of coal, is not banded and exhibits a waxy lustre and a conchoidal fracture. The varying nature of these components are reflected in their carbonizing properties both with respect to their behavior when heated and the yields of products. Conclusions from a correlation of the petrographic compositions of coals and the results of certain carbonizing tests have been presented by R. Thiessen and Sprunk 3 in which one of the coals considered is an Illinois coal from the No. 6 bed in Franklin County. Carbonizing properties. — Of the four components, fusain alone does not form a coke. Splint coal will form coke but the individual pieces of coal tend to form individual pieces of coke. Bright coal is the only satisfactory coke-forming coal. Tests show that of the two ingredients of bright coal, the anthraxylon (or vitrain) by itself tends to form a frothy, thin-walled coke. Its separation in a high state of purity does not appear to be very desirable for coke making purposes. The carbonizing properties of the bright attritus or clarain depend on its composition in terms of plant fragments, such as resins, waxy, or humic materials. In using fine sized coal for coke making, the questions of ash and fusain content are far more important than are the proportions of anthraxylon and attritus (vitrain and clarain). Fusain tends to make the coke more blocky in structure, but should not be present in sufficient quantity to make the coke soft and friable. Composition of Illinois coals in terms of banded components. — Illinois coals are predominantly bright coals. Occurrences of splint and cannel coals are localized and may be neglected in a consideration of Illinois coals for coke making purposes. The average petrographic composition of the No. 6 bed in Illinois as determined from a megascopic examination of 21 columns, each of which represented the entire seam and which averaged 6 feet 8 inches in thickness, was found by McCabe to be 75.9 per cent clarain (bright attritus), 13.8 per cent vitrain (anthraxylon), 5.6 per cent fusain (mineral charcoal) and 4.7 per cent non-coal impurities. 4 The petrographic composition of No. 6 coal from Orient No. 6 mine, West Frankfort, Franklin County, is discussed in detail in the report of the United States Bureau of Mines, American Gas Association cooper- ative investigation of that coal. 5 This coal is quite uniform from top to bottom. •'! Sprunk, G. C, and Thiessen, R., Relation of microscopic composition of coal to chemical, coking and by-product properties: Ind. P]ng. Them. vol. 27, No. 4, pp. 44fi-451. April 1935. 4 McCabe, L. C, The lithology of coal No. 6: Trans. Illinois State Acad. Sci. vol. 2, r >, No. 4, pp. 149-50, 1934. 5 Fieldner, A. C, and others, Carbonizing properties and constitution of No. 6 bed coal from West Frankfort, Franklin County, 111.: IT. S. Bureau of Mines Tech. Paper 524, 1932. EFFECT OF IMPURITIES ON COKE STRUCTURE 117 Compared to other coals from the United States studied in that investigation it contained much anthraxylon (vitrain). McCabe, Mitchell, and Cady G have reported the petrographic composition of other Illinois coals. 7 Because coal components differ considerably in their resistance to crushing and breakage, the various components tend to concentrate in different sized fractions of the coal. Fusain, being the most friable, is found concentrated in the very fine dust, the minus 200-mesh material containing 50 per cent or more of it. Vitrain is the next most friable material and is also found concentrated in the fine sizes. 8 Because vitrain (anthraxylon) tends to have a lower specific gravity because of its freedom from ash forming material, it is also found con- centrated in the light float-and-sink fractions. The concentration of fusain in the very fine sizes of coal recovered from dedusting plants has been studied by G. Thiessen 9 who found that the minus 200-mesh fraction was composed largely of fusain. IMPURITIES The impurities which seriously affect the value of a coal for the manufacture of coke are ash, sulfur, moisture, and in some cases phosphorus. Moisture is undesirable because of the heat required to evaporate and drive it out of the coal and because the resulting liquor must be condensed and handled in the by-product recovery system. The ash is objectionable because it becomes con- centrated in the coke, decreasing its fuel value, and because it may impair coke structure. Sulfur is objectionable not only because part of the sulfur remains in the coke and affects the value of the coke for practically all of its uses but also because part of the sulfur is volatilized and is found partly in the tar, where it is relatively innocuous, and partly in the gas where its presence is serious and from which it must be removed if the gas is to be sold. Phosphorus is important only in certain metallurgical applications of coke. The effects of the impurities on the coking process and on the character of the resulting coke and by-products, and the methods which may be employed to eliminate those impurities or to minimize their harmful effects are discussed. MOISTURE Moisture is an undesirable component of coking coal because of the large amounts of heat required to vaporize it. The high latent heat of vaporization of water keeps the temperature beyond the plastic zone in a coking charge of coal down to the boiling point of water. Much of the heat used to vaporize this 6 McCabe, L. C, Concentration of the banded ingredients of Illinois coals by screen sizing and washing: Amer. Inst. Min. Met. Eng. Tech. Pub. 684, Class P, Coal Division No. 73, Table 2, 1936. 7 McCabe, L. C, Mitchell, D. R., and Cady, G. H., Contributions to the Study of Coal- Banded ingredients of No. 6 coal and their heating values as related to washability characteristics: Illinois State Geol. Survey Rept. Inv. 34, 1934. 8 McCabe, L. C, Op. cit. o Thiessen, G., Fusain content of coal dust from an Illinois dedusting plant: Amer. Inst. Min. Met. Eng. Tech. Pub. 664, Class P Coal Div. No. 70, 1936. 118 EXPERIMENTAL INVESTIGATIONS moisture comes from hot products of carbonization. The presence of moisture does, however, require the addition of an increased quantity of heat to the charge, and this heat, coming through the oven walls is expensive and decreases the oven capacity by increasing coking time. Koppers 10 states that each two per cent moisture corresponds to an increase of one hour in coking time. A large excess of water may cool the oven refractories sufficiently to damage them. The evapo- rated moisture must later be condensed and provided for in by-product recovery system capacity and finally disposed of without creating a nuisance. The moisture contents of Illinois coals (Table 11, pp. 62-3; fig. 17, p. 69) are relatively high when compared with the eastern coals commonly used for coke making. Wet coal-cleaning processes add more water to the fine sized coal and introduce further difficulties. Drying processes commonly employed which use heated gases may seriously decrease the coke forming properties of these coals through oxidation. The natural tendency of these coals to oxidize is accelerated by the increased temperature, the small size of the coal particles, and the rapid circulation of air through the mass of coal. ASH OR MINERAL MATTER Ash is objectionable in metallurgical coke because its presence reduces the furnace capacity and increases the quantity of slag. For domestic or industrial fuel, ash is objectionable in coke for the same reason that it is objectionable in coal, with the added fact that coke ashes have a tendency to be more dusty when not clinkered. Furthermore the large clinkers which may form due to the higher temperatures attained in a bed of burning coke are undesirable. The particle size of coal as charged into a coke oven is usually rather small, less than Yi inch in most cases, the small size being obtained either by crushing larger coal or through the use of small screenings. The small size of the coal particles limits the size of the pieces of mineral matter or ash forming streaks in the coke and makes the ash distribution more homogeneous. The swelling and flowage of the coal during its plastic stage further disrupts the ash forming materials. The result is that the ash residue is powdery and frequently fluffy when coke is burned at a temperature below the fusion point of the ash. Although the more uniform distribution of the ash forming material in coke as compared with coal ordinarily results in less frequent formation of clinkers, 11 such clinkers as do form tend to be large, vitreous, and relatively strong. Cokes made from coals having ashes with low fusion temperatures are most likely to product such clinkers, since the ash forming components are quite uniformly distributed and the attainable temperatures in the fuel bed are high. lo Koppers Mitteilungen 1921, No. 2, p. 58— quoted in Gluud, W., (.Tacobson, P. L.): Internal ional Handbook of the By-product Coke Industry. American Edition, New York, 1932, p. 96. ii Thiessen, ::<;. EFFECT OF IMPURITIES ON COKE STRUCTURE 119 PQ ■* 00 M3 ro Oi CO O 00 rO 00 r-l y-t 00 N r*5 CO O 'O S"|. co O u X CO u u o •** .s.s.s \i-0 o c — i A co ° c p -O M0 o " w I © r-t r^ to CN tO uo io lO lo iO "J IT) to LT) lO to UUUUUU £; 00 y-t rv) os o to u-) vO O io O "2 io io io lo io UUUUUU 120 EXPERIMENTAL INVESTIGATIONS EFFECT OF IMPURITIES ON COKE STRUCTURE 121 The mineral matter in coal is further troublesome because of the effect it may have on coke structure. Free mineral matter is non-caking material which must be consolidated into the coke mass if the coke is to be strong. If the agglutinating strength of the coal is already low, sufficient quantities of such inert matter may weaken the coke. Even in small amounts the effects of such free mineral matter are apparent when a piece of coke is examined and compared with a piece of coke made from the same coal from which such inert matter has been removed. This is illustrated by figure 43 in which two pieces of coke are pictured side by side, one made from raw ~Y\ inch screenings, the other from screenings which had been cleaned at a specific gravity of 1.5. It will be noted that, in the coke from the raw screenings, the cracks in the coke either begin, or end, at a piece of loosely held shale. That these larger pieces of mineral matter are only loosely held in the coke is evident from a comparison of the ash content of the larger sized ( + ^> inch) and smaller sized (-!/> inch) pieces resulting from breakage during the shatter test (Table 17). Ash forming materials tend to segragate in the naturally produced finer sized screenings from Illinois coals. 12 This tendency is more pronounced in some coals than in others but must always be considered as a possible serious problem if small sized screenings are to be used as raw material for coke manu- facture. As examples of how the ash content of the various sized fractions from coals may vary, the ash content of the face samples and of sized fractions from naturally formed screenings from three of the mines studied by McCabe and his co-workers are presented in Table 18. Table 18. — Variations in Ash Content with Size in Naturally Produced Screenings from Illinois Coal Ash content, per cent on dry basis Mine B a Mine G a Mine J a Face sample 1 x /i x 0-mesh 1 34 x ^i inches 6.5 19.4 11.0 14.3 19.7 30.5 36.5 11.4 18.1 15.5 16.9 19.3 25.8 25.2 8.4 11.8 9.7 Y^ x z /% inches % inch x 10-mesh 10.2 11.7 10x48-mesh 18.5 Minus 48-mesh 16.8 Mine B. Northern Illinois No. 2 coal. Mine G. St. Clair County No. 6 coal. Mine J. Saline County, Harrisburg, No. 5 coal. The available ash analyses of face samples of coal from Illinois mines, mine and county averages, and available ash softening temperature values have 12 McCabe, L. C, Mitchell, D. R., and Cady, G. H., Proximate analyses and screen tests of coal mine screenings produced in Illinois: Illinois State Geol. Survey Rent. Inv. 38, 1935. 122 EXPERIMENTAL INVESTIGATIONS been presented by Cady. 13 County average ash values have previously been presented in this report in figure 18 and Table 11. Since the quality of coke produced from Illinois coal is materially affected by the ash content of the coal used, it is very important that the ash content be reduced as much as possible. The cleaning of coals as a means of reducing the mineral matter even for coal used as a fuel in the raw state is becoming important in the Eastern Interior coal field and much attention is being given to the problems involved. Only the larger pieces of free mineral matter or the higher-ash-content pieces of coal can be removed by physical means. Finer crushing frequently permits a more complete removal of the ash forming material. The lower limit to which the ash content of a coal may be reduced is the so-called inherent ash content of the coal. This "inherent ash" is the ash which is finely disseminated through the coal substance or in chemical combination with it and which entered the coal during its formative peat stage. Studies on the washability of Illinois coals have been made by Mitchell, 14 by Callen and Mitchell, 15 and by McCabe, Mitchell, and Cady. 16 An early account of coal washing in Illinois has been given by Lincoln. 17 Statistics covering recent coal washing developments in Illinois as well as in the rest of the United States have recently been published by Plein of the United States Bureau of Mines. 1S Descriptions of coal cleaning plants in Illinois may be found in the technical literature. Regarding the cleaning of coal in general, it may be said that each coal offers its own problems which must be solved by experimentation. Coal cleaning is a problem by itself, the consideration of which does not fall within the field of the present investigation. SULFUR Sulfur is an especially undesirable impurity to the coke and gas industries. It distributes itself during coal carbonization between the volatile materials, gas, light oils and tar, and the solid coke residue and is undesirable in all of them. Sulfur in blast furnace and foundry coke is easily picked up by iron. If the sulfur content of blast furnace coke is unusually high very close watch must be kept of the slag composition and the slag must be made more basic by the addition 13 Cady, G. H., Illinois State Geol. Survey Bull. 62, 1935. 14 Mitchell, D. R., The possible production of low ash and sulphur coal in Illinois as shown by float-and-sink tests: Univ. of Illinois Eng. Exp. Sta. Bull. 258, 1933. L5 ('alien, A. G, and Mitchell, D. R., Washability tests of Illinois coals: Univ. of Illinois Eng, Exp. Sta. Bull. 217, 1930. L6 McCabe, L. G, Mitchell, D. R., and Cady, G. H., Banded ingredients of No. 6 coal and their heating values as related to washability characteristics: Illinois State Geol. Survey Rept. Inv. 34, 1934. it Lincoln, P, C, Coal washing in Illinois: Univ. of Illinois Eng. Exp. Sta. Bull. 69, 1913. L8 Plein, L. N., Statistical analysis of the progress in mechanical cleaning of bituminous coal from 1927 to 1934: U. S. Bureau of Mines Economic Paper IS, 1936. EFFECT OF IMPURITIES ON COKE STRUCTURE 123 of limestone, which reduces seriously the capacity of the furnace. If the coke is used for burning limestone, the sulfur is absorbed by the lime, contaminating it for chemical use. If the coke is burned for domestic or commercial heating the problem of sulfur is not so serious, but even here it is undesirable from the standpoint of atmospheric pollution. The ashes from a coke high in sulfur may also give off enough sulfurous gases after removal from the furnace ash pit to be a nuisance and even a danger to health. If the gas is to be sold for domestic use and for many industrial uses the sulfur compounds in the gas must be thoroughly removed. Sulfur enters the gas mainly as hydrogen sulfide and in organic compounds such as carbon bisulfide and mercaptans. All of these have unpleasant odors, tarnish silver and many other metals, and yield sulfur dioxide when burned. Hydrogen sulfide is a deadly poison. The cost of purification is an important item in gas manufacture. Occurrence in Illinois coals. — Sulfur occurs in coals as the iron disul- fides, pyrite and marcasite, as part of the organic coal substance, and as the sulfate minerals which occur in appreciable quantities only in weathered coals. Pyrites may be found in all sizes from small, finely disseminated particles to large masses, lenses, or bands. Pyritic sulfur may be partly removed during coal preparation. Organic sulfur cannot be removed without destruction of the coal. The county average sulfur contents of Illinois coals and the low-sulfur coal area have already been discussed. (Table 11, figs. 19 and 20, pp. 62, 71, 72). The available values for pyritic and organic sulfur contents of Illinois coals as determined on face samples have been presented by Cady 19 who has also discussed 20 in detail the distribution of sulfur in Illinois coals, and the low sulfur area in Southern Illinois. 21 The distribution of sulfur in Illinois coals between varieties has also been studied by Yancey and Fraser. 22 As stated before, Illinois coals in general, and except for small areas, have sulfur contents higher than is thought generally to be desirable in coals for the manufacture of metallurgical coke. Removal from Illinois coals. — Coal cleaning or preparation procedures remove only the inorganic forms of sulfur, more particularly the pyritic sulfur occurring in larger aggregations. The problem of sulfur reduction has received considerable attention because of its economic importance. Callen and Mitchell, 2 '' 19 Cady, G. H., Illinois State Geol. Survey Bull. 62, Table 6, 1935. 20 Cady, G. H., Distribution of sulfur in Illinois coals and its geological implications- Illinois State Geol. Survey Rept. Inv. 35, p. 25, 1935. 21 Cady, G. H., Low sulfur coal in Illinois: Illinois State Geol. Survey Bull. 38, pn 432-434, 1922. 22 Yancey, H. F., and Fraser, T., The distribution of the forms of sulphur in the coal bed: Univ. of Illinois Eng. Exp. Sta. Bull. 125, 1921. 23 Callen, A. C, and Mitchell, D. R., Washability tests of Illinois coals: Univ of Illinois Eng. Exp. Sta. Bull. 217, 1930. 124 EXPERIMENTAL INVESTIGATIONS Mitchell, 24 and McCabe, Mitchell, and Cady 2 '" 1 have presented results of wash- ability studies of Illinois coals which include work on the effect of gravity separation on the reduction of the sulfur content of the coal. These tests show that, with some exceptions, not much reduction in sulfur can be expected through washing Illinois coals. This is due, first to the relatively high organic sulfur content of these coals and, second to the fact that much of the pyritic sulfur is often found in a finely disseminated form 26 too small to be removed except after grinding the coal to extreme fineness. The possible reduction of the sulfur content of the coal produced at each mine or group of contiguous mines presents a new problem which must be solved experimentally. Behavior of coal sulfur during coal carbonization. — The conclusions which could be drawn from a review of the pertinent literature 27 were that : ( 1 ) about half the sulfur in the coal appears in the coke but the percentage of sulfur in the coke will be about 80 per cent of that in the coal because of the loss of volatile matter during coke formation; (2) there is no unanimity of opinion as to the relative importance of organic and pyritic sulfur as a source of sulfur in the coke; and (3) the sulfur found in the coke appears to be directly pro- portional to the total sulfur content of the coal from which the coke was made. The conclusions from this review of the literature indicated that it would be desirable to carry out investigations on the distribution of coal sulfur between the volatile matter and the residual coke or fixed carbon and to correlate this distribution with the distribution of sulfur in the coal between pyritic and organic forms in an attempt to find a way to estimate the probable sulfur content of a coke when the sulfur content and distribution of the sulfur between the two varities in the coal is known. In the experimental work as carried out, the coke was prepared in platinum crucibles under conditions prescribed for the standard volatile matter test for 24 Mitchell, D. R., The possible production of low ash and sulphur coal in Illinois as shown by float-and-sink tests: Univ. of Illinois Eng\ Exp. Sta. Bull. 258, 1933. 25 McCabe, L. C, Mitchell, D. R., and Cady, G. H., Banded ingredients of No. 6 coal and their heating- values as related to washability characteristics: Illinois State Geol. Survey Rept. Inv. 34, 1934. 2G Thiessen, R., Occurrence and origin of finely disseminated sulfur compounds in coal: Amer. Inst. Min. Eng. Bull. 153, pp. 2431-2444, 1919. Also discussed in Univ. of Illinois Eng. Exp. Sla. Bull. 125, pp. 54-63, figures 18-24, 1921. 27 Thiessen, G., Behavior of sulfur during coal carbonization: Ind. Eng. Chem. vol. 27, No. 4, pp. 473-478, 1935. Parr, S. W., The coals of Illinois: their composition and analysis: Univ. of Illinois Bull. vol. 1, No. 20, July 15, 1904; The University Studies Vol. 1, No. 7. Powell, A. R., A study of the reactions of coal sulfur in the coking process: Ind. Eng. Chem. vol. 12, No. 11, p. 1069, 1920. Campbell, J. R., Effect of aeration and "watering out" on sulfur content of coke: Bull. Amer. Inst. Min. Met. Eng. No. 109, pp. 177-80, 1916. McCallum, A. L., The action of organic sulphur in coal during the coking process: The Chemical Kngineer vol. 11, No. 1, pp. 27-28, Jan. 1910. Sperr, F, W., Gas purification in relation to coal sulphur: Proc. Second int. Conf. Otl I'.if. Coal, pt. Ill 1 , pp. 37-60, 1928. EFFECT OF IMPURITIES ON COKE STRUCTURE 125 5.6 4.8 z 2 a. u §3.2 '2.4 • % / • -• / f • # *_ \ f 30t • • t • l*T k'-'w F Fe D a c ry •|.3 o C fo ?& • LEGEND O FACE SAMPLES O COMPONENT SAMPLES V VITRAIN C CLARAIN F FUSAIN 1.3,1.4 SPECIFIC GRAVITY SEPARATION SAMPLES WD WEATHERED SAMPLES FeD SAMPLES WITH ADDED IRON X MC CALLUM'S DATA 4 \& O 5Ji * ^ JT / OF 3 8 i 6 2 .4 3 2 4 4.8 56 TOTAL SULFUR IN COAL Fig. 44. — Relation of Sulfur Content of Coke to Sulfur Content of Coal. (From "Behavior of Sulfur during Coal Carbonization," by G. Thiessen, Ind. and Eng. Chem. vol. 27, April, 1935.) < u 2 1.2 o u LEGEND • O COMPONENT SAMPLES V VITRAIN C CLARAIN F FUSAIN 1.3,1.4 SPECIFIC GRAVITY SEPARATION SAMPLES WD WEATHERED SAMPLES FeD SAMPLES WITH ADDED IRON X MC CALLUM'S DATA F O • ■^ i ) • •* A £ • < 1 F F 1» 'J ^OC F o F. D W 3 V )W^ ~£* 1 co C o F & o 1.3 r FO 1.3 V° C 2.4 TOTAL 3.2 4.0 SULFUR IN COAL 4.8 5.8 Fig. 45. — Sulfur Content of Coke, Expressed on Basis of Original Coal, as Function of Total Sulfur Content of Coal. (From "Behavior of Sulfur during Coal Carbonization," by G. Thiessen, Ind. and Eng. Chem. Vol. 27, April, 1935.) 126 EXPERIMENTAL INVESTIGATIONS coal. Four buttons were prepared from each sample except where rechecks were required. Analyses for pyritic and sulfate sulfur were made according to the method of Powell and Parr, 28 for other values by standard A.S.T.M. methods. A total of 82 samples was studied, including 55 mine face samples, 17 samples of banded coal components comprising four samples of anthraxylon (vitrain), seven samples of attritus (clarain), and six samples of fusain, three samples of coal treated by float-and-sink procedure, five samples of the same bed taken at points successively nearer the outcrop to determine the effect of weathering, and two samples of coal to which iron oxide had been added in order to simulate one of the characteristics of weathered coals. «D 2.8 _i < 8 * I. o o t p ^i *■&> 'W ^ ^ \x> &- 2 1* ^ CC N * ^ F.D firs t¥ f? LEGEND #FACE SAMPLES O COMPONENT SAMPLES V VITRAIN C CLARAIN F FUSAIN 1 .3 SPECIFIC GRAVITY SEPARATION SAMPLES WD WEATHERED SAMPLES FeD SAMPLES WITH ADDED IRON CDrj F« C & m < ifi & t? • fl * 7 *& X \AC CA LLUM 'S DATA 1.6 2.4 3.2 PYRITIC SULFUR IN COAL Fig. 46. — Sulfur Content of Coke, Expressed on Basis of Original Coal, as Function of Pyritic and Organic Sulfur Contents of Coal. (From "Behavior of Sulfur during Coal Carbonization," by G. Thiessen, Ind. and Eng. Chem. vol. 27, April, 1935.) Studies on Face Samples. — The data pertaining to the face samples, including the percentage of organic, pyritic, and sulfate sulfur in the original coal samples are presented in Table 19. Better to visualize relationships which might exist, the data were plotted as follows : ( 1 ) Sulfur in the coke, as determined, against the total sulfur content of the coal (fig. 44) ; (2) the sulfur content of the coke expressed as a percentage of the original coal against the 28 Powell, A. It., with Parr, S. W., A study of the forms in which sulphur occurs in coal: Univ. of Illinois Eng. Exp. Sta. Bull. Ill, 1919. Powell, A. It., The analysis of sulphur forms in coal: U. S. Bureau of Mines Tech. Paper 254, 3921. Powell, A. It., Determination of sulfur forms in coal: Ind. Erxg. Chem. vol. 12, No. 9, p. S87, 1920. EFFECT OF IMPURITIES ON COKE STRUCTURE 127 total sulfur content of the coal (fig. 45) ; and (3) the sulfur content of the coke, on the original coal basis, against the pyritic sulfur content of the coal (fig. 46). With few exceptions the points in all three cases fall close to straight lines. The two curves for which the total sulfur content of the coal is one of the variables (figs. 44 and 45) pass through the origin. Sulfur in coke and total sulfur in coal. — When values for the sulfur in the coke were plotted against the values for the total sulfur content of the cor- responding coal (fig. 44), it was found that the points fell reasonably well along a straight line whose equation is : S , as determined=0.83 S , coke coal In figures 44 to 46 only those points representing face samples were used in determining the curves. These values are represented by solid dots whose radius is equal to the permissible error between laboratories for sulfur determi- nations on coke (0.05 per cent.) 29 Table 19 shows the individual values for the ratio of S coke to S , These values range from 0.695 to 0.982 and average 0.831. This value is in agreement with the statement of Sperr 30 : 'With coals containing from 0.5 to 3.0 per cent sulfur, the per cent sulfur in the furnace coke averages about 80 per cent of that in the coal." The agreement and the general similarity of figure 44 with the graph illustrating the relationship between coke and coal sulfur presented by Sperr permit the conclusion that the results of this laboratory stud} are directly applicable to standard by-product coke-oven practice. Sulfur in coke (on basis of original coal) and total sulfur in coal. — By cal- culating the sulfur content of the coke to the basis of the original coal, it is possible to find the proportion of coal sulfur which remained in the coke and the proportion which entered the volatile matter. For the face samples studied it was found (fig. 45) that 51.9 per cent of the coal sulfur remained in the coke, the values ranging from 42.3 to 62 per cent. These points also lay reasonably close along a straight line whose equation was: S coke («> al basis)=0.519 S coal The value given by Parr 31 was 48.5 per cent. The general statement that approximately half of the total coal sulfur remains in the coke is borne out by the present results. 29 Amer. Soc. Testing- Materials, Standards, Pt. II, pp. 387-426, Method D271-30, 1936, Amer. Standards Assoc. No. K18-1930, especially p. 426. 30 Sperr, S. W., Op. cit., especially p. 51. 3i Parr, S. W., The coals of Illinois: Op. cit. 128 EXPERIMENTAL INVESTIGATIONS t^ }U90 J9d s 00 vd LO "tf [BOO-g 0^ •p^9p 9J[00-g P sisBq [boo UO [BOO-g 0% axoo-g oi^y cni ^U90 J9d [boo uo aaoo-g - ^U90 J9d 3300-g O !11I90 J9d Hsy + '0 "tf On ^U90 J9d qsy 00 }U90 J9d '0M t- |U90 J9d OS o JU90 J9d <*s LO JU90 J9d tosg Th ^U90 J9d is co •0J^ [BOO puB A^unoQ CM •on °l (I,m!S F>0 - ofv[ ajdnrBS 33{OQ OOO O OO H ^H ^ • O vO CO OO o^aoooo.HNoooNH^ NO tH OC co OOOM J>-^ OCOOO lo J>- lo On hi^O 00 lo >H00iO h^OO\>0 nO lo no lo co io^fO \OiO>OM3 OOOOOOOOOOOO OO OOO OO OOO oooo lONOOfOOoOONfOfOOOOOt^ Ol co ONIO LO "* 00 CS 00 Nt>OlO Or-rr) NO l-H lOMM o^^oo Ol> O NO UO 00 On Th \ON>Oifl O O O t-h i-H O O O t-h O O O OO OOO O i-i O O i-i oooo i^rON>0'OiOOoO(Nro(NN i— 1 1-~ co i— i r^ On On 00 00 oo lo O i— i O -hh On cn r-~ OnOn novo oo ^ oo On r~~ no t^-t^t^ oooooooo OOOOOOOOOOOO OO OOO OO OOO oooo rOrONOOi-ii-iCsloOLOOONOO no -hh rotNO i^ro hioio OOONOON t^^t 1 LOCNro O0C0 J>- t^ no CM On CN co lo lo lo lo lo -^ tH "^^^i io^t'O'o OOOOOOOOOOOO OO OOO OO OOO oooo ONOHi^ONac^NH 00 lo ONio Oh t^ t^ o ^ V£ ,_l "° NOOiOOOOO^iOOO roiO i-hoOOn t^- CN i-Hi-ho co O CM cn) OOOOOOOi-hOOOO CMNtNt^rO'HMO^OsO^rO OOh nO^lo OOaNOOONNaO 1 * fOr- 1 T^ i— I CN CNCsl CN04CN CNCNCNCN H -H O O-rH OH ONi^NOO^t^CNjr^ro LO^h 1 rororo i* rO fO CO fO C) CO CO ro •HH i-i no ^ cm r- O NNO^O O vO NO ^l^fNCOOOuOinOO-HO^ON OCN OIOni-h CM On O NO CM O^"^^ QO\rHTHfoONOt^O\OOCiO 1>- lO CNCOCO vO(N lo lo -HH ^OCjCO oiO^t^^rsi^ONONi-irONO ^ On Tf io iO i— i t— i NO O J>- ' r ~ ' ^ On coO'H^^ioi^co^mDnO'* lolololo^lolOLOLOLOlolo t-~Oi— i O°°lo0n v O00C0cO00 >OmDn "^ ^t 1 iHH LO 00 NO NO ,-jH H H ^ t^ ^J>- On OOOOOOOOOOOO CN(NCN i—l O NO 00 00 NO l-H CM i-H CO l-H NO LO 00 CO lOf^ON "* IO ON ON ^ ^O <£ f2 i? 00 NO LO CO NO lo i-~ J>- ^ NO NO O LO ^t 1 CO ON i—i CO 00 CO i-i lo i>- CO LO i>- ooooooo-^ooo- CNJLOLOCO^r^i— I i-H ' oooooooo- CM CM i-l CM ?N CM CM lo "* ^ "* o o o o OOOOOOOOOOOO OO OOO OO OOO o © o o q> h i* i-i 00 CO CM 00 r-H 00 CM ro co On O 00 ^ On CM i-h ,-h o r^ h VO On OJCNIi-iOncOi-h^coOi-hcNno 00nO OOlono nO CM nOloco "* h(n _ h ■> ^ *# Tt ^ ^ ^ ^ lo ^ CO co co ta W CO On O i-i ON rvi ro -H >0 l— 00 On -fl-fllO'OOONONONONOO co CO -t -+ -V i+ >0 IO i-H no i- CO -t <0 00 On i— i O 00 t-h r^i -t -t -t -t •+ -+ -t CO IO IO IO IO oooooococooooooooooooooo -t o O '— t cs ro co co 1--NO i+NOlO OHCNTji (NO c^irsio ^'^■^•^r CO CO i-H i-l i-H •H cntJ< Oi-hco io no 1^ o OO OOO co co CO co On On On On On "^f ^f ^ ^ EFFECT OF IMPURITIES ON COKE STRUCTURE 129 O th OOOO -OOt-h no h 00 00 On "0 ■>* NO t— On t-h O t— On lOro^d'^'OO'OO "* ro lo ** ioionOlolonOlolo ^OO ro i-H t-H ro On lo O v© lo oo io LO ^T 1 ** ^ ^ ro Tf lo O O t— ro t-h t— On ro NO no tJh rt NO O ro On t— i^N T-H On NO NO Th NO T^ LO oo oooooooo OO ooooo ooo oo oo ooo O rO O^NCNiHCNONfO ro On t— -^ rjn On O On 00 00 t— ON LOt^LOLOOOTHOONO t— On fOt^iOt^O OCMN LO NO ** OO *H CO ON NO t-H r<5 t— ro 1-H CO LO LO acNfono CN LO LO ,-H O tj-i t— t-H O LO t-H OO t-H©000000 OO OrHHrtH OO HH O NO On t— odd t— On ONOOONOt— rONOiO Onlo ■"* O •"* ^ On 00CS t— ©ONT-HONNOOrO LO'— I '^'HOO'H^ t^h no ro no t)h O ON t-H On NO O CN LO t-h "* t-h 00 t-H -rp "Ot^tN lo 00 oooo t— oooooot— ooooOn t— on oo oo t— oo t— oo t— on r— t— ro O OOfO ro NO ro M(N00 00 00 00 On 00 00 On OO OOOOOOOO OO OOOOO OOO oo oo ooo rot— oONHHtono LO TfH LO LO LO LO LO LO LO NO oo oooooooo oo ooooo ooo oo oo ooo o NO'* OiONNCNONNi^ -^ rjn 00 t-H lo NO NO ""* "* ro 00 On NO t— tot^io *-* ro tHh -rH ^Jh 00 lo 00 t-h NO NO ro ro NO tJh ro <* t— ONN t— O O lo t— t— ro 00 ■<* ro LO O O lo On ro no on 00 NO NO LO LO O LO "* LO LO O 00 t— 00 1— iHMN ro t— CM On ro t— © LO Tf LO LO Tf "* -HH LO NO NO LO ■HH LO LO "* LO Th LO LO tH tH r-i t-h O CO r0 t-H t-H t-H t-H t-H ro t-H ro i-H ON O On ^h ro 00 NO ro Th 1 On t^ 00 y* 00 lo lo rf O 00 NO On ro NO ro On r^i ro On O On t*» 00 00 ro ro t^ O t> On t— co ■»* TH(N]iNCN)iHiHrHO roro M M (N cn ro Hr-itN roco roro ^ ro ro ro t-h t— ©NO»--LONOro 00 rfi N^OOO rooOLO Onlo ooo t>> ro ro lo On no lo ro tHh ro ^ t-h tHh lo On O On On O On 00 t-h ro ro t— t- O On O^t^ no lolo OnOnOnOnOnONOnO lonO lolonOlolo nOnOnO lolo nolo iololo lo oono rot—ONOOOOt—NOro -rt^ro rooNNOO\ro ^ ^ O On ^h O On -hh O ro "* no --h ro ■«* ro o t— oo oo 00 ro ro t-h co t-» lo lo o ro ro t-h ro ro t— i t— O\rot— OnooOt— i ^ ro io io ** t— oo t— i *hh ro *— i ro t— no roroO *~- ONro i— iioOn ONroOro 00 LO On h h O no t- no tHh oo ro no i— i ro i— i ro no roro roio •rfroNO ro lonOnOnOO OnOnOn hco conO ONOOOO O tHt-h,-h,-h©©©© ro OOO HHMrtOHOO i-H t-H t-H t-H r0 O T-H OO OOOOOOOO OO OOOOO OOO OO OO OOO O On t— no 00 ** 00 "* On t— ro tH ro "* t— t-h vO On lo On no ro rot— ** t— NO lo y* t-H oo ro o lo ro ro O ONro ro On OO O O On lo On O no On t-h ro t— «hh Tfio rorororoT-iroT-H^H -^ ■"* ro ro ro ro ** ■^f ^ ro ro lo ro ^ en — i OnO no ro ro lo LO 3 o LO C o t— t— C3 LO B 3 13 c bfl 0 NO "* "* -* t* 1— t^. t- ^H 00 t- "* tHH t« t+ t| t-H oo oo oo oo oo oo oo oo oooo <* ^ ^ ^ oo oo oo 10 oooo •* ^ -^ -* -^ on ° « I* >tH CO "3 rt hqO . 1-S-S8 b m KltH 3tH^ W §32 .S-S c b b § 3 3,5 33° mm ra o P< o o ,rt o Cmofi IHVUHIH oooo .2 .2 .2 .2 PhPhPhPh MTf«ino ci c _bj M T ■- bD--j o .So° l!o J3 o-m o^ c Tl^J) T,g W «Oh ^^3 ^ w ci3 p-S O . S'S o 2h 3^3^ Og >»ET. O0 3kim-*.n I "*< LCS CO C^ 0O CT> O 130 EXPERIMENTAL INVESTIGATIONS Sulfur in coke (on basis of original coal) and pyritic sulfur in coal. — When the values for the sulfur in the coke, calculated to the coal basis, were plotted against the pyritic sulfur content of the original coals, it appeared (fig. 46) that there were two groups of points, each defining a straight line. Reference to analyses of the coals corresponding to the points in the lower group showed that the organic sulfur contents of all these samples was remarkably uniform at about 0.5 per cent and that for the upper group, especially those which lay closest to the straight line, the organic sulfur contents were about 1.5 per cent (values from 0.76 to 2.0 averaged). This leads to the conclusion that the organic and pyritic forms of sulfur in the coal contribute proportionately to the sulfur in the coke. The equations for the curves were found by calculation from the individual values to be : Lower curve: S coke =0.61 S + 0.26 Upper curve: S coke =0.63 S p + 0.58 The sulfur in the coke may be represented by the equation : S coke (coal basis) =a S p + b S o The constant (a) is 0.61 and 0.63 in the foregoing equations, and (b) is found to be 0.52 or 0.37, respectively, on the assumption that the organic sulfur contents were 0.5 and 1.5, respectively. The average equation would be: ScoKe =0.62 S p + 0.45 S If the organic sulfur value, S , is varied by finite steps, this equation may be represented on Cartesian coordinates by a series of straight lines, since there is a straight line for each value of S chosen. Three such lines have been drawn => on figure 46 at S = 0.5, 1.5, and 2.5 per cent, respectively. The almost direct proportionality between coke sulfur and total coal sulfur is therefore probably due to the fact that approximately constant proportions of the pyritic and the organic sulfurs remain in the coke; the proportions are as shown in the foregoing — apparently 62 per cent of the pyritic and 45 per cent of the organic sulfur. From a practical standpoint these results are of interest in showing that it is the total sulfur reduction which is important in the coal cleaning, and that, since it is impossible to remove the organic sulfur content of the coal by coal cleaning processes, the organic sulfur content determines the minimum approachable sulfur value. To test the validity of the above relationship, the sulfur contents of the cokes, on the coal basis, were calculated from the pyritic and organic sulfur contents of the coals by means of the equation given above. The results of these calculations are shown in Table 20. EFFECT OF IMPURITIES ON COKE STRUCTURE 13 Table 20. — Comparison of Determined and Calculated Coke Sulfur Contents Coal No. Sp per cent So per cent 0.62 Sp -f 0.45 So S-coke on coal basis detd. Difference Face Samples 11 0.80 0.47 0.71 0.70 +0.01 48 0.66 0.50 0.63 0.67 —0.04 49 0.58 0.51 0.59 0.60 —0.01 50 0.38 0.50 0.46 0.51 -O.05 51 0.66 0.68 0.72 0.65 +0.07 309 0.51 0.45 0.52 0.60 —0.08 492 0.72 0.69 0.76 0.67 +0.09 493 1.71 0.66 1.36 1.09 +0.27 494 0.43 0.58 0.53 0.49 +0.04 495 0.61 0.53 0.62 0.59 +0.03 507 0.66 0.53 0.65 0.67 —0.02 508 1.05 0.58 0.91 0.91 0.00 139 4.58 1.17 3.37 3.38 —0.01 145 3.43 1.18 2.66 2.55 +0.11 135 2.35 1.41 2.09 2.10 —0.01 136 1.97 1.53 1.91 1.87 +0.04 137 2.19 1.39 1.98 1.95 +0.03 27 3.34 2.22 3.07 2.76 +0.31 306 1.85 2.80 2.41 2.21 +0.20 24 2.39 2.05 2.40 2.17 +0.23 26 2.19 2.02 2.27 2.14 +0.13 305 1.54 2.20 1.94 2.00 —0.06 140 2.75 1.67 2.46 2.36 +0.10 141 2.33 1.74 2.23 2.05 +0.18 142 2.53 1.67 2.32 2.21 +0.11 144 2.76 1.39 2.34 2.26 +0.08 369 2.59 1.89 2.46 2.36 +0.10 520 2.55 2.53 2.72 2.44 +0.28 29 1.21 1.91 1.61 1.10 +0.51 30 1.54 1.15 1.47 1.45 +0.02 31 2.03 1.10 1.75 1.87 —0.12 32 1.81 1.07 1.60 1.57 +0.03 37 0.86 0.69 0.84 0.82 +0.02 192 1.47 0.72 1.24 1.19 +0.05 193 0.68 0.60 0.69 0.67 +0.02 194 0.63 0.43 0.58 0.65 —0.07 39 3.16 1.73 2.74 2.24 +0.50 40 2.63 1.55 2.33 2.34 —0.01 149 1.73 1.58 1.78 1.68 +0.10 150 1.31 1.62 1.54 1.41 +0.13 151 1.19 1.66 1.48 1.35 +0.13 152 1.38 1.61 1.58 1.46 +0.12 631 2.01 2.03 2.16 1.76 +0.40 199 0.74 0.91 0.87 1.04 —0.17 361 1.59 0.92 1.40 1.24 +0.16 132 EXPERIMENTAL INVESTIGATIONS Table 20, — Comparison of Determined and Calculated Coke Sulfur Contents (concluded) S-coke Coal No. Sp So 0.62 Sp + on coal Difference per cent per cent 0.45 So basis detd. 518 1.84 0.96 1.57 1.73 —0.16 733 1.91 2.13 2.14 1.78 +0.36 734 2.27 2.33 2.46 2.09 +0.37 153 2.57 1.32 2.19 2.06 +0.13 154 1.49 1.65 1.67 1.57 +0.10 146 3.21 1.94 2.86 2.75 +0.11 147 1.88 1.83 1.99 1.77 +0.22 148 2.53 1.86 2.41 2.25 +0.16 599 2.19 1.02 1.82 1.81 Average +0.01 i 0.13 Weathered Samples 542 1.79 1.91 1.97 1.78 +0.19 543 1.28 1.87 1.64 1.65 —0.01 544 1.54 1.98 1.85 1.68 +0.27 545 0.20 1.99 1.02 1.41 —0.39 546 0.15 1.78 0.89 1.40 —0.51 385 0.71 2.30 1.48 1.07 +0.41 385+Fe 2 3 0.69 2.24 1.44 1.72 —0.28 356-516 0.35 1.40 0.85 0.75 +0.10 356-516+ Fe 2 3 0.34 1.37 0.83 1.36 Average —0.53 : 0.30 Component Samples 48 0.66 0.50 0.68 0.67 +0.01 66 0.23 0.62 0.42 0.60 —0.18 508 1.05 0.58 0.91 0.91 0.00 503 0.23 0.52 0.38 0.30 +0.08 506 0.27 0.17 0.24 0.22 +0.02 350 1.75 0.53 1.32 1.59 —0.27 354 0.49 1.42 0.94 0.78 +0.16 355 0.77 1.68 1.23 1.17 +0.06 356-516 0.35 1.40 0.85 0.75 +0.10 135 2.35 1.41 2.09 2.10 —0.10 134 0.69 0.27 0.55 0.86 —0.31 369 2.59 1.89 2.46 2.36 +0.10 385 0.71 2.30 1.48 1.07 +0.41 387 1.53 2.07 1.88 1.49 +0.39 372 0.16 0.50 0.32 0.70 —0.48 376 0.96 0.41 0.78 1.39 —0.61 377 7.72 0.75 5.12 6.50 —1.38 370 1.51 2.50 2.06 1.65 +0.41 373 1.57 2.60 2.14 1.72 +0.42 374 2.10 2.54 2.45 2.10 +0.35 378 1.36 2.12 1.82 1.50 +0.32 381 1.11 2.02 1.60 1.38 +0.22 Average 0.29 EFFECT OF IMPURITIES ON COKE STRUCTURE 133 Coal Components. — Samples of vitrain (anthraxylon, fusain, and clarain (attritus) were included in this study for the purpose of determining whether or not the sulfur in any of the components behaved differently from that in the others. The values for these samples are presented in Table 21 and are represented on the curves by open circles with letters (V, F, C) designating the component. Although the points representing these samples are reasonably close to the curves indicated by the face samples, certain general trends are evident. The cokes from fusain have higher sulfur contents and cokes from vitrain and clarain have lower sulfur contents than do cokes from face samples of the same sulfur content. The high sulfur retention of the fusain can be explained by the frequent high calcite and pyrite content of fusain. Similarly, because the vitrain and clarain samples were hand-picked, it would be probable that these samples would be low in adventitious mineral matter such as cleats, partings, and pyrite nodules. Samples of coal of low specific gravity obtained by float-and-sink procedure are also included in this group of samples, and, although not sufficient in number or variety to permit an accurate judgment, the results appear to conform with those for face samples. Effect of Weathering. — The samples used in this study were made available in the course of studies on the weathering of coal under other super- vision and were peculiar in the quite wide variation in the proportion of organic, pyritic, and sulfate sulfur. Analysis of the coke buttons yielded anomalous results. The anomalies are thought to be the result of variations in the effects of weathering to which the coal had been subjected, the samples having been taken at points 82, 60, 40, 25, and 15 feet from the entrance of the drift. The analyses show that the organic sulfur content of these samples (Table 22) has probably not been seriously altered, the pyrite, on the contrary, has been oxidized to an increasing extent as the outcrop is approached, and is almost completely eliminated in the sample closest to the surface. The sulfate content of these samples, while unusually high for normal coals, is not sufficiently high to account for the sulfur from oxidized pyrite. As the outcrop is approached, the sulfate sulfur content of the coal increases and then decreases. The ferrous sulfate which was formed by the oxidation of the pyrite has been leached or hydrolyzed to sulfuric acid and hydrated iron oxide. Since the high sulfur content of the cokes from these weathered coals would be unusual if the coals were normal, it was thought possible that the high sulfur retention was due to the iron oxides left in the coal from hydrolysis of the pyrite oxidation products. To test this, 2.5 per cent of ferric oxide (corresponding approximately to the iron associated with 2 per cent of pyritic sulfur) was added to each of two samples of coal, and the coke buttons from the mixture were carbonized. The sulfur contents of the cokes were materially increased by the iron oxide. 134 EXPERIMENTAL INVESTIGATIONS r-~ ^uao jad 7 -00 SO LO dS/0S H /•ppp a^OO-g CO sisBq l^oo uo i^oo-g CN ^U80 J9d siSBq jboo uo ksoo-g - •juao jad aaoo-g o ^uao jad H S V + '0 '£ ON }U80 J8d qsy oo !}uao J9d t^ •JU30 J9d og so ^uao jad •IS u\ !)uaa jad »-osg H !JU80 Jad XS co UOUdUDSSQ CN •o^j JBOO - [609 - o^i o[duit:g o o o o o oo so oi i-i O OMOMO CO '— i *"» CO CO h On r— lo co oo © oo so lo roMOOOO^ lo 00 lo © 00 ooooo NO'OMtN t~~ O On co CO On h 00 t^ so O H so lO ON u co u fjj'rt O^-tNO ■ 00 lO CO t- 00 1— (NO^OO • tJh io i-i On h On no CN no co sO -t^OlcOi— i SO OoOcoOOOOO sOiOfOOvOSOOsOO^M co 00 -^ co locnhc^nOOOcocohcoco lo r-» lo co oooo OO ooooooooooo oooo ■ CO 00 CN O O i-l ■O^MO O On • CO 00 H O so co OCvfOiONNOOOsOO 1-hsOOnoO COCOLOCNCNONLOLOi— IONCN Tt*fCOt-> t^CNcOi— i rfi O sO SO (N i^ oo QOcoOnsO © CN oi h OO O co i— i co O O CN CO — H LO i— I O r- On oi cn i-i co ■ oo so oo r^ On O !>• csi © i-i t"- O i-i On lo r-~ on On OO h 00 00 00 H J>- On O OJ On O On H co sO H o 00\Ot^f5iHON00t^00Nl^ On © t^- t~~ OOOO OOOOOO O i-h o o OnsOsOi-h coon HTfi4LOH©r^-Hi-i,-iLoccs OOCOOLO^t^LOSOLOi-lO H 0O O so \ONN-* OOOOO •OH tH tH h© HtN(NOOONNNtN[Nl H O H *H so co lo co t^ • lo On 1>- lo • t^. H t^ CO lo on CO vo ONi-icosOsOOJi-it—OsOi-i LOt^LOi— lONt^-LOLOi— 1 CO H J2 -D „D ,0 l^NI>N On co 00 00 OO thOO ■h ©OO 04 O rNOHOOr^i-i^HOi^^-i HCSHO lo co O O i-i OOOOO • NO — 1 i-H CO •oooo 33 (NiHCN^lO^iHiHfOr-IM OOOOOcniOOOOO OOOOO •oooo Oo OOOOOOOOOOO i-i 00 co lo lo r» 00 so f- h "*CNOOO ■ co On h t-~ O on 00 on © CN CN © CN i— lOlOOt^-HHLO iOCON^^Oh\OiOh On O *-- co LO I-H LO CO (NrHlNrH co © H CO co © h 00 r^ H H CO co co co co Ol oo lo c c c . rt 3 dJS d fe fe > U > rt 3 c c c c C C CO so CO co so -tsOOOO lO irj lO i^ O N CO On -t i-i "O On On CO On CO CO CO O H io lo >0 'T LO lO iO I co CO CO CO CO sO i-i T-t © 00 On © 1^ co J^- I— O0 i— i -t O0 OO 00 'O OMO N - On O CM On 00 -* NO i-t lo co CM LO N CM OOOO-rH ■^ ** Th 1 LO NO CO LO ^N OOOOO oo oo 00 LO 00 H o N CM ON LO NO n co ^^^^^ ^^ O ^H NOOOl^"* Ot^OOCSCN NO On O co ON 00 CN y-t CO CM CM CM CM CN co T-H CM CO LO O O NO On 00 NO 00 ON O <0 CN LO LO NO NO NO CM O LO LO N CM LO NO CO "^t 1 On On co 00 N CO 00 O O On On CM CN CM 1-H CO O '-H i-H y-t CO CN O LO 00 O0 On O CO O On 00 NO 00 Ttl-^LOLOLO "tf 1 ^ lolo On "tf co OOOr^OiO ho LO-^fi N(NU)(NH N NO coco ^ -H HOO Si S> £> £i & o o o o ^ "-< h o o 00 00 CO 00 CN > 1— i nOnQnOnOnO OO HH 0000000000 OnOn loon 0"^ n3 cm cd ctf rH O O . 1-S-S8 o w •-' o ao rra O -' fj 3 !< «i i C3 -i-i CD tq m o p, gs s ^-i o . o ™ O o C^J Or B 3 g W CD o £j EFFECT OF IMPURITIES ON COKE STRUCTURE 137 cent of sulfur in the coke and coal was also determined. The coke sulfur content was calculated both neglecting the sulfate sulfur content using the expression Coke sulfur = (0.60 S p + 0.45 SJ/coke yield and considering the sulfate sulfur content using the expression Coke sulfur = (0.60 S p + 0.45 S o + 1.00 Sso 4 )/coke yield The data and results of the calculation are presented in Table 23. Inspection of the table shows that the calculated values are low as compared with experi- mental determinations, the values obtained by including the sulfate values checking most closely with the experimental values. The fact that the coke sulfur content is high may be due to the great opportunity present in the apparatus for sulfur compounds which as hydrogen sulfide and mercaptans to crack and yield free sulfur which would combine with the very reactive low temperature coke first formed. The coals had also been dried in air in a finely divided state at a temperature of 105°C, conditions which are favorable to oxidation of pyritic sulfur. Oxidation of pyritic sulfur to sulfate sulfur would cause a much greater retention of sulfur by the coke. Sampling and analytical errors also can account for a considerable proportion of the differences. Removal of sulfur from coke — No practical method of removal of sulfur from coke has yet been discovered. Those proposed involve either ( 1 ) the reaction of the sulfur in the hot coke with steam, hydrogen, or coke oven gas to form gaseous sulfur compounds, mainly hydrogen sulfide, or (2) the addition of chemicals to the charge. The problem of the removal of sulfur from coke is complicated by the fact that the sulfur exists in coke in very intimate organic combination with or in solution in the carbon as well as in the form of sulfides. These intimately combined forms of sulfur are exceedingly difficult to remove, practically necessitating the destruction of the coke. These methods have been discussed and reviewed in the technical literature. 33 GENERAL CONCLUSIONS Illinois coals are of the common bright banded type of coal and are com- posed of jet black, vitreous appearing material derived from larger pieces of woody materials interspersed in a lustrous, striated mass derived from highly fragmented plant material, together with a minor amount of charcoal-like material. Because of the differing friabilities of these components their pro- portions in the sized fractions of naturally formed screenings differ from their proportions in the coal as it existed in the face. Such segregation is not especially 33 Gluud, W. (Jacobson, D. L.), International handbook of the by-product coke industry: American Edition, New York, 1932, pp. 433-4. Schellenberg, A., Uber den Scbwefel in der Steinkohle und die Entschwefelung des Kokses: Brenstoff-Chemie 2 (22), Nov. 15, 1921. Powell, A. R., and Thompson, J. H , A study of the desulphurization of coke by steam: Carnegie Institute of Technology Cooperative Coal Mining Investigations Bull. 7, 1923. 138 EXPERIMENTAL INVESTIGATIONS CO W W $» » CO < CO Ratio S-coke to S-coal OOOO'-iO^hOOO-hO Differ- ence per cent t^fO^OOONCN'^rO'-i( N lfOf N l OOOOOOOOOOOO Actual sulfur content of coke per cent ,-H so lOOONiOCNfO^COOOiSO ^^OtOHHrtNHHOO si CO ^^^t«3HHH(NCNcNbO CJ '5 o lO^t^T^00t>-^-'*'HS0O'-l ■r— i lo ■« — isOiO-^fiOCSOssOsOsO (MCNCN'^hOOO'-iOOOO CJ On co^OfOCNOfC)^'-iO'^ lOCNrHr-I^COSOM3I>NHT-i (NNNCNOOO'-lOHOO CU OS "3 CO o^soaocNMio^H^o -lOOO-HOOOroOOO OOOOOOOOOOOO cu CO tON oo On rO t~~ On OOOn-^OnOIOnOnOnO uuuuuuuuuuuu c c X. (- u HfOiONOOOH(NTt 'a 53° clO^^N00 00 HONHHHHNTf Oj > +-> u n ^ «-< c 3 33 ONiHO\MiOO\ l*H (J O^O-HlOOr-lMDlO 3 i* CO "53 rt *-> o ■ — ! rt . ro fO "* •>— i iO ^f 1 t-~ *— i vO < fO fO ro fO fO fO M '— I jj OJOOCO^ONfOOOrOt^- <% ONNOvoooaaw ft 4_, 3 C «3 U fOHO^NNVO(NN OOO^OCMNCNOVON " S 53 MT3 o .2 rt " o JJ ft »-, \00 — CO "C o &o vC vO < 1 ^Ol g £ Q U Ph -> ■^ -d £ -O o .3 oi i? lr— t— t— oooooo X; : '., .-.: On O OnOnOnOOOOOO'>-H'<-H'— ih ^^ ^^ ^^ ^^ ^"* ^^ ^i^ ^t^ *^d ^ ^f Tt ^ ^d- TtHTjjH-^LOLOLOLOLOLOLOLOLOLO tOOtNO^HOC LO O 00 iflNi- 'I c OONOO •CNONOrHNrHm^lO CN co CN re co *> co CO Cs CN CO CO c- co u- CO LO LO • LO IO CO 00 CO "* 00 CO CN LO 00 -^ ■HCNOMOCNCNOO TjHCMOtNi^a HQ^ ii- H C tH C LO ->• NO •NNHlOHOCN^il OOON^vOX t^NONC On On l> NO OC 001^00 • 00 On 00 00 On OO 00 On 00 >> >» -(J +J c 3 3 C __— o U U ,£ rHVOfOfOOO OOO" 00 00 CN NO 1> X a THHtxHocN^oN^aO'-i 2 rHT^ONH OOO o to c ON OC o rHO^CCOOOONCOHOOHiO s T-H i-IHH i-H i-H i-H H 1-H HH H ti 3 to «5 E o u «4-l o u 13 88888 OOC ooc 8 O "- O -OOcncncnoO o O On ^h -h -h ^HO OC ^hooOoOthOnOOOOOOO u 1-1 ^ ^H ^ ^H T-l ,- •rt >^- T" 1 ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ PQ - W 4© «o © aj N 6 fc aj w CO « CO IO b b u as aS rt t\ ts rt E Cu 4-1 cS CJ IN CN CN CN -^ CN CN CN CN 1—1 CN lO 13 13 aS 13 aS "a 'aflf'a 13 13 13 13 'a aSaSaSaSaSaSaSaSaSaSaSaSaS OOOOOO OOC o o o O C OOOOOOOOOOOOO CJ u u u u u CJ CJ CJ CJ CJ CJ CJ CJ CJCJCJCJCJCJCJCJCJUUCJCJ NO no no nO nO no NO NO NO NO NO NO NO NO nOnOnOnOnOnOnOnOOnOnOnOnO c* 6 6 6 o" c 6 6c 6 fc^W ^ ^ J &^ 3 c^^c^c^^^^^^^^c^^ o o o o o c o o IT o CO iO O LO OOOOCOCOlololoOO OCOOOOOOOOOt^ONONOO O O 00 OO no 00 GO O0 1^ t- t^- On On On 1 — 1 TH HHHrlrl *— 1 1— I CO i— 1 Th NO 00 CN to t— CO o On no t^ LO lOCOt^CN-^OOi— iOnOOOOnnOC " C^ CN o C^ rH e> i— i CN i— CN OJ CN •<— Tf SMALL SCALE TESTS 145 oi rsi 01 «m LO LO lO LO LO LO ^ O r^ rfn Ol ro U OJ 00 LO On PS cc o o oo O LO O LO O oo O oo o o o o o a (j o LO LO LO LO o o o o y LO = s o a; p £ ^ o a -v — i o U 146 EXPERIMENTAL INVESTIGATIONS Table 26. -Size Distribution of Cokes as Recovered from the Retorts and Results of Shatter Tests Size in inches Percent- age of each Percent- S lze distribution (per cent) after shatter test Coke size as re- covered age based on + 1 inch drop No. from retort as 100 1 2 3 4 1 +2 28.1 30.3 5.4 -2 + 1 64.6 69.7 75.0 70.0 61.2 49.0 -1 + H 3.3 17.2 26.1 33.7 44.3 -Yi 4.0 2.4 3.9 5.1 6.7 2 + 2 3.9 4.5 -2 +1 84.0 95.5 86.7 76.5 68.5 62.5 -1 +K 8.4 12.2 21.4 29.2 34.2 -H 2.7 1.1 2.1 2.3 3.3 3 + 2 3.8 4.0 -2 +1 90.5 96.0 78.8 76.8 67.8 63.8 -1 +H 5.3 20.6 21.7 30.1 33.9 -Vi 0.4 0.6 1.5 2.1 2.3 4 +2 34.5 35.7 3.1 -2 +1 63.0 64.3 84.8 80.0 80.0 74.7 -1 +H 1.3 11.3 18.5 18.1 23.1 -¥i 1.2 0.8 1.5 1.9 2.2 5 +2 36.1 39.3 3.2 -2 +1 55.7 60.7 85.7 82.7 76.6 70.5 -1 +^ 5.2 9.9 15.8 21.8 26.0 -Vi 3.0 1.2 1.5 2.6 3.5 6 +2 21.4 22.9 -2 +1 72.5 77.1 84.5 70.5 50.3 41.8 -i + A A 2.3 14.6 27.4 46.5 54.3 -Vi 3.8 0.9 2.1 3.2 3.9 7 + 2 20.2 26.7 -2 +1 74.0 78.3 79.8 70.5 60.7 60.2 -1 +V2 2.5 18.3 27.6 36.9 36.8 -Vi 3.3 1.9 1.9 2.4 3.0 8 + 2 26.3 28.2 -2 +1 66.8 71.8 84.2 75.7 73.3 64.7 -i + l A 4.1 14.6 22.1 23.7 32.2 -Vi 2.8 1.2 2.2 3.0 3.1 9 + 2 56 . 1 57.2 7.4 -2 +1 42.0 42.8 88.0 93.5 92.3 90.7 -1 +M 0.7 3.2 4.5 5.3 5.6 -Vi 1.2 1.4 2.0 3.4 3.7 10 + 2 95.3 98.1 63.5 46.3 45 . 5 45.2 -2 +1 1.9 1.9 34.7 50.1 50.3 49.7 -i +3^ 0.5 0.8 1.8 1.5 1.6 -Yl 2.3 1.0 1.8 2.7 3.5 small scale tests Table 26. — Continued 147 Size in inches Percent- age of each Percent- S lze distribution (per cent) after shatter test Coke size as re- covered age based on -4- 1 inch drop No. from retort as 100 1 2 3 4 11 + 2 65.0 66.6 8.4 4.6 4.6 -2 +1 32.7 33.4 86.3 88.6 86.0 88.8 -1 + Yl 1.0 4.0 4.9 6.7 8.2 -Vi 1.3 1.3 1.9 2.7 3.0 12 + 2 89.0 98.2 81.2 45.2 44.3 31.8 -2 +1 1.7 1.8 16.3 48.8 46.7 57.8 -1 + Y 1.1 0.6 1.5 2.5 2.6 — Yi 8.2 1.9 4.5 6.5 7.8 13 + 2 49.5 74.5 50.8 31.2 31.2 30.0 -2 +1 16.9 25.5 36.7 44.2 37.4 33.7 -l +3^ 5.0 5.2 10.2 11.0 13.5 - A A 28.6 7.3 14.4 20.4 22.8 14 + 2 100.0 100.0 48.3 49.2 37.2 31.1 -2 +1 >1 piece 45.1 43.7 52.7 56.0 -1 +^ 3.5 4.9 5.9 7.7 -Yi 3.1 2.2 4.2 5.2 15 + 2 54.6 57.9 -2 +1 39.6 42.1 94.5 90.2 86.2 85.5 -i + l A 3.8 4.5 8.0 11.4 11.5 — Yl 2.0 1.0 1.8 2.4 3.0 16 + 2 34.3 36.1 4.6 4.6 -2 +1 60.8 63.9 92.5 91.0 94.5 94.0 -1 + Y 1.4 1.4 2.2 2.8 2.6 — Yi 3.5 1.5 2.2 2.7 3.4 17 + 2 93.5 94.8 77.5 46.8 29.1 20.6 -2 +1 5.2 5.2 18.0 44.5 60.4 65.8 -i +3^ 0.3 2.8 5.4 6.4 7.8 — M; 1.0 1.7 3.3 4.1 5.8 18 +2 96.7 100.0 55.4 45.5 36.3 30.8 -2 +1 41.8 50.0 58.1 61.4 -1 +M 0.3 0.8 1.4 1.4 2.8 -Vi 3.0 2.0 3.1 4.2 5.0 19 +2 94.5 97.0 79.3 62.7 64.0 54.4 -2 +1 2.9 3.0 13.0 26.3 21.6 31.3 -1 -YVi 0.3 4.7 5.7 7.3 6.2 — Yi 2.3 3.0 5.3 7.1 8.1 20 +2 56.5 57.5 15.0 4.8 -2 +1 41.7 42.5 83.0 90.5 94.6 91.5 -1 +V2 0.2 0.8 2.4 2.4 4.8 —Yi 1.6 1.2 2.3 3.0 3.7 148 experimental investigations Table 26. — Continued Size in inches Percent- age of each Percent- S ze distribution (per cen after shatter test t) Coke size as re- covered age based on + 1 inch drop No. from retort as 100 1 2 3 4 21 +2 23.2 25.4 -2 +1 68.3 74.6 93.5 86.5 85.7 81.0 -1 +V2 3.9 5.3 11.1 11.3 15.4 -Vi 4.6 1.2 2.4 3.0 3.6 23 +2 69.0 72.9 17.2 -2 +1 25.7 27.1 76.8 89.0 82.2 77.8 -1 + l A 1.9 4.0 7.1 12.7 16.3 -Vi 3.4 2.0 3.9 5.1 5.9 24 +2 2.7 3.2 -2 +1 82.0 96.8 86.8 79.7 71.3 68.2 -i +y 2 11.9 12.5 18.9 26.7 29.4 -V2 3.4 — 0.7 1.4 2.0 2.4 25 +2 42.7 29.6 -2 +1 64.3 70.4 92.7 87.0 82.8 79.5 -i +3^ 3.0 5.5 10.5 13.2 15.5 -Vt 5.7 1.8 2.5 4.0 5.0 26 +2 42.7 45.4 -2 +1 51.4 54.6 84.3 76.6 71.0 70.0 -1 +3^ 1.2 14.6 21.2 26.5 26.9 - l A 4.7 1.1 2.2 2.5 3.1 27 +2 3.7 4.0 -2 +1 89.7 96.0 87.0 71.6 68.2 61.3 -1 +V2 4.1 12.1 26.2 29.2 35.6 -Vi 2.5 0.9 2.2 2.6 3.1 28 +2 45.6 49.6 -2 +1 46.3 50.4 90.7 87.5 82.0 80.2 -1 +V2 3.3 8.1 10.3 14.7 15.7 -Vi 4.8 1.2 2.2 3.3 4.1 29 +2 90.7 93.5 65.0 26.7 36.4 19.2 -2 +1 6.3 6.5 29.2 63.3 51.0 63.0 -1 +V2 0.7 3.5 5.2 6.3 9.5 -y 2 2.3 2.3 4.8 6.3 8.3 30 +2 80.5 82.8 56.4 27.0 23.2 18.0 -2 +1 16.7 17.2 40.7 65.2 65.3 67.1 -1 +^ .6 2.1 5.2 8.1 10.6 - l A 2.2 .8 2.6 3.4 4.3 31 +2 93.8 96.5 63.8 57.0 47.0 34.9 -2 +1 3.4 3.5 31.7 35.8 42.5 53.2 -1 +^ .5 3.3 4.5 7.2 7.8 -H 2.0 1.5 2.7 3.3 4.1 small scale tests Table 26. — Concluded 149 Coke Size in inches Percent- age of each size as re- covered Percent- age based on 4- 1 inch S ize distribution (per cent) after shatter test drop No. from retort as 100 1 2 3 4 32 +2 56.4 57.5 11.1 -2 +1 41.7 42.5 73.7 65.8 60.5 50.5 -1 + l A 1.3 13.1 29.3 33.1 41.1 -Vi .6 2.1 4.9 6.4 8.4 33 +2 85.0 86.7 30.1 6.9 -2 +1 13.0 13.3 64.5 79.2 78.4 73.8 -1 +V2 .6 3.4 9.9 15.8 19.3 -H 1.4 2.0 4.0 5.8 6.9 34 +2 85.9 87.8 42.9 16.9 11.7 -2 +1 11.9 12.2 48.5 69.9 68.2 72.0 -i +y 2 1.3 6.3 9.3 14.4 19.9 -H .9 2.3 3.9 5.7 7.2 35 +2 40.1 41.9 -2 +1 55.6 58.1 89.1 77.4 70.0 69.7 -i + l A 2.2 9.4 20.1 26.5 25.8 -Vi 2.1 1.5 2.5 3.5 4.5 36 +2 73.9 76.2 31.8 16.3 11.5 9.4 -2 +1 23.0 23.76 64.6 74.3 76.6 75.0 -i + l A 1.5 2.2 7.2 8.3 11.1 -¥i 1.6 1.4 2.2 3.6 4.5 37 +2 83.3 84.0 31.2 15.9 7.7 7.7 -2 +1 15.9 16.0 66.2 80.9 85.6 85.6 -1 +H .2 2.0 1.9 4.7 4.3 - X A .6 .6 1.3 2.0 2.4 38 +2 63.9 64.7 16.3 4.3 4.3 -2 +1 34.8 35.3 79.2 86.6 82.9 81.4 -1 +H .1 3.1 6.8 9.8 15.0 -Y2 1.2 1.4 2.3 3.0 3.6 39 +2 23.9 25.3 -2 +1 70.4 74.7 89.4 82.5 73.7 64.8 -i +3^ 3.8 10.5 15.8 23.7 31.9 -y* 1.9 .1 1.7 2.6 3.3 40 +2 58.9 59.6 4.7 4.7 -2 +1 40.0 40.4 90.2 88.0 86.3 84.6 -1 +^ .7 4.5 6.1 11.7 12.9 1 -Vi .4 .6 1.2 2.0 2.5 150 EXPERIMENTAL INVESTIGATIONS Effect of the addition of non-caking carbonaceous material. — Cokes from Illinois coals shrink excessively when devolatilized due to the high volatile matter content of the coal. This shrinkage results in the formation of fingery and much fractured coke which yields an excessive amount of breeze. The addition of low volatile content, non-caking carbonaceous material has been found an effective remedy in certain cases. The present tests show that the addition of coke or dedusting plant dust to Illinois coals tends to give more hlocky and less fractured coke with a higher resistance to shattering; but that, because of the relatively poor caking power of Illinois coals, the coke is more friable and may even crumble quite readily, when appreciable quantities of the non-caking materials are added. Compare figure 47a (coke 3) with figures 47f, 48a, 48b, 48c (cokes 12, 5, 7, and 13) and figure 49c (coke 15) with figure 50e (coke 19). The addition of a strongly caking, low volatile coal to the mixture of coal and non-caking material did not provide any considerable improvement, probably because the Illinois coal and the low-volatile coal became plastic during different temperature ranges. See figures 48d, 48e, and 50f (cokes 10, 9, and 20). The addition of non-caking material such as coke dust or dedusting plant dust (fusain) is recommended only if the coal has sufficient caking power to permit complete incorporation of the inert material in the coke. Effect of low volatile coal. — The addition of low volatile content coal is another way in which excessive shrinkage of coke is sometimes minimized. Brewer and Atkinson 2 have pointed out, however, that the difference in softening point of low and high rank coals may make them non-compatible and that coke from a mixture of two coals of widely different plastic temperature ranges may be weaker than cokes made from either of the coals alone. This was found to be the case with the coals studied in these tests, although there was indication of improvement in some cases. Compare figure 47a (coke 3) with figures 47c, 47d and 47e (cokes 4, 6, 8) ; figure 49c (coke 15) with figure 50d (coke 21) ; figure 49d (coke 23) with figure 50c (coke 18) ; figures 52a and b (cokes 35 and 36) with figures 52c and d (cokes 37 and 38). It is possible that a more compatible low volatile coal can be found, but no attempt to do so was made during the course of this work. Effect of the addition of tar or petroleum oil. — The addition of coal tar or petroleum oil to the coal in an attempt to increase the quantity of cement- ing materials was not encouraging. The addition of petroleum oil gave a poorly- structured, abradable coke with much fines, figure 49b (coke 25), probably because of the well known incompatibility between petroleum oils and coal tars. The addition of coal tar did not improve the structure of the cokes probably because the greater part of the tar was volatilized before the coal became plastic 2 r.rewor, It. E., and Atkinson, R. G., Plasticity of coals: End. Eng\ Chem. Anal. .I'M. vol. 8, No. (i, pp. 483-9, Nov. 1936. SMALL SCALE TESTS 151 and coked. Compare figure 47a (coke 3) with figures 48f and 49a (cokes 26 and 27; figure 49c (coke 15) with figure 51a (coke 28; and figure 49e (coke 17) with figure 51b (coke 29). Influence of the rate of heating and final temperature. — The use of lower rates of heating may cause better coke structure by preventing over- coking and fingering of the coke already formed, but this is not definitely estab- lished. A low final temperature does not tend to prevent much of the fracturing. Compare figures 47a (coke 3) with figure 47b (coke 11) ; figure 52a (coke 35) with figure 52b (coke 36) ; figure 52c (coke 37) with figure 52d (coke 38) ; and figure 49c (coke 15) with figure 49d (coke 23). Effect of added moisture. — The effect of moisture added to prevent segregation of certain of the mixtures was apparently deleterious, unless the effect noted was actually caused by the added materials which were prevented from segregating or being blown out of the charge by the evolved gases. Compare figure 47d (coke 4) with figure 47e (coke 6) ; figure 48a (coke 5) with figure 48b (coke 7) ; and figure 49c (coke 15) with figure 51e (coke 16). Effect of removal of mineral matter. — The effects of reduction of mineral matter on the coke structure did not show up sufficiently in these tests to warrant conclusions. One reason for this was the few tests in which this factor could be directly evaluated; another was the small amount of coal used in most of the tests. See figure 5 Id (coke 40), figure 53a (coke 31), and figure 53b (coke 33). 152 EXPERIMENTAL INVESTIGATIONS Fig. 47. — Cokes from Franklin County, Illinois, No. 6 Coal, Made in Six-inch Retort (3-inch grid). a. 100% No. 6 coal, 20-mesh; 11.1°C. per minute to 1000°C. (Coke No. 3). b. 100% No. G coal, 20-mesh; 4.6°C. per minute to 900°C. (coke No. 11). C. 80% No. G coal, 20% Pocahontas; 10.8°C. per minute to 1100° C. (coke No. 4). SMALL SCALE TESTS 153 Fig. 47 {con't.). — Cokes from Franklin County, Illinois, No. 6 coal, Made in Six-inch Retort (3-inch grid). d. 80% No. 6 coal, 20% Pocahontas, wetted; 12.3°C. per minute to 1100°C. (coke No. 6). e. 00% No. 6 coal, 40% Pocahontas, wetted; 11.8° C. per minute to 1100° C. (coke No. 8). /. 80% No. 6 coal, 20% coke breeze, wetted; (no temp, data) (coke No. 12). 154 EXPERIMENTAL INVESTIGATIONS Fig. 48. — Cokes from Franklin County, Illinois, No. 6 coal, Made in Six-inch Retort (3-inch grid). a. 80% No. 6 coal, 20% dust; 10.0° C. per minute to 1100° C. (coke No. 5). b. 80% No. 6 coal, 20% dust, wetted; 10.0°C. per minute to 1100°C. (coke No. 7). c. 75% No. 6 coal, 20% coke breeze, 5% dust; 9.9°C. per minute to 1100°C. (coke No. 13). SMALL SCALE TESTS 155 Fig. 48 {con't.). — Cokes from Franklin County, Illinois, No. 6 coal, Madi<: in Six-inch Retort (3-inch grid). d. 70% No. 6 coal, 10% Pocahontas, 20% coke breeze, wetted; 10.8°C. per minute to 1100° C. (coke No. 10). e. 73% No. 6 coal, 20% Pocahontas, 7% dust, wetted; 5.8°C. per minute to 1100°C. (coke No. 9). /. 1)5% No. 6 coal, 5% tar; 10.2°C. per minute to 1025°C. (coke No. 20). 156 EXPERIMENTAL INVESTIGATIONS Fig. 49. — Cokes from Franklin and Randolph counties, Illinois, No. 6 coal, Made in Six-inch Retort (3-inch grid). a. 00% Franklin Co. No. G coal, 10% tar; 9.0°C. per minute to 1050°C. (coke No. 27). b. 95% Franklin Co. No. 6 coal, r>% petroleum oil; 8.7°C. per minute to 1000° C. (coke No. 25). SMALL SCALE TESTS 157 Fig. 49 (con't.). — Cokes from Franklin and Randolph counties, Illinois, No. 6 coal, Made in Six-inch Retort (3-inch grid). c. 100% Randolph Co. No. 6 coal, 20-mesh ; 11.1° C. per minute to 1100° C. (coke No. 15). d. 100% Randolph Co. No. 6 coal, 20-mesh; 10.1°C. per minute to 800° C. (coke No. 23). e. 100% Randolph Co. No. 6 coal, 4-mesh ; 4.8°C. per minute to 1000°C. (coke No. 17). 158 EXPERIMENTAL INVESTIGATIONS a V •••■>■ ■'• •:■ * :"> - AnBIIa ^ Fig. 50. — Cokes from Randolph County, Illinois, No. 6 coal, Made in Six-inch Retort (3-inch grid). a. 100% No. coal, 4-mesh; S.1°C. per minute to S00°C. (coke No. 22). 7>. 100% No. 6 coal, 4-mesh; C>.9°C. per minute to 1100°C. (coke No. 14). c. 80% No. G coal, 20% Pocahontas; 10.2° C. per minute to 975° C. (coke No. 18). SMALL SCALE TESTS is 1 ; Fig. 50 (con't.). — Cokes from Randolph County, Illinois, No. 6 coal, Made in Six-inch Retort (3-inch grid). d. S0% No. G coal, 20% (coke No. 21). e. 80% No. 6 coal, 20% (coke No. 19). /. 75% No. 6 coal, 20% Pocahontas, 5 1025° C. (coke No. 20). Pocahontas; 8.4°C. per minute to 1000°C. coke breeze; 7.9°C. per minute to 1000° C. dust; 13.7° C. per minute to 160 EXPERIMENTAL INVESTIGATIONS Fig. 51. — Cokes from Randolph County, Illinois No. 6 coal, Made in Six-inch Retort (3-inch grid). a. 95% No. 6 coal, 20-mesh, 5% tar; 11.3°C. per minute to 1025°C. (coke No. 28). b. 95% No. (J coal, 4-niosh, 5% tar ; 8.9°C. per minute to 1025°C. (coke No. 29). SMALL SCALE TESTS 161 Fig. 51 (con't.). — Cokes from Randolph County, Illinois, No. 6 coal, Made in Six-inch Retort (3-inch grid). c. 100% No. G coal, wetted; 11.0°C. per minute to 1000°C. (coke No. 16). d. 100% No. G coal cleaned by floating on liquid of 1.5 sp. gv. : 6.1°C. per minute to 1000° C. (coke No. -10). 162 EXPERIMENTAL INVESTIGATIONS Fig. 52. — Cokes from Saline County, Illinois, No. Six-inch Retort (3 -inch grid). 5 coal, Made in 100% No. 100% No. >al ; 12.7°( 1 . oal; 8.4° C. per per minute minute to L000°-C. to 850° C. (cok SDK 36) . SMALL SCALE TESTS 16': Fig. 52 ( co n't.). — Cokes from Saline County, Illinois, No. 5 coal, Made in Six-inch Retort (3-inch grid). c. 85% No. 5 coal, 15% Pocahontas; 5.4° C. per minute to 1000° C. (coke No. -°>7). d. 85% No. 5 coal. 15% Pocahontas : 9.2° C. per minute to 850° C. (coke No. 38). 164 EXPERIMENTAL INVESTIGATIONS Fig. 53. — Cokes from Various Coals, Made in Six-inch Retort (3-inch grid). a. 100% Franklin Co. Illinois No. coal, minus %-inch size, washed: 7.2° C. per minute to 1000 °C. (coke No. 31). b. 100% Perry Co. Illinois No. (►coal, minus %-inch size, washed: 5.6°C. per minute to 825° C. (coke No. 33). SMALL SCALE TESTS 165 ■K m z? :: ■■ ■ ::::.ifi»Iil Fig. 53 (con't.). — Cokes from Various Coals, Made in Six-inch Retort (3-INCH GRID). c. 100% Woodford Co. Illinois No. 2 coal, 20-mesh ; 11.1°C. per minute to 1100° C. (coke No. 2). (1. 100% West Virginia Pocahontas coal. 40-niesh ; 9.3°C. per minute to 1100° C. (coke No. 24). 166 EXPERIMENTAL INVESTIGATIONS BOX TESTS IN KNOWLES SOLE-FLUE OVENS While making other tests in connection with the operation of sole-flue ovens, advantage was taken of the opportunity to make basket tests on prepared screenings from Franklin County No. 6 coal. In order to show the effect of different methods of treating or preparing coal screenings the baskets used to hold the coal samples were made of angle iron welded to form a frame 18 inches long, 12 inches wide and 10 inches high, covered with expanded metal lath (fig. 54). The filled baskets were placed on the floor of the oven in a space cleared of the regular oven charge and the oven charge then filled in around them. The experimental charges were thus coked under conditions as close as possible to those to which the normal oven charge was subjected. Description of the coals, analyses of the coals and cokes, and results of shatter tests on the coke are presented in Table 27. Photographs of specimens of the cokes obtained are shown in figures 55 and 56. Photographs of cross-sections prepared according to the method of Rose are shown natural size in figure 57 and at a magnification of seven times in figures 58, 59 and 60. Fig. 54. — Metal Basket Used in Coking Tests. Size 18 inches long, 12 inches wide, and 10 inches high) Results of box tests in Knowles ovens. — In comparing the coke made in the baskets with that made at the same time outside of the baskets as the usual oven charge, the apparent effect of including the coal in the basket must be taken into consideration. The cokes made in the baskets had a much greater tendency to form fingery pieces, as is evident from inspection of the photographs of coke 71 (figure 55b), made in a basket, and coke 72 (figure 55c) made in the oven as part of the regular charge from the same coal. These fingers radiated from the central part of the top of the charge to the lower corners and edges. This tendency of the coke made in the baskets to acquire a fingery structure as a result of its isolation from the rest of the oven charge must be taken into consideration in judging the experimental results on coals varying in character from the usual charge. SMALL SCALE TESTS .167 OH w * £ c- c < £ K ^ o c< i ^^ 1 CN OJ ON 00X-~ O ^ OJ CN CN ■>—( CN CN CN 3 o ^ bo £ U^CQ tH H ,— ' t-l 1 OO o OO nO On On no 4^1 <-*-, CO ^H (>] OO^ro 3 cu 'd *-> 3 O CN ^H t—t T- 1 T^ T— 1 .3 (U CJ CO re ■}-> £ S 5J ^ 3 I-H CO Ph LO CN ^ ] X^- lo lo On co x-£ ^ o co cn X^ oo O O O S u °co oo oo CO 00 oo 00 oo -^. cu u i— ( t~^ CM NO CO o o O -H ^f -fl O CON ^h LO NO NO NO co cn fO CO CN CN CN o^« i—i i i ' r-< 1— 1 1— I r-H T-H CJ LO O , O CN NO NO LO ^O NO NO LO r^ xf^ E fl "3 oo o O O O O JV o M .y ^ oo On On On On no no Ph 4j .31 ^5 LO CN VO \0 cn On On £ s CU ^~3 O — < O O ^h O O co CJ Dh w "rt O CJ cu — H S-, -* o CM O t-O c-O CO ^ +J <_£ ■<-< On ro CO 00 ^h r^ CO O ' — ' O re H g T—t T— 1 ' ' ^ ^ ^ ^ CO , T3 g Ci re M CJ i-H -^ -^ tH O On On "rt Q ■^ LO l-~ 00 no co co 3 < LO LO LO LO LO LO LO CD ^ r3 dj t-H ^ CO J>~J.-- O O > 6 NO -^ LO lO T-l TjH T^H ro ro CO co co co co _3 oo 04 00 CNCO'HrH co < On O NO NO CN CN CN i—i i— LO • re t3 ,— ' TZi J=. J £ o rt ^ cu CO re % . ^ V* • CO ■St |.| C '£ .JJ 3 in ■ - 1 ST 1 S CU CU re ^ 3 co 4 " J '3 C CJ •3 -^ CO O rt t4_ sC O ,_ c .§£ ' u u CO CI c a ^ K TO o u b £ O D x b P .3 8 c c C (U CJ , o > > co o O cu £ s-h" ci g ft.' LO -3 »- >-h Q bfl-'- co-C \-t . o o \y co X X A WW CJ LO 3 xj 3 3 3 re 7 ^ &" \00 — L — i ^*±!<*. ■ l^N t— 1 rf 1 ro ro ro rt O £ r a On CO On On On On On <-C ^ c U pq (U a oo -^ ^H O 00 00 o £< ti rt 2 X J^ CO O fO NO f— rHPO J^ rt ^u 00 !>■ 00 00t-> f— !>■ 8| ^Q In "o rt Q ^ iH .,- X rt t; o u . ^ u ^_ TT M xj u C Oh • S £ <- o o n XXT r- ^J= J- (U i- & V M_q be 1 o C/) NO t— O 00 On tH ro NO MO !>■ no no t-; !>■ Ut toto U* toto Pi SMALL SCALE TESTS 169 The coal used in the box tests was very similar except for size to the coal regularly charged into the oven at the time the tests were made. It was taken from one of the mines which supplied coal to the coke ovens. The influence of cleaning on the strength of the coke is shown by the results of the shatter tests on cokes 67 and 70, Table 27. The amount of coke remaining on the one-inch screen at the end of the test increased from 68 per cent to 82 per cent while the amount passing the one-half inch screen dropped from 12.0 per cent to 8.4 per cent. That much of this minus half-inch material is composed of shale or other high-ash material which is only loosely held by the coke mass is indicated both by the decreased amount in the coke from the washed coal and also by the lower ash content of the minus-half-inch material from the washed coal and by the smaller difference between the ash content of the larger and smaller coke from the washed coal. Comparison of the magnified Rose section of cokes 67 and 70 (figs. 59a and 59b) shows the effect of the reduction of mineral matter by coal cleaning on the amount of non-fusing material in the coke. Many non-cellular dark areas may be noticed in figure 59a which are absent in 59b. These dark, non-cellular areas represent pieces of shale or high-ash material. The washed coal showed a greater tendency to produce a fingery coke (coke 70, fig. 56b) than did the raw coal (coke 67, fig. 56a), probably because of the greater amount of moisture in the washed coal. These effects of reducing the mineral matter content of coal on the resulting coke are also evident from a comparison of cokes 68 and 71 which, while not from the same coal, are made from material of similar size. The coke from the washed coal, No. 68, had 90 per cent remaining on the one-inch screen compared with 80 per cent for the coke from the raw coal, No. 71. The ash content of the minus one-half inch material was 8 per cent higher than the ash content of the plus half-inch material in coke from raw coal and only 1.8 per cent higher in coke from washed coal. The coke made from the minus 10-mesh material (No. 69, fig. 56d) is typical of coke made from fine coal containing much inert matter, such as fusain. It has good resistance to breakage by shattering but is soft and friable. Its friability was more apparent on handling than is indicated by the amount of minus half-inch material formed on dropping. Its blocky structure is striking when compared with the other cokes in this series. The large amount of fine fusain in this coke shows up in the Rose sections of it in figures 57 and 60b as small dark non-cellular fragments. The coke No. 66 from the 2 X ^4-inch washed screenings which had been crushed in a hammer mill so that the entire sample passed a screen with 14 -inch openings gave a small celled, rather uniform coke with a tendency toward cross fracturing, figures 55a, 57 and 58a. The coke from the ^-inch by 10-mesh coal (No. 68) showed a greater tendency to fracture than did the other cokes, probably because of the removal of the minus 10-mesh material which contained a high proportion of non-shrinking fusain, figures 56c, 57 and 60a. 170 EXPERIMENTAL INVESTIGATIONS GENERAL CONCLUSIONS ON THE USE OF ILLINOIS COALS FOR MAKING COKE Because of the relatively high volatile matter content of Illinois coals, cokes made from them tend to form slender fingery pieces. The formation of the fingery pieces is caused by shrinkage due to loss of volatile matter after the plastic coal mass has become solid. The formation of shrinkage cracks can be greatly decreased by adding low volatile matter content, non-fusing carbonaceous material to the coal to be coked. However, Illinois coals have too low caking powers to hold this non-fusing material firmly. Consequently cokes from such mixtures are soft and abradable even though they may resist shattering. Because of the tendency of ash forming material to concentrate in the naturally formed fine sizes of coal commonly used for coke making and because this ash forming material decreases the value of the coke, it is usually desirable that the cokes be made from properly cleaned coals. The coking process should be so regulated that the coke already formed in the oven is not overcoked before the remainder of the charge is coked. Illinois coals should be coked as soon as possible after removal from the mine face. The coal should not be ground too fine and the oven charge should have as great a bulk density as possible. The coal charge should not be too wet when charged into the oven. With reasonable care, coke which is satisfactory for domestic fuel, many industrial and some metallurgical uses, can be made from most of the coals found in Illinois. Exceptions would be those coals which have lost their caking properties due to weathering, either because they are now, or have been at some time past, too close to the surface. SMALL SCALE TESTS 171 Fig. 55. — Basket Test Coke and Plant-run Coke (3- INCH GRID, Coke No. 66, made in basket test from 2 X % inch washed slack, crushed. Coke No. 71, made in basket test from regular plant feed, minus 5/16-inch screenings. a. b. c. Plant-run coke No. 72. 172 EXPERIMENTAL INVESTIGATIONS Fig. 56.— Cokes Made in Basket Tests (3-inch grid). a. Coke No. G7 made from % x inch raw screenings. h. Coke No. 70 made from % X inch washed screenings. SMALL SCALE TESTS 173 Fig. 56 (con't.). — Cokes Made in Basket Tests (3-inch grid). o. Coke No. (58 made from % inch X 10-inesh washed screenings. d. Coke No. G'J made from dried minus 10-mesh screenings. 174 EXPERIMENTAL INVESTIGATIONS mmirm ll£§3sf ImS&I *th *> & ■re*-; -^ ; !^ Fig'. 57. — Rose Cross-sections, Naturae size, of Cokes Made in Basket Tests a. (Joke No. <»<; (fig. 55a), from 2 x % inch washed slack: crushed. />. Coke No. 71 (fig. 55b), from minus 5/16-inch screenfngs. <•. Coke No. <>7 (fig. 56a), from % x inch raw screenings. (I. Coke No. 7<> (fig. 56b), from % x <> inch washed screenings. e. Coke No. 68 (fig. 56c), from % inch X 10-mesh washed screenings. /. Coke No. 69 (fig. . r >(!d) from dried minus 10-mesh screenings. SMALL SCALE TESTS 175 Fig. 58. — Rose Cross-sections at a Magnification of Seven Diameters. washed screenings. SMALL SCALE TESTS 177 Xft •Kg**. ^Z" *& •HtfV n' ■'*. **""%. ^ *-*©» >^ •aa.* v t« ^/^ Fig. 60. — Rose Cross-sections at a Magnification of Seven Diameters. a. Coke No. 68 (figs. 56c and 57e), from % X 10-mesh washed screenings. &. Coke No. 69 (figs. 56d and 57f), from dried minus 10-mesh screenings. CHAPTER XVI— AN INDUSTRIAL COKING PLANT UTILIZING ILLINOIS COAL Coke is produced commercially at the present time from Illinois coal only at an installation of 26 Knowles type, sole-flue ovens at West Frankfort, Frank- lin County. Because the Knowles coke oven is a relatively new type of carbon- ization equipment of which descriptions are not commonly available and because it is now being used with apparent success in the production of a domestic smokeless fuel (coke), a description of its construction and operation seems pertinent. For satisfactory operation of the slot-type coke oven the coal charge must have certain definite properties. The charge must not swell excessively or develop any great pressure against the side walls or serious damage to the oven results ; the coke must shrink away from the walls in the final stages of coking so that the charge may be pushed from the oven without damaging the walls and, finally, the coke must have a coherent structure so that it will move as a single block under the influence of the pusher ram, and not crush and jam against the walls. This last requirement offers the most serious difficulty encountered when coking coals of the Eastern Interior basin in ovens of this type. Because of their high volatile matter and moisture content, these coals have a tendency to form slender, fingery pieces during the final devolatilization and shrinkage period. These fingery pieces tend to cause the charge of coke to crumble in front of the pusher ram and to jam in the oven, making it necessary to maintain very close control over the coking operation. The difficulties are further aggravated by the fact that these coals rapidly lose their coking properties upon storage. It was to overcome these difficulties that the Knowles oven in which the coal is heated from below in a wide, flat layer was applied to Illinois coals (fig. 9). The charge is more easily removed from the oven even if it should crush in front of the pusher ram. A coke more suited to domestic use in which the coke grades from a low-volatile-matter content, high-temperature coke at the bottom to a high-volatile-matter content, low-temperature coke at the top can be made if desired and pushed from the oven. The present installation consists of three batteries, one of ten and two of eight ovens each, 1 shown in panoramic view in figure 61. i Anon., Knowles oven widens market lor No. 5 Illinois screenings by turning fines into domestic coke: Coal Age vol. 39, pp. 421-423, Nov. 1934. McBride, R. S., Processing coal in Knowles coke oven. Chem. and Met. vol. 42, No. 6, pp. 300-3, June 1935. [179] :* • , - if »!•«.:'.■ .%v--t. i.r: *- AN INDUSTRIAL COKING PLANT 181 These ovens are built in the form of an arched rectangular chamber, approxi- mately M) feet long, 7]/> feet wide, and 4 feet high at the top of the arch and 2 feet high at the sides. Heat is applied through Hues which form the floor or sole of the oven. These Hues are connected to regenerators built into the oven founda- tions. Figure 62 shows construction details of the ovens in cross and horizontal section. Figure 63 shows a view of the coke side of the ovens, while figure 64 shows another view of the coke side with coke quenching car in position as an oven is being discharged. The ovens are charged with 5 tons of "No. 5 carbon" (5/16-inch) screenings making a layer 10 to 12 inches thick. Coking time has varied from 8]/> to 1 V > hours. Temperatures in the flues are about 1360°C. (2480°F.) with the floor temperature at the completion of a coking cycle at about 1100°C. (2000°F.) Historically, this type of oven is derived from the recuperative beehive oven in which the products of combustion of the volatile part of the coal are brought under the oven flooi in a flue system. 2 The sole-flue oven has been developed both as a recuperative'' and as a regenerative 4 by-product recoverj oven. The ovens which have found commercial use have been regenerative, although some non-heat recovery ovens of this type are in use in the petroleum industry for the production of a dense, low volatile- matter content petroleum coke from petroleum residues.' OPERATION OF KNOWLES SOLE-FLUE OVENS In the investigation of the operation of these ovens the temperature conditions in the body of the charge of coal being coked were considered especially important in view of differences in the manner of applying the heat to the coal in these ovens compared to other coal carbonizing equipment. The importance of the rate of coking to the production of good coke from high-volatile content, poorly caking coals is mentioned in the coke literature.' 1 Temperature conditions inside of coal charges being coked in slot-type by-product ovens and in vertical continuous gas retorts have been studied by various investigators and are available in the liter- ■-' Bone, w. A., Coal and its scientific uses: Longmans, I^ondon, 1910, p. 302. Sole-flue beehive oven patented by Breckon and Dixon, 1858. Rectangular ovens mentioned. Fulton, J.. Coke: international Textbook Co., Scranton, Pa., 1906, p. 173-7. The Ramsay Patent Beehive Coke Oven. [bid., p. 177-8, I>aiiiic's Economic Down-Draft Coke Oven. Moss. It. S., Improved Heminway Process: Mines and Minerals, vol. 21, No. (;, pp. 412-4, April 1901. 3 Zwillinger, B. U. S. Patent 1656617, Jan. 17, 1928. Zwillinger, B. U. S. Patenl L428621, Sept. 12, 1922. von Bauer and Zwillinger, B., U. S. Patent 1478410, Dec. 2.",, 1923. 4 Knowles, A. S., and Mclntire, C. V., U. S. Patent 1635280, July 12, 1927. Knowles. A. S., U. S. Patent 1745996, Feb. 4, 1930. a Campbell, O. F., Petroleum coke makes ideal household fuel when its properties are understood: Oil and Gas Journ. vol. 33, No. 45, pp. 68-69, March 28, 1935. Knowles, A. S., U. S. Patent 1717884. June 18, 1929. Ziegenhain, W. T., Tidal refining company making high quality coke in Knowles unit: Oil and Gas Jour. vol. 30, No. 17, pp. 16-17, 100, Sept. 10, 1931. o Ovitz, F. K., Coking of Illinois coals: U. S. Bureau of Mines Bull. 138, p. 17, 1917 Roberts, A., Art of coking coal: U. S. Patent 1352696, Sept. 14, 1920. 182 EXPERIMENTAL INVESTIGATIONS AN INDUSTRIAL COKING PLANT is: ature for purposes of comparison. 7 Through the courtesy of the Radiant Fuel Corporation, owners of the Knowles ovens at West Frankfort, it was possible to study the temperature conditions inside a charge being coked in this type of oven. Fig. 63. — Coke Side of Knowles Sole-flue Ovens (From "Coke from Illinois Coal," by G. Thiessen, Ind. and Eng. Chcm. vol. 29, May 1937.) OBJECTIVES OF THE INVESTIGATIONS The objectives of these investigations were to determine the temperature conditions inside a charge of Illinois coal being coked in the sole-flue type of coke oven, including a determination of the rate of travel of the plastic zone, the temperature gradients, and the final temperature achieved in the charge at selected points ; to measure regenerator and flue temperatures ; to determine the volatile-matter gradient from bottom to top in the finished coke ; and to determine average gas composition. 7 Cooper, G. S., The by-product coking- industry and its relation to the manufacture of iron and steel. Appendix II. Temperature Measurements in a Koppers Oven: Jour. Iron and Steel Inst. vol. 90, pt. II, pp. 17-47, 1914. Hilgenstock, R. "W., Uber Destillations-Cokerei: Journal fur Gas Beleuchtung 45, pp. 617-21, 1902. McBride, R. S., and Selvig, W. A., Coking- of Illinois coal in Koppers type oven: U. S. Bureau of Standards Tech. Paper 137, 1919. Simmersbach, O., Berg- und Htittenmannische Rundschau lf>7, 1913. 184 EXPERIMENTAL INVESTIGATIONS Fig. 64. — Coke Side of the Knowles Ovens with the Quenching Car in Position. (From "Coke from Illinois Coal" by G. Thiessen. Ind. and Eng. Chem. vol. 29, May, 1937.) TEMPERATURE MEASURING EQUIPMENT Temperatures in the charge being coked in the oven selected for the investi- gation were measured at the center line of the oven 6i/? feet from the door, directly below 7 the outer charging hole line at seven points spaced above the floor by means of the following arrangement : Seven chromel-alumel thermo- couples, made of number 22-gauge wire, 12 feet long, were insulated with 2-hole sillimanite insulators and enclosed in i/J-inch steel pipes welded shut at the end for protection. These protecting pipes were welded to steel strips in such a way that they lay flat on the floor of the oven as they passed under the door and were then directed upwards toward the center line of the oven where they were supported in a vertical sequence by a supporting strip with the lowest couple on the oven floor and the others successively 1, 3, 5, 7, 9, and 10i/? inches above the floor (fig. 65), the last about at the surface of the charge. The pipes contain- ing the wires extended about a foot in a horizontal line beyond the supporting strip in order to minimize possible errors caused by the conduction of heat along the pipes. In figure 65 the location of the couples with respect to the ovens is as follows: The end of the pipes extend 8 inches beyond the edge of the 4-foot AN INDUSTRIAL COKING PLANT 185 PLAN VIEW ELEVATION walkway extending along the ends of the ovens, pass over the walkway and under the 10-inch thick door, and then gradually curve upward and toward the center line of the oven. Oven number nine was selected for this investigation because its operation was typical of those in the battery ; it was not an end oven and yet was near to the end of the battery for convenient loca- tion of test equipment out of the way of operating equipment. The temperatures measured by the couples located 1, 3, 5, and 7 inches from the floor were automatically recorded by a Brown Instrument Co. four- point recording potentiometer pyrometer, whereas the temperatures measured by the remaining couples were manually read every five minutes with a Leeds and Northrup portable potentiometer. Temperatures in the regenerator outlet pass and in the stack were measured with chromel-alumel couples and the portable potentiometer. Flue temperatures were measured with a Leeds and Northrup optical pyrometer. In addition to the thermocouple protection tubes, four similar open-ended tubes were included in the assembly for measurement of gas pressure in the charge. The pyrometers and other test equipment were mounted on a framework at the end of the battery and leads were so placed that connections could be made quickly after the thermocouples were in place. - 8'+ 4' f I O'f Fig. 65. — Thermocouple Well Assembly. (From "Coke from Illinois Coal," by G. Thiessen. Ind. and Eng. Chem. vol. 29, May, 1937.) PROCEDURE FOR TEMPERATURE MEASUREMENTS The thermocouple assembly was lifted into position on the floor of the empty oven, from which the charge had just been pushed in such a position as to clear the base of the charge-leveller shoes. The oven was then charged and levelled with care so as not to disturb the thermocouples. The doors of the oven were then closed and luted and the connections to the measuring instru- ments made. The automatically recording potentiometer connected to the couples at 1, 3, 5, and 7 inches from the floor was started, recording temperatures about ten minutes after the oven was charged. Manual readings of the tempera- tures indicated by the thermocouples on the floor, at the top of the charge and 9 inches from the floor were not started until almost 45 minutes after the oven was charged, due to activities of all observers in connection with the leveling and door closing procedure. Thereafter these temperatures were read every five minutes. Upon return to the laboratory, temperatures were read off the recorder chart at corresponding five-minute intervals and combined with the manual readings in the form of a table and were plotted. 186 EXPERIMENTAL INVESTIGATIONS SAMPLING AND TESTING OF COAL, COKE, AND GAS Samples of the coal being coked at the time the tests were made were taken from the belt conveyor carrying the coal from the railroad cars to the storage hopper on the oven. Laboratory samples were quartered from the bulk samples, the remainder being used in one of the box coking tests conducted. A sample of / / / / / 1 V y i / 2 i / / / i t i / / / // / / 1 / / // /- / '/ A / 1 // / i I / V / // I 1 / / / / / i // // 4 6 A % 7 l( i j / \V y 200 400 600 800 1000 1200 TEMPERATURE- DEGREES C Fig. 66. — Temperature-Time Relations at Various Points in the Sole-flue Oven Charge. (From "Coke from Illinois Coal," by G. Thiessen. Ind. and Eng. Chem. vol. 29, May, 1937.) 1. On oven floor. 2. One inch from oven floor. 3. Three inches from oven floor. 4. Five inches from oven floor. 5. Seven inches from oven floor. 6. Nine inches from oven floor. 7. Ten and a half inches from oven floor. coke representing the oven charge made during the temperature measurements was taken from the coke conveyor belt. Individual pieces representing the coke at the end of the thermocouples were carefully removed as the coke quenching car was being emptied. The piece of coke directly at the ends of the thermo- couples was recovered in one large piece. The analyses were made in the labora- tories of the geochemical section of the Illinois State Geological Survey using A.S.T.M. standard methods. AN INDUSTRIAL COKING PLANT 187 Gas samples were taken over a period of 40 minutes, from the main gas line before and after sulfur removal, by displacement of water in 5-gallon glass bottles. Gas analyses were made both with the Orsat-type gas analysis equip- ment and with the Podbielniak gas fractionating equipment. RESULTS OF TEMPERATURE MEASUREMENTS Temperatures in charge. — Table 28 presents the temperatures measured inside the charge at five-minute intervals from a time shortly after the coal was introduced into the oven to a time just a few minutes before the coke was pushed out of the oven. The temperatures at the seven places in the charge are shown plotted against time in figure 66 as smooth curves through average values. The actual curves for points 1 and 2 and the later part of the curve for point 3 show a regular wave form caused by temperature changes in the combustion flues brought about by the half-hourly reversal of regenerators and burners. Being closest to the flues, the thermocouple resting on the oven floor very definitely showed this effect. In figure 67 the isothermal lines for each 50°C. temperature interval have been plotted and in figure 68 the isochronal lines for the same data have been plotted for hourly intervals, to assist in making the significance of the results apparent. EXPERIMENTAL INVESTIGATIONS Table 28. — Temperatures in Charge of Illinois Coal Coked in Oven 9, Knowles Oven- Installation of Radiant Fuel Corporation, West Frankfort, Illinois Noon to 8 :45 P. M., Oct. 10, 1935 Time Thermocouple 21 1 2 3 4 5 6 7 12:12 ,.°C 575°C 240°C 200°C 220°C ,°C ...°C 12:15 600 220 187 207 12:20 642 198 170 190 12:25 770 186 159 175 12:30 700 180 152 167 12:35 717 178 149 161 12:40 735 178 146 157 12:45 12:50 (12:44)* 507 750 762 180 182 145 144 155 153 (12:44)162 (12:44)264 12:55 1:00 1:05 (12:56)* (1:01)* * 527 770 $27 774 532 785 190 198 210 144 144 144 151 150 148 (12:56)i57 (12:56)266 (1:01)157 (1:01)264 157 268 1:10 * ,39 793 223 144 148 162 268 1:15 * >49 797 240 144 150 156 283 1:20 * 556 805 256 147 147 157 286 1:25 * >66 813 276 147 147 162 292 1:30 * ,76 822 297 146 146 159 300 1:35 * >86 827 315 147 147 164 307 1:40 * ,91 833 340 150 148 169 310 1:45 £ >96 837 360 151 149 167 314 1:50 * ,91 838 380 152 149 172 314 1:55 * ,91 840 396 154 150 172 319 2:00 * >91 841 411 156 150 174 324 2:05 * ,91 843 427 157 151 177 326 2:10 * >94 846 446 159 152 179 326 2:15 * ,98 850 466 161 153 184 331 2:20 c >04 858 491 163 155 184 334 2:25 c >11 867 514 165 156 187 336 2:30 9 28 876 543 167 157 192 33S 2:35 c >34 886 568 170 158 194 341 2:40 c >38 895 593 174 160 194 341 2:45 c >41 900 620 177 161 197 341 2:50 <; 44 905 645 182 162 202 343 2:55 9 >44 907 664 187 164 202 343 3:00 910 680 193 166 3:05 c )44 912 695 200 167 209 350 3:10 5 ►46 915 708 207 171 212 353 3:15 c >51 918 717 216 174 214 350 3:20 c >56 922 729 225 176 217 353 3:25 5 64 927 738 235 178 219 355 3:30 c >74 935 750 247 180 224 358 3:35 5 >79 942 760 257 182 227 355 3:40 c >87 950 770 270 184 229 360 3:45 i >87 952 780 280 186 232 360 3:50 c >87 954 788 292 190 236 362 3:55 c )87 953 793 308 192 239 364 4:00 c >84 953 800 323 195 244 367 4:05 c )76 955 805 339 198 242 362 4:10 c >79 956 808 355 200 242 362 4:15 c )84 961 812 373 204 246 364 4:20 c )92 966 817 388 207 246 364 4:25 5 97 971 824 408 211 249 367 4:30 1( )02 977 830 430 214 254 367 4:35 1( )02 980 835 450 215 259 369 an industrial coking plant Table 28. — Concluded 189 Thermocouple a Time 1 2 3 4 5 6 7 4:40 1002°C 980°C 840°C 467°C 220°C 261°C 369°C 4:45 (4:46)1002 980 845 485 222 (4:46) 266 (4:46)374 4:50 1000 979 848 503 225 268 376 4:55 1000 980 852 522 228 273 376 5:00 1005 983 855 540 234 276 376 5:05 1007 990 860 554 240 280 379 5:10 1020 993 865 567 245 286 379 5:15 1023 1003 868 580 250 288 381 5:20 1034 1007 875 595 254 292 384 5:25 1034 1016 882 609 258 295 386 5:30 1044 1020 890 624 263 300 388 5:35 1046 1023 895 637 267 305 391 5:40 1046 1025 898 650 273 310 391 5:45 1044 1024 903 663 278 312 393 5:50 1041 1022 908 675 285 317 396 5:55 1017 910 686 291 322 6:00 1041 1016 913 696 297 324 400 6:05 1041 1020 915 708 303 331 403 6:10 1038 1022 917 719 310 334 405 6:15 1052 1027 920 728 317 338 405 6:20 1062 1033 924 736 324 343 407 6:25 1065 1041 928 745 330 348 410 6:30 1075 1046 935 753 336 353 410 6:35 1078 1052 942 762 343 358 412 6:40 (6:42)1083 1060 948 770 351 (6:42)364 (6:42)417 6:45 1086 1065 955 778 359 369 419 6:50 (6:51)1094 1070 962 787 367 374 419 6:55 (6:56)1102 1076 968 796 375 379 419 7:00 1099 1082 974 805 385 381 419 7:05 1083 979 813 393 7:10 1096 1082 984 820 400 391 422 7:15 1094 1083 986 827 410 396 424 7:20 (7:21)1099 1085 988 835 422 405 426 7:25 1099 1087 991 842 436 410 426 7:30 1107 1090 995 849 450 417 428 7:35 1110 1092 999 855 462 426 431 7:40 1115 1098 1004 860 475 431 433 7:45 1115 1102 1009 865 486 438 436 7:50 1115 1105 1013 872 497 448 438 7:55 (7:56)1112 1100 1012 875 512 455 440 8:00 1107 1095 1012 878 523 462 443 8:05 1110 1093 1012 882 535 471 445 8:10 1107 1092 1014 886 553 466 450 8:15 1107 1091 1014 892 565 492 452 8:20 1110 1091 1015 896 580 504 455 8:25 1110 1093 1018 902 595 516 457 8:30 1112 1096 1021 907 608 527 462 8:35 1115 1100 1026 910 624 539 466 8:40 1112 1100 1029 914 637 555 471 8:45 1100 1032 917 650 Position of thermocouples a No. 1 — On oven floor. No. 2 — 1 inch from oven floor. No. 3 — 3 inches from oven floor. No. 4 — 5 inches from oven floor. No. 5 — 7 inches from oven floor. No. 6 — 9 inches from oven floor. No. 7—10V 2 inches from oven floor, on top of charge. 190 EXPERIMENTAL INVESTIGATIONS 10.5 13 5 7 9 10.5 DISTANCE IN CHARGE - I NCH ES FROM FLOOR 'itf. 67 (Above). — Isothermal Relations, and Fig. 68 (Below). — Isochronal Curves of Temperature Distribution Through Sole-Flue Oven Charge. (Both from Jnd. and Eng. Chem. vol. 29, May, 1937.) AN INDUSTRIAL COKING PLANT 191 Relation of time and temperature. — The isochronal chart (fig. 68) shows graphically the temperature in any part of the charge, at the position of the thermocouple column, at any time during the coking period. It demon- strates the relative effect on the charge of the heat received from the floor and that received from the roof and gases above the charge. The similarity of the curves on the left hand side of the diagram up to the line of 7 inches indicates that the temperature conditions in this part of the charge are similar and that the transfer of heat is from the floor. The upper 3^ inches of the charge received heat mainly from above during the early part of the coking period, then gradually more from below until after eight hours the rise in temperature of the top of the charge due to heat from below completely overshadowed the effect due to heat from above, and it is evident that the charge lost heat to the space above it in the last stages of the coking cycle. Fig. 69. — Coke Taken from End of the Thermocouple Assembly (3-inch grid). (From "Coke from Illinois Coal," by G. Thiessen. Ind. and Eng. Chem. vol. 29, May, 1937.) The heat absorbed from above caused no apparent temperature effects below ?>]/z inches from the upper thermocouple. The charge may, therefore, conveniently be regarded as consisting of two parts on the basis of its thermal behavior, a lower part coked at relatively high temperature by heat transfer from below, and an upper part coked at relatively low temperature mainly by down- ward heating. In the lower zone the plastic zone travels regularly upward and the charge is heated relatively rapidly ; the upper part of the charge is heated gradually and more or less uniformly but to a lower temperature, first 192 EXPERIMENTAL INVESTIGATIONS by gases in the oven chamber and eventually by heat transmitted through the lower part of the charge. In the upper zone the heat gradients are lower, the final temperature is lower, and the mass cokes throughout within a relatively short time interval after the plastic temperature of the coal is attained. The gases are probably released throughout the mass rather than at a plastic zone of limited extent as in the lower part of the charge. The difference in the progress of heat- ing in the upper and lower portions of the charge is equally well shown by plotting the determinations on a time-distance (isothermal) chart (fig. 67). This shows clearly the relatively higher temperature of the upper 3 to 4 inches of the charge as compared with the middle part, at least during the first 6 to 7 hours. It also shows that this part of the charge never attains the temperature of the lower part. The effect of this manner of heating the charge upon the appearance of the coke is demonstrated by the accompanying photographs (fig. 69) of pieces of coke taken from the oven from near the end of the thermocouples after the observations were made. Travel of the plastic zone. — A charge of coal being coked in a high temperature coke oven consists of ( 1 ) a zone of coke extending from the hot wall or floor to (2) a plastic zone which is quite narrow and which separates the coke from (3) the unchanged part of the coal charge. It is common knowledge that there is a large temperature gradient across the plastic zone, on one side of which there is hot coke and on the other side coal at a temperature a little above that of boiling water. The rate of coking is determined by the rate of travel of this plastic zone. The movement of the plastic zone can be estimated ( 1 ) from measure- ments of gas pressure at various points along the line of travel of the plastic zone, since a high gas pressure occurs at the plastic zone due to its impermeability, (2) from the movement of the point of maximum temperature gradient in the charge, since there is a high heat gradient across the plastic zone due to its low heat conductivity, or (3) from the movement of the temperature zone corresponding to the plastic range of the coal. 8 Our attempt to follow the movement of the plastic zone through pressure measurements was unsuccessful due to plugging by tarry material of the tubes placed in the oven for pressure measurements. In order to determine the movement of the plastic zone by following the point of maximum heat gradient in the charge, this point was determined graphically using values read from a large scale plot of the isochronal lines (fig. 68). Temperature differences over one-half inch intervals were plotted against the position of the mid-point of the interval to obtain the approximate position of the point of maximum gradient. Temperature differences across smaller intervals in this region were then taken to determine the position of the point more s Ryan, W. P., Rate of travel of fusion zone in coke oven: Proc. A.G.A. vol. 7, pp. 801-878, 1925. AN INDUSTRIAL COKING PLANT 193 900 700 600 500 ISOC HRONAL CURVE 1 DIFFERENTIA ISOCHRONAL L OF CURVE 1 \ N K -« DISTANCE IN CHARGE - INCHES Fig. 70. — Method of Determining Position and Magnitude of Maximum Heat Gradient from an Isochronal Curve. accurately. Figure 70 shows the curve obtained for the one-hour point, plotted together with the isochronal curve for one hour to illustrate how the values were obtained. The values obtained for the position of the point of maximum gradient, the temperature gradient, and the temperature at that point are represented in Table 29 ; the movement of the point and the temperature at the point are shown graphically in figure 71, the broken line indicating the movement of the point through the charge. The relatively large heat gradient compared to the temperature and the relatively constant temperature at the point of maximum gradient should be observed. Table 29. — Location of Points of Maximum Heat Gradient at Hourly Intervals after Charging, and Heat Gradient over Distance One-fourth Inch Each Side of that Point Time, hours after charging Location of point of maximum gradient inches from floor a Heat gradient degrees C. per half inch at point of maximum gradient Temperature at point of maximum gradient 1 2 3 4 5 6 7 8 2.125 2.875 3.688 4.375 5.250 6.125 6.375 6.500 290 210 170 150 140 140 150 140 480 475 470 490 470 450 545 630 to nearest eighth inch 194 EXPERIMENTAL INVESTIGATIONS Determination of the position of the plastic zone by determining the position of the temperature interval during which the coal was plastic was accomplished by first determining the plastic temperature range of the coal by means of the Agde-Damm apparatus. 9 It was found that the coal started to soften at 350°C. and that the decomposing plastic mass set to coke at 450°C. 3 4 DISTANCE, INCHES Fig. 71.— Travel of Plastic Zone, 350° C. to 450° C, and Position of Point at Maximum Gradient Temperature. (Modified after "Coke from Illinois Coal," by G. Thiessen. Ind. and Eng. Chem. vol. 29, May, 1937.) The position of these temperature (350-450°C.) limits have been included in figure 71. The close parallelism of the line of maximum heat gradient and the upper limit of the plastic temperature range (450°C.) leaves no doubt that the maximum heat gradient occurs at the plastic zone. Due to the influence of heat absorbed from above, the conditions in the upper 3 or 4 inches of the charge are not the same as those in the lower part. The heat gradients are greatly different in these two parts. o Fieldner, A. C. and others, Methods and apparatus used in determining the gas, coke and by -product making properties of American coals: U. S. Bureau of Mines Bull. d44, )>. 15, 1931. AN INDUSTRIAL COKING PLANT 195 The rate of travel of the position of maximum gradient, which may also be considered the rate of travel of the plastic zone, was found to be 0.78 inch per hour over the lower six inches of the charge. This is somewhat greater than the rates reported as occurring in standard slot-type coke ovens. It must not be overlooked that the coking time in inches per hour as frequently given for slot-type ovens is found by dividing the oven width by the coking time. The rate so obtained is twice the rate of coking from each wall to the center which would correspond to the rate considered here. The combustion flue temperatures were 1360°C. (2480°F. ) during the test. 2 4 6 8 DISTANCE, INCHES FROM FLOOR 10 Fig. 72. — Relation of Heat Gradient and Volatile Matter Gradient at Completion of Coking. (After "Coke from Illinois Coal," by G. Thiesscn. Ind. and Eng. Chcm. vol. 29, May, 1397.) Temperature in flues, regenerators, and stack. — Temperatures of the side walls of the sole flues in various ovens in the battery as measured with the optical pyrometer ranged from 1300°C. (2372°F.) to 1420°C. (2588°F.), with most of the flues at around 1350°C. (2462°F.). The side walls of the flues in oven No. 9 at the time of the test were at temperatures averaging 1360°C. (2480°F.). 196 EXPERIMENTAL INVESTIGATIONS The temperature of the flue gases leaving the last regenerator pass and entering the stack flue ranged between 230°C. (446°F.) and 590°C. (1094°F.) for ovens in normal operation. The temperature regularly increased during a regeneration cycle. Stack temperature ranged between 300° C. (572°F.) and 340 °C. (644°F.) also reflecting the regeneration cycle but to a lesser extent. ANALYSES OF COAL COKED AND OF COKE PRODUCED The proximate analysis of the minus 5/16-inch screenings which were being coked in the ovens at the time the temperature measurements were being made and of the coke produced are given in Table 30. The coke analyzed is material remaining on a one-half inch screen after the original sample had been subjected to four drops in the standard shatter test. The minus one-half inch material removed is composed largely of uncoked material and of mineral matter which came away when the coke broke. The plus one-half inch coke corresponds roughly to the coke as sold after being screened to remove the fine materials. Table 30. ■Analyses of Coal Coked and of Coke Produced During Test (Dry Basis) Sample of Sample No. Date sampled Coal C-1545 10/10/35 Coke+M" C-1563 10/11/35 Coke-i^" C-1556 10/11/35 Ash, per cent 12.1 34.0 53.9 1.43 0.01 0.96 0.46 12,650 16.1 3.0 80.9 1.36 12,100 93.6 20.4 Volatile matter, per cent Fixed carbon, per cent Total sulfur, per cent 6.6 73.0 1.44 Sulfate sulfur, per cent Pyritic sulfur, per cent Organic sulfur, per cent Heating value, B.t.u./lb Proportion of total sample, per cent. . . 11,540 6.4 Table 31. — Analyses of Coals Coked and of Coke Produced at Various Times in the Sole-Flue Ovens (Dry Basis) Sample of Sample No. Date sampled Coke C-811 3/10/34 Coal C-1439 7/29/35 Coke C-1453 7/29/35 Coke C-1454 7/29/35 Ash, per cent 17.9 5.3 76.8 1.39 11,615 12.0 33.8 54.2 1.73 0.04 1.20 0.49 12,680 17.6 5.1 77.3 1.44 11,870 2063 1128 15.5 Volatile matter, per cent 4.2 Fixed carbon, per cent 80.3 Total sulfur, per cent 1.21 Sulfate sulfur, per cent Pyritic sulfur, per cent Organic sulfur, per cent Heating value, B.t.u./lb Ash softening temperature °F Ash softening temperature °C 12,180 2087 1142 AN INDUSTRIAL COKING PLANT 197 This procedure was adopted since in the sampling of this particular sample special care was taken to keep the sample in as large pieces as possible, resulting in the inclusion of the unconsolidated top layer. The analysis of this minus one-half inch material is included in Table 30. In Table 31 are presented the analyses of samples of coke taken at other times from the main coke conveyor. INFLUENCE OF OVEN CONDITIONS ON COKE CHARACTER At the completion of coking about two-thirds of the charge had the nature of high temperature coke while the top one-third being coked under conditions of low temperature gradient and being finished at a low final temperature was more like low temperature coke. The 8 :30 line on figure 68 shows the distribu- tion of temperature in the charge at the time the charge was removed from the oven. To investigate the influence of these conditions pieces of coke represent- ing the entire thickness of the charge were taken from a region close to the ends of the thermocouples (fig. 69). One of these was sectioned horizontally at approximately 1.75 inch intervals and the individual pieces were analyzed (Table 32) ; and cross sections at approximately 1.25 inch intervals and a longitudinal section through an entire thickness of other pieces also taken from near the thermocouples were made for a study of cell structure, according to the method of H. J. Rose. 10 The analyses of the sections are presented in Table 32. The volatile matter content of the sections has been plotted in figure 72 on which also appears the final temperature distribution in the oven. Table 32. — Analyses of Cross Sections of Coke Specimens Made During Test Section No. 1 2 3 C-1564 C-1565 C-1566 10.5-9.5 9.5-8.0 8.0-6.75 1.0 1.5 1.25 12.9 15.4 15.7 20.5 9.7 4.7 66.6 74.9 79.6 1.23 1.18 1.31 23.5 11.5 5.6 76.5 88.5 94.4 490 530 590 Lab. Sample Number. Location, inches from floor Thickness of section, inches Ash,a (%) Volatile matter, a (%) . Fixed carbon, a (%) . . . Sulfur,* (%) Volatile matter, b (%) . Fixed carbon, b (%) . . . Final temp, at mid- point of section °C. 6.75-3.75 3.0 15.5 2.4 82.1 1.34 2.9 97.1 890 C-1568 3.75-1.75 2.0 17.0 2.9 80.1 1.41 3.4 96.6 1040 C-1569 1.75-0 1 15 2 82 1 2.8 97.2 1100 ! c-1563 0-10.5 10.5 15.5 5.5 79.0 1.31 6.5 93.5 10.5 16.1 3.0 80.9 1.36 3.5 96.5 a dry basis b moisture- and ash -free io Rose, H. J., The study of coke macrostrueture: Ind. Eng. Chem. vol. 17, No. 9, pp. 895 ff., 1925. 198 EXPERIMENTAL INVESTIGATIONS Fig. 73. — Rose Longitudinal Section, Natural Size, of Specimen of Sole-flue Oven Coke (No. 72, figs. 55c and 69). Bottom of photograph illustrates structure near the end of the specimen resting on the oven floor. AN INDUSTRIAL COKING PLANT 199 Fig. 74. — Rose Cross-sections, Natural Size, of Sole-flue Oven Coke (No. 72) Repre- senting Horizontal Cuts in the Same Specimen (Fig. 73). a. Section through top of specimen. h through (j. Sections at successive 1% inch levels from top toward bottom of specimen //. Section through bottom of specimen which rested on the floor of the oven. 200 EXPERIMENTAL INVESTIGATIONS • -• , •• mm'*'*, 'w,m ft; . S Fig. 75. — Stalagmite Carbon on Surface of Coke Made in Sole-flue Oven. AN INDUSTRIAL COKING PLANT 201 Longitudinal and cross sections of one of the blocks shown in figure 69, made according to the method of H. J. Rose, are shown in figures 73 and 74, respectively. The regular gradation of coke structure from a small celled, much shrunken, highly devolatilized, high temperature coke on the bottom (fig. 74/z) to a large celled, very thin walled coke on the top (fig. 74a) is plainly visible. : - ■ , .-A- •••••-■ «t: ! --'- >•;*-•.%■ ' ■■„■■* II, : :<^r ; s^«-;: : r^^^;f:p'- Fig. 76. — Hair Carbon on Surface of Coke Made in Sole-flue Oven. Carbon deposited in the form of stalagmites (fig. 75) and hair carbon (fig. 76) may be seen by close inspection of the surfaces formed during coking in the lower part of the charge. Such deposits are characteristic of the upper portion of the coke formed in beehive ovens and have been discussed at length by Roberts and Jenkner, 11 who believe that they are formed by the cracking of volatile matter as it passes from the downward travelling plastic zone up through the hot coke. The presence of such deposits in the lower part of the coke made in sole-flue ovens indicates that shrinkage cracks had formed in the coke in the lower part of the charge while the upper parts of the charge were still evolving volatile matter and may also indicate a too rapid coking rate. ii Roberts, J., and Jenkner, A., International coal carbonization: 1934. 453 pp. See Chapter V, especially pp. 103-108. Pitman, London 202 EXPERIMENTAL INVESTIGATIONS Volatile matter gradient in coke made during 10 l / 2 hours coking- time. — Since the time the temperature measurements were made two new batteries of eight ovens each have been put into operation. All ovens are now operating on a ten and one-half hour coking cycle. Believing that it may be of interest to show the influence of a longer coking time upon the analysis of the coke, pieces of coke from a charge coked in one of the new ovens were sctioned horizontally at one-inch intervals and analyzed for moisture, ash, and volatile matter. (Table 33). The volatile matter content of the upper sections of this coke is much lower than those of corresponding sections from the eight-hour coke (Tables 30 and 31). Table 33. — Analyses of Cross Sections of Coke Specimens Taken from New Ovens 10.5 Hours Coking Time, Sampled July 22, 1936 Section No. 1 (top) 2 3 4 5 6 7 CI 865 CI 866 C1867 CI 868 CI 869 C1870 C1871 7-8 6-7 5-6 4-5 3-4 2-3 1-2 14.2 15.3 14.4 15.5 14.9 15.4 16.3 5.0 4.4 3.4 2.8 2.4 1.6 2.3 80.8 80.3 82.2 81.7 82.7 83.0 81.4 5.8 5.2 3.9 3.4 2.8 2.0 2.7 (bottom) Sample No Location, inches from floor. Ash, a per cent Volatile matter, a per cent. Fixed carbon/ 1 per cent. . . Volatile matter, b per cent. C1872 0-1 16.2 2.5 81.3 2.9 Dry Basis. Moisture -and -ash -free basis. GAS ANALYSES Five-gallon gas samples were taken at three different times and analyzed. The samples consisted of: (1) a pair of samples taken July 29, 1935 from the main gas line, one just before the gas entered, the other just after the gas left the sulfur removal equipment; (2) a sample taken Oct. 11, 1935 just before the gas entered the sulfur removal equipment; and (3) two samples taken July 22, 1936 from the suction mains coming from the two new batteries and from the old battery, both just before the gas entered the suction pump. The samples taken at all times were analyzed in a modified Shepherd-Orsat apparatus 12 which had been equipped for the determination of hydrogen by combustion over copper oxide, 13 and for the determination of illuminants by absorption in fuming sulfuric acid. The sample taken at the second time was also analyzed by fractional distillation in a Podbielniak. Model A precision fractioning unit. 14 12 Shepherd, M., An improved apparatus and method for the analysis of gas mixtures by combustion and absorption: Bureau of Standards Jour, of Research vol. 6, No. 1, pp. 121-167, Jan. 1931. Research Paper No. 266. 13 Burrell, G. L., Seibert, F. M., rev. by Jones, G .W., Sampling- and examination of mines gases and natural gas: U. S. Bureau of Mines Bull. 197, pp. 46-S, 1926. 14 Podbielniak, W. J., Apparatus and methods for precise fractional-distillation analy- sis: Ind. Eng. Chem., Anal. Edition, vol. 3, pp. 177-188, April If), 1931. AN INDUSTRIAL COKING PLANT 203 The results of these analyses together with published analyses of coke oven gas 15 - 16 are presented in Table 34. The heats of combustion are calculated from the analyses of the gas as taken from the sample bottles in the laboratory. At the time the samples were taken in the plant an automatic recording calorimeter indicated a heat of combustion for the gas of 480 B.t.u. per cubic foot. The lower temperatures in the oven are reflected in a lower content of methane and probably a lower hydrogen content, greatly increased by cracking. Table 34. — Analyses of Coke Oven Gas Produced in Sole-Flue Ovens Orsat Analysis Description Sole-flue oven gas Standard coke oven gas a a a b b Sample No. G-lll G-112 G-113 G-128 G-129 ( 14 ) ( la) Date of sampling 7/29/35 7/29/35 10/11/35 7/22/36 7/22/36 Before After Before Main Main Sampling position H 2 S H 2 S H 2 S from new from old removal removal removal batteries battery C0 2 , per cent. . . . 3.7 4.4 3.4 5.4 4.1 2.2 2.6 1.4 2 , per cent 1.3 1.2 0.7 0.6 1.3 0.8 0.6 0.5 Illuminants, per cent 2.6 3.3 2.2 2.8 1.9 4.0 5.2 2.9 CO, per cent 14.2 14.9 15.7 12.2 14.4 6.3 6.1 5.1 H 2 , per cent 44.0 40.0 46.5 45.7 44.3 46.5 47.9 57.4 N 2 , per cent 17.0 17.3 15.2 11.7 17.7 8.1 3.7 4.2 CH 4 , per cent. . . . 15.1 17.6 16.1 20.0 15.1 32.1 33.9 28.5 C 2 H H , per cent. . . 1.7 1.3 0.2 1.6 1.2 H 2 S, per cent. . . . 0.4 negative to lead acetate paper not deter- mined not deter- mined not deter- mined B.t.u./cu. ft. gross, calc 423 441 407 471 400 574 600 536 a Eight and one -half hours coking time. tJ Ten and one-half hours coking time. Fractionation Analysis G-113 Methane and lighter H 2 , CH 4 , CO, 2 , N 2 , per cent 96.3 c Ethane and ethylene, per cent 1.8 Pentanes and higher, per cent 1.9 e Too much methane and lighter constituents for satisfactory fractionation of con- stituents higher than methane fraction. 15 Combustion, a reference book on theory and practice: American Gas Association, Ed. 3, 1932, p. 95. ifi Haslam, R. T., and Russell, R. P., Fuels and their combustion: New York, McGraw- Hill, p. 282, 1926. Part V — Conclusions and Recommendations CHAPTER XVII The present report must be considered largely as a progress report ; it by no means provides all of the information necessary or desirable for the com- mercial carbonization of Illinois and similar coals. Certain conclusions may be drawn from the experimental work reported here and from the review of the previous available information and some recommendations for additional desirable work may be made. SCOPE OF THE REPORT This report includes a study of the economic factors involved in the production of coke from Illinois coal, a consideration of the possible market for such a coke, a review of the fuel requirements for the Illinois coal market area and a review of the production and sale of coke in that area. The history of the production of coke from Illinois coals is briefly sketched and the difficulties which were encountered are reviewed. Considering the question as to whether Illinois coals are coking coals, various criteria which have been proposed for the identification of coking coals have been reviewed and applied to the coals of Illinois. These criteria are based largely upon the behavior of coals from which satisfactory metallurgical cokes have been made and are probably too stringent in identifying coals from which satisfactory domestic coke can be made. The section of the report covering the experimental work carried on in an investigation of the possibilities of making coke, particularly domestic coke, from Illinois coals includes the determinations of the coke, gas, and by-product yields of various Illinois coals, the temperature range during which they are plastic, their agglutinating or caking values, the influence of geographical location, size, admixtures, preparation, and the temperature of completion of coking on coke quality. The influence of impurities such as moisture, ash and sulfur on coke quality are discussed and an experimental study of the influence of the organic and pyritic forms of sulfur in the coal on the sulfur content of the coke is reported. The petrographic composition of Illinois coals is briefly described and the minor significance of petrographic composition on coke quality in the case of the majority of Illinois coals is discussed. [205] 206 CONCLUSIONS AND RECOMMENDATIONS An installation of Knowles sole-flue coke ovens at West Frankfort, the only plant at which coke is at present being made from Illinois coal, is described and some details of its operation, especially of the temperature conditions in the oven, are reported. CONCLUSIONS As a result of the experimental work and review of previous information, presented in this report, the following general conclusions may be made. ( 1 ) Under favorable conditions coke suitable for domestic and some industrial fuel uses can be made from most of the coals occurring in Illinois. (2) The domestic fuel market appears to be the best and to be a growing outlet for coke made from coals from the Eastern Interior coal basin, of which most come from Illinois. (3) Previous attempts to produce coke commercially from Illinois coals have failed, probably because its sale was attempted in the metallurgical coke market, the market for domestic coke having not yet been foreseen or developed. (4) Illinois coals should be coked with as little delay as possible after having been removed from the mine face, say within two weeks at the most. (5) Mineral impurities in the coal should be kept at a minimum, not only to obtain a coke with a low content of impurities but in order to obtain a stronger and better structured coke. (6) The organic sulfur content of the coal sets the lower limit to the sulfur content which can be attained in the coke by processes which treat only the coal. (7) Because of the high volatile matter content of Illinois coals, cokes made from them shrink excessively when completely devolatilized and develop a highly fractured, fingery structure. The addition of non-coking, non-shrinking carbonaceous material to the coal before it is coked as a means of minimizing this shrinkage is not practical in the case of Illinois coals because of their small caking power or ability to cement non-coking material into a homogeneous mass. (8) Compared with the range of variation in carbonization yields and behavior under the influence of heat possible among coking coals, Illinois coals are relatively uniform, with any variations which do occur taking place gradually as the coal fields are traversed and may in general be correlated with changes in rank as indicated by the moist mineral-matter-free calorific value. (9) Coke is being made commercially from Illinois coal screenings in ovens of a type which have not previously been used commercially. In these ovens the coal is heated in a wide flat layer, mainly from below. No conclusions are to be made concerning the commercial success of this plant from the data presented. (10) Before commercial production of coke from a given coal by a given process is attempted, tests should be made to determine the effects of the various factors which may be encountered, such as the time elapsed after the coal was RECOMMENDATIONS 207 removed from the face, the influence of the coal preparation, a cleaning process to which the coal might be subjected, and the effects of segregation of components of the coal if screenings are to be used. RECOMMENDATIONS FOR ADDITIONAL WORK ( 1 ) The addition of minor quantities of strongly caking low volatile coals to Illinois coals in order to improve the coke made from them has not been particularly successful. Some of the results presented in this report indicate that this may be due to the differences in the temperature range over which the two coals become plastic. Additional study of this point seems to be desirable, both to either establish or disprove this theory, and to find, if possible, a suitable coal for blending which will materially improve the quality of the coke without prohibitively increasing the cost. (2) Additional coals, particularly from Northern and Central Illinois, should be investigated as to the quality of coke which can be made from them under the most favorable conditions. (3) The influence of coal cleaning as now carried out in the various types of coal cleaning plants on the coke-making properties of the coal, especially any tendency towards increasing the rate at which the coking property is lost during the period between cleaning and coking, should be studied. Appendices By Gilbert Thiessen and Paul E. Grotts APPENDIX A DESCRIPTION OF THE BY-PRODUCT ANALYSIS METHOD AS APPLIED TO ILLINOIS COALS The test adopted for the by-product analysis method was the dry distillation test described by the chemists of the United States Steel Corporation. 1 A fused quartz tube was used in these experiments in place of the Jena glass distillation tube specified by the Steel Corporation method. To distinguish between the two methods, which are identical in other respects, the test employed by this laboratory has been designated as the coal carbonization assay in the following discussion. Method of the Coal Carbonization Assay A 20-gram sample of coal is progressively carbonized at 900° C. and the volatile products are passed through crushed silica brick, maintained at 720° C, to absorption bulbs, a freezing tube, and a gasometer. The apparatus and procedure provide for the separation and subsequent estimation of the gross fractions of the distillation products which are coke, tar, ammonia, liquor (water), hydrogen sulfide, carbon dioxide, light oils, and fixed gas. Character of the Assay The test is carried out in laboratory scale with the 20-gram portion of coal being subjected to controlled heat effects and cracking surfaces which approximate the cor- responding effects existing under slot-type coking plant operation. The results obtained by the analysis are empirical and the yields do not conform absolutely to yields from any one plant. However, an estimate is obtained, by means of the test, of the practical coke and by-product yields from a coal. The non-conformity is due to the impossibility of accurately reproducing the action of a coking plant in small laboratory scale. To arrive at a quantitative prediction of the behavior of a coal in commercial carbonization relationships between the assay and the plant yields must be evaluated. A correlation may be established which may be used in interpreting the assay analysis. Inasmuch as the yields indicated by the assay are relative it is not of immediate importance to establish the correlation in a study of the coking possibilities in a series of coals. Thus a variation in the amounts of tar obtained in two specific coals would be approxi- mately the same in both the laboratory test and in plant yield. In the assay the coal charge is heated progressively. Heat is first applied to the discharge end of the test charge corresponding to coal next to the oven wall in a freshly i Methods of the Chemists of the United States Steel Corporation for the Sampling- and Analysis of Coal, Coke and By-Products, 3rd Ed. pp. 130-143, 1929. [ 209 ] 210 APPENDIX A ISOCHRONAL CURVES OF TEMPERATURE AT THREE POINTS IOOO °C '4'40 4:27 4117 4.07 355 3!42 332 322 3M0 3!00 2:50 2J35 » i i ! ! GAS EVOLUTION I N K i < >\ ( N^ v *»*» ^ =a— A Fig. I. — Travel of Heat and Evolution of Fixed Gas in a Carbonization Assay Test. (Carbonization No. 44.) BY-PRODUCTS ANALYSIS METHOD 211 charged coke oven. The evolved gases pass outward through a section of coarse silica particles which is held constantly at a temperature of 720° C. This provision is made to expose the coal gases to the same kind of cracking surface found in the coke oven and to such an extent that equivalent cracking takes place. The first heating of the test charge occurs at the end corresponding to the oven wall face and when that portion has coked, an additional section of uncoked coal is heated. Continuing in this way the whole test charge is carbonized. Although the manipulation of the burners heating the charge proceeds in a step-wise fashion the temperatures in the coal charge grade continuously, but not linearly, from the hot end to the cold. At the end of the test all of the coal charge has been carbonized and the coke has reached 900° C. Throughout the heating, however, the silica portion of the tube packing has been held at 720° C. Table I — Carbonization Assay Test No. 44 Sample: R-24 (Franklin Co. No. 6 Coal) (Plastic Range: 360-408°C.) Thermocouple Time I 11" from open end of charge II 3" from open end of charge III Open end of charge 2:35 35°C 50 50 100 200 630 875 35°C 100 160 500 750 870 875 35°C 2:50 550 3:00 650 3:20 750 3:42 775 4:17 875 Final 4:40 880 The temperature experience of the coal and coke in the laboratory assay may be graphically represented. Data from one of the tests (Table I) has been used to fix the temperature curves in figure I which represents temperatures along the coal charge a* Fig. II. — Carbonization Assay Equipment. 212 APPENDIX A specific times. A separate measurement by the use of the Agde-Damm plastic range test has shown the plastic range of this sample of coal to have been from 360° to 408°C The curves show that the face end of the test charge had softened and solidified within a few minutes after heating was started. However, this position in the charge did not reach a temperature of 900° C. until late in the coking period. This was due to the cooling effect of the evolved gases which had to pass through and over the carbon- ized coal. The cooling effects of the volatile products of the coal when passed over the portions of the charge already carbonized would be due in part to the absorption of heat in the cracking of the gaseous materials. The impossibility of raising these parts of the charge to 900° early in the test was found to be typical of the test. Fig. III. — Specimens of Silica, Coal, and Coke. The plastic range traveled the length of the charge and then disappeared before the burners had been lighted under the last J4 °f the charge. These effects are at variance with the conditions found in a coke oven. First the coking period is much shorter, but this is dependent on the* second effect: that the uncoked coal is heated by radiation and convection arising both outside and in the tube. In the coke oven, heat must invariably pass through the fused plastic zone before the uncoked coal may be raised to its plastic temperature. The plastic zone is a most efficient heat insulator. Thus the coking period in the coke oven depends on the travel of the plastic zone through the charge, which in turn depends on the heat applied and the nature of the coal. The situation is different in the assay since the plastic zone does not exist in the same form as in the coke oven. The manner of applying the heat to the test charge precludes the existence of the thin layer of plastic coal with a high temperature gradient through it. Instead heat reaches the uncoked coal from almost all directions. Preparation of the Sample Samples of coal were prepared by the standard method of the American Society for Testing Materials for the preparation of 60-mesh samples of coal for analysis except that the samples were ground to pass a 40-mesh U. S. Standard Series wire mesh sieve (35-mesh Tyler Standard series). The 40-mesh coal is dried at 105 °C. for 1 hour and cooled in a desiccator over calcium chloride. BY-PRODUCTS ANALYSIS METHOD 213 Apparatus and Procedure The assembled test equipment consists of the carbonization tube and a gas-fired combustion type of furnace exactly as described in "The Method of the Chemists of the U. S. Steel Corporation for the Sampling and Analysis of Coal, Coke and By-Products," the tar piece, the tar filter enclosed by a steam jacket, a Geissler bulb to contain sulfuric acid and which is immersed in an ice-water bath, a drying U-tube, a second Geissler bulb to contain potassium hydroxide solution and which is also immersed in an ice-water bath, another drying tube, the freezing tube and finally the gasometer. Fig. IV. — Distillation Tube, Tar Piece, and Tar Filter for Carbonization Assay. Equipment: (1) Gas-fired combustion furnace (2) Steam jacket (3) Battery jars (two) for ice-water baths (4) Wide mouth quart size Dewar flask (5) Gasometer (see figure II) (6) Fused silica distillation tubes (three) (7) Fused silica tar pieces (three) (8) Pyrex tar filters (three) (9) Geissler bulbs (four) (10) Glass stoppered U-tubes (four) (11) Freezing tube (two) (see figure V) specifications figure IV (12) Auxiliary equipment included the automatic four-point temperature recorder, and sampling and analytical equipment. The furnace. — The special gas-fired combustion furnace used in the test is equipped with 25 individually controlled Bunsen type burners. The furnace supplied by several scientific supply houses does not meet the specifications as set forth in the description referred to above. The method described as used in the laboratories of the U. S. Bureau of Mines in the Bureau of Mines, American Gas Association cooperative studies of the coal, coke, and by-product making properties of Illinois coals also does not meet these specifications, the furnace used being one supplied by a number of apparatus supply houses with a smaller number of Meeker type burners. The method specifies a type of flame and rate of heating which is obtainable only when artificial gas having a rapid rate of combustion and a short hot flame is used. The laboratories of the Illinois State Geological Survey were supplied with such a gas when this equipment was first put into use. Shortly thereafter the gas supply was changed at various times to natural gas 214 APPENDIX A or to a mixture of natural and manufactured or reformed natural gases. Even with burners of different design it was impossible to obtain the required type of flame and rate of heating when gas alone was supplied to the burners. This problem was solved by supplying a mixture of gas and air to the gas inlet of the burners. Both gas and air lines were equipped with flow meters and the gas line with a manometer preceding the flow meter in order that suitable flames could easily be reproduced. One specified condition of the test is that the portion of the distillation tube containing the crushed silica is to be maintained at 720° C. To facilitate the control of that part of the tube the first four gas burners were replaced by a small electric tube furnace fitted with a thermocouple connected to the recorder. Three other thermocouples were installed to measure the temperatures of the tube at 0, 3, and 11 inches from the "face", or open end of the leveled coal charge. Temperatures were recorded at four points in the tube throughout the test. The distillation tube and its preparation. — The tube is fused silica, 51 cm. in length and 18.5 mm. in outer diameter. In the test it contains a 20-gram coal sample, asbestos packing, and sized silica particles. The cleaned, dried distillation tube is weighed and a weighed sample of coal is placed in it. Previously ignited asbestos, rolled into a plug 13 mm. long is inserted to a position 292 mm. from the closed end of the tube. Another 13 mm. plug of asbestos is pushed into the tube to a position 6 mm. from the first plug. Crushed silica brick sized to pass the 10-mesh and be retained by the 20-mesh screen is poured into the tube until the column of silica particles is 76 mm. in length, held in place by another asbestos plug 13 mm. in length which completes the packing of the tube. The tube and its contents in place ready for a test is shown in figure IV. After weighing the packed tube and leveling the coal charge by rolling, the tube is ready for the test. At the end of the test it is reweighed and the loss of volatile matter is determined. The yield of coke is determined as the difference between the weighed sample and the amount of the volatile matter driven off. The tar piece. — A fused silica sleeve, 128 mm. in length and with an outside diameter of 14 mm., accurately weighed, is inserted in the end of the packed and weighed distilla- tion tube to receive the tar condensate that forms in the cooler end of the distillation tube. The amount of this tar is combined with the amount of tar found in the succeeding tar filter. In order that the high temperature tar condense in the tar piece it is necessary that it fit snugly in the distillation tube. Its length is such that one end touches the asbestos plug (fig. IV) and the other protrudes from the distillation tube. The tar filter. — Also shown in figure IV is the Pyrex glass tar filter. It steps down from a tube of 1 inch inner diameter to the *4 mcn outlet tube, and is packed loosely with dried absorbent cotton. The function of the tar filter is to remove suspended tar from the evolved gases. It is weighed when packed and is connected to the distillation tube by means of a gas-tight cork (faces of cork, 1]4 inches and 1 inch, length 1 inch). From its weight at the end of the test the amount of tar condensed at 100° C. is computed. During the test the tar filter is enclosed by a steam jacket consisting of a cylindrical copper can, figure II, fitted with a steam inlet and thermometer. The outlet end of the tar filter which is connected to the first Geissler bulb is kept hot by extending a small cylindrical sleeve from the steam jacket. This provision is necessary to prevent the collection of crude naphthalene in the small tubes. BY-PRODUCTS ANALYSIS METHOD 215 The absorption train. — The first Geissler bulb which contains 8 cc. of normal sulfuric acid and one drop of methyl orange indicator solution added as an indicator is connected to the tar filter. This bulb collects water, naphthalene, and free ammonia. Connected to the outlet of the bulb is a glass stoppered U-tube containing calcium chloride previously saturated with C0 2 to absorb any water which the gas may pick up from the acid in the bulb. The two are weighed together. Fig. V. — Freezing Tube for the Collection of Light Oil. The Inlet is to the Left. The second Geissler bulb contains 13 cc. of 1:1 potassium hydroxide solution and absorbs hydrogen sulfide and carbon dioxide. A calcium chloride U-tube is also con- nected to the outlet of this Geissler bulb. It is connected to the inlet of the freezing tube. As in the first Geissler unit the potassium hydroxide bulb is weighed with its drying tube. Both Geissler bulbs are immersed in an ice bath during the test. The freezing tube. — The light oils are condensed in the freezing tube, figure V, which is immersed in a mixture of dry ice and acetone. The freezing mixture is con- tained in a wide mouth quart DeWar flask which is supported in cotton packing. The 216 APPENDIX A amount of light oils is determined by the increase in weight of the tube. They are removed and discarded by sweeping the tube with dry air at room temperature. The gasometer. — The apparatus (fig. II) used to collect the fixed gases leaving the absorption train consists of two 12-liter bottles suspended on movable cages for con- venient raising or lowering. The bottle used to receive the gas is graduated in 200 ml. divisions, the other bottle used for leveling the solution over which the gas is collected is not graduated. The gasometer bottles contain 13 liters of a saturated solution of sodium sulfate to which was added 5 per cent by volume of sulfuric acid. The solution was saturated with the fixed gases before starting the assay tests. The gas-receiving bottle is fitted with a rubber stopper carrying a thermometer, manometer, outlet tube (for sampling), inlet tube and siphon connecting the two bottles. At the start of the test the receiving bottle is filled with the gasometer solution and during the test the height of the cage is adjusted so that a pull of about 6 inches of water is imposed on the system. At the end of the test the pressure of the collected gas is brought to atmospheric pressure by adjusting the relative positions of the bottles, and the temperature, barometric pressure, and gas volume are recorded. Manipulation. — The packed and weighed distillation tube containing the tar piece and with the tar filter attached, is rolled in the hands and gently tapped to level the coal charge to give a free and open passage for the gases evolved during the test and to adjust the coal level uniformly for observation of the coking qualities. The tube is placed in position in the asbestos-lined trough for supporting it in the furnace, so that the first asbestos plug is immediately above the first burner. The copper steam bath is next placed in position and the small sleeve is placed over the outlet end of the tar- filter to keep the small end of the filter and the inlet tube of the first Geissler bulb hot to prevent condensation of naphthalene during the test. The first Geissler bulb, contain- ing the sulfuric acid, is now attached to the tar filter with seamless rubber tubing and the outlet of the bulb is attached to its accompanying drying tube containing granular calcium chloride previously saturated with C0 2 for the absorption of water. The potassium hydroxide bulb is connected to the drying tube of the first Geissler bulb and it in turn is connected to its drying tube which is connected to the freezing tube. The freezing tube is provided with its freezing jacket containing acetone and dry ice, and below each Geissler bulb is placed a battery jar ready for the subsequent addition of ice-water. The purified gas is lead from the freezing tube to the 12-liter gasometer, in which it is collected, with about three feet of l^-inch rubber tubing. The train, connected as described above, is tested for possible leaks by lowering the leveling bottle to produce suction and successively opening the stopcocks on the gasometer and the absorption train, starting at the last tube and working toward the furnace. Heating the combustion train. — When the train is found to be air tight, and after the silica brick becomes heated to a constant temperature of 720° C, which must be maintained for this section throughout the test, the gas inlet stopcock on the gasometer is closed, and the gasometer is refilled with solution. The leveling bottle is then set at a position such that there will be a suction of 6 inches of water from the gasometer at the end of the distillation. The fifth burner, that is, the one under the asbestos plug, is lighted immediately, and the stopcock on the gasometer is opened for the collection of gas. In a few minutes the gas pressure in the system increases so that ice and water may be safely placed in the battery jars containing the absorption bulbs. To cool the BY-PRODUCTS ANALYSIS METHOD 217 evolved gases so that a maximum absorption may be obtained, it is necessary to keep all of the absorbing bulbs in ice water. At the end of 10 minutes after the fifth burner is lighted, the sixth burner, or the first one under the coal charge, is lighted. The rest of the burners are lighted successively at approximately 10-minute intervals, the exact time being regulated by the speed of evolution of the gases. After all burners have been lighted, the heating is continued until there is practically no further evolution of gas. Disconnection of the train at the end of the test. — When the evolution of gas has ceased the burners of the furnace and the steam jet are turned off, and the suction of the gasometer is left on the train. If no leaks have developed during the distillation, the solutions in the Geissler bulbs will start to draw back as the furnace cools. The stop- cocks on the gasometer and on the first drying tube are then closed. The apparatus is allowed to cool, after which the absorption train is disconnected from the tar filter and freezing tube. Two liters of air free from carbon dioxide and moisture are slowly drawn through the absorption bulbs and drying tubes to displace the gas to bring them back to the same conditions under which they were first weighed. The bulbs with their accompanying drying tubes are then removed from the ice bath and allowed to stand until they have reached room temperature, when they are dried and weighed. The freezing tube which was removed from the train before air was drawn through the train is treated in a manner identical with the treatment before its first weighing. To avoid losses from oxidation due to air coming in contact with the hot tar and coke, the distillation tube and tar filter are allowed to cool nearly to room temperature before they are separated. The tar filter is then disconnected and weighed, and the distillation tube is allowed to stand in a vertical position for 20 minutes, so that it will fill with air, when it is also weighed. Analytical Procedure The separation and estimation of naphthalene. — The contents of the sulfuric acid bulb are transferred after the final weighing to a separatory funnel. The bulb is washed thoroughly with water and then with pentane and the washings are added to the separa- tory funnel. In washing the bulb with pentane care is taken to dissolve all of the naphthalene. After agitation the water layer is drawn off and reserved for the ammonia determination. The naphthalene-pentane layer in the separatory funnel is washed several times with water and the washings are added to the acid solution. The pentane solution is poured through a dry filter paper, this procedure is repeated, and in each step the filter paper and funnel are washed with fresh pentane. In this way water droplets are removed from the pentane solution. The last filtration is made into a weighed glass-stoppered 125 cc. Erlenmeyer flask and the pentane is evaporated in a stream of dry air at 0°C. When the evaporation is complete the flask is reweighed and the weight of naphthalene determined. The apparatus is pictured in figure VI. Determination of free ammonia. — As distinguished from the ammonia present in the tars referred to as combined ammonia, the gaseous ammonia is absorbed in the acid bulb and may be determined by the usual method of evolution and absorption in a known amount of standard acid as follows. The ammonia-acid liquor from the naphthalene separation is set aside in a Kjeldahl flask and is then ready for the ammonia determination. To the dilute acid solution approximately 1 gm. of granulated zinc and 20 cc. of 1:1 potassium hydroxide are added and the flask is connected at once to the ammonia distilling head. The distillate is collected in a dilute solution of a known amount of N/10 sulfuric acid. At the end of 218 APPENDIX A the distillation the excess acid is titrated with N/10 sodium hydroxide. Blank determi- nations are made on the reagents used. From the data obtained the amount of ammonia is computed. Fig. VI. — Apparatus for Filtering Pentane-Naphthalene Solution. Determination of combined ammonia.— The tars obtained in the coal carbonization assay contain combined ammonia. The contents of the tar piece and tar filter are extracted by placing them in boiling distilled water for ten minutes, which is then filtered. BY-PRODUCTS ANALYSIS METHOD 219 The determination of the ammonia in the water used for the extraction is carried out as in the free ammonia determination. Less acid may be used in the flask receiving the ammonia distillate since the amount of combined ammonia occurring is much less than the amount of free ammonia. Blanks are also run. Determination of hydrogen sulfide. — The potassium hydroxide in the second Geissler bulb is transferred into a 100 ml. volumetric flask, the bulb is washed several times with water and the washings are added to the volumetric flask. The solution is diluted to 100 ml. and portions are removed for the hydrogen sulfide determination. An aliquot part of the potassium hydroxide solution is added to an acidified known excess of N/10 iodine solution diluted with 300 cc. of water. The excess iodine is determined by titrating with N/10 sodium thiosulfate. Calculation of Results The following is a guide to the computation of the weights in grams of the various products of the distillation. The conversion of the computed weights of each to weight per cent, on the basis of the dry coal as 100 per cent, is made by multiplying by 5. (1) Volatile matter equals the loss in the carbonization tube and includes all the gaseous products and the tar. (2) Coke equals the remainder obtained by subtracting the weight of the volatile matter from 20, the weight of the sample of coal. (3) Ammonia equals the ml. of acid multiplied by the factor: equivalents per ml. X 17.03. (4) Tar equals the increase in weight of the tar filter and of the tar piece, plus the naphthalene, minus the combined ammonia. (5) Water equals the increase in weight of the sulfuric acid bulb and drying tube minus the naphthalene and free ammonia. (6) Hydrogen sulfide equals the ml. of N/10 iodine required for the entire bulb multiplied by .0017. The liters of hydrogen sulfide equal its weight divided by 1.4564 which is the weight of one liter at 60° F. and 30" Hg. (7) Carbon dioxide equals the increase in weight of the potassium hydroxide bulb and drying tube, minus the hydrogen sulfide. The liters of carbon dioxide equals its weight divided by 1.873 which is the weight of a liter at 60° F. and 30" Hg. (8) Light oil equals the increase in weight of the freezing tube. (9) Gas. — Grams of gas equal the grams of volatile matter minus the sum of the tar (including naphthalene), ammonia, water, hydrogen sulfide, carbon dioxide, and light oils. (10) Computations of the gas volumes. — To convert the observed volume of gas to a dry basis at 30" Hg and 60° F., formula No. 2, which is derived from formula 1, was used. Y x = observed volume of gas 459.6 +60 P ~ P i_ V 2 = corrected volume (30" Hg and 60° F, dry) en AO ~ ~~ P = observed barometric pressure in inches 459.6 + t 30 f mercury 17.32 (P — P ) P] = vapor pressure of water 1 V; X = V 2 (2) t = temperature of gas at the time of reading the 4595 _[_ t gas volume (temperatures are all Fahrenheit) i The vapor pressure of water is used in the computations. Although the gas is collected over the acidified sodium sulfate solution the vapor pressure lowering due t< the salt was not great enough to alter the corrected volumes as recorded. The magni- tude of the effect was investigated by analysis of the gasometer solution and by calcu- lations based on vapor pressures found in International Critical Tables. 220 APPENDIX A The gas yield in cubic feet per ton of coal is found by multiplying the corrected liters of gas into the factor 1600. 2 (11) Specific gravity of the gas. — The specific gravity of the gas is computed using the gas analysis data corrected for air and by the use of the following values: Constituent Sp. Gr. CH 4 0.5543 CO 0.9671 Illuminanrs 1.1520 H, 0.0695 N 2 0.9673 2 1 . 1053 In addition to specific gravities determined by calculation from the gas analysis the specific gravities of the gases obtained in this study were determined experimentally. The determined values were corrected for air by subtracting the value: (per cent N 2 X Sp. Gr. N 2 + per cent 2 X Sp. Gr. 2 ) and by adding the value: (per cent [N 2 + 2 ] X Sp. Gr. H 2 ). At the beginning of the test the system is filled with air. This air is displaced by the carbonization gases during the test. At the end of the test the gas remaining in the train is nearly pure hydrogen. Since this hydrogen is not added to the collected gas, correction is made for it and the air it displaced which is in the collected gas. (12) The calorific value of the gas. — The gross and net B.t.u. values of the gas are computed using the data: B.t.u. per cu. ft. B.t.u. per cu. ft. Gross Net CH 4 1012 911 CO 323 323 Illuminants 1853 1740 H 2 325 275 (13) Therms are calculated on gross and net bases: Total B.t.u. of the gas per ton of Coal X 1/10 5 , or cu. ft. gas per ton X B.t.u. per cu. ft. X 1/10 5 = therms per ton. (14) The practical yield per ton of coal. Gallons of light oil = 2.72 X per cent light oil Gallons of tar = 2.0 X per cent tar Pounds of sulphate = 77.6 X per cent total NH 3 (15) Example: To illustrate the assay the experimental data for a single test is given herewith: Carbonization test No. 44 Date: August 17, 1932 Sample: R-24 2 The volumes of gas are exclusive of carbon dioxide and hydrogen sulfide. If it is desired to increase the gas volume by the volum.es of these gases it may be done using the following relationships: Cu. ft. of CO2 per ton of coal = per cent CO2 X 170.8 Cu. ft. of H 2 S per ton of coal = per cent H 2 S X 219.8 by-products analysis method Carbonization Data 221 Before test Afte Difference Sample (dried at 105° C) Tar piece Distillation tube + sample + packing Distillation tube + sample + packing + tar piece Tar filter Sulphuric acid bulb + drying tube Potassium hydroxide bulb + drying tube Freezing tube 20.0000 16.6988 114.5668 131.2656 a 31.7306 136.2779 148.1874 82 . 2809 17.1385 a 108.6487 125.7872 32.1259 138.0778 148.7559 82.5052 + .4397 -5.9181 5.4784 .3953 1 . 7999 .5685 .2243 Coke: strongly caked, no swelling. Average temperature: of coil, 720° C. b ; of coke, 880° C. Observations on the gas Gas volume 6 . 90 1 . Gas temperature 36.4° C. Barometric pressure 29 . 15" Hg. Gas Density Temperature gasometer 29.7°C Temperature water bath 26.4 °C Barometric pressure 29 . 29" Hg Weight bulb 102.9540 Weight bulb + gas 103 . 1939 Volume of bulb 519 . 74 cc. Analytical data Naphthalene Flask 59 + naphthalene 56.0384 gm. Flask 59 55.8619 Weight of naphthalene 1765 Free ammonia Combined ammonia 0.1196 Normality of acid ,. 0.1196 41 .75 Acid neutralized by ammonia 3 . 10 .0850 Gms. ammonia 0063 Hydrogen sulfide Normality of iodine . 1048 Iodine equivalent to 1/10 H 2 S. .... .3.59 (1/10 titrated) Gms. H 2 S 0641 The gas analysis Gas laboratory number: Per cent C0 2 0.2 2 ... 1.0 Illuminants 4.1 CO 10.4 H 2 52.3 N 2 3.2 CH 4 28.8 100.0 G-82 a Calculated. b Observed temperatures in Centigrade, subsequent Fahrenheit. ^as volumes are corrected to 60" 222 APPENDIX A Preliminary Results Product Grams Per cent Remarks Coke. Tar Volatile matter 14.0819 1.0052 5.9181 2.4904 0.193 .5044 .0641 1.5384 .2243 70.41 5.03 29.59 12.45 .46 2.52 .32 7.69 1.12 (includes naphthalene, excludes combined NH 3 (includes tar) (after passing absorption train) Gas (fixed) Total NH 3 Carbon dioxide Hydrogen sulfide Water Light oil Gas corrected volume, liters 5 .87 Gas yield, dry cu. ft. per ton 9392 Gas + C0 2 + H 2 S, dry cu. ft. per ton 9893 Standardization of the Assay Before the examination of the selected coals was started a number of runs were made to standardize the laboratory manipulation. The first tests were made using Jena glass distillation tubes which required the removal of the open end of the tube at the end of the test. The tube was cut at the end of the first asbestos plug and this 4 to 5 inch section of glass contained the tar deposit corresponding to that found in the tar piece as used in the silica equipment. The Jena tubes were soon replaced by fused quartz tubes and the separate tar piece for the reason that the latter could be used repeatedly while the Jena tubes could be used only once. Furthermore, tar percentages were found to be less difficult to reproduce when using silica tubes. Table II — Data for Comparative Assays Products Illinois Steel Company Per cent State Geological Survey Average of two tests Per cent Coke Tar Free ammonia Combined ammonia Water Carbon dioxide Hydrogen sulfide Light oils Gas (by weight difference) Total volatile matter Gas analysis: Nitrogen Hydrogen Methane Illuminants Carbon monoxide B.t.u. per cu. ft. of gas — gross B.t.u. per cu. ft. of gas — net. . 73.57 4.50 .29 .01 5.49 1.36 .15 1.38 13.25 26.43 1.2 58.0 29.1 4.7 7.0 593 529 73.58 4.25 .31 .01 5.73 1.45 .19 1.24 13.22 26.42 54.4 33.2 5.2 7.3 633 566 BY-PRODUCTS ANALYSIS METHOD 223 Preliminary tests were made using the fused quartz tubes and they were used in all the subsequent by-product analyses. Through the courtesy of the Illinois Steel Company it was possible to compare the by-product analysis of two laboratories on a single sample of coal. The Central Laboratory of the company supplied the Survey laboratory with a sample previously analyzed by them, and also a complete report of the analysis. The by-product analysis by the Central Laboratory was made by the use of the dry-distillation test method of the chemists of the United States Steel Corporation. Duplicate assays were made by us on this sample and pertinent items of the analyses are compared in Table II. APPENDIX B DESCRIPTIOM OF I Ml IGDE-DAMM PLASTIC RANGE TEST IS APPLIED TO ILLINOIS COALS MetM A lamp I e of com| ss red coal is heated at a constant rate of temperature ction of the sample is observed and temperatures are -h<»w the initial softening, active decomposition, and solidification temperatures in the plastic range of the coal. Some exceptions are found uratiu and Procedure \ cylindrica I inches in diameter and 7 inches long is fitted into an I tu- copper block is bored with two \ j-inch holes which with the dimension l > mm. inside diameter 03 75 mm. in length. sily-movable hollow rod is mounted vertically so that rhe lower end of the rod is hollow to receive a thermo- couple. [*he weight "t th< - ljusted to one pound and during the test a micrometer mounted to bear on the top of the rod. A Brown recording potentiometer ruin thermocouple are used for temperature measurements. Ilu- tuxiliai ttus consists of two carbon-pile resistances and an In the prescribed two test specimens are prepared, one of which is placed under the one-pound the other is allowed to expand or contract freely in the le of the copper block. A sample • • i gram of 100-mesh coal is compressed in the small glass test tube under a weight oi 5.8 kilos. 1 he length of the briquet is recorded and the specimen is in the coppei block at room temperature. The one-pound rod is placed so that it n the coal, and the thermocouple is adjusted to fit into the bottom of the rod. The is mounted with the pointer at an intermediate position on the scale. The furnace is heate 1 at a rate of 4\2 (A per minute and heating is terminated at Temperature, distance, and time are recorded. The data are plotted with temperatures as abscissae and distance readings as ordinates. Figure 54 p. 90 shows the furnace, with distance gage in position, as well as auxiliary apparatus. Corrections and Interpretation Since the temperatures recorded are those for a position immediately above the coal it is necessarx to correct the temperature data. To determine the amount of the correction i Procedure as described in the U. S. Bureau of .Mints Bulletin :U4. i>. 16, L931. [225] 226 APPENDIX B two thermocouples were mounted in the furnace: one, in the prescribed position in the plunger-rod, and the second inserted directly into a compressed sample of the coal. Temperatures in the coal were higher than the temperatures recorded and corrections were accordingly applied. Other corrections in connection with the differential expansion of the iron parts of the apparatus were not applied since the small uniform tendencies involved do not interfere with the interpretation of the data. This is true because the test is not concerned with the absolute expansion of the coal but rather with the temperatures at which the changes take place. When the distance versus temperature data are plotted the following temperatures may be read from the curve: (1) The initial softening temperature is the temperature at which the sample starts to contract due to fusion and softening of the coal particles (2) The initial active decomposition temperature is the temperature at which active rapid decomposition of the coal commences, and this is indicated by the sudden expansion of the coal. In the case of some samples the expansion does not occur. If the particles of coal do not coalesce, the gases escape without causing swelling (3) The solidification temperature is the point at which expansion ceases abruptly and at which the material sets into a rigid cellular mass (4) The plastic range is the temperature interval between the initial softening temperature and the solidification point Shrinkage of the solidified sample is also indicated by the curve. APPENDIX C THE TEST FOR AGGLUTINATING VALUE AS APPLIED TO ILLINOIS COALS Method The apparatus and procedure used in the test closely adhere to those proposed by the American Society for Testing Materials. 1 A sample of 200-mesh coal is mixed intimately with sized Ottawa sand 2 in the proportion of 14 parts of sand to 1 part of coal and carbonized at 950° C. for 20 minutes. The coke-sand button thus obtained is crushed in a compression machine which applies the load at a constant rate, and the crushing strength of the button is determined. Six buttons are tested for one sand-coal ratio and for each sample of coal. If one or more buttons have crushing strengths deviating more than 10 per cent from the average a new set of buttons is prepared. Apparatus (a) Furnace. — The furnace used in the standard volatile matter determination is used to carbonize the coal-sand buttons. It is regulated to maintain 950° C. in the crucible zone, and a calibrated chromel-alumel thermocouple is used to measure the temperature. (b) Compression machine. — A Scott testing machine 3 ordinarily used for tensile tests of rubber or fabrics was adapted to the work by the use of a cage designed to provide compression. A constant-speed motor and reversible gear arrangement gave a constant rate of compression of one inch per minute, and the crushing strength of the specimen was observed in pounds. Machines designed specifically for the agglutination test are on the market. (c) Sieves. — (1) For the preparation of the coal: U. S. Standard Series sieve No. 200 (74 micron opening). (2) For the preparation of the sand: U. S. Standard Series sieve No. 45 and No. 60 (350 micron and 250 micron respectively). (d) Porcelain crucibles. — (1) For moistening sand, and for mixing coal and sand: Coors high-form crucible, size 2: 52 mm. top diameter; 25 mm. bottom diameter; 43 mm. height; 55 ml. capacity. i Proposed Draft, June 1934. Amer. Soc. Test. Material, Standards on Coal and Coke, pp. 89-94, Sept. 1936. 2 Varying results are frequently obtained from duplicate tests using- different batches of sand. This is no doubt due to the fact that the surface of the sand grains has much to do with the values obtained, and apparently no two batches are identical. Recently an attempt has been made to substitute carborundum as the inert material for this determi- nation. Although it holds some promise, the difficulties have not yet been overcome. 3 Henry L. Scott and Co., Providence, R. I. [227] 228 APPENDIX C (2) For carbonizing the coal-sand mixtures: Coors cylindrical crucible (with covers) No. 390: 28 mm. inside diameter; 30 mm. inside height; 18 ml. capacity. These crucibles must have circular cross-section, and must be free from irregularities on the inner surface. If a lip is on the inside of the rim it may be removed by grinding. Covers are used. Since their outside diameter is nearly equal to that of the tube furnace two notches are ground in the sides of the cover where it comes in contact with the wires supporting the crucible. (e) Compressing device. — The coal-sand mixture is compressed in the cylindrical crucible by a weight of 3.5 kg., and the apparatus was made in the laboratory. An easily movable piston with a machined face fits snugly, without binding, into the cylindrical crucibles, and the piston is weighted to make a total load of 3.5 kg. Procedure (a) Calibration of the furnace. — The temperature of the standard volatile matter furnace is controlled by means of a rheostat, and the temperature in the furnace is measured by a thermocouple installed permanently through the bottom of the furnace. However, the equipment is tested to ascertain the temperatures attained inside the crucible. A cylindrical crucible is rilled with sand and a hole is ground in the center of a porcelain cover to admit a thermocouple. The crucible and sand are placed in position in the furnace and the standardizing thermocouple is inserted through the hole in the cover. The fused junction is fixed at the geometric center of the sand and crucible and the furnace rheostat is adjusted until the prescribed temperature of 950°C. is maintained inside the crucible. At this setting the temperature read on the permanent couple is recorded and used subsequently since it corresponds to the desired temperature in the crucible. (b) Preparation of the sand. — The sand used for mixing the sand-coal mixtures is natural silica sand obtained from Ottawa, Illinois. 4 It is carefully graded to — 45 -{-60 in particle size and this portion is washed with water. The turbid water is decanted and the washing is repeated 5 times, after which it is boiled in dilute HC1 (1:1) for 30 minutes. It is then washed thoroughly (until free from chloride) and dried. (c) The sand-coal mixture. — The fresh coal is ground to 200-mesh, and the weighed sample is mixed with the weighed portion of sand which previously has been moistened with glycerine. The 14:1 sand-coal ratio was used in the work on Illinois coals but other ratios, which have been tested, were as follows: Ratio sand Weight'of Weight"of to coal sand, gm. coal, gm. 10:1 18.182 1.818 12:1 18.462 1.538 14:1 18.667 1.333 15:1 18.750 1.250 20:1 19.048 0.952 25:1 19.231 0.769 The sand is weighed into the high-form No. 2 crucible and moistened with a drop of glycerine weighing 0.07 gm. 5 The sand is thoroughly mixed with a spatula for 1 minute. The weighed portion of coal is added to the crucible and mixed into the sand for 2 minutes. Ottawa Silica Company, Ottawa, Illinois. TEST FOR AGGLUTINATING VALUE 229 The mixture is carefully transferred into the cylindrical crucible and the top is leveled by means of a square-end spatula. It is compressed for 30 seconds under a load of 3.5 kg. by means of the compression device. After compression the empty space at the top of the crucible is filled with sand-coke mixture from buttons previously crushed. With the cover in place the button is car- bonized at 950° ± 10° C. for exactly 20 minutes. It is allowed to cool in air, and then it is removed by inverting the crucible. If the faces of the button have irregularities they are removed by the use of No. 00 sandpaper. The prepared button is placed in the center of the ^-inch rubber pad on the bottom of the crushing cage and the crushing strength is determined. 5 The drop of glycerine weighing .07 gm. is obtained from a suitable burette, or by use of a heavy-wall capillary tube ground plane on the end. With additional grinding to reduce the area of the plane face the weight of the drop may be reduced. Index Agde-Damm plastic range test, Appendix B, 225, 226 Adge-Damm test, 90, 225, 226 Agglutinating value, test for, 92; Appendix C, 227-229 of Illinois coals, 111-112 Significance of, 92 Appendix A, by-products analysis method, 209-223 Appendix B, Agde-Damm plastic range test, 225, 226 Appendix C, test for agglutinating value, 227-229 Ash in coal, effect on coke structure, 118 effect of agglutinating value, 113 Atmospheric pollution, cost of, 37 Ash content of coke breeze, 118-119 B Beehive ovens, 43 coking test in, for 111. coals, 50 in Illinois, 49, 50, 51 Behavior of 111. coal under heat, results of tests agglutinating values, 111, 112 characteristics of coal affecting aggluti- nating values mineral impurities, 113 freshness, 113 fusain content, 113 plastic ranges, 109, 110, 111, 112 significance of plastic range, 111 softening temperatures, 109, 110, 111, 112 Bituminous coal, disadvantages for blast furnace, 41 markets for, in Illinois area, 18, 19 Briquets, coal, consumed in 111. market area, 20 By-product coke disposal of, by states in 111. market area, By-product coke {Cont.) distribution by percentage in U. S. 27 distribution by uses in U. S., 24, 25 distribution by uses in 111., 32, 35 market for coal, 13 production in 111., 23 production in U. S., 23 source of coal used for, in 111. market area, 32 By-products analysis method, Appendix A, 209-223 Carbonization assay of coal, method of, Appendix A, 209-223 Carbonization assay results, 104-107 Carbonization, low-temperature Greene-Laucks process, 55 Parr process, 54 reasons for, 46 Carbonization, medium-temperature, 47 Coal agglutinating or caking power, test for, 92 bituminous, disadvantages for blast fur- nace, 41 cost, 21 coking, resources in 111., 21 consumed, 1917-1936 by public electric utilities, 14 for beehive coke, 14, 15 for by-product coke, 14, 15 fields of U. S., 79, 81 liberation of volatile matter from, test for, 91 market for small size as coke, 42 petrographic components, 115 petrographic components, behavior of sulfur content during carbonization, 133, 134 reasons for coking, 41 softening characteristics, test for, 90, 225-226 [231] 232 INDEX Coal (Con't.) St. Louis, origin, 21 swelling characteristics, test for, 90, 225- 226 used in Knowles oven tests, sampling and analysis of, 186 weathering effect on sulfur content, 133, 136 Coals coked in sole-flue ovens, analyses of, 196 Coals, coking analysis, A.S.T.M., 60 areas of U. S. in which produced, 79, 81 ash requirements of, 61 carbon ratio of, 60 chemical properties of, A.S.T.M., 60 moisture requirements of, 60 physical properties of, A.S.T.M., 60 requirements for slot type ovens, 179 sampling, A.S.T.M., 60 special requirements of, 61 special requirements for gas in, 61 specifications for, A.S.T.M., 60 sulfur requirements of, 61 Coals, Illinois analyses of, 62, 64 ash content, 64, 70 carbon ratios, 62, 64, 65, 66, 67, 68 carbonization tests, small scale, American Gas Association, 56 carbonization tests, small scale, Parr, 55 carbonization tests, small scale, U. S. Bur. Mines, 56 carbonizing properties of, 116 classification, by location, 76, 77; by rank, 75; for coking, 73; on basis of specific volatile index, 82, 83 coke making from, historical, 49 coke making possibilities, 21 coking classification by ultimate analyses, 79,80 coking in Knowles sole-flue ovens, 54 coking tests by Parr process, 54 in beehive ovens, 50 in beehive ovens, Sparta, Randolph Co., 51 in by-product ovens, Canal Dover, 52 in by-product ovens, St. Paul, 53 in Koppers ovens, 53 in Roberts ovens, 53 Coals, Illinois (Con't.) composition in terms of banded com- ponents, 116 conclusions as to coking suitability, 83 industrial coking plant for, 179 market area, definition, 17 moisture content, 64, 69 occurrence of, organic sulfur in, 123; pyritic sulfur in, 123 petrographic constituents of, 115 rank indexes, 73 samples used in experimental work, 95, 96 specific volatile index, 83 sulfur content, 62, 64, 71, 72 tabulation of tests, 95 ultimate analyses, 79, 80 Coals, United States coking classification by ultimate analyses, 77, 78, 79 Classification chart, 74 Coke beehive distribution by percentage in U. S., 27 distribution by uses in U. S., 24, 25 market for coal, 13 production in U. S., 23, 24 by-product disposal of, by states in 111. market area, 22, 32, 34 distribution by, percentage in U. S., 27 distribution by, uses in U. S., 24, 25 distribution by, uses in 111., 32, 35 market for coal, 13 product in 111., 23 production in U. S., 23 source*of coalused for, in 111. market area, 32 cleanliness as domestic fuel, 38 domestic market in Illinois, 18 produced and total sales in 111., 33 effect, of ash on structure, 118; of banded components on structure, 115; of moisture on coking time, 117; of sulfur on, 122 efficiency as domestic fuel, 37 firing methods as domestic fuel, 39 from 111. coals Coke {Cont.) box tests in Knowles sole-flue ovens and results, 166-177 general conclusions of tests, 170 from 111. coals and blends tests in six-inch retort, 141; coals used tests, 141, 142; results of tests, 143- 165 general conclusions of tests, 170 history of manufacture in 111., 57 industrial plant for production in 111., 179 low temperature, advantages and dis- advantages, 47 marketing problems, 21 markets, metallurgical in 111., 18; new, 22 methods of making, 43 produced in Knowles oven tests, sampling and analysis, 186 produced in sole-flue ovens, analyses, 196 production from 111. coals; experimental work, 85 quality, influence in box tests of coal cleaning, fusain, inert matter, reduc- tion of mineral matter, size, 169 quality, influence in six-inch retort tests of added moisture, 151 addition of non-caking carbonaceous material, 150 addition of tar or petroleum oil, 150 kind and size of coal, 143 low volatile coal, 150 rate of heating and final temperature, 151 removal of mineral matter, 151 quality, test for, 92 reasons for manufacture, 41 relative cost as domestic fuel, 38 removal of sulfur from, 137 shatter tests for, 93 sulfur content as related to organic sulfur in coal, 126, 128, 129, 130 pyritic sulfur in coal, 126, 128, 129, 130, 131, 132 sulfate sulfur in coal, 137 total sulfur in coal, 125, 126, 127, 128, 129 sulfur in, 124 use as domestic fuel, 37 Coke ovens beehive Coke ovens {Cont.) Beehive {Cont) in Gallatin Co., 49 in Illinois, 49 in Jackson Co., 49 in South Chicago, 50 in Sparta, Randolph Co., 51 tests of Illinois coal in, 49 Becker — see Koppers by-product, 43, 44, 45 location of, in 111. coal market area, 26, 28-31 by-p'roduct using 111. coals coked in Canal Dover, O., 52 coked in Granite City, 111., 53, 54 coked in St. Paul, 53 Knowles sole-flue, 179-181 coking 111. coals in, 54 construction of, 182 operation of, 181 Knowles sole-flue, tests in analyses of coals coked, 196 analyses of coke produced, 196 analyses of gas produced, 202, 203 equipment for temperature measure- ment in, 184 heat gradient, maximum at hourly in- tervals, 193, 194 heat gradient, relation to volatile matter gradient, 195 isochronal relations in charge, 187, 190, 191 isothermal relations in charge, 187, 190 oven conditions, influence on coke character, 197-201 plastic zone, estimation of, 192, 194 plastic zone, travel of, 192, 195 procedure for temperature measurement in, 185 results of temperature measurements, 186-190 sampling and testing of coal, coke, and gas, 186 temperature in flues, 195 temperature in regenerators, 195 temperature in stack, 195 temperatures in charge in, 187 temperature- time relations in charge, 186, 187 234 Coke ovens {Cont.) time and temperature relation of, 191 volatile matter gradient, 195, 202 Koppers, coking tests of 111. coals in, 53 plants, 23, 26, 28 relative sizes, 45, 46 Roberts type, 33 coking tests for 111. coal in, 53 sole-flue, 44, 179, 180, 182 Coking classification of 111. coals by use of ultimate analyses, 79, 80 Coking coals requirements, 59 sources in 111. market area, 32 Conclusions of report, 206 D-L Domestic fuels, relative cost and average efficiencies, 38 y 39 Experimental work introduction, 85 methods for chemical analysis, 87 methods used, 87 selection of methods, 87 small scale tests, value of, 85 test for agglutinating or caking power, 92 test for gas, coke, and by-product yields, 89, 209-223 test for liberation of volatile matter from coal, 91 test for plastic range, 90, 225, 226 tests for influence of heat on coal, 88 Fuel oil consumed in 111. market area, 19 Fuel requirements in 111. coal market area, 18 Fuels consumed in 111. coal market area, 18 Gas, manufacture of illuminating, 41 Gas, natural, consumed in 111. market area, 20 Gas, produced in Knowles oven tests, sampling and analyses, 186 Gas produced in sole-flue ovens, analyses, 202, 203 Gas retorts, 43 Heat gradient in sole-flue oven charge, 193, 194 Illinois coal market area, 17 Isochronal relations in sole-flue oven charge, 187, 190, 191 Isothermal relations in sole-flue oven charge, 187, 190 Low-temperature carbonization, reasons for, 46 M-Y Market for small size coal as coke, 42 Mineral matter, in fine sizes of 111. coals, 121; effect of on coke structure, 119-120; removal of, by coal washing, 122 Moisture in coal, effect on coking time, 117 Oven conditions, sole-flue ovens, influence on coke character, 197-201 (See also coke ovens). Oxidation, effect on coking properties of coal, 109, 110, 113. Parr process for coking tests of 111. coal, 54 Plastic range test, Agde-Damm, Appendix B, 225-226 Plastic zone estimation in sole-flue oven charge, 192, 194 heat gradient across, 192-193 travel in sole-flue oven charge, 192, 195 Purposes of the report, 13 Recommendations for additional work, 207 Report, conclusions of, 206 scope of, 15, 205 Scope of the report, 15, 205 Small scale tests method of, 87 value of, 85 Smoke pollution, cost of, 37 Sole-flue ovens, influence of oven conditions on coke character, 197-201 Specific volatile index, use in classifying coals, 82 Sulfur, behavior during carbonization, 124; conclusions, 135 behavior in U. S. Steel Corporation car- bonization, 135, 138 content of 111. coal, 62, 71, 72 effect on coke quality, 122 in coke, 124 occurrence in 111. coal, 123 organic, in coal, 123 pyritic, in coal, 123 removal from coke, 137 removal from 111. coal, 123 235 Temperature in sole-flue oven, 195 equipment for measurement, 184 in flues, 195 in regenerators, 195 in stack, 195 results of measurements, 186-190 procedure for measurement, 185 Temperature and time relations in sole- flue oven charge, 186, 187, 191 Tests applied to 111. coals, key to, 95 Time and temperature, relation of in sole- flue ovens, 186, 187, 191 U. S. Steel Corporation test, 89, 209-223 U. S. coals, see Coals, U. S. Volatile matter gradient in sole-flue oven coke, 195, 202 Washability of 111. coals, 122 Washing 111. coals, effect on sulfur content, 123-124 Yields, coke, gas, and by-product according to geographic location, 99, 100, 104-107 comparison of results from several labor- atories, 101, 108 comparison of results on coals from other states, 103 comparison to rank indices, 101, 102, 103