LIBRARY OF THE UNIVERSITY OF CALIFORNIA. Class EUGENE ALLEN SMITH. PH. D.. Director. IRON MAKING IN ALABAMA, s SECOND EDITION BY WILLIAM BATTLE PHILLIPS, PH. D, Consulting Cneiist ana" Metallurgist ! MONTGOMERY, ALA.: II ROEMER PRINTING CO., STATE PRINTERS AND BINDERS. 1898. ALABAMA GEOLOGICAL SURVEY EUGENE ALLEN SMITH, PH. D,, Director. IRON MAKING IN ALABAMA SECOND EDITION BY WILLIAM BATTLE PHILLIPS, PH, D. f Consnltins demist and ffletallnrgist- MONTGOMERY, ALA.: ROEMER PRINTING CO., STATE PRINTERS AND BINDERS. 1898. ftr . To His Excellency, JOSEPH F. JOHNSTON, Governor of Alabama : DEAR SIK : I have the honor to transmit herewith a Second Edition of Dr. Phillips' Report on Iron Making in Alabama. Very respectfully, EUGENE A. SMITH, State Geologist. University of Alabama, October 1st, .1898. TABLE OF CONTENTS. Letter of Transmifctal, 1-2. Introduction to First Edition, 3-11. Introduction to Second Edition, 12-15. CHAPTER I. THE ORES GENERAL DISCUSSION. Kinds of ore used ; no known deposits of Bessemer ore ; phosphorus in the ores used ; production of the dif- ferent varieties of iron ore in the United States; produc- tion of pig iron in the United States ; purchase of ore on analyst s ; improvement of the ores : production and valu of the ores in the United States and in Alabama. Pages 16-34. CHAPTER II. THE HEMATITE ORES SPECIAL DISCUSSION. Classification ; the soft red ore ; vertical section of the seam of soft red ore ; analysis of the soft red ore ; the hard red ore , or limy ore ; analysis of the limy ore ; the brown ore, or limonite ; occurrence and mining of the brown ore : analysis of the brown ore ; valuation of the brown ore ; mill cinder ; blue billy, or purple ore. Pages 35-61. CHAPTER III. THE FLUXES. Limestone ; analysis of tbe limestone ; limestone being replaced by dolomite ^analysis of dolomite ; use of dolo- mite in the furnace. , -Pages 62-75. VI CHAPTER IV. FUEL. Classification ; chemical composition ; physical struc- ture; composition of ash; comparison with other cokes of the country ; statistics of coke ovens built and building ; coking in a bee-hive oven ; analysis of gas from a bee- hive oven ; changes undergone by coal in coking; yield of coke in a bee-hive oven; Alabama coal in Otto-Hoffman by-product ovens ; the Semet-Solvay by-product oven ; bee-hive and by-product coke. Pages 76-138. CHAPTER V. COKE FURNACES. Coke furnace practice on different burdens ; burdens of soft and hard red ore; burdens of soft, hard and brown ore ; comparisons of the various results ; consumption of raw materials ; consumption of coke ; cost of the raw materials per ton of iron; various data in regard to blast furnace practice with coke and charcoal. Pages 139-165. CHAPTER VI. PIG IRON. Ordinary grades made ; some Bessemer pig iron has been made but the supply of Bessemer ore is limited and the composition variable ; grades recognized ; nor- mal composition of the different grades ; new system of grading suggested. Pages 166-186. CHAPTER VII. COST OF PRODUCING PIG IRON IN ALABAMA. 9 . * Returns from the United States Labor Bureau ; inde- pdhcLdnt returns ; ddfc&ils of cost ; comparisons for three rears. Pages VII CHAPTER VIII. COAL AND COAL WASHING. Area of coal-fields ; coal production by counties ; avert* age price of coal at mines; statistics of labor employed and working time ; bituminous coal product of the United States; various data in regard to the coal mines of the State ; coal washing ; list of coal washing plants ; results of washing coal ; Jeremiah Head on the Birmingham district ; calorific power of coals; Landredth's results : independent results ; comparison with other coals . Pages 200-246. CHAPTER IX. CONCENTRATION OF LOW GRADE ORBS. Magnetization and concentration ; use of Hoffman concentrator; use of the Payne concentrator; use of the Wetherill concentrator on non-magnetic ore ; excellence of results reached on low grade soft and limy ore , need of some system of using the large deposits of low grade ore ; observations on the situation in Alabama. Pages 247-289. CHAPTER X. BASIC STEBJL AND BASIC IRON, The first production of basic steel in the State ; early experiments of the Henderson Steel and Manufacturing Company at North Birmingham ; results of the work in 1888 ; chemical and physical qualities of the first basic steel made ; furnace charges; steel at Fort Payne ; steel made by the Birmingham Rolling Mill Company ; com- parison of the Birmingham steel with similar steels of northern make. Pages 290-344. VIII CHAPTER XI. Coke furnaces in Alabama ; periods of greatest activity in construction ; production of coke iron ; charcoal fur- naces jn Alabama ; periods of greatest activity in con* struction ; production of charcoal iron ; statistics of hot blast stoves ; rolling mills, steel works, pipe works, car works; statistics of production of pig iron, coal and coke ; freight tariffs on pig iron, etc. Pages 345-371. LETTER OF TRANSMITTAL. (First Edition.) Dr. Eugene A. Smith. Director, Ala. Geol. Survey, University, Ala. SIR I beg to transmit herewith a report on Iron Mak- ing in Alabama, prepared for the Geological Survey. No systematic attempt has yet been made to bring this industry to the attention of the general public. Numer- ous article, have appeared in the technical papers in this and other countries during the last ten years, deal- ing with special phases of the subject, and many of them possess great merit. In particular may be mentioned the following : The Iron Ores and Coals of Alabama, Georgia, and Tennessee. Jno. B. Porter, Trans. Amer. Inst. Min. Engrs., vol. xv, 1886-87, pp. 170-208. Comparison of some Southern Cokes and Iron Ores. A. S. McCreath and E. V. D'Invilliers. Trans. Amer. Inst. Min. Engrs., vol. xv, 1886-87, pp. 734-756. General Description of the Ores used in the Chatta- nooga District. H. S. Fleming. Trans. Amer. In<4. Min. Engrs., vol. xv, 1886-1887, pp. 757-761. The Pratt Mines of the Tennessee Coal, Iron and Rail- way Co. Erskine Ramsay. Trans. Amer. Inst. Min. Engrs., vol. xix, 1890-91, pp. 296-313. Notes on the Magnetization and Concentration of Iron Ore. Wm. B. Phillips, Trans. Amer. Inst. Min. Engrs., vol. xxv, 1895-1896. A series of articles by E. C. Pechin, in the Iron Trade Review in 1888, and by the same author in the Engineer- Z GEOLOGICAL SURVEY OF ALABAMA. ing & Mining Journal, vol. Iviii, 1894. The Proceeding of the Alabama Industrial and Scientific Society, 1891-1897, contain many valuable papers, as also the files of the Engi- neering wrought out the Tennessee. They were as full of pro- gressive ideas regarding steam and armor as their master, and nothing but the scanty means at their dis- posal prevented a much more formidable iron-clad than the Tennessee from being set afloat. ' 'Car-wheel makers are the exclusive users of our iron." It is interesting to note in connection with the Con- federate States foundry at Selma, that it used coke made from the Gholson seam mined at Thompson's Lower Mine, on Pine Island branch, in Sec. 10, T. 24, R. 10 E., Bibb County, and elsewhere in the vicinity, as we are informed by Eugene A. Smith (Ala. Geol. Survey Re- port of Progress for 1875, pp. 32 and 33.) This was about 1863, and is probably the first use of Alabama coke for foundry purposes. "In 1863-64 Capt. Schultz of the Confederate army made a large quantity of coke from seams in the Coosa coal field, getting it to market by floating it down the river in flats to the railroad bridge across the Coosa o River, whence it was carried by rail to Montgomery and Selma. The coke was said to be the finest ever made in the State, and to equal the very best English cokes."' (Smith ut supra, p. 38.) In 1825, there was a bloomary near Montevallo, Shel- by County ; several in Bibb County in 1830-1840 ; one in Talladega County in 1842; two in Calhoun County in 1842. In 1856 there were enumerated 17 forges and bloomaries, about one-half being in operation and pro- ducing 202 tons of blooms and bar iron. The total product of charcoal pig iron in 1856 was 1,495 gross tons. IRON MAKING IN ALABAMA; INTRODUCTION. 11 In 1876 the Eureka Coke Furnace was built at Ox- moor, Jefferson County, by Col. J. W. Sloss, one of the most active iron-masters in the State, and the founder of the coke iron industry. This was the first furnace to go in on coke, and was followed in 1880 by the Alice furnace, built at Birmingham in 1879-80, by H. F. DeBardeleben, another noted name in the history of the iron trade in Alabama. Then followed the first of the Sloss furnaces at Birmingham, built by Col. J. W. Sloss in 1881-82. and put in blast April 12th, 1882. Space would fail us to enumerate the names of those concerned in the early history of the coke iron trade in Alabama, but J. W. Sloss '(who died in 1890), H. F. DeBardeleben. T. T. Hillman, and Geo. L. Morris, who are still enjoying the fruits of their foresight and en- ergy, will always be first called to mind by the histo- rian of the days, not long past as we measure years, but removed from us by a continuous series of splendid achievements. Si monumentum quseris, circumspice. WM. B. PHILLIPS. BIRMINGMAM, ALA., May 1896. 12 GEOLOGICAL SURVEY OF ALABAMA. INTRODUCTION TO THE SECOND EDITION. The very kind reception this little book has met with from the public has so nearly exhausted the first edition that a second is now thought necessary. It was the first systematic attempt to set in order the conditions under which the manufacture of pig iron has been pos- sible in Alabama, and although none realized its imper- fections more keenly than the author, yet. they were errors of judgment and not of fact. With the exception of the introduction of double and in one case treble the usual number of tuyeres, the blastfurnace practice has not altered, materially, since the publication of the first edition in 1896. The same ores are being used, and the same coke. The use of dolomite, as flux, has steadily increased, so that it has now become the main fluxing material in the Birmingham district. The cheap soft red ore is becoming notably scarcer, and there is more interest felt in" deposits of brown ore, and in the possi- bility of employing larger amounts of hard, or limy ore. At least one new brown ore deposit has been opened in the Birmingham district, by the Sloss Iron and Steel Co., and is of great promise. The brown ore deposits near Russellville, Franklin County, now supply the Sheffield furnaces, and excellent results have been reached by Mr. J. J. Gray in the use of these ores. The brown ore deposits near Anniston have been reopened, and the Woodstock furnaces have been making a good record. The Pioneer Mining and Manufacturing Com- pany, with two furnaces at Thomas, have opened a new soft ore mine on Red Mountain, near Bessemer, and IRON MAKING IN ALABAMA ; INTRODUCTION. 13" new openings on Red Mountain have been made by J. W. Worthington & Co. This latter company con- tinues to mine excellent dolomite from the Dolcito quarry, six miles from Birmingham. The Jefferson' Mining and Quarrying Co., also mines excellent dolo~ mite, somewhat nearer the city, and the dolomite at North Birmingham has also been opened by The Sloss Iron and Steel Co. The Solvay Process Company, Syracuse, N. Y., is building 120 Sernet-Solvay Recovery coke ovens at Ensley, to be operated in connection with the blast furnaces there, owned by the Tennessee Coal, Iron and Railway Co. It is thought that they will be in operation by the close of 1898. Messrs. Stein 5 % , and to make proportionally as good a showing for the low grade limy ores, and the low grade brown ores. No attempt at concentrating the red ores is now being made, but it seems to be not out of place to detail what was done. That these ores will some day come into use through some method of concentration seems probable. The chapter on Fuels has been entirely recast, and a large amount of information gathered by the writer in his own laboratory in regard to the chemical and physi- cal quality of the various cokes has been added. A new chapter on Pig Iron has also been added, and many analyses of the various grades have been inserted, and anew chapter als? on Coal Washing, additional in- formation as to the coal industry, compiled from the reports of Mr. James D. Hillhouse, State Mine Inspector, has also been inserted, including the number of mines operated, the number employees, &c., &c. Having been consulting chemist for the Birmingham Kolling Mill and Steel Works since the building of their first basic open hearth steel furnace, the opportunity of adding a chapter on Steel Making has been presented. IRON MAKING IN ALABAMA ; INTRODUCTION. 15 My sincere acknowledgments are due to the above company for its kindness in permitting the publication of information not hitherto given to the public. Mr. David Hancock has been associated with me in the steel laboratory, and has been of the utmost assistance. In connection with the making of steel there will be found a full description of the manufacture of basic iron in Alabama, so far as concerns its chemical aspect, which is republished from The Mineral Industry, Vol. V, through the courtesy of the Scientific Publishing Co., New York. There has also been added a chapter on the cost of making pig iron in the State. This has been done to correct an impression that iron is made here for less than $5.00 a ton. It is no longer a question that the cheapest pig iron made in the world is made in Alabama, and it has been thought that a brief statement of facts in regard to the matter would not be out of place. The exportation of 218,633 tons of iron to England, Continental Europe, Japan &c, during 1897, as against 65,000 tons in 1896 marks a new and hopeful develop- ment of outside markets for Alabama iron. WM. B. PHILLIPS, Birmingham, Ala., May 1898. 16 GEOLOGICAL SURVEY OF ALABAMA. IRON MAKING IN ALABAMA. CHAPTER I. THE ORES : GENERAL DISCUSSION. The ores used in the production of pig iron in Ala- bama fall naturally into two classes, and for convenience of reference the local names will be used with full ex- planations under each. They are either limonites, the so-called " brown ores," or hematites, the so-called soft, and hard ores. There are deposits of blackband ores and of magnetites, none of which, however, come into use. Efforts have been made to use the more or less bituminous blackband ores, both raw and calcined, but they were not successful. Several years ago an attempt was made at one of the coke furnaces to employ the raw blackband ore found in association with one of the coal seams in the northern part of Jefferson County, but the furnace worked badly, probably owing to the very bitu- minous nature of the ore, and the experiment was dis- continued. The same ore was afterwards calcined in, piles in the open air and a portion of the resulting ma- terial was of fair quality. But owing, it is thought, to the lack of care in the management of the business there was a gooddeal of trouble from the caking of the ore. In places it resembled impure iron and was almost malle- able. Nothing has been done in this direction for some time, as the available supply of ores that do not need such treatment is still very large. Practically all of the iron made in the State has been produced from Jimonite,- hematite, or a mixture of the two. IRON MAKING IN ALABAMA ; THE ORES. 17 For special purposes, as for instance, car wheel iron or some particular kind of iron destined for the pipe works, brown ores alone are used, although at times some admixture of hematite is permitted even then. For ordinary foundry and mill irons, and of late for basic iron, the common practice is to use a mixture of brown and hematite ores, the proportion of brown ore being for the most part about 20 per cent, of the ore bur- den, although there are some important exceptions to this rule. It seems best to take up the ores under separate head- ings, that a fuller understanding of the subject may be reached, but before doing so some observations on the ores in general may not be out of place. In Alabama a vast deal of prospecting has been car- ried on for more than twenty years to ascertain if ife were possible to find' richer ores or ores of cheaper ac- cessibility. During the flush times several chemical laboratories were in active operation in more than one town and thousands of analyses were made of almost every known deposit. In many cases the samples were taken by interested persons and in many others by per- sons wholly unacquainted with the first principles of sampling ore seams. In the writer's own experience it has happened many times that a single piece of ore, not larger than the fist, would be brought in as representing the seam. In one case of the kind it happened that the ore showed a comparatively small amount of phosphorus t with some 46 per cent, of iron. Whereupon the report was circulated that a large deposit of Bessemer ore had been discovered and for a while speculators were busy. If there be any large deposit of Bessemer ore in the State it has not yet been found. There are places where some of the brown ores show phosphorus below the Bessemer limit, but fifty feet away they are liable to carry from 2 18 GEOLOGICAL SURVEY OF ALABAMA. 0.20 per cent, to 0.50 per cent, of this element. The same observation applies to certain seams of fine grained soft red hematite. Many seams have been carefully sampled and many analyses made in the search for ore that would not show phosphorus above the Bessemer limit, i. e., not over 0.05 per cent, for 50 per cent, of iron. But the conclusion has finally been reached that for the present we shall have to confine ourselves to ores that contain from 0.10 to 0.40 per cent of phosphorus per 50 per cent, of iron, and in many of the brown ores we may expect a considerable increase over these fig- ures. It will not be denied that for a small furnace and with great care in the selection of the ore, the chemist being constantly employed in analyzing for phosphorus, it might be possible to make Bessemer iron in this State from some of the brown ores, but no one could be ad- yised to undertake the project with present lights. The attempt has been made and several thousand tons of iron with less than 0.10 per cent, of phosphorus were pro- duced, but the enterprise languished and has not been revived . The treacherous nature of brown ore with respect to the continuit y of the deposit, is enough to forbid reason- able hope of success. The hematite ores, on the other hand, carry phos- phorus much above the Bessemer limit. They carry generally from 0.30 % to 0.40 % of phosphorus, al- though there is in -the district contiguous to Birming- ham a small seam of red hematite that carries 5.41 % of phosphorus and another 2'.31 %, the metallic iron being about 38 % . In the early* days of iron making in the Birmingham district It was the rule, according to one of the contract- ors, " to mine anything that was red," and what was mined went into the furnace. The difference between IRON MAKING IN ALABAMA : THE ORES. 19 good, bad, and indifferent may have been known, but was not a factor with the contractor or with the fur- nace manager. The. following table page 20 taken from the excellent report of Mr. John Birkinbine on "The Pr:duction of Iron Ores in 1897, U. S. Geological Survey, Division of Mineral Resources," shows the production and valuation of iron ore by states in 1896 and 1897. From this table it will be seen that Alabama ranked third in the production of iron ore. When one considers that Alabama converts practically all of her ore into pig iron she is easily first among the states in the local consumption of her product. The amount of iron made in the state from outside ore is in- significant. Michigan, the largest producer of ore, made in 1897 only 132,578 tons of pig iron, and Min- nesota, the second largest producer made none at all. Alabama is also third in the production of red hema- like ore, Michigan and Minnesota being first and second. Virginia is the first in the production of brown hema- tite, and Alabama second, with Tennessee a close third. No magnetic ore or Jcarbornate ore is mined in the State, although there are considerable deposits of both these varieties. It is of special interest to know that the group of mines on Red Mountain between Grace's Gap and Reeder's Gap, including the Alice, Fossil, Muscoda, Redding, and Ware's was the largest single producer in the United States in 1896 with 945,805 tons. In connection with this table it would be of interest to give one showing the pig iron produced in 1896 and 1897 by states, from the report of Mr. James Swank, manager American Iron and Steel Association, 1897. 20 GEOLOGICAL SURVEY OF ALABAMA. GQ COO1' ICCCMiOt^tOtO^OSt^- CO ^ CO Cij CM CM CM "f lO OS CM CO I CO rH iO CO t- T "*l CO rH t^ rH Tfrl t^ CM CO CO ^ OS H/l O IRON MAKING IN ALABAMA ; TIE ORES. 21 TABLE II. PRODUCTION OF PIG IRON IN 1896 and 1897, BY STATES. TONS OF 2,240 LBS. 1896 1897. Pennsylvania . . 4,024,166 4,631,634 Ohio. . . 1,196.326.. ..1,372,889 Illinois 925,239.. ..1,117,239 Alabama 922,170. . . . 947,831 Virginia 386,277.... 307,610 Tennessee 248,338 272,130 New York 206,075 .... 253,304 Wisconsin 158,484. . . . 103,909 Michigan 149,511 .... 132,578 West Virginia 108,569. . . . 132,907 Maryland! 79,472. . . . 193,702 Kentucky! 70,660. . . . 35,899 New Jersey 59,163 .... 95,696 Colorado 45 V 104. . . . 6,582 -Georgia 15,593 17,092 Missouri 12,548. . . . 23,883 Connecticut 10,187. . . . 8,336 North Carolina 2,151. . . . Massachusetts 1,873.:.. 3,384 Texas . 1,221.... 6,175 Total. 8,623,127. . . .9,652,680 The largest production of pig iron in any one year was in 1897. The principles underlying the valuation of iron ores are but little used in the State, the old system of pur- chasing by the ton still being maintained. The value 22 GEOLOGICAL SURVEY OF~ALABAMA. of an ore is the price at the mine, for, unless the miner also pays the freight, he has already added to the cost of mining all the legitimate costs that should apply to a ton, including royalty. If his contract require that he pay the freight, he cannot reasonably add the freight to the value of the ore, for this varies with the distance it has to be transported. With the exception of some brown ores, which are purchased on the unit basis, but which constitute a small part of the ore used, and some special contracts relating to hematite, the ores in Alabama are bought by the ton without regard to their composition. The price is so much per ton, whether they carry forty, or forty-three, or forty- seven, or fifty per cent, of iron. This system has but little to recommend it, except a mistaken notion of economy in the saving of laboratory expenses and sampling. A close inspection may be kept on the ore as received and daily reports made as to its composition, but unless there is a penalty attached to the shipping of poor ore, there is really no way in which it can be stopped. The price is uniform, no matter what the ore may be. It may be improperly mined, it may contain unusual amounts of water, or clay, or chert, but the price is the same to the furnace. A car load of ore may contain 47 % of iron to-day, to-morrow the ore from the same mine may contain only 43 % , yet the price i^ the same. A brown ore may reach the furnace with its customary 7 % of water, to-morrow it may have 13 % , yet the ore is sold by the ton and the water is counted as ore. There are two main results from this system : First, the contractor is not impelled to furnish ore any better than would be accepted. His sole aim is to avoid disputes with the furnaceman by sending ore that indeed could be better but still will pass muster. There may arise under IRON MAKING IN ALABAMA ; THE ORES. 23 this condition of affairs a tendency towards careless mining, and if the line between acceptable ore and bad ore be an arbitrary one, as is frequently the case, there is a temptation to " put the shot down" a little bit deeper than the line of separation. In the mining of the soft red ores by open cut, the over-burden having been removed, it is practically impossible to distinguish be- tween ore of 46 % iron and ore of 40 % simply by the eye. The chemist alone can decide the question. It is a fortunate circumstance, in the Birmingham district, that for the most part the contractors are fully alive to the advantages of shipping ore that will cause no dis- pute. Under the present system- it is difficult to see how they could ship better ore than they do. But the system itself is wrong in principle. The administration of it may be as fair to the contractor as to the furnace, but this does not do away with the main objection to it, which is, that the same price is paid for ore that is barely usable as for ore that is really good. It cannot be denied that this objection is valid and that until it is removed the true principle underlying the valuation of ores can not be put into practice. The second result from the system of purchasing ore by the ton and not on analysis is that the furnaceman cannot know that his ore to-day is of the same com- position as it was yesterday and will be to-morrow. The purchase of ore on analysis does not necessarily condition regularity of stock, but it is a long step to- wards this most desirable end. It is more than prob- able that under it there would be a tendency towards the higher grades of ore, for these would be more profitable to the contractor than the lower grades. The irregularity in the stock is one of the most serious obstacles with which the Alabama iron master has to contend, especially when he is using Red Mountain ores. 24 GEOLOGICAL SURVEY OF ALABAMA. The most untiring vigilance is demanded in order that the entire make of the furnace shall not be injuriously affected. It is of course the fact that a great dbal of ex- cellent iron has been made in the State without calling into constant requisition the services of a chemist. Bat this is no more than saying that many a case of illness has been cured without the care of a regular physician. We venture the assertion that even under the present insufficient; system a lower cost, account for the making of iron would be shown by the companies employing chemists than by the others. By far the greater amount of iron now made in Alabama is the product of com- panies with well equipped laboratories, and some of the most important sales of iron ever consummated in the State were, to a great degree, brought about by the fact that the laboratory could be depended upon not only for the inspection of the product, but also and particularly for ihe inspection of the stock. Uniformly good iron can not be made at a uniformly low cost with irregular stock, and variations in the cost of the iron are to a considerable extent due to variations in the composition of the raw materials. Pay close at- tention to what goes into the furnace and capping hole will take care of itself. It is a poor policy to fill the furnace with almost anything that may be to hand and trust Providence to look after the cast-house. There is nothing in the nature of the ores used that forbids their sale on analysis, and as this system is al- ready applied to nearly all the flux used, and to a not inconsiderable quantity of coke and ore, the extension of it would not appear to offer insurmountable diffi- culties. The greater part of the cost of making iron is borne by raw materials. The quality of these materials, therefore, and their regularity of composition are of vital importance. As respects composition, there is a 25 point beyond which it is not possible to make iron profit- ably, no matter what the price of the materials may be. How" low this point may be will depend, ceteris paribus, upon the difference between the cost of the iron and its selling price. When this difference is considerable, as was the case in this State ten or fifteen years ago, iron may be made at a profit from very inferior materials. But when the margin of profit is narrow, as has been the case of late years, the use of inferior materials be- comes impossible. With increasing competition and a narrowing selvage of profits, the necessity for using better and better ore becomes more and more pressing. To keep the furnaces in blast and avert disaster from the district, it may happen that the price of ore will fall below the figures at which it can be mined profitably, unless the operations be conducted on a very large scale and long time contracts can be made, assuring a steady output for a number of years. Under such conditions some concessions may be made by the furnacemen in respect to quality, but at the same time they would be warrant- ed in holding out for uniformity of composition. One would be inclined to consider the uniformity of compo- sition as more important than the quality, provided al- ways that this would not entail too much handling of stock per ton of iron made. When ore is sold for stock- house delivery at a fraction over a cent per unit of iron, it would seem that no further reduction in price could be expected. Under all circumstances, except such as embody the sale of the ore at so much per unit of iron, there will be complaint by the furnaceman that the ore is not as good as it might be, and it will be met by the miner with the assertion that it is as good as it can be at the price paid for it. This may, indeed, be true, but at the same time at is not to be hastily concluded that for more money the 26 GEOLOGICAL SURVEY OF ALABAMA. miner is willing to guarantee better ore. For the most part his endeavor is to get the largest possible returns from the smallest possible outlay, a resolution in the highest degree laudable but apt, at times, to cause more or less friction as to shipments. To him a ton of ore is a ton of ore. It weighs 2,240 pounds, and whether it contains fifty per cent, of iron or forty- five he receives the same pay. But to the furnaceman, who has to con- sider the amount of iron he can get from that ton and the ease with which he can do it, the question is of an- other kind. There is a side of the matter not yet touched upon r but which can not be neglected. If the higher grade ore only be mined, the exhaustion of the deposit is cer- tainly set forward. It rarely happens that all of a deposit is high grade ore. and if only the best be in demand one has to cut his cloth to suit the pattern. The miner may have incurred large expense in opening the mine and in equipping it with proper machinery under the expecta- tion that his output would be profitable to him. If he be restricted to a certain portion of the ore and this be below the amount required to yield a profit on the in- vestment, he would be subjected to hardships not toler- able under ordinary conditions. He is quite willing to encourage the belief that it is cheaper to use a large amount of low priced, low grade ore than to pay more for better ore of which not so much is used. In the minds of some whose opinions should be worthy of con- sideration the value of a fifty per cent, ore is propor- tional to the value of a forty-five per cent, ore, and they argue that as the lower grade material can be bought for fifty cents per ton, or 1.11 cents per unit of iron, the better grade material is worth proportionally more, or 55.5 cents per ten. They forget that the value of an ore increases very rapidly as one nears the fifty per cent.. IRON MAKING IN ALABAMA J THE ORES. 27 mark. As a matter of fact, if a forty-five per cent, ore be worth fifty cents, a fifty per cent, ore is worth 83 cents, that is, it will cost as much to make a ton of iron from the one at 50 cents as from the other at 83 cents. Above fifty per cent, the difference becomes even more striking. Attempts at improving the quality of the ores used in the State have been confined so far almost entirely to the brown ores, although it is possible to better the soft red ores to a very considerable extent also. A description of the methods in use will appear under each kind of ore, so that it is merely necessary here to direct atten- tion to the matter in a general way. The ore that most readily lends itself to processes of beneficiation, without any very heavy expense, is the limonite or brown ore. Occurring, as it does, as more or less isolated masses imbedded in clay, it was compar- atively easy to devise machinery that would treat the entire mass of stuff, removing the clay by suspension in water and passing the cleaned ore over screens of ap- propriate sizes. In this matter the clay, unless it was of a very plastic^nature, was removed from tire ore, the wash water being collected in settling dams and again used, after the clay haa been deposited. The process was crude at first and the ore was insufficiently cleansed, but of late years it had been much improved and can now be depended on to furnish fairly good ore from even the more tenacious clays. At some establishments it has been customary to im- prove the brown ores still further by calcining the washed ore in open piles with wood or charcoal ''breeze' ' as fuel, and, later, in gas fired kilns. In this manner the ordinary water is completely removed, and the com- bined water, which does not go off under a full red heat, to an extent depending on the temperature and the dur- 28 GEOLOGICAL SURVEY OF ALABAMA. ation of the firing. Washed brown ore carrying 44 per cent, of iron has been greatly improved by calcining, the iron in the calcined ore being as high as 54 to 56 per cent, over a period of several months. While it is now customary to wash nearly all the brown ore used in the State, but little calcining is done. The reasons for this practice will appear under the discus- sion of the brown ores, and it will be shown that unless the deposit is known to be large or the demands upon it not very exacting as to quantity, the erection of calcin- ing kilns could not be expected to yield much return 011 the investment. For improving the soft red ores several plans have been proposed, but none of them have worked their way into actual use on a large scale, although at least one of them may now be said to have passed the experimental stage. It was proposed to wash the lower grade soft red ores in such a manner as to remove the more ferru- ginous material from the more sandy portion and to re- cover the ore in setting dams. Some experiments were very successful as regards the possibility of concentrat- ing the ore, but the large amount of water required at points where it was expensive to get and the impracti- cability of handling large quantities of damp ore that would certainly fall into the finest powder as soon as it was charged into the furnace have caused the investiga- tion to be postponed. During the last two or three years extensive experi- ments have been made with the hope of concentrating these ores magnetically. Two plans have been propos- ed. First, to render the ore magnetic by raising it to a full red heat in a properly constructed kiln and then passing a reducing gas over it so as to convert the ferric oxide into the magnetic oxide. Subsequent crushing and sizing would bring the ore into a condition in which IRON MAKING IN ALABAMA J THE ORES. 29 it could be treated over a magnetic separator, the sand, etc., being removed by centrifugal action. The other plan for magnetic concentration of these low grade soft ores is to dry them thoroughly, crush and size and pass over a magnetic belt which will pick up the more ferruginous portions and allow the more sandy portions to fall away into suitable receptacles. Both there processes will be described in the chapter on The Concentration of Ores. On the whole, therefore, it may be said that in actual practice the only ores subjected to a process of beneficia- tion on a large scale are the brown ores. Practically all of the pig iron made in Alabama is obtained from native ores. In this respect the situation is quite the reverse of that found in Ohio, which with a pig iron production of 1,463,789 tons in 1895, and 1,196,326 tons in 1896, probably did not derive more than 3 % of it from native ore. The only ores brought into Alabama for any purpose are some brown ore from Georgia, a little " spathite " ore from Tennessee, and Lake ore for use as ' ' fix " in the rolling mills. The production and value of the ore mined in the State, so far as canjnow be ascertained, are given in the following table, compiled from the reports of Mr. John Birkinbine to the United States Geological Survey, Di- vision of Mineral Resources, from the census returns and from independent sources. 30 GEOLOGICAL SURVEY OF ALABAMA. TABLE III. PRODUCTION AND VALUE OF IRON ORES IN ALABAMA AND THE UNITED STATES. ALABAMA. UNITED STATES. Value. Per- cent. Value. Tons. of Pro- Tons. . Per Total. duc- Per Total. Ton tion. ; ion. 1850 1,838 $3.68 $ 6,770 0.12 1,579, 31S $ 4.23! $ 6.98L.679 1860 3,720 5.31 19,765 0.15 2,401,485 5.31 12,757,848 1870 11,350 2.66 30,175 021 5,302,952 5.63 29.843,420 1880 171,189 1.18 201,865 2.3 7,497,509 300 23,156,955 1881 220,000 1.30 286,000 2.4 9,094,369 2.97 27,000,000 1882 250,000 1.20 300.000 2.8 9,000,000 360 ::2,400,000 1883 385,000 1.20 462.000 4.6 8,240,594 3.00 24,750,000 1884| 420,000 1.00 320.000 5.1 8,200,000 2.75 22,550,000 1885 505.000 1.00 505,000 6.6 7.600.000 2.50 19,000,000 188" 650,000 0.96 624.000 6.5 10,000,000! 2 80 28,000.000 1887 675.000 0.96 648,000 6.0 11.300.000 3 00 33,900,000 1888 1,000,000 0.9H 960,000 8.3 12,060,000 2.40 28,944,000 1889 ,570,000 0.96 1,507,200110.9 14,518.041 2.30 33,351,978 1890 ,897,815 1.00 1,897,815 11.8 16,0360431 2.20 35,279,394 1891 ,986.830 1.00 1.986,830 13.6 14,591,178| 210 30,641,473 1892 ,312,071 1.06 2,442,575 14.2 16,296.666 2.04 33,204,896 1898 ,742,410 1.86} 1,490,259 15.0 11.587.629 1.66 19,265,973! 1894 493,086 0.83 1.240,895 12.6 LI, 879,679 1.14 13,577,325 1895 2,199,390 0.80 1,759,512 13.8 15,957,614 1.14 18,191,679 3896 2,041,7931 0.69 1.417,451 12.8 16,005,449 1.42 22,788,069 1897 2.098.621 074 1.546.543111.9 17.518.046 1.08 18.953.221 For a number of years Michigan has held the first place as a producer of iron ore, Minnesota coming up from the 6th place in 1890 to the second place in 1894, 1895 and 1896. It is not likely that Alabama's rank as third in the production of iron ore will be interfered with for some years. She held the second place from 1889 till 1894, when she was surpassed by Minnesota, and Pennsylvania the third place until 1892 when Minnesota came up to the second place. It is not likely that the relative positions IRON MAKING IN ALABAMA ; THE ORES. 31 will be changed for some years. The immensity of the Mesabi ore deposits and the cheapness with which they are mined will, perhaps, keep Minnesota in the second place for the next two years, if indeed she does not push Michigan for first place within that time. Michigan does not produce much pig iron, the output being 132,- 578 tons in 1897. Minnesota made no iron-in 1894, nor in 1895, nor 1896. The difficulty of procuring good coke at that distance from the coal fields has hitherto pre- vented these States from converting their ore into iron, and the tendency seems to be more and more to reduce the cost of these ores to Illinois, Ohio, and Pennsylvania furnaces. But it is a wise man who prophesies concern, ing the iron trade in this day of rapid industrial changes. It would appear, however, that Alabama will have to face competition from furnaces much nearer than Michi- gan and Minnesota. It is just here that questions of transportation play the really vital part. So long as the rich Lake ores can be hauled to Ohio and Pennsylvania furnaces and converted into pig iron which can be sold profitably for half a cent per pound, the situation in Alabama will be one in which the cost of transporting the iron to market after it is made is the main question. With the Northern and Eastern furnaces the great question is the cost of gathering the raw materials into the stockhouse. In Alabama the great question is the cost of marketing the pig iron. With better ore, better coke, and better furnace practice it may be possible even in Alabama to reduce the cost of making iron, but the transportation companies will control the situation then as they do now, unless a closer union can be effected be- tween the two interests. 32 GEOLOGICAL SURVEY OF ALABAMA. According to the Iron Trade Re view, Cleveland, Ohio r the Lake shipments of iron ore in 1892, were 8,545,313 tons ; in 1893, 5,836,749 tons ; in 1894, 7,621,620 tons ; in 1895, 10,234,910 tons; in 1896, 9,916,035 tons, and in 1897, 12,457,002 tons. These figures mean that consid- erably more than half of the total amount of iron ore mined in the United States is transported by water to the vicinity of the furnaces using it. Were it not for this fact the enormous development that has been reached in the Lake regions, with respect to the mining of iron ore, could not have been attained within so short a time, if at all. In order to exhibit the relation that Alabama sustains to the other iron ore producing states, in respect to the value of the ore mined, the following table taken from the reports of Mr. John Birkinbine to the U. S. Geo- logical Survey, Division of Mineral Resources, is ap- pended. IRON MAKING IN ALABAMA ; THE ORES. 33 oo ; 10^ rH^ CM lO^ lO ^ C^ CO^ CO t^ ^^ lO T-^ CM CO CO^ "H i~^ O^ C^ ^H CO CO d lO CO GO IT-- C^l T~H *"* CM !> ~^< tO !> OO GO O5 CD '"^ '"^ t~OO5'-HI>-OS(M^CM(MI>-CDCO Sbs -.2 ;l IllllllillllllllS 34 GEOLOGICAL SURVEY OP ALABAMA. TABLE IV Continued. Total Valuation and Average Value Per Ton of Iron Ore Produced in the United States in 1889, 1892, 1892, 8893, 1894, 1896, 1897. . . - . : . ' d v< o O5 GO 3 "si || rn 05 05 i-H O O t- CM OO CO'-HOCGOOiOt-t- f~iO^OOOCCOt-Or-05C5.-lr- l t-OCCOCM OSMCMOT iCMr-lO- "T-I-HT IT ti lOOl ir-i S r- 1 CM CM co" 00 jn & Idaho, e. In- d'gNevada& Wy. j 1 Tt^ O* ^ O> CD C^ *~ t^- - OC I>" CO !> 1C O r- lt~-TTiOCOOCDa5CCaGCMOTfrlOOOt CD LO CD CD CM CO O5 t" ^^ CM ^^ * ^ O5 CO "^ CO 1O^ t 1 '"' ' COO Tfi CD O0'^ |t ~'o5t^ r-T 00* -^ Z.-. me. Valuation. f-t d O5 CO CM CC 3 I C7i> OC CO CM CL? C^ CO CO CO ^^ lO C^l CD CM egon. d. Includ'g Oreg ud'g Vermont. j.Inclu s- 33 -4-s o H i^isis^Ioo^^is^iisS 05 o oo~ oo- r-" f co o"cc co'co'cb^cM CD"GO O'CM'O CM"CO O*^^ ^ to i i '" ^ i ' CN! ib t- O5 .TJH c^jO^ 9^ 1894. Valuation. l| OD ^ a O 7M CM C tt0JRi9S^^8^8 T 1 3'8 Jto* ! . a i o'?> * T5 3 1C i ' ' OC C5 C5 IO CM lQ OC CD' CD CM O5 O I 1C OC CO iO ^ I s - t^-CDtOi l^fCDOCD CD O5 CD "* OO T 1 t^ CM CM CD r i CC T ' r-i iO CO CD CM OO CO i*- STATES. ..... \ ; : : : '. - i H a Including Maine b. Includ'g Oregon Washington & I clud'g Nevada, f. Includ'g Delaware, g. Includ'g Wyoming '; ; : : : : : : j ; i : : : ~ Alabama Colorado . Conn, and Mass Georgia and N. Car. . . . Kentucky Maryland MiVhifrnn Minnesota Missouri Mont.. N. Mex., and Utt New Jersey NAW Vnvlc 9 ^ a 2| ^ a OP- i a rj T T l a <;. a E- Texas Virginia Wisctn in IRON MAtflNk IN ALABAMA ; THE HEMATITES . 35 CHAPTER II. THE ORES : SPECIAL DISCUSSION; THE HEMATITES. In the discussion of the hematite ores we shall have to exclude the brown hematites as they properly belong, to the limonites, although often mis-called by the former name. The limonites are locally termed "brown ores" and the output is about 25 per cent, of the total ore pro- duction of the State. They will be discussed under their proper heading. The hematite ores are, for convenience, classed under two heads : First, the soft red ores, carrying but little lime and Second, the " hard red " ores carrying from 12 to 20 per cent, of lime and in many cases self-fluxing, that is, they carry enough lime to flux the silica contained in them . In order that a clear understanding of the matter may be had at the outset the following brief description of the geological and topographical feature of the deposit of hematite ores so largely used in the State is given here. They belong to the Clinton formation of the Silurian, which extends with some breaks, from the middle por- tion of Alabama to the northern part of Maine. They are overlaid by chert, sandstones, and clays, the over- burden at places reaching a depth of forty and fifty feet. The seams now worked vary in thickness from 3 to 25 feet, run in a north-east directipn and dip towards the south-east at angles varying f:om 15 to 22 degrees, the dip increasing as one goes towards the south-west! For 36 GEOLOGICAL SURVEY OF ALABAMA, the most part they occupy the crests of the hills, the outcrop forming a striking and persistent feature of the landscape for several miles in the vicinity of Bir- mingham. The Soft Ores. As a rule, to which, however, there are some import- ant exceptions, the outcrop is " soft red," a term of com- parative significance only as the ore]is quite firm and has to be won by regular blasting operations. It is soft as compared with the limey or "hard red" ore. The soft ore may extend from the outcrop for a distance of 300 feet on the dip, depending on the thickness and imper- viousness of the cover, although the hard ore conies to the surface at more than one place. In winning the soft red, the overburden is removed and the ore mined, at day, by benches. Under cover the ore becomes limey and hard and is mined from in- clines on the dip by drifts and slopes. The soft ore is the hard ore with the lime removed by atmospheric influences and is richer in iron the poorer it is in lime, When the overbruden is stripped off there is found a seam of ore quite soft and seemingly disin- tegrated, of a deep red or purple color, the so-called " gouge. 5> It may be only a few inches thick but often runs to 24 and even 36 inches, and comprises generally the best part of the ore. Underneath this begins the more solid ore diminishing in content of iron according to the vertical depth. The best quality of " gouge" will carry 52 per cent, of iron while ten feet below its line of demarcation the iron falls to about 46 per cent. Be- tween the " gouge " and the ore .proper there is often a, thin seam of yellowish clay, which, however, is by no means constant in strike. In the more solid ore, be- IRON MAKING IN ALABAMA ; THE HEMATITES 3? neath the " gouge," there are seams of the same clay, sometimes as much as two inches thick but for the most part not above half an inch thick. In the early days of iron making in the Birmingham district it was the cus- tom to mine 15 to 20 feet of the soft ore and to send the whole material to the furnace. Of late years, however, the mining has been restricted to ten feet, including the 'V gouge " as it was found that below this depth the ore rapidly became siliceous and unfit for use. Taking the -content of metallic iron in the "gouge" at fifty per cent, as mined, the loss in iron according to vertical depth, is about one-half of one per cent, per foot. This would bring the iron in the first ten feet of the seam to forty-five per cent, and in the next ten feet to forty per cent. A large number of analyses extending over sev- eral years show that when the mining is limited to the ten foot mark the iron content is a little over 47 per cent, in the ore as mined, i. e., with seven per cent, of water, and including the " gouge." The rapid increase of the silica in the ore below the ten foot mark is shown by the fact that to get even 47 per cent, of iron in the upper ten feet from one-fifth to one-third of it must be composed of the "gouge," with its 50 per cent, of iron. The following successions of materials has been ob- served at East No. 2 mine on Red Mt. south of Grace's Gap, and about 4 miles from Birmingham. The over- burden here was 24i feet thick, and was removed before .any of the ore was mined. In places the overburden is not so heavy, and in other places it is heavier. There' is no general rule in regard to the thickness of the over burden or its nature. Within two miles of this locality towards the southwest the overburden is a great deal thicker, and of a different character. 38 GEOLOGICAL SURVEY OF ALABAMA. SECTION I. VERTICAL SECTION AT EAST NO. -2 MINE, -RED MT. SOFT RED ORE. Ft. In. Soil and red clay 6 Sandstone 3 Clay 1 Sandstone 1 Clay 2 Ore 6 Clay 2 Ore 31 Clay 1 Ore 4 lay 4 Ore 4 Clay ."..' i Ore 1 1 Clay ..., 2 Ore...; 10 Clay ,.... 1 Ore 2i Clay ......... i Ore.... i increase in the iron 44 27.00 | the insoluble matter 45 25.00 I falls 2 per cent. 46 23.00 J 47 .22.00") 48 20.50 49 19.00 For each 1 per cent. 50 17.50 (^ increase in the iron 51 16.00 f the insoluble matter 52 14.50 falls 1.50 per cent. 53 13.00 54 1 46 GEOLOGICAL SURVEY OF ALABAMA. It is not necessary to carry the list further, as the sup- ply of fifty-four per cent, soft red ore is limited. It is not claimed that this ratio is absolutely correct, but a large number of analyses substantiate its reliability for ordinary purposes. The ratio from 40 iron through 46 iron is as 1 :2. Beginning with iron 47 and insoluble 22, the ratio appears to be nearer 1 :1.50 than 1 :2, for with iron 48 the insoluble matter- is about 20.50. It may, therefore, be said with a fair degree of accuracy that a soft red ore carrying 40 per cent, of iron may be expected to contain 35 per cent, one with 45 per cent, of iron 25 per cent, and one with 50 per cent, of iron 17.50 per cent, of insolute matter. There are, of course exceptions to this rule and it does some times occur that an ore with 46 per cent, of iron will be found to carry 22 per cent, and one with 48 per cent, of iron will'have 21 or 22 per cent, of insoluble matter. But on the whole the 1 fact remains that an ore with 45 per cent, of iron will carry 25 percent, of insoluble, and one with 50 per cent, of iron from 17 to 18 per cent., and the lisft may be used as an approximation to the truth. In texture, the soft red ore is a mass of minute silic- eous pebbles held in a ferruginous cement. The pebbles are seldom larger than a No. 4 shot, and are frequently much smaller. They are all more or less rounded and stained reddish-brown. The cementing material is softer tban the pebbles, and on sizing even a very lean ore the material passing a screen of fifty meshes per linear inch is much richer in iron than the material remaining on a 10 or a 20, mesh screen. A soft red ore of 40 per cent, iron, on being ground to pass a ten mesh screen, will yield through a fifty mesh 53 per cent, of iron, and the amount passing the 50 mesh screen is from 25 to 30 per cent, of the ore, by weight. So far as concerns their physical structure, this is one IJEION MAKING' IN ALABAMA ; THE HEMATITES. 47 of the points of differentiation between the soft red and the so-called brown ores, for these, on being sized, show-- a steady loss of iron the finer the screen. The fact of increasing richness in iron the finer the screen renders the concentration of the low grade soft red < ores much simpler than would otherwise be the case, as the "fines' 7 can be briquetted without further treatment, and the troublesome question of handling them becomes com- paratively easy . The rounded form of the more siliceous pebbles also occasions less wear on the shutes, screens, and conveyors ; a p^oint of no little moment in ^concentra- ting works. The better grades of the soft red ore do n ; ot occur at every point on Red Mountain, nor is it possible to mine even ten feet profitably everywhere along the ridge. It is frequently the case that the inferior ore sets in, as the saying is, "at the grass roots," and even the richer * 'gouge" is sometimes absent. Mining operations can not be undertaken without careful prospecting and many analyses , for the difference between a fairly good ore and one that is not passable is often so slight as to deceive" even the most experienced man who grades merely by the eye. After having become accustomed to a partic- ular kind of ore, one may judge of its quality by the ap- pearance with a reasonable degree of accuracy: While for the most part the soft ores are of the same general texture and color, it not infrequently happens that serious mistakes may be made unless the services of a chemist are called into requisition . When freshly mined the ore is of a deep red color, inclining to purplish' red in the richer portions, but on drying there is assumed something.of a brownish tint. For ordinary stockhouse delivery the ore contains on the average 7 per- cent, of hygroscopic water, which, owing to the coarse-grained nature, soon dries out under cover. 48 * GEOLOGICAL SURVEY OF ALABAMA. In the early days of iron making in the Birmingham district, before the real value of the limy or hard ores was generally accepted, the furnace burden was com- posed almost entirely of the soft ores. Of late years, however, the tendency is decidedly towards a greater and greater proportion of the limy ore, the proportion rising at times to above 90 per cent, of the ore burden. It is still to some extent a mooted question as to the relative reducibility of the two ores, but a careful investigation of the subject would, we think, show that in this respect the limy ore has the advantage. When the soft ore descends into the zone of reduction in the furnace, it does so without losing its firmness of texture. Even after it has become red hot. or white hot, it maintains its shape, except as this may be changed by friction during the descent. The reducing gases act upon it in the lump, and if the lumps be of considerable size the reduction to metallic iron may be delayed and the ore may appear before the tuyeres. The case is quite otherwise with the limy ore. The lime is present as carbonate, (except such as may be combined with the phosphorus as phosphate of lime, an amount rarely exceeding 0.50% ,) and when this reaches a point in the furnace at which its carbonic acid begins to come off, the ore begins to fall to pieces . The friction of the other materials aids this tendency quite as much as, and perhaps, more than in the case of the soft ore. The reducing gases can and do have a greater ore sur- face to work on and the result is that for a given weight of coke and a given composition of the gas there is greater reducing action. The soft ore is more fusible than the limy ore, but this does not necessarily mean that it is more easily penetrated by the reducing gases within the furnace. On the contrary a fused crust on IRON MAKING IN ALABAMA ; THE HEMATITES. 49 the outside of a piece of soft ore interposes considerable opposition to the passage of the gases, and as this crust becomes thicker and thicker the gases penetrate with more and more difficulty. In the case of the limy ore as soon as it begins to part with its carbonic acid it be- gins to disintegrate, and this very fact of disintegration enables it to receive to better advantage the reducing power of the gases. In comparing the two ores another circumstance must not be lost sight of, and that is the intimate comming- ling of the ore and the lime that is to flux it. This is a distinguishing characteristic of the lime ores. It would be impracticable to effect by artificial means such an in- timate mixture of ore and lime as Nature has already provided in these o*res. This circumstance is of the greatest importance in any discussion of the relative value of the soft and the lime ores, for while these latter require a higher heat for fusion they are not therefore to be considered less easily reducible. The reducibility of an ore depends far more upon its permeability or porosity than upon its fusing point. For the most part the loss of energy in a furnace is chargea- ble to lack of reducing power rather than to lack of fus- ing power. The tendency now is more and more towards the use of the limy ores; for the enormous demand that has been made on the better quality of the soft ore within the im- mediate vicinity of Birmingham has begun to make itself felt. Three courses of action may be open : First, the in- creasing proportion of limy ore in the burden may in- duce the furnacemen to look towards the use of eighty or ninety per cent, of it, the difference being made up with soft and brown ore. Second, other sources of soft ore may be utilized. Third, the lower grades of the soft 4 50 GEOLOGICAL SURVEY OF ALABAMA. ore, now remaining in the ground, may be concentrated and made to take the place of the ore that has been re- moved. It is not thought that the proportion of brown ore used will be materially increased. Under the existing conditions it would appear advisa- ble to begin at once to increase the proportion of limy ore used, so as to establish on the basis of wider experi- ence the economic relation that this burden would sus- tain to former practice, or to push the work of concen- trating the lower grades of soft ore to some definite re- sult. The experiments on concentrating soft ore, to which some allusion has already been made, showed the possi- bility of taking an ore of 40% iron and 35% silica and .bringing the ore to 57% and the silica to 15%, on the average. In this process two tons of raw ore were re- quired to make one ton of concentrates. The matter is fully discussed in Chapter VII on the Concentration of Low-grade Ores. The Limy, or so-called Hard Ore. The ore sets in sometimes at the outcrop but much more frequently it is found only under cover and is the continuation of the soft ore in the direction of the dip. -For distances varying from nothing to 300 feet on the dip the ore is soft, then the hard ore begins and con- tinues to depths not yet ascertained but certainly very considerable. In other words, as has been already stat- ed, the hard ore, which originally appeared at the sur- face, has been deprived of its carbonic acid by atmos- pheric influences and converted into soft ore along the dip to varying depths, the lime having been removed by .leaching. Relatively the same differences that are to be observed in the soft ore from various places are also found in the hard ores. There are points along the IRON MAKING IN ALABAMA ; INTRODUCTION. 51 mountain where the minable seam of soft ore is better than at others, and there are places where the hard ore is better than at others. On a vertical section of the soft ore the content in iron decreases downward, the rate b^ing about one-half of one per cent, per foot. The rale holds good for the hard ore on a vertical section. The mining on the big seam of soft ore is now confined for the most part to the upper ten feet, the mining on the hard ore is also the same, and below the ten-foot mark the hard ore also be- comes too siliceous for economic use. The hard ore de- rives its value from two circumstances, first there is a great deal more of it than of the soft ore, because it ex- tends to very considerable depths, and second because of the intimate admixture of carbonate of lime with the ferruginous material. The best hard ore carries more lime than is required to flax its silica, while in the ordi- nary grades the ratio of one of silica to one of lime is generally conserved. When this is the case the ore is termed "self fluxing" and in burdening a furnace ex- clusively with hard ore of this type it is not necessary to add limestone to flux the ore. When the burden is com- posed of hard and soft ore, or of hard and brown, or of hard, soft, and brown the amount of limestone to be added is calculated from the silica of the ore other than hard, the silica of the fuel and of the stone itself. The increase in the use of hard ore would tend to diminish the consump- tion of limestone by an amount represented by the lime- stone in the ore and if a strictly self-fluxing ore were used the consumption of limestone would be greatly dimin- ished. There is a kind of hard ore, termed semi-hard, which contains from one-third to one-half of the lime in typical hard ore, but of this sort very. little is used, and it is not mined regularly. Within the last three years the use of crushed hard ore 52 GEOLOGICAL SURVEY OF ALABAMA. has become quite common in the Birmingham district^ The soft ore does not lend itself readily to crushing un- less thoroughly dry. With the amount of water it usu- ally contains it becomes somewhat like clay in the crush- er, i. e. more or less gummy, and the machine soon be- comes choked. A general average of the hard ore used shows : PER CENT. Water 0.50 Metallic Iron . 37.00 Silica .13.44 Lime 16.20 Alumina 3.18 Phosphorus . . . 37 Sulphur 0.07 Carbonic acid 12.24 Adding the alumina and the silica together we have for silica plus alumina 16.62%, the lime is 16.20%, and the ore may be termed self-fluxing. It cannot be said that all of the hard ore used is self-fluxing, <*s some of it contains 5% more of lime than of silica plus alumina. Taking a general average, however, of analyses of all kinds of hard ore extending over several years this ore carries enough lime to flux the silica plus alumina. It may be urged that aluminous soft ore needs silica as a flux for the alumina, and this is indeed true. But we have to flux the silicate of alumina with lime, and it is merely a question as to whether all the bases of the burden shall be calculated as lime, and all the acids as silica, or whether we shall regard the silica plus alumina as requiring so much lime. In either case the type of slag to be made has to be considered, and for any one type the two calculations lead to the same result so far as concerns the consumption of limestone per ton of iron. IRON MAKING IN ALABAMA ; THE HEMATITES. 53 The question has been raised as to whether the hard ore, on the dip, may not gradually lose its content of iron and become GEOLOGICAL SURVEY OF ALABAMA. ore of far less content of lime and therefore requiring the addition of flux. At the first glance it would appear that it is better to avail ones self of whatever advan- tages Nature herself has conferred upon us in the way of an ore carrying its own lime. But the matter can not be settled out of hand and without careful investiga- tion of all the data bearing upon it. From the stand- point of the furnace man, if he could depend on secur- ing self-fluxing ore regularly, the matter resolves itself into the simple consideration as to whether he can make as much iron and as cheap iron in the one way as in the other. He may, indeed, go a step farther and ask if he make iron more cheaply in the one way than in the other. Having settled this, he has no further concern with the matter. If he can make iron more cheaply by iising a greater and greater proportion of hard ore than oy using an ore which requires the addition of extrane- ous flux, it is his duty to do it. This, however, is a one- sided view. There are other investments in the State that must be regarded as well as investments in furnaces. How is it with the contractor for ore and flux? Would Ms business be hindered by the substitution of hard ore for stone? Tf his profit on the ore were the same as his profit on the stone, no great hardship would follow the increase in the use of the one and the decrease in the -use of the other. But if it should happen that his profit in mining stone were greater than his profit in mining liard ore, and there should be such an increase in the consumption of hard or3 as to destroy the value of his stone quarry, he would not be apt to appreciate the ad- vantages of the change. In this respect this iron dis- trict differs from any other in the country, and the rela- tions of stone to ore burden vary perhaps more widely than elsewhere. The ability of the furnaces to dimin- ish at will the consumption of limestone, places them in THE FLUXES. a very independent position. If the price of stone be too high, they can run on increased proportions of hard ore. If they succeed in obtaining the stone at reasona- ble cost, they take off hard ore and put on soft or brown. For instance, a certain coke furnace during a certain month in 1895 made about 5,000 tons of iron with an ore burden composed of 50.9 per cent, hard, and 49.1 per cent, soft ore. The total burden was as follows : Hard ore 27.7 per cent. Soft ore 26.7 Limestone 15.5 " Coke 30.1 100.00 " The consumption per ton of iron was : Ore 2.36 tons (2240 Ibs.) Stone : 0.67 " Coke 1.32 4.54 And the cost per ton of iron was, for raw materials : Ore $1.32 Stone 0.34 Coke 1.83 $3.49 The consumption of coke per pound of iron made was 1.32 Ibs., and practically all of the iron was of foundry grades. Shortly before, the same furnace was running on 33.4 per cent, hard, 65.3 per cent, soft, and 1.3 per cent, brown ore. The total burden was : 68 GEOLOGICAL SURVEY of ALABAMA. Hard ore 17.0 per cent- Soft ore 33.1 ' Brown ore 0.6 " Limestone.. 16.9 " Coke 32.4 '< 100.00 The consumption per ton of iron, of which something" over 4,600 tons were made, was, in tons of 2,240 Ibs. : Ore 2.20 Limestone .0.73 Coke 1.41 4.34 The cost per ton of iron was, for raw materials : Ore $1.26 Stone 0.43 Coke... . .'. 1.83 $3.52 The consumption of coke per pound of iron was 1.41 Ibs., and in this case also practically all of the iron made was of foundry grades. In these two cases there was a saving of nine cents per ton of iron by increasing the proportion of hard ore and lessening the amount of limestone added. The ore cost six cents a ton of iron more than when the ^larger proportion of soft ore was used, so that the net gain was three cents per ton of iron,. $3.49 for the hard ore burden, ^and $3.52 for the other. But with the lesser amount of hard ore the furnace made 358 tons of iron more than with the greater amount. This has to be set to the credit of the soft ore burden. Perhaps no positive conclusions can be drawn from THE FLUXES. 69 one or two instances, and as the whole matter will be fully discussed under Furnace Burdens, it may be best to defer any further remarks. Enough, however, has been said in this chapter on the fluxes to direct attention to the importance of the considerations advanced. The future of the iron industry in the State depends not on any one circumstance or con- dition, howsoever vital it may seem, but upon the result- ant of a number of forces, some of whose effects may be .at the present but dimly foreseen. It is possible that the relation between hard ore and limestone, or dolomite, is one of these. Mr. C. A. Meissner w-is the first furnace manager in the Birmingham district to make use of dolomite regu- larly and systematically. While manager of the Van- derbilt furnace he began to prospect for workable de- posits of dolomite, and succeeded in locating and opening the quarries now belonging to the Jefferson Mining & Quarrying Co., about 2 miles from North Birmingham. This was the first quarry of dolomite opened and the iirst shipment- were made in 1890 to the Sloss Iron and Steel Company. This quarry is still in active operation, and yields ex- cellent; stone. All of the output is taken by the Sloss Iron & Steel Company. Following the successful operation of this quarry J. W. Worthington & Co. opened the Dolcito dolomite quarry along the same deposit towards the North-east. The Dolcito quarry was opened in July 1895, the first shipments being made about August 1st. to the Tennessee Coal , Iron & Railroad Co. This quarry furnished 425,000 tons of stone to the close of 1897. and has a fine equip- ment, power drills, wire-rope transmission Irom the face to the crusher^&c. It has a daily capacity of 500 tons of crushed stone and 500 tons of lump stone. The aver- 70 GEOLOGICAL SURVEY OF ALABAMA. age analysis of the Dolcito dolomite has already been given. From its North Birmingham quarry the Sloss Iron & Steel Company is now obtaining dolomite that averages less than 0.50% of silica. After Mr. Meissner had shown that good dolomite could be obtained within the immediate vicinity of Birmingham and in almost any quantity. Mr. E. A. Uehling, manager of the Sloss Iron & Steel Company, took the matter up. In an article written for the Alabama Industrial & Scientific Society (see Proc. Vol. iv, 1894, p. 24) Mr. Uehling described at some length the nature of this stone, and compared its value with that of the ordinary limestone of the district. This paper was published in full in the first edition,, ana from it is taken the following : "In determining the value of a stone as a flux, it is not only necessary to deduct the impurities it contains r but in addition to that, as much of the base as is neces- sary to flux these impurities. What remains only can be considered as available flux, and has value id the blast furnace. To get at the available flux, we must de- duct 2 per cent, from the carbonate of lime for each unit per cent, impurity in the stone. Taking the limestone at 96 per cent, of carbonate of lime and deducting from this 8 per cent, to take care of its own impurities, we have left for available flux 88 per cent, of carbonate of lime. "As the average dolomite contains only 2 per cent, of impurities and 43 per cent, of carbonate of magnesia with 55 per cent, of carbonate of lime, we will have, after deducting 4 per cent, from the carbonate of lime,. 51 per cent, of this material, and 43 per cent, of carbon- ate of magnesia. THE FLUXES. ' 71 Reducing the carbonate of magnesia to its equivalent in fluxing power of carbonate of lime, we have, because, the fluxing powers of the two carbonates are to each other as 84 to 100, 43x100 x51=102.19. 84 The relatiue values of the two available fluxing ma- terials of the district are, therefore, to each other as 88 is to 102.19. That means that 88 tons of dolomite will do as much work in the Hast furnace as 102.19 tons of limestone. Put into dollars and cents, this means that if dolomite c;m be bought for 69 cents a ton, limestone is worth only 52 cents a ton ; or if limestone costs 60 cents, dolomite is worth 69.5 cents a ton. There Js only one valid objection that can be brought- up against the use of dolomite as a flux in the blast fur- liases, and that is that magnesium has less affinity for sulphur than calcium has, and dolomite is therefore less efficient as :i d< sulphurizer than limestone, to the extent that caustic limp is displaced by magnesia. This objfvt, however, becomes quite insignificant where thu- ores are free from sulphur, as is the ca-e in the Birmingham district. When a considerable propor- tion of hard ore is used in the mixture, its lime, in con- nection with what is contained in the dolomite itself, is ample to take care of the sulphur contained in the coke. One-quarter to one-half dolomite has been regularly used in the Sloss furnaces for nearly two years, and, at intervals, as high as three-fourths have been put on with the best results. The ore mixture being half hard and half Irondale (soft) at the city furnaces, and from one- fourth to one-third brown with, generally, equal propor- 72 GEOLOGICAL SURVEY OF ALABAMA. tions of Irondale (soft) and liar d at the North Binning* ham furnaces. The coke used contained considerably above the aver- age amount of sulphur found in the coke of the district. The iron was of as good quality as could have been produced with all limestone as a flux, and the furnaces have worked more regularly than they did prior to the use of dolomite. The assertion that the use of dolomite has a tendency to make light colored iron is not sustain- ed by fact. Some of the most celebrated foundry irons are made with all dolomite as a flux.- The writer had used it for years, while in charge of the blast furnaces of the Bethlehem Iron Co rnp my. prior to coming down here, and experienced no difficulty in keeping the sul- phur within the required limits, even with ores contain- ing as high as 1.5 per cent, of that element. The Illinois Steel Co. are also using dolomite ex- clusively in their Joliet Works. They are doing very good work, and have no trouble with the sulphur what- ever. The deficiency of dolomite to carry off sulphur is probably very much exagerated. There are impure dolomites as well as impure limestones ; but when of good quality and used intelligently and without preju- dice, it always gives good satisfaction. In addition to its superior fluxing power there is decidedly less ten- dency to 'hanging' with dolomite than with carbonate of lime." We can not agree with Mr.Uehling that dolomite is a less efficient desulphurizer than limestone. Experience here with all kinds of burdens in the manufacture of basic iron, in which 't was required that the maximum sulphur should be 0.050%, has shown the contrary. When limestone was used exclusively it was with diffi- culty that the specifications as to sulphur were met, and THE FLUXES. 73 the percentage of casts with maximum sulphur 0.050% was very much less than when dolomite was used ex- clusively. These conclusions are based on the analysis of some 1500 casts. Speaking with reference to the man- ufacture of low-silicon, low-sulphur iron if any one thing was abundantly proved it was that limestone failed to give any thing like such good results as dolo- mite, not only with respect to silicon, but also and es- pecially with respect to sulphur. This whole matter was carefully worked over by the writer in an article on "The Manufacture of Basic Iron in Alabama," pub- lished in The Mineral Industry, Vol. V. 1896. It may be regarded as practically settled that as a de- sulphurizer in the blast furnace dolomite is quite as efficient as limestone for ordinary grades of iron, and much more efficient for basic iron requiring unusually low-sulphur. With r-espect'to the effect of dolomite on the silicon of the silvery and the soft irons we are not prepared to make a positive statement at this time. By some the low silicon that has characterized these irons during the last two years has been attributed to the prevailing use of dolomite. And yet some furnaces that do not use dolomite at all are troubled in the same way. The low silicon in the 'hot' irons may be due in part, at least, to the increasing amount of limy ore that is 'being used. A very basic slag requires a very high heat for perfect fusion, and it makes no great difference whether the lime is in the ore, or is added in the shape of limestone. The more basic the slag the greater the heat required to melt it, and the more pronounced the tendency towards exceeding the point at which the sili- con enters the iron. 1 here is a point at which silicon fails to combine with iron because the temperature is not sufficient for the re- 74 GELOGICAL SURVEY OF ALARAMA. duction of silica. May there not also be a point at which silicon fails to enter the iron because ( a) the tem- peratuie is too high, or (b) the slag is too basic? If silica is deoxidized the resulting metalloid alloys with iron, and in the measure in which the deoxidation goes on in the same measure will high-silicon irons be pro- duced. Ten years ago when there was less limy ore used than is the case now, there was no special difficulty in making silvery and soft irons. The difficulty was in keeping the silicon down, for Mr. Kenneth Robertson informs us. (Trans. Amer. Inst. Min. Engrs., Vol. XVII, 1888- 18S9, P. 94 et seq.) that No. 1 Foundry iron carried 3.6o% of silicon, while No. 1 Mill, which was also called No. 3 Foundry, carried 2.tt7%. Unfortunately Mr, Robertson does not give the analysis of the irons according to the burden on which they were made. He does say that 54.81 % of the total make was foundry- iron, but analyses are what- is needed for a discussion of this kind, not calculations as to the proportion of foun- dry grades made, for the grades included under this, classification are not the same now as they were then. It would require a great deal of labor to look over the records of those days with a view to .ascertaining the ef- fect of the burden on the silicon content of the iron, if indeed the investigation would lead to anv definite in- formation, for chemical analyses were not then carried on with this purpose. There has nofc been very much improvement in this respect of more recent years, and even today chemists here are not expected to exa nine the furnace records. Still, enough information has been gathered to warrant one in saying that the tendency of limy ore burdens is towards decrease of silicon. The subject is referred to here because it is not a mat- THE FLUXES. 75 ter of indifference as to whether the flux shall go in with the ore or be added as limestone. In the light of the experience of the last two years it begins to look as if furnace managers would do well to examine into the effect of dolomite on the content of sil- icon, and to cultivate the laboratory more systematically. A chemist who is made to feel that he has nothing to do with the burdening of the furnace soon restricts himself to the merest routine work, and regards the questions of more lime or less lime, more hard ore or less hard ore, limestone or dolomite with an indifference born of re- peated rebuffs. In the chapter on Furnace Burdens there is given a blank form which has proved to be extremely useful. It may, of course, be modified to suit any emergencies. Properly filled out with additional information as to the amount and heat of the blast, silicon content of the irons &c., it would enable the chemist to be of far greater value to the furnace than he can ever be if regarded merely as an analyst whose business begins and ends with the grinding out of a certain number of results every day. If a chemist is worth anything at all he is worth trusting. If he can not be trusted with all kinds of information as to the working of the furnace he should not be trusted to make analyses, and unless he can know what goes into the furnace his knowledge of what comes out is of no use to him, and but little to any one else. 76 GEOLOGICAL SURVEY OF ALABAMA. CHAPTER IV. FUEL. The fuel used in the blast furnaces of the State is coke and charcoal. There are no known seams of coal that could be used without coking, as is done in Ohio in this country, and in Scotland, particularly, abroad. Coke. There is, perhaps, no subject connected with the iron business that gives rise to more discussion than that of coke. There are so many different kinds made, and so great diversity among them in respect of chemical and physical properties, that it is almost a hopeless task to attempt to set the matter forward in a manner satis- factory to all concerned. In this State, which produces about 10 per cent, of the coke made in the United States, there is a very considerable difference in quality between the various grades of this fuel. This chapter is not a treatise on coke, nor is it neces- sary to enter upon the subject beyond what is required to explain the situation. Three kinds of coke are made here, from lump coal, run of mines, and washed slack, and each of these three may be 48 hr. or 72 hr. coke. Regarded in this way, and excluding mixtures, of which there may be endless variety, we have six different kinds to-wit : 48 hour 72 hour Lump, Lump, Run of mines, Run of mines, Washed slack, Washed slack. FUELS. 77 The ordinary practice is to use 48 hr. coke, and per- haps 90 per cent, of the coke is of this kind. The chief difference between the 48 hr. coke and the 72 hr. coke is in the strength, or the ability to resist abrasion and crushing, the latter having somewhat the advantage in this respect. The following table gives the results of some experi- ments undertaken to establish the crushing strain of some of the principal cokes used for blast furnace and foundry purposes. The table given in the first edition of this book gave tests on coke made here in 1891-92. Since that time there have been many improvements, and it has been thought best to substitute a new table for the old one. The later results represent the pres- ent composition of the cokes, and in addition the com- position of the ash. It is much to be regretted that all the Alabama cpkes are not represented, but it has been impossible to secure the proper samples. Enough, how- ever, is given to show the quality of some of the chief varieties of this fuel now used in the State. 78 GEOLOGICAL SURVEY OF ALABAMA. > *1 '^JA'B.IQ CD OC r O5 rM lO -M ! CC O T T CC tO CO -f 1C OC GO OC GO C5 CT, OC' OC ocodddd d rH W CO 1 CO I 30 lO CD CD Ow ~H CO CC "^ ^^ C^ ^ C5OO *O - O5O' O5|O 05 o q < p O~HtHr-Ji-!i-iO'-i'-iO|!-H FUELS. 7^ TABLE VII. 72 HR. BEE-HIVE COKE, MADE FROM WASHED PRATT SLACK. iu o cc 5 a -g a> _> ;' c ' OQ t^ O p^j u 5 ^~* ^^ jJ '^ rO " 4 No. Is d> 1? S:S "& (B i^ C x" 1 3 A cs ^ ** o> "Q <^2 S > ?'5 ^ *s 'p o. y 'O - >i . *^J & r^ o *3 i' 02 5*1 s>S P* "*" a Q DQ S. ]8 1.153 1.848 3760 32.60 550 13.10 0.90 J9 1.131 1.881 3980 35.30 575 13.20 0.95 20 1.057 1.821 41.98 39.72 537 8.70 131 21 1.013 1.855 4540 44.78 , 575 1130 0.92 22 0.701 1.810 4H.4L 48.00 725 9.10 1.10 23 1.217 t.861 3463 2728 550 6.50 0.98 24 0.971 1.891 4863 50.06 675 8.80 1.05 25 1.155 1.890 38.60 33.40 450 7.40 1.15 26 0.967 1.810 4654 48.11 700 7.70 1.00 27 1.071 1.813 4094 38.22 600 9.40 1.04 28 0862 1.850 53.30 61.80 431 8.50 0.90 29 0.840 1-.765 5420 67.60 420 9.00 094 30 0910 1.805 50.25 5530 460 9.25 0,97 Aver. 1.003 1.838 44.48 | 4%. 77 558 9.34 1.02 48 h: . Disintegrated Pratt Nut. Not Washed. 31 32 0866 0.812 ' 1.667* 1.359 48.00 4055 55.34 325 49/62 , 400 11.55 14.02 1.30 1.40 Aver. 0.839 1.513 4.27 52.48 1 362 12. 7H 1 1.35 Tl hr. Disintegrated Pratt Nut. Not Washed. 33 34 1.355 1.351 2593 2542 47.77 4700 3425 5310 550 587 1050 10.30 1.20 1.25 Aver. 1 358 2.567 47.38 53.67 568 10.40 1.22 GEOLOGICAL SURVEY OF ALABAMA. TABLE VII Continued. 48 hr. Washed and Disintegrated Pratt Slack. i ^ en . *-4 O OQ- 1 11 -t. Oi CX-*" S 2 5 "^ "s p,ca S-^ "^ 6 a 35 0996 1.850 46.10 46.20 650 10.10 0.98 36 0.861 ) 839 56.20 69.40 400 1030 1.00 37 1.000 1805 4440 44.10 575 1030 1.03 38 0.862 1.695 49 12 56.96 700 10.70 1.06 39 0.828 1.818 46.00 45.50 500 9.00 0.96 40 1.100 1850 40.30 36.60 675 1120 1 10 41 0.920 1.630 4420 48.80 625 9.90 1.00 Aver. 0.938 1.784 | 46 62 49.65 589 10 12 ) 1.02 72 hr. Washed and Disintegrated Pratt Slack. 42 43 1.330 0.956 1.850 1.822 38.20 47.60 41.00 50.00 750 750 9 30 9.25 1.04 1.02 Aver. 1.143 1.836 4290 45.50 750 9.27 1.03 Black Creek 48 hr. 44 I 0.900 1.84 46.20 52.00 4( 400 ! 3.90 Milldale (Standard C. & C. Co.,) 72 hr. Jefferson C. & C. Co., Lewisburg. 0.79 45 0.961 1.88 47.00 52.50 545 7.60 0.80 46 i 0.84 1.764 52.46 6254 531 10.20 0.68 Gas Carbon. 47 1.25 2.10 40.50 43.00 600 5.90 1.23; FUELS. 81 The analyses here given show that these cokes fall naturally into two main groups, characterized by the porosity (cell space) and the size of the cells. With regard to this principle of classification we have coke in which the percentage of cells by volume is just above 50, and coke in which the percentage of cells by volume ie just above 40. To the first group belongs the Blue Creek coke, and to the second the Pratt coke. The figures given are in each case averaged from a number of determinations on separate prices, not less than 5. r and in most cases 10. For 48 hr. Blue Creek coke made from washed slack the averages are : Apparent specific gravity 0.85& True " 1.764 Per cent, of cells by volume 52.18 Volume of cells in 100 parts by weight 61.59 Compressive strain 474 Ibs. Ash 11.05 Sulphur 0.94 For 48 hr. Pratt coke made from washed slack the averages are : Apparent specific gravity 1.046 True " " 1.839 Per cent, of cells by volume 42.96 Volume of cells in 100 parts by weight 41.49 Compressive strain 464 Ibs. Ash 9.16 Sulphur ; 0.95 There is a marked difference in these two cokes, the one exhibiting a large cell, and the other a small cell,, while in strength they are about equal. For determin- ing the specific gravities, and the cell space the method 6 82 GEOLOGICAL SURVEY OF ALABAMA. first proposed by Dr. T. S terry Hunt in 1863, and modi- fied by Dr. F. P. Dewey was used with certain changes. Instead of using the air pump, which is indeed is not necessary, the samples were boiled in water for 16 hrs. and allowed to stand 16 hrs. in water before weighing. For the compressive strain one inch cubes were accu- rately cut from sound pieces of coke with a hack-saw, and a Star blade, using a miter-box. This was very tedious, and some .of the cubes required as many as 6 blades. Coke is very destructive to steel saws. Per- haps the best tool would be one of the diamond wheels used in preparing specimens of rocks for the micros- cope. In each case at least 3 cubes were cut, and care was taken to have them free of cracks and pieces of slate. They were crushed in a standard Riehle Testing Ma- chine, operated by hand, and reading to 3,000 Ibs. It was observed that now and then some cubes of 72 hr. and 96 hr. coke withstood more than 3,000 Ibs. stress, but this does not often occur, and is immaterial as a coke testing 3,000 Ibs. is certainly strong enough. It must be understood in all discussions of the physi- cal qualities of coke that great differences may be found in samples from the same oven, and indeed in samples from the same part of the oven. Too much importance should not, therefore, be laid upon such investigations, for one may very easily be mislead, and draw entirely erroneous conclusions. In connection with chemical analyses physical tests may be relied upon, in compar- ing one coke with another, to give fairly accurate data, but they should be accepted only if based upon a long series of determinations in which the conditions of manu- facture are positively known. The size of the coal coked , the amount of water it holds, the rapidity of the coking process and its duration, the amount of water used in FUELS. 83 quenching, and whether inside or outside watering is used are some of the factors to be considered. Within certain limits the chemical composition of the coal ap- pears to be of less influence upon the physical qualities of the coke than the factors just mentioned. The 72 hr. bee-hive coke made from washed Pratt slack does not differ materially from the 48 hr. except as to its strength, giving 558 Ibs. as against 464 Ibs. It is especially adapted for foundry purposes, the in- crease of strength being of greater benefit here than in the blast furnace. In respect of strength the 48 hr. un- washed, but disintegrated Pratt nut is much inferior to the 72 hr. The disintegration of Pratt nut coal, un- washed, and subsequent coking, whether for 48 hrs. or 72 hrs. appear to yield a coke of about the same percent- age of cells by volume as the 48 hr. and 72 hr. washed slack, but the volume of the cells is much larger, viz., as 53 to 41. The strength of the unwas-hed disintegrated Pratt nut of 48- hrs. is inferior to that of the 48 hr. washed slack, while that of the 72 hr. unwashed, disin- tegrated nut is somewhat above the strength of the 72 hr. washed slack. In other words, disintegrating the unwashed nut coal gave a coke of about the same per- centage of cells by volume, "and increased the size of the cells, but failed to better the coke with respect to crush- ing strain. Washed and disintegrated Pratt slack, whether coked for 48 hrs. or 72 hrs,, makes a fine coke in every respect. In order to compare these cokes with standard Pennsyl- vania and Virginia cokes we append results obtained by Mr. John Fulton, and given in his excellent Treatise on Coke. The average standard Connellsville coke shows : True specific gravity , 1.77 84 GEOLOGICAL SUKVAY OF ALABAMA. Per cent, of cells by volume 45 .8T Volume of cells in 100 parts by weight 54.13 Compressive strain . 279 Ibs.. Ash 10.58 Sulphur .81 Two cokes from Big Stone Gap, Va., showed on the average : True specific gravity 1 .64 Per cent, of cells by volume 44.78 Volume of cells in 100 parts by weight 55 .22 Compressive strain 285 Ibs^ Ash 5.61 Sulphur 0.87 Pocahontas coke gave : Apparent specific gravity 1 .83 Per cent, of cells by volume 52.07 Volume of cells in 100 parts by weight ........ 47.93 Compressive strain , . . . . 236 Ibs. Ash 5.88 Sulphur 0.73 Mr. Fulton gives also the results from an examination of Blocton coke. Ala., as follows : True specific gravity 1.75 Per cent, of cells by volume 49.97 Volume of cells in 100 parts by weight 50.03 Compressive strain 409 Ibs. Ash 6.94 Sulphur 0.74 The writer found the average of two samples of Bloc- ton coke, 48 hr. bee-hive : FUELS. 85 True specific gravity 1 .65 Per cent, of cells by volume 44.46 "Volume of cells in 100 parts by weight 45.98 Oompressive strain 737 Ibs. Ash 5.80 --Sulphur 1.35 And a sample of Pocahontas (Stonega) coke gave : True specific gravity 1.84 Per cent, of cells by volume , 53 .83 Volume of cells in 100 parts by weight 63 01 Oompressive strain 588 Ibs. Ash , 6.50 Sulphur r. 0.75 Coke made at Earlington, Kentucky, by the Sc. Ber- nard Coal Company, gave for 48 hr. bee-hive : True specific gravity 1.69 Per cent, of cells by volume 53.47 Volume of cells in 100 parts by weight. 67.67 Compressive strain 275 Hb Ash 14.60 Sulphur 1 .74 Not to protract this matter further, although many more tests could be given, we close with a 72 hr, bee hive coke, made at Brookside, Ala., by the Sloss Steel & Iron Company, of washed slack : True specific gravity 1.87 Per cent, of cells by volume 55.00 Volume of cells in 100 parts by weight. . 65.20 Compressive strain 320 Ibs. Ash 10. Sulphur 1.25 86 GEOLOGICAL SURVEY OF ALABAMA. Some investigations have been made as to the effect of carbonic acid on red hob coke, but other and more press ing work prevented their completion. Some high au- thorities, among them Sir I. Lowthian Bell, have recom- mended that the action of carbonic acid on red hot coke be included among the determining factors in the valua- tion of coke for blast furnace purposes. But the writer is not disposed to think that such data are of much, if any, importance. The gases in a blast furnace are mix- tures of carbonic acid, carbonic oxide, hydrogen, oxygen, and nitrogen, with aqueous vapor also. It is not known what effect, if any, is produced by carbonic acid in the presence of these other gases. It is not as if carbonic acid were the only gas that would or could act upon the coke, for as a matter of fact it is always accompanied by other gases in greater or less quantities. Doubtless the dissolving action of carbonic acid upon red hot coke is an important phenomenon, and one well worthy of study, but until it is known whether the other gases exert a neutral, an accelerating or a deterring effect upon this dissolving tendency it does not appear that much prac- tical information is gained . The matter of zone reactions in the furnace also complicates the question, as well as that of the occlusion of gases in coke. The average composition of the cokes used in the State is as follows : Coke from Run of Mines Coal. PER CENT. Moisture 0.75 Volatile and combustible matter 0.75 Fixed carbon 84.50 Ash 14.00 100.00 Sulphur 0.901.60 per cent. FUELS. 87 Coke from Washed Slack. Moisture 0.75 Volatile and combustible matter 0.75 Fixed carbon 88.50 Ash. 10.00 100.00 Sulphur 0.801.10 per cent. Coke from Lump Coal. Moisture : 0.75 Volatile and combustible matter 0.75 Fixed carbon 87.00 Ash 11.50 100.00 Sulphur 1.001.30 per cent. In chemical composition there does not seem to be any material difference between the 48 hr. and the 72 hr. coke. The composition of the ash of the various cokes in use may be given as follows : Run of Mines. Silica 47.03 Ferric oxide 12 46 Alumina 33.62 Lime 1.53 Magnesia 1 *>9 Sulphur 0.15 88 GEOLOGICAL SURVEY OF ALABAMA. Washed Slack. Silica . 45.10 Ferric oxide 12.32 Alumina 31.60 Li me 1 .50 Magnesia Trace. Sulphur 0.14 Lvmp. Silica 46.00 Ferric oxide 12.00 Alumina.. 32.00 Lime 1.00 Magnesia .50 Sulphur 0.16 It would be interesting to know if the amount of ash and its composition influenced the strength of the coke, or whether the treatment of the coal, prior to charging the ovens, and the duration and temperature of the pro- cess should alone be looked to in explanation of this point. It does not seem probable that the amount of ash or its composition, per se, would influence the strength of the coke as much as the distribution of the ash constitu- ents in the coal. Thai is, if the coal was finely pulverized before charg- ing there would be a more equable distribution of the ash-constituents with consequent uniformity of composi- tion in the coke. But uniformity of composition, how- ever desirable, does not necessarily imply increase in strength. Granting that there would be increase in strength is this effect beneficial when the coke is already strong enough? If the coke made from any coal, with- out pulverizing, were already strong enough, the only advantage in pulverizing would be in the greater uni- formity of composition. But some coals do not yield FUELS. 89 strong coke unless they are pulverized. Whether this is due to the irregularity of the distribution of the ash, or the bituminous matter, or the relation between the Coking and the non-coking constituents of the coal, is not known. When, however, such coals are pulverized they often make excellent coke. The composition of the ash of coke, by affecting its fusibility, may affect also its strength, the size and shape of the cells and the thickness of the cell walls. But of such matters very little is known. It requires a great deal of time to make such investi- gations, as well as skill and perseverance. The composition of the ash of coal, whatever effect it may have on the quality of the coke made from it, cer- -tainly has an important bearing on furnace practice. It must influence the fusibility of the burden, and to a greater or Usser degree affect the consumption of lime- stone, whether this be the carbonate of lime in the hard ore, or t-xtra stone. The more acid the ash the more base is required for fluxing. The amount of coke used per ton of iron varies, of course, with the nature of the coke, and of the other constituents of the burden ; with the kind of iron made, the shape and size of the furnace, the rate of driving, and other circumstances grouped generally under the term ''furnace practice." The range is from 1.16 to .1.72 tons of 2240 pounds. From an rxamination of 150,000 tons of iron made from 1890 to 1895 under vary- ing conditions the lowest consumption for a period of one month was 1.16 tons per ton of iron. In this par- ticular case the furnace was working on all brown ore, the burden being composed of brown ore 52.9, limestone 20.4, and coke 26.7. The tons of iron made per charge was 1.53 tons, number of charges 1802, total iron made 2766 tons, of which 99.1 per cent, was of foundry grades. 90 GEOLOGICAL SURVEY OF ALABAMA. The consumption of materials per ton of iron made was ore 2.31 tons, stone 0.89 ton, and coke 1.16. The particular case in which 1.72 tons of coke were- used per ton of iron made was when a furnace was run- ning on the following mixture, stated as percentages, hard ore 53.7, soft ore 34.2, brown ore 12.1. The entire burden was composed as follows, in percentages, hard ore 28.5, soft ore 18.2, brown ore 6.3, limestone 10.6, coke 36.4. The iron made per charge was 1-88 tons, number of charges 1819, total iron made 3418 tons, of which 92 per cent, was of foundry grades. The con- sumption of material in tons per ton of iron was as fol- lows : Ore 2.51 Stone 0.49 Coke .1.71 The average consumption of coke per, ton of iron may be taken ai 1.41 tons of 2240 pounds. This would mean that for producing the 835,851 tons of coke iron in 1895- there were used 1 ,179,375 tons of coke and that 250,000' tons of coke made in the State during that year were- diverted to same other purpose. The average for the best coke made in the State may- be taken at 1.30 tons of 2240 pounds for a ton of iron of 2240 pounds. A pound of iron has been made in the- State with less than a pound of coke, but for a very lim- ited period. This matter will be taken up more fully in the chap- ter on Furnace Burdens, as tables have been prepared based on more than 83,000 charges and an iron produc- tion of nearly 150,000 tons over a period of several years. There has been a notable decrease in the consumption? FUELS. 91 of coke per ton of iron since the introduction of coke made from washed slack coal. It is much superior to ordinary coke both in structure and composition, and might be still further improved by pulverizing the coal before charging the oven, as in this way a better distri- bution of the ash is rendered possible as well as a stronger coke. No constituent of the burden responds as readily to variations, in furnace practice as coke. It forms gener- ally more than a third of the burdon, and always more than half of the total cost of the materials entering into a ton of iron is chargeable to coke. It is not only the most costly single ingredient, it is more costly than the ore and the stone taken together. Economy in the use of coke is, therefore, the most important economy that can be set on foot and carried out in connection with the manufacture of pig iron in this State. Better ore and better stone are needed if there is to be no better coke. To improve the ore and the stone is to increase the yield of iron per charge, and to decrease the consumption of the most costly material entering the furnace, i. e. coke. The following table gives a bird's eye view of the coke industry in Alabama from 1880 to the close.of 1897, and is compiled from the reports of Joseph D. Weeks to the United States Geological Survey, Division Mineral Re- sources, with additional statistics for 1896 and 1897. The greatly lamented death of Mr. Weeks, on the 26th of December, 1896, removed from the industrial world one of the best statisticians, and one whose contributions to the manufacture of coke were always especially recog- nized and appreciated. GEOLOGICAL SURVEY OF ALABAMA TABLE VIII. COKE OVENS IN ALABAMA. 00 00 "O c c Ovens. 1 OJ 3 Value of Coke. .s si 1 ^ 2 | pj g z> Built. Building 13 ^ isl 93 o | 0) 'fl 5 i O H o^ r* -=3 ' Q K*^ . P"! I 1880 4 316 100 106,283 60,781 57 $ f83,063 $3.01 1881 4 416 120 184,881 109,033 59 326,819 3.00 1882 - 536 261,839 152,940 58 425,940 2 79 1883 6 767 122 :>59,699 217,531 60 598^,473 2^75 1884 8 976 242 413,184 244,009 60 609,185 2.50 1885 11 1,075 16 507,934 301,180 59 755,645 2.50 1886 14 1,301 1,012 635,120 375,054 59 993,302 2.65 1887 15 1,555 1,32 550.047 325,020 59 775.090 2.39 1888 18 2,475| 406 848,608 508,511 60 1,189,5791 2.34 1889 19 3,9441 427 1,746,277 1,030,510! 59 2,372,417 2.30 1890 2C 4,805 j 371 1,809,964 1,072,942 59 2,589.447 2.41 1891 21 5,068! 50 2.144,277 1,282,496 60 2,986,242 2.33 1892|20 5,320 90 2,585,966 1,501,571 58 3,464,623 2.31 1893 23 5,548| 60 2.015,398 1.168,08o 58 2,64K,632 2.27 1894 22 5,551 50 1,574,245 923,817 58.7 1,871,348 2 25 1895 22 5,658 50 2,459,465 1,444.339 ! 58. 7 3,033,521 2.10 1896 24 5,363 1,769,^20 1,038,707^58.7 2.181.284 2.10 1897)25 5,365 120 2,451,475 1,443,01715.8.8 3,094,461 2.14 The average value of the coal used in making coke in 1895 was 87-J- cents per ton; in 1896, 79 <>-lO cents; and in 1897, 88i cents. There were no new ovens built in 1896 or 1897.' Mr. Jas. D. Hillhouse, State Mine Inspector, makes the production of coke in 1896, 1,689,307 tons, and the number of ovens 4,494. As of interest in connection with coke and coking operations, there is given here an article, by the author, on coking in a Bee-hive oven, published in the Engi- neering and Mining Journal, N. Y., and also part of a report made to the Sloss I. & S. Co. on the use of Pratt washed slack coal in the Otto Hoffman oven, published in the American Manufacturer and Iron World. FUELS. 93 COKING IN A BEE-HIVE OVEN. (The Engineering and Mining Journal, Vol. LXIV, Nos. 25 and 26, and Vol. LXV, No. 3.) It has for several years been of interest to me to ob- serve the progressive changes that took place in a bee- hive oven from the moment of charging 'the coal to the withdrawal of the coke. The opportunity of observing and noting these changes from hour to hour was pre- sented lately, and gladly accepted, and for nearly 48 hours the oven was closely watched. The observations were taken in person. The coal used was washed slack, from the Pratt seam. The oven was of the usual bee-hive type, of 12 feet diameter, the spring of the arch beginning at 26 in. from the floor. The door was 2 ft. wide and 3 ft. high. The trunnel head was 14 in. deep and 14 in. in diameter. The weight of washed slack charged was 11,575 Ibs., but as it contained 5% of moisture the drv weight was / ^ o 11,024 Ibs. The oven was charged at 11 :50 a. m., and, after leveling, the top of the coal was 4 ft. below the bottom of the trunnel head. The door was bricked up at once. A charge of coke had been drawn from the oven during the morning, so that it was hot. Within a few minutes after charging there was an odor of light hydro- carbons from the door and from the trunnel head, and in 20 minutes, after charging, this odor became quite per- ceptible. For the first two hours there was no flame, but the evolution of a grayish-black smoke became more and more intense. At 2 :30 p. m-, 2 hours and 40 min- utes after charging, -the first flame appeared and burned with a decided reddish tinge until 3 :30, or one hour, when it became yellowish. For the next two hours the flame from the trunnel head was yellowish and smoky. On top of the coal the flame was yellowish, streaked 94 GEOLOGICAL SURVEY OF ALABAMA. with grayish- black bands of smoke, which seemed to lie rather closely to the coal. By six o'clock, six hours after charging and 3-J- hours after the first ignition, the flame from the trunnel head was 4 ft. high and of a de- cided yellowish color. At seven o'clock, 4i hours after ignition, the oven was perceptibly hotter, the flame was burning fiercely, and there were wisps of blackish-gray smoke in the oven. There were but few signs of fritting, although the smoke in the oven might have obscured them had they been present. Shortly after seven o'clock I was unfortunately called away and could not return for two hours, o there were no observations until at 10 o'clock, 7i hours after ignition ; the flame had then loat its distinctive yellowish cast and was decidedly whitish. It was still 4 ft. out of the trunnel-head and the oven was much hotter. The top of the coal was fritted, cracks of considerable size had appeared : there was not much smoke in the oven, but white flames were issuing from the cracks and burning in a flickering, lambent manner. There was no perceptible swelling up of the coal, but on top it was uneven and jagged. The cracks did not seem to lie in any special direction, nor to be of any uniform size or depth. The play of the flames from the cracks was most beautiful. None of them burned steadily, al- though none went out. There was no appearance of "blows" of gas or any sudden outburst at any spot. Now and then a white flame would seem to be sucked back into the depths of a crack and to vanish, but at no time did any of them go out entirely. There were no wisps of smoke in the oven. The flames seemed to burn with about the same intensity and there was a remark- able uniformity in their height and general appearance. Nine hours after ignition. The flame from the trunnel head was still from 3 to 4 ft. high, but had not changed much in appearance, being still decidedly whitish ; it FUELS. 9 5 was thinner than before. Inside the oven the cracks in the coal were wider and deeper and the coal was much more broken and jagged. In several places, noticeably beneath the trunnel head, the coal had sunk, and there were crater-like depressions, from which flickering white flames issued and had a slightly bluish tinge. The oven was much hotter than at the last observation. Bright white flimes burned in jets over the surface of the coal, the so-called "candles" of the coke burner. They were distributed irregularly over the surface of the coal, burned intermittently, died down and came up again from the same place, or close by. About 12 inches of the coal from the top seemed to be burning, as the door was hot for this depth, but cool below. Ten hours after ignition. No apparent change beyond the further development of cracks in the coal, and its further subsidence. The oven was hotter. Eleven hours after ignition. No apparem change except that the oven was much hotter, approaching a white heat. The bluish tinge of the flame inside was entirely gone. There was' no specially noticeable change at the 12th and 13th hours after ignition, but at the 14th hour the oven was of a clear white heat, the inside flames were thin and white, and the flames from the trunnel head had begun to drop. The cracks in the coal were larger and more numerous. The coal had burned down to the 24-inch mark on the door. Fifteen hours after ignition. Flames from the trunnel head much thinner, burning fiercely and swiftly in a somewhat streaked fashion. Within the oven the heat was very intense , the cracks in the coal were larger and white flames of a slightly bluish tinge played irregularly over the surface. At the 16th, 17th, 18th, 19th and 20th hours after ig~ 96 GEOLOGICAL SURVEY OF ALABAMA. nition there was not much apparent change ; but at the 21st hour the flame from the trunnel head was much thinner than at the 15th hour, and had receded much more. By the 22d hour the flame was decidedly thinner than at the 21st hour, and from this until the 28th hour it gradually became thinner and thinner, and burned swiftly with a striated appearance. Inside the oven the cracks were still developing, and white flames played over the top of the mass. The heat was now well along toward the bottom of the oven. Thirty-fourth hour after ignition. . There were no special changes in the flame from the 28th to the 34th hour, except that it became thinner all the while, and at the 34th hour was just out of the trunnel head. From this time to the 40th hour the flame gradually drew back into the oven, until it could no longer be seen. But when the oven was opened for drawing, at the end of the 46th hour, there were thin jets of bluish white flame now and then on tof of the coke. The door of the oven was taken down at the end of the 46th hour after igni- tion, and the coke watered inside the oven for 18 min- utes. The oven was drawn by two men in one hour. The yield of coke over a fork of 14 tines, 21 inches wide, with spaces H inches in the clear, was 5,875.80 Ibs., or 58.78% of the weight of the dry coal. The weight of the dry breeze through the fork was 322 Ibs., or 5.13% of the weight of the coke over the fork. The proximate analysis of the coal used was, on a dry basis : Volatile and combustible matter, 32.43 % ; fixed carbon, 60.91 % ; ash, 6.66%. The sulphur was 1.91%. The composi- tion of the coke over the fork was, on dry basis : Vola- tile and combustible matter, 1.51%; fixed carbon, 88.90% ; ash, 9.59%. The sulphur was 1.37%. The composition of the breeze and ashes passing the fork was, on dry basis: Volatile and combustible matter, FUELS. 97 1.47% ; fixed carbon, 56% ; ash, 42.53%. The sulphur was 1.14 per cent. The composition of the black ends of the coke, the so-called " black-jack," was on a dry basis: Volatile and combustible matter, 1.82 per cent ; fixed carbon, 89 per cent; ash, 9.18 per cent. The sulphur, 1.29 per cent. By screening the breeze and ashes over a 1-inch screen there was recovered 25 Ibs., or 8 per cent of material that had the following composition, on a dry basis : Volatile and combustible matter, 1.25 per cent; fixed carbon, 88.40 per cent; ash, 10.35 per cent ; sulphur y 1.30 per cent ; while the 297 Ibs., or 92 per cent, passing the 1-inch screen was of the following composition on & dry basis: Volatile and combustible matter, 1.25 per cent ; fixed carbon, 61.40 per cent ; ash, 37.35 per cent; sulphur, 0.85 per cent. Passing the breeze and ashes over a i-inch screen gave 35 per cent over and 65 per cent through. The material over the i-inch screen gave, on dry basis: Volatile and combustible matter, 1.20 per cent; fixed carbon, 80.80 per cent; ash, 18 per cent; sulphur, 1 percent ; while the material 'passing the i-incli screen gave, on dry basis : Volatile and combustible matter, 0.80 per cent ; fixed carbon, 51.90 per cent ; ash, 47.30 per cent ; sulphur, 0.80 per cent. It is usual in the Birmingham district to fork coke over a l|-inch opening, and the amount of breeze and ashes left is often a considerable item. It depends to a great extent upon the coal itself, but also upon the skill of the coke-drawer, the manner in which the oven is watered having a great deal to do with it. Coke made of washed coal gives much less breeze than the same coal unwashed, the difference at times rising to 50 per cent in favor of the washed coal. Irrespective of the difference in the quality of the coke made from un- 7 98 GEOLOGICAL SURVEY OF ALABAMA. washed and from washed coal, which of course is the most important matter, the difference in the yield of furnace coke, as between the two, is well worth consid- ering. A second oven was charged with a similar coal on the same day, and was operated for 96-hour coke. Weight of dry coal charged, 11,024 Ibs., the coal containing 5 per cent of moisture. The yield of dry coke over a li-inch fork was 6,350 Ibs., or 57.51 per cent of the dry coal, or 54.86 per cent of the coal as charged. Time of watering, 20 minutes; time of drawing, one man, 1 hour 57 minutes weight of breeze and ashes, dry, 240 Ibs., or 2.17 per cent of the dry coal charged. The analysis on dry basis was : Breeze and Coal. Coke. ashes. Vol. and combust, matter. 32.46 1.06 2.68 Fixed carbon . . 60 . 86 89 . 63 69 . 79 Ash fiS fl 31 97 *2 Sulohur. . 100.00 100.00 100.00 1.89 1.34 1.23 Over a 1-inch screen there was recovered from the breeze and ashes 14 Ibs. (=5.8 per cent) of material of the following composition, dry : Volatile and combusti- ble matter, 1.56; fixed carbon, 8655; ash, 11,89. The sulphur was 1.20 per cent. The material passing the 1-inch screen was not analyzed . A third oven was charged on the same day with a similar coal, and operated for 72-hour coke. Time of watering, 17 minutes; time of drawing, two men, 55 minutes; weight of dry coal charged, 11,024 Ibs., or with 5 per cent moisture, coal charged, 11,575 Ibs.: weight of dry coke, 6,590 Ibs., over a H-inch fork, or 59.7 per cent by weight of the dry coal and 56.93 per cent of the coal as charged ; weight of breeze and ashes, 285 Ibs. diy, or 2.58 per cent of the weight of the dry FUELS. 001 99 coal charged and 4.33 per cent of the weight of the coke over a li-inch fork . The analysis was as follows : Vol. and combust, matter. Fixed carbon Coal. . 82.55 . . . 60 64 Coke. 1.71 88.35 Breeze and ashes. 1.09 79.97 Ash R -1 9 94 18 94 r:'j '../<>'!'*"'>& f O^O.'J t*1 Sulnhur . 100.00 1.93 100.00 1.31 100.00 1.21 V / The composition of the black ends of the coke was : Volatile and combustible matter, 2.26 ; fixed carbon, 86.52 ; ash, 11.22. The sulphur was 1.28 per cent. Screening the breeze and ashes over a 1-inch screen gave 34 Ibs. (11.9 per cent) of material of the following composition, dry : Volatile and combustible matter, 0.80 ; fixed car'- on 87.64 ; ash 11.56 ; sulphur was 1.28 per cent. The material passing the 1-inch screen was of the follow- ing composition, dry : Volatile and combustible matter, 1.00; fixed carbon, 69.90; ash, 29.10; sulphur, 1,10 per cent. The coal used in these three ovens was the same, washed slack, and was of practically the same composi- tion. Each buggy of coal was sampled as it was dis- charging into the oven. In the following table, which embodies the results, the composition of the coal is the average of the three analyses, and all the calculations are based on dry ma- terial : TABLE IX. SHOWING CHEMICAL CHANGES FROM COAL TO COKE. PROXIMATE ANALYSES. ^ "3 pi I* ' ,1 _~ X2 Su a 6 I 8 ** !. Vol and c- mtte | Sulphur. Viela oi dr iu coke. *1 2 [m reaseof from coal t< 1? Lecrese o ma ter fro t< >ke ueciease < ],hur Iron to coke Per Per Per Per Per Per Per Per I'er Per cent cent cent. cent. cent ' ceut cent. rent. Ctllt cent- Coal. 32.48 60.80 '6.72 1.91 48 hr coke 72-hr, coke 1.51 1.71 88.901 --9.50 88.35 9.94 1.37 1.31 58 78 59.77 2 92 2.58 46 21 J2..71 45.31 48.51 95.35 91.73 28.27 31.41 % nr- <-<>k 1.06 89.631 W.31 .34 57.51 2.17 47.41 3^.54 96.73 29.S4 100 GEOLOGICAL SURVEY OF ALABAMA. The average yield of dry coke over a H-inch fork, from dry coal, was 58.69 per cent. The average increase of the fixed carbon was 46.31 per cent and of the ash 43.25 per cent. The average decrease of the volatile matter was 95.94 per cent, and of the sulphur 29.84 per cent. As a further contribution to this study, I give the ulti- mate analyses of the coal and of the coke, averaged dry basis : TABLE X. ULTIMATE ANALYSES OF GOAL AND COKE. Coal. Dense coke. "Needle" coke. Carbon 78.23 84.55 97.55 Hydrogen '.. 4.51 1.33 1.12 Oxygen $.98 4.33 1. Nitrogen 1.56 0.18 0.00 Ash.. 6.72 9.61 0.10 10C.OO 100.00 10000 Sulphur 1.90 1.31 0.27 The analysis of the needle coke will be commented' upon later. By comparing the proximate composition of the coal and of the coke with the ultimate composition several very interesting things are observable. What is termed " fixed carbon " in the proximate analysis of coal is a very different thing from the carbon obtained on com- bustion, being in the one case 60.80 % .and in the other 78.23 % . In the proximate analysis the fixed carbon is the difference between the sum of the volatile matter and the ash and 100, on a dry basis. If the volatile matter is 32.48, and the ash 6.72, the fixed carbon is 100 - (32.48 plus 6.72) .equals 60.80. But in driving. FUELS. 101 off the volatile matter, even iri a covered platinum cru- cible enclosed within another covered crucible, there is a serious loss of carbon because the volatile matter itself is largely composed of gaseous hydrocarbons together with more or less solid carbon going off in the smoke. ^The soot is not pure carbon, but contains some hydro- carbon compounds whose nature varies according to cir- cumstances, such as the rapidity of the heating, the dur- ation of the heating and the nature of the coal itself. But the question at once arises. Can any of these vola- tile hydrocarbons, reckoned as such in the ordinary proximate analysis, be used in the coke oven, during the coking process, as a source of carbon ? The answer to this depends upon the nature of tiie hydrocarbons, the temperature of the oven and the thickness of the bed of coke over the still burning coal. It is well known that certain hydrocarbon gases evolved from coal at a comparatively low temperature are decom- posed at a higher temperature with deposition of carbon ; for example, olefiant gas, C 2 H 4 , and acetylene, C 2 H 2f this latter gas, indeed, decomposing under certain con- ditions, at ordinary temperatures. But olefiant gas and acetylene do not occur in the destructive distillation of coal beyond a few tenths of 1 per cent., as shown by Dr. Fyfe several years ago in the Journal of Gaslighting. and Ebelmen found that after being in the oven 7i hours coal gave only 1.667 per cent, of carburetted hydrogen in the gases collected . It is possible that reactions going on within the mass of burning coal and the mass of red-hot coke are of such a nature as to allow some of the hydro- carbons evolved to deposit carbon ; but it is almost im- possible to calculate just how much of this deposited carbon there is in any one oven of coke. The very bright silvery needles and blades of coke found on bee-hive -coke are composed of almost pure carbon, the combustion. 102 GEOLOGICAL SURVEY OF ALABAMA. giving 97.55 per cent. But these blades and needles form an insignificant proportion of the coke, and a very thin coating of this silvery deposited carbon serves to improve the appearance of the coke. Deposited carbon may and probably does increase the yield of the coke, but to a very slight extent, and appears to enhance the appearance of the coke without adding much to its weight. Some time ago I had an opportunity of secur- ing some very fine specimens of deposited carbon from a bee-hive oven. Some large lumps of limestone were thrown in on top of a charge to make lime. When the coke was ready to water these lumps were taken out be- fore the water touched them. I examined them closely and found that in the cracks of the lower lumps, and indeed upon the surface of some of the smaller pieces,, which, however, may have come from the larger lumps, there were sheets of almost pure carbon, the analysis giving about 98 per cent. The sheets were as thick as ordinary letter paper, and were somewhat flexible. There were countless little globules of bright, silvery carbon scattered all over the sheets, and under a i inch objective these globules were seen to be covered with a network of fine lines, running hither and thither. On illuminating these globules with a focussing glass in bright sunlight, they presented a most beautiful appear, ance under the microscope, resembling great globes of silver floatrag in blackness. I have never seen a more beautiful sight than they exhibited. It is a curious circumstance that the appearance pre- sented by these globules under the microscope closely resembles that given by botryoidal limonite. There is the same network of fine lines, dividing the surface into many irregular shaped patches. Hair coke, the so-called " whiskers " of the coke burner, is also composed of almost pure carbon, and under a 1-12 inch objective are FUELS. 10B often found to be covered with little globules, adhering to the sides of the ' hair " and looking like pearls strung on a silver wire. Percy (Metallurgy, Fuel, Etc., pp. 421-422) speaks of the hair-like form of coke and gives an explan- ation of its origin, through deposition of carbon in tha inner surface of carbon tubes blown out by escaping gas. The hairs are sometimes completely filled with car- bon, but at other times are hollow, as I have myself observed. A study of this hair-coke by a competent microscopist would certainly be interesting. Now and then the hairs are covered with little curved projections, while again they resemble a thread partially untwisted so that the separate strands are visible. Occasionally they are pierced through by minute holes, a high magni- fying power showing several holes in lines across the hair. I have amused myself mounting many specimens of coke, deposited carbon, hair-coke, etc., for the microscope and in observing their peculiarities of structure and their exceeding beauty when finely illuminated. Dull and uninteresting as coke may seem to the naked eye, whpn properly mounted in balsam and the balsam from the upper part removed with gasoline there are few ob- jects more 1 eautiful under a -J inch objective, or even a i inch. It might be that a microscopic study of coke, and especially of the various forms of deposited carbon found on coke, would give us soms valuable information, and I did begin such a study, but the pressure of other mat- ters forced me to abandon the investigation at the time, and since then I have been unable to resume it. My excuse for this degression must be that in those forms of carbon, whether sheets, or blades, or needles, 104 GEOLOGICAL SURVEY OF ALABAMA. or hair, we seem to have nearly pare forms of deposited carbon. Percy (ut supra) , has more or less to say about deposited carbon, and Fulton in his excellent book on Coke also speaks of it. But although all authorities agree that such action may and probably does take place in a coke oven the amount of carbon thus gained is not and cannot be stated with accuracy. As before re- marked, a very thin coating of bright silvery carbon may serve to better the appearance of the coke without adding materially to the weight. The 48-hour, 72-hour and 96-hour cokes from this in- ve^tigation were examined for specific gravity, cell space and strength. The results were as follows : TABLE XI. SPECIFIC GRAVITY, CELL SPACE AND STRENGTH OF COKES. Per cent. Volume of cells Compressive strain % 48-hour . Appar. specific gravity. ....1.029 True specific gravity. '1.913 of cells by volume. 4658 in 100 parts by weight. 46.29 ultimate strength 1 in. cube. 440 Ibs. 72- hour . ....0.875 1.785 52.22 61.45 550 " 96-hour . ....0.921 1.839 4884 54.30 660 " L may be remarked in regard to the porosity of coke, as cSerermined by the percentage of cells by volume and thi- v >lume of' cells in 100 parts by weight, that single estimations are rarely of any value. During the last few years I have made many su^h estimations, and the variations in samples from the same oven are often verv considerable, confirming Dr. Dewey's observa- tions. One would naturally expect variations between the dense, well-bodied coke and the black ends, whether FUELS. 105 from top to bottom, but the variations I refer to are to be found in even the best coke from the same ovens. The results given in Table XI are aver- ages from two samples taken from the best-looking coke. In determining the apparent and the true specific gravity, the percentage of cells by volume, and the volume of cells by 100 parts by weight, I have used the method first suggested by Dr. Sterry Hunt (" Can- ada Geological Survey, 1863, 1866, " pp. 281-283), and afterward improved by Dr. F. P. Dewey. (" Trans- actions. American Institute Mining Engineers, Vol. XII, p. Ill) . But not having a goqd air pump, I boiled the samples for 12 hours and allowed them to stand in the water for 12 hours more. The formu- las used are as follows : a = weight of dry coke ; b = weight of water absorbed ; c = loss of weight in water of the saturated coke. Then: c : a = 100: x = Apparent specific gravity, c b : a= 100 :x = True specific gravity. c :b =100 :x = Per cent, of cells by volume a :b^lOO :x = Volume of cells in 100 parts by weight. The determination of the ultimate strength, which, divided by four, gave the compressive strain, was made in a Riehle Standard Testing Machine on 1-in. cubes. The cubes were carefully sawed from the coke, and were cut so as not to include any cracks. An 8-in. jhack-saw with " Star " blades does very well, although the destruction of the blades proceeds with distressing rapidity. A diamond saw, such as is used for prepar- ing sections of minerals for microscopic examination, would doubtless be an excellent tool for this work. I have used as many as six and eight ''Star " blades in sawing out a sing'.e cube. Coke is very destructive t steel saws even the very best soon becoming utterly use- 106 GEOLOGICAL SURVEY OF ALABAMA. less, as might be expected from the nature of the mate- rial. Objection has been raised to this method of pre- paring coke samples for crushing, and Dr. Thoerner recommends cylindrical test pieces. But I have ob- tained closely concordant results by careful sawing out of 1-inch cubes, and the advantage is that the ultimate strength is given directly from the beam, the compres- sive strain being taken as one-quarter of the ultimate strength. Dr. Dewey recommends taking as many as 15 sepa- rate samples from each oven, for determining specific gravity, etc., and in view of the wide variations in coke from the same oven, perhaps this number is not too large. Speaking generally, Alabama cokes fall into two main divisions, so far as concerns the porosity, large-celled and small-celled, and the duration of the coking process does not seem to affect the principle of the classification seriously. With the exception of a few Thomas ovens in operation, all of the coke now made in the ^tate is the product of bee-hive ovens. The Solvay Process Com- pany, of Syracuse, N. Y., is building 120 by-product ovens at Ensley. The coal to be used will be similar to- the coal of these experiments. As a rule, 48-hour coke is used by the blast furnaces,, the 72-hour coke going for foundry purposes. The chief difference between them is in the superior density and strength of the 72-hour product. There is also les& breeze from the ovens. Referring now to Tables IX and X : If all the so-called volatile matter should escape without depositing any of its carbon, and none of the so-called fixed carbon should be burned, but be changed to coke, one might expect to find in the coke itself 90.04 per cent, of carbon. Excluding the volatile matter of the 48-hour coke, 1.51 percent, the FUELS. 107 actual amount of fixed carbon found in the coke was 90 26 per cent. It would thus appear that the carbon burned in the oven is counterbalanced by the carbon deposited from hydrocarbons. But the difficulty of ascertaining, by analysis of the escaping gases, just what, amount of carbon is burned is so complicated that there is but little hope of arriving at even approximate accuracy. For instance, what are the products of the combustion of carbon, under the conditions maintaining in a bee- hive oven ? The entire consumption of the carbon would, of course, imply the free entrance of air, but the air is to a great extent excluded. Ebelmen found on collecting gas at three different times from cylindrical ovens, not recovering the by-products, the following, the figures being from Groves & Thorps " Chemical Tech- nology," Volume I, " Fuels," by Mill & Rowan. TABLE XII. COMPOSITION OF COKE OVEN GAS ELBELMEN. After After After 2 hours. 7% hours. 14 hours. Mean- Carbon ic acid. .: 10.13 9.60 1306 10.93 Carburetted hydrogen. . . . 1.44 1.66 40 1 17 Hydrogen 6.28 3.67 110 368 Carbonic oxide 417 3.91 2.19 3.42 Nitrogen 77.98 8116 8325 80.80 The composition of these gases varies widely, accord- ing to the period of coking, and there are doubtless other circumstances, apart also from the composition of the coal itself, which would cause variations the rapidity of the firing, the thickness of the bed of coal and coke, the size of the coal charged, the quantity of air entering the oven, etc., etc. 108 GELOGICAL SURVEY OF ALABAMA. Furthermore, changes are continual going on in the oven from the time the coal gets hot and begins to evolve gases until the coke is watered and drawn, and these changes are not necessarily the same in kind or in degree throughout the coking mass. At one point decomposa- ble gases are being evolved, at another they are deposit- ing carbon, at a third non-decomposable gases non-de- positing gases are coming off, at a fourth gases are being evolved that under proper conditions would de- posit carbon, but which, in fact, are escaping into the air. It has been said above that the deposited carbon counterbalanced the carbon that was burned in the oven. This presupposes that the fixed carbon of the coal is of the same nature as the -fixed carbon of the coke ; a sup- position not always tenable. When the volatile matter is driven off from coal in a platinum crucible at the highest temperature of a blast lamp, it is certainly pos- sible that some carbon is deposited in the mass of the coke thus formed. A closed platinum crucible within another closed crucible is a miniature coke oven, and if carbon is deposited in the large oven it should, also, other things being equal, be deposited in the very small one. Under the microscope carbon left in the crucible does exhibit evidences of the existence of deposited car- bon, for the fine globules of bright silvery luster with the reticulated markings, so characteristic of deposited carbon, are sometimes observable under a i-inch object- ive, and now and then, but more rarely, under a i inch objective. But the conditions favorable to the deposition of car- bon are more abundant and more pronounced in a coke oven than in a crucible, so that it is likely that the coke from an oven has relatively much more deposited (and therefore very pure) carbon than the residue in a cruci- ble after driving off the volatile matter. Taking every- FUELS. thing into consideration, it would appear that the fixed carbon, as determined in the ordinary method of analy- sis, is not of the same nature as the fixed carbon of the coke. But, practically, the difference is not of any mo- ment and the subject has merely a scientific interest. Deposited carbon, in pieces of considerable size, is sometimes obtained from the arch of recovery (by pro- duct) ovens. Under this discussion it might be of interest to con- struct a table from Table X, which would show the changes in ultimate composition between the coal and the coke, as Table IX does for the ingredients determined by proximate analysis. Table XIII. Changes in Ultimate Composition from Coal to Coke. M M f-t 4-l *H o o o 0,0 1 1 i 1 1. , i! ii it if h O K0^^5 Q^fi ft ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ Coal ....... 78.23 4.51 8.98 1.56 6.72 .... .................... Coke ...... 84.55 1.33 4.33 0.18 9 61 8.07 70.51 51.78 88.46 43.00 Coke is very far from being pure carbon and ash- forming ingredients, as it is sometimes taken to be. Aside from the ash in this analysis there are present hydrogen, oxygen and nitrogen, the sulphur not being considered. Of these there may be nearly 6 per cent. Parry made some determinations of the nature of the gases occluded in coke, and found that both carbonic acid and methane, were present, and he remarked that the carbonic acid probably arose from the oxidation of the carbon after the coke was made, and that an ap- preciable loss of carbon might result in this way. But this is a subject of which little is known. It presents. 110 GFLOGICAL SURVEY OF ALABAMA. many interesting questions to the metallurgical chemist, and is deserving of further investigation. No investigations were made on these cokes as to the action of carbonic acid, as recommended by Sir I. Low- thian Bell, or of hydrogen, as recommended by Dr. Thoerner. In the employment of each of these reagents considerable loss takes place, and it has been proposed to use this loss as one of the elements entering into the valuation of coke for blast furnaces. There are many questions arising in connection with coke, and to exam- ine into ail of them would take much more time than is at the disposal of most metallurgical chemists. The following results were obtained too late for incor- poration in the body of this article. They relate to the yield of coke, over a H-inch fork, and 'ashes' (breeze and ashes) from Pratt coal in a bee-hive oven. The moist- ure in the coal was not given, but would be about 6%. Washed Pratt slack: charged coal 12,650 Ibs ; ob- tained 72-hr, coke 7,080 Ibs '(=55.96%) , and 'ashes' 348 Ibs. (=2.75%). Washed Pratt slack: charged coal 13,150 Ibs.; ob- tained 72-hr, coke 7,725 Ibs. (=58.74%), and 'ashes' 346 Ibs. (=2.63%). Disintegrated washed Pratt slack ; charged coal 11,000 Ibs; obtained 72-hr, coke 6,715 Ibs. (=61.04%), and 'ashes' 271 Ibs. (=2.4(>%). Disintegrated washed Pratt slack : charged coal 11,300 Ibs.; obtained 72-hr, coke 7,275 Ibs. (=64.38%), and 'ashes' 230 Ibs. (=2.04%). In these experiments the disintegration of the coal was followed by a considerable increase in the yield of coke, and the waste in ashes fell off, in one case, from 2.63% to 2.04%. The quality of the coke made from the disintegrated coal was in no wise inferior to that made from ordinary FUELS. Ill washed slack, and in fact the coke was stronger and denser than under ordinary circumstances. Disintegration of coal, previous to coking, is not car- ried on to much extent in Alabama. ALABAMA COAL IN BY-PRODUCT OVENS, BY WILLIAM B. PHILLIPS ALABAMA COALS IN BY-PRODUCT OVENS. I115 ALABAMA COAL IN BY-PRODUCT ; v , ^ EXTRACT FROM A REPORT MADE TO THE SLOSS IRON AND STEEL CO. (American Manufacturer, Vol. LXI1, p 446.) BY WILLIAM B. PHILLIPS. It is proposed to give in this paper an account of the testing of 54,000 pounds of Alabama coal at the Otto- Hoffman by-product ovens of the Pittsburg Gas and Coke Co., near Glassport, Penn., undertaken with the view of ascertaining to what extent this coal would lend itself to the recovery of by-products, and the production in a by-product oven of coke suitable for use in the blast furnace. The works at Glassport have been in successful opera- tion for more than a year under the superintendence of Dr. F. Schniewind, who introduced the system into this country. They consist of 120 ovens, in four batteries of 30 ovens each. The capacity of each oven, when fully charged, is about 7.5 tons of coal. The works are well provided with condensing chambers, ammonia apparat- us, exhaust pumps, etc. The testing of this coal did not in any wise interfere with the usual operations there, except in so far as it was necessary to weigh and meas- ure the products obtained. The conditions of the test did not vary materially from those under which large and regular operations are carried on every day. The . GEOLOGICAL SBRVBY OF ALABAMA. results, therefore, do not represent what might be ob- tained from a special test under special conditions, but it is believed, that t&ey can safety be used as the basis of calculations as to future work. . Th$ coal used? was slack from- the mines of the S loss Iron and Steel Co., Jefferson County, Ala., washed in a Robinson-Ramsay washer. It came from the Pratt seam and was of the usual quality of this coal, when washed, as the following average analysis will show : Proximate analysis of Pratt washed slack coal. Drs. Mason and Luthy, Pittsburg Gas and Coke Co. Moisture 5 .95 Volatile matter 32.69 Fixed carbon 54.33 ... 7.03 100.00 Sulphur 0.94 Phosphorus 0.0117 . The ultimate analysis of this coal, as made in the Phillips Testing laboratory, Birmingham, is as follows : Ultimate analysis of washed Pratt slack coal, made by the Phillips Testing Laboratory, Birmingham. Analysis on dry basis : Carbon 76 .50 Hydrogen 4.90 Oxygen , 10.15 Nitrogen 1 25 Ash.! 7.20 100.00 This amount of nitrogen is equivalent to 1.15 per cent, of ammonia, and the disposable hydrogen would be 3.61 per cent. The -coal, as charged, contained, on the aver- age, 5.95 per cent, of moisture, but for convenience it ALASAlVfA COALS IN BY-PRODUCT OVENS. 117 ~will be best to consider it as dry and to base all the re- sults and calculations on dry coal. Four separate charges were tried in an oven fitted up for the purpose and used in testing various coals that have been sent to the works. A very careful watch was maintained over the entire operation and especial thanks are due, not only to Dr. Schniewind, and to Mr. W. P. Parsons, Assistent Superintendent, but also to the gentlemen comprising the laboratory force, and to Messrs. Thos. G. Littlehales and Wm. Speakman for the very kind and unremitting attention given to the test throughout its entire duration. The first charge contained 13,067 pounds of dry coal, and the coking time was 34 hours and 35 minutes. Tiie coke was pushed in 1 minute after taking down the doors and was watered on the outside. When ready to load the coke contained 1.80 percent of moisture. The yield of dry coke, over a H inch fork, was 8,490 pounds or 64.9 per cent, of the weight of, dry coal. The dry breeze weighed 320 pounds, or 2.45 per cent of the dry coal, so that the total weight of coke obtained was 8,810 pounds, or 67.3 per cent, of the weight of the dry coal. The yield of sulphate of ammonia was 19.2 pounds, per ton of dry coal. It was decided not to weigh the tar from each separate charge, but to wait until the test was completed. The highest candle powerobserved during the first test was 18.8, and the average was JL3.2. The average specific gravity of the gas was 0.471. The highest calories were 6649, equivalent to 748.0 British Thermal Units, which will be referred to hereafter in this paper as B. T. U. The average heat units during the first tests were 651.8, or 5794 cals. The second charge of coal represented 13,509 pounds of dry coal. The coking time was 29 hours and 30 118 GEOLOGICAL SURVEY OF ALABAMA. minutes. The yield of dry, forked coke 9,275 pounds, or 68.6 per cent, and of breeze 582 pounds, or 4.3 per cent., a total yield of coke of 9,857 pounds, or 72.9 per cent. The yield of sulphate of ammonia was equiva- lent to 23.9 pounds per ton of dry coal. The highest candle power observed was 16.3, the average being 11.1. The highest calories were 6617, equivalent to 744.4 B. T. U., the average being 5352 cals., or 602.1 B. T.U. The specific gravity of the gas was 0.411. The amount of moisture in the coke, when ready to load, was 3 per cent. The third char ore represented 13,882 pounds of dry coal, and the coking time was 30 hours and 5 minutes. The yield of dry, forked coke wis 9,02) p Kinds, or 65.4 per cent., and of breeze 600 pounds, or 4.3 percent., a total yield of coke of 69.7 per cent. When ready to load the coke contained 3.0 per cent, of moisture. The highest candle powQr ob3erved was 17.8, the average be- ing 11. The specific gravity of the gas was, on the av- erage, 0.451. The highest calories were 6330, equiva- lent to 782.1 B. T. U., the average being 5429 cals., or 618.1 B. T. U. The yield of sulphate of ammonia was 26.9 pounds per ton of dry coal. The fourth charge represented 14,171 piunds of dry coal, and the coking time was 32 hours and 20 minutes. The yield of dry, forked coke was 9,608 pounds, or 67 8 per cent., and of breeze 708 pounds, or 5.0 per cent , a total yield of coke of 72.8 per cent. When ready to load the coke contained -..30 per cent, of moisture. The highest candle power observed was 12.6, the average being 10.3. The average specific gravity of the gas was 0.426. The highest calories were 6576, equivalent to 739.6 B. T. U., the average being 5765 cals., or (348.8 ALDBAMA COALS IN BY-PRODUCT OVENS. 119 B. T. U. The yield of sulphate of ammonia was 25.5 pounds per ton of dry coal. The average amount of moisture in the coke, when ready to load, was 3.02 per cent. The average yield of dry, forked coke from dry coal was 66.7 per cent., and of breeze 4.0 per cent., a total yield of 70.7 per cent. The average yield of sulphate of ammonia was 23.9 pounds per ton of dry coal. The yield of tar was 90 pounds per ton of dry coal. The average quality of the tar from the seal-pot may be stated as follows : Moisture 3.93 Oil 1.52 Specific gravity 1.211 The average quality of the tar from the exhauster was as follows : Moisture 2.04 Oil 2.04 Specific gravity 1.211 The average analysis of the coke (Drs. Mason and Luthy) was as follows, on a dry basis : Volatile matter 0.98 Fixed carbon < 90.22 Ash. 8.80 100.00 Sulphur 1.28 I will not, at this time, enter upon the subject of the adaptibility of this coke for blast furnaces . There is no in- formation to hand respecting its use in the Alabama fur- naces, but experiences elsewhere has shown that per unit 120 GEOLOGICAL SURVEY OF ALABAMA. of carbon it has nothing to fear from the competition of bee-hive coke, especially when to the making of coke is added the saving of by-products. If it were merely a question of coke making, without reference to by-pro- ducts, perhaps there is no system any better than the bee-hive. In structure, the bee-hive coke has the adv-an- tage over by-product coke in that it is more uniform, but in carbon duty there is not much if any thing to choose between them. By-product coke is apt to contain more moisture and to be somewhat more brittle than bee-hive coke, but under conditions allowing of the utiliza- tion of the gas, tar and ammonia, the loss in quality of the coke is more than counterbalanced by the profits accruing from the sale of these substances. The average yield of this coal in the bee hive oven is somewhat below 60 per cent, counting breeze as coke, so that in this respect the by-product oven has an advant- age of 10 per cent to 12 per cent, greater yield. It is doubtful if the difference will amount to 15 per cent, or 16 per cent, as some would claim. If we accept the statement of Sir Lowthian Bell that bee-hive coke is 10 percent, more useful in the furnace than by-product coke, or that Mr. John Fulton that it is 7 per cent, more use- ful, we should be prepared to anticipate a balancing of carbon duty against greater yield, one off-setting the other. But so much depends upon the class and condi- tion of the stock, and the actual furnace practice that generalization is hazardous. The gas from this coal is well adapted for illumina- ting purposes, but would have to be enriched in some carburetting apparatus to bring it up to the require- ments ordinarilv made in regard to candle power. Du- ring the first 24 hours of the process the candle power did not fall below 8, and went as high as 18.8. The average candle power during the first period of 12^ hours ALABAMA COAL IN BY-PRODUCT OVENS. 121 was 13.1, and during the second period of 12 hours it was 10.8. By operating a sufficient number of ovens in series, so as to keep the candle power at about the same figure, it would doubtless be possible to reach 15 or 16, leaving from to 7 candle powers to be added by the carburizer. The gas is well adapted for fuel purposes, the heat units ranging from 141,379,700, per 1,000 cubic feet, in the second period of 12 hours to 165,178,770 in the first period. These figures are less than for natural gas, as this may go to 209,979,000 heat units, per 1000 cubic feet. The yield of gas was 9,600 cubic feet per ton of dry coal, of which 3.000 cubic feet would be surplus gas. At the meeting of the Alabama Industrial and Scien- tific Society, held in Birmingham, December 21st, 1897, Mr. W. H Blauvelt, Engineer for the Solvay Process Company, Syracuse, N. Y. , read a paper on The Seraet- Solvay Coke Oven and Its Products, Mr. Blauvelt had previously discussed the subject in The Mineral Indus- try, Vol. IV., 1896. He has recently republished the important parts of these two articles in pamphlet form. Considering that the Solvay Process Company is now erecting 120 of the Semet-Solvay ovens at Ensley, Ala., .and will soon have them in operation, and that Mr. Blauvelt is thoroughly versed in the construction and conduct of this oven, it does not seem to be out of place to introduce here his latest remarks upon the subject. Aside from the utilization of the hot air from some bee- hive ovens for raising steam under boilers usually fired with coal, the only product Irom the coke ovens in this State has been the coke. But besides the coke there are other valuable products to be obtained from coal as it is being changed into coke, as, for instance, ammoniacal compounds, tar, and gas suitable for heating and illumi- nating purposes. These can be and in many places are 122 GEOLOGICAL SURVEY OF ALABAMA, now recovered from the coal without prejudice to the coke. Their recovery and subsequent utilization marks one of the great and beneficent departures from the former way of making coke. The system is peculiarly adapted to Alabama coals, as they are rich in tar, am- moniacal compounds, and gas, all of which can be re- covered and used. . From the tar maybe made pitch and ordinary light and ''dead" oils, and a great number of products now recognized as coal-tar products, number- less dyes and flavoring extracts and medicines. From the atnmoniacal compounds sulphate of ammonia is made, a very valuable material used in the manufact- ure of fertilizers, anhydrous ammonia now so largely used in the South in ice-making establishments, and other substances more or less largely employed in the arts. The surplus gas may be used for all kinds of heat- ing purposes, for cooking, and heating residences ; and, by enrichment, for lighting purposes. Instead, there- fore, of throwing away these by-products they will be utilized, and the plant at Ensley the first of the kind in the South will enter upon this work within the present year. Mr. Blauvelt's pamphlet which is here republished.. by permission is as follows : ALABAMA COAL IN BY-PRODUCT CVENS. 123 THE SEMET-SOLVAY COKE OVEN AND ITS PRODUCTS.* I BY WILLIAM H. BLAUVELT. [Extract from proceedings of the winter meeting of the Alabama In- dustrial and Scientific Society, held in Birmingham, Ala., De- cember 21, 1897.1 Gentlemen of the Alabama Industrial and Scientific Society: The plant of by-product retort ovens, which is being erected at Ensley, is only the sixth installation of by- product ovens in this country. In Continental Europe such ovens have become quite an old story, and, in fact, practically no bee-hives are built there, except in small or isolated plants. So few years have passed since by- product ovens were first introduced in America, that they are still a novelty to very many, even of those who are well acquainted with the use of coke and its manu- facture in the old fashioned way. It has, therefore, been suggested that a brief description of the plant at Ensley, and a comparison of these new ovens and their products with the old bee-hive type, will be of interest to your Society. The plant of oven c now under construction at Ensley will consist of 120 retort ovens, with their accompany- ing apparatus for collecting the by-products from the distillation of the coal. It is probably unnecessary to say that retort ovens are essentially different in shape from the bee-hive oven, the coking chamber being usually about 30 feet long and 6 feet high, and varying in width from 15 inches to 30 inches or more, depending upon the coal to be coked, and the type of oven. The coal is *Portions of this paper are taken from an article by the writer, which was published in "The Mineral Industry," Vol. IV., 1896, which is a copyrighted work, and such extracts are here used by the special permission of the Scientific Publishing Company, the proprietors of "The Mineral Industry." 124 G-EO LOGICAL SU RVE Y OF ALA BAM A. charged through three or more holes in the top, in the same manner as in a bee-hive oven, except that the oven is filled with coal to within about eight inches of the top. The coal is heated and the volatile matter driven off by means of the heat generated by the combustion of gas in the flues or passages in the side walls of the ovens. A fourth opening in the roof of the oven is connected with a pipe or main, which carries the gas, as it comes off from the coal, to the by-product apparatus. The Ensley ovens are of the Semet-Solvay design. This oven is the principal exponent of what is known as the horizontal flue type, in contradistinction to the vertical flue type, the principal representative of which is the Otto-Hoffman oven. In the vertical flue type the gas is burned in two horizontal flues, or combustion chambers, at each side of the ovens at the bottom, which extend half way toward the other end. The products of combustion ascend through some sixteen small vertical flues, which reach to the top of the oven, where they deliver into another horizontal flue, which reaches the whole length. This connects with a similar set of small flues, which deliver the hot gases into a horizontal flue, or combustion chamber at the bottom, like the first, and thence to a regenerator of the familiar Siemens type. Every hour the travel of the gases is reversed, hot air being supplied for the combustion of the gas from the regenerators, as in an ordinary Siemens furnace. In the horizontal flue ovens there are three horizontal flues, one above another, on each side of each oven, ex- tending the full length of the oven, and connected with each other at the ends, so as to form a continuous flue for the gas and flame. The travel of the gases is from above downward; that is, through the top flue, then backward through the second, etc., the bottom flues "being connected with a passage to -the chimney. A ALABAMA COAL IN BY-PRODUCT OVENS. small amount of gas is introduced at the ends of the- top and. second flue, along with a sufficient amount of air for its combustion. This air is preheated by a simple arrangement in the bottom of the ovens, and the com- bustion goes forward continuously without any attention , often for weeks at a time, it being only necessary to see that the proportions of gas and air remain the same, and are of sufficient quantity to keep up the necessary heat in the ovens. The gases after leaving the ovens are carried under boilers, and supply steam for operating the machinery of : the plant. These gases go to the stack at a temperature of not much over 200 6 C, so that from the point of view of heat economics these ovens are very efficient. The Semet-Solvay ovens are usually about 16 inches wide, and contain about 4 tons of coal per charge. This charge is coked in about twenty-four hours, and when the gases are all driven off, the doors at each end of the ovens are opened, and the whole- charge of coke is pushed out with a steam pusher, or ram, in a minute or two. As soon as the ram has been withdrawn and the doors are closed, the oven is ready for another charge, and practically no heat has been, lost, as the quenching is all done on the outside of the oven. The whole pro- cess of discharging and recharging an oven can readily be completed in fifteen minutes. As the gas which is distilled from the coal leaves the ovens it enters a large flue known as the hydraulic main. This extends the whole length of the block of ovens, and is partially filled with water. The gas bubbles through the water, and a portion of the tar and ammonia is con- densed out. From the main the gas passes to the con- densers. These are large vertical cylinders filled with tubes through which water is made to circulate. The gas passing around these tubes is cooled, and a further 126 GELOGlCAi; SURVEY OF ALARAMA. portion of the tar and ammonia condenses. Rotary 'ex- hausters occupy the next place in the series of apparatus, their use being to draw the gas from the ovens through the pipes and condensers, and to discharge it into the next following apparatus, which is the ammonia washer. In this vessel the final traces of ammonia are removed, and the gas thus cooled and washed is free from con- densable matter and ready to be used for heating or lighting. A portion of it is usually withdrawn at this point and used to heat the flues of the ovens, but if there is sufficient demand for the oven gas for other pur- poses, ordinary producer gas may be substituted for it, and the whole amount produced will be available for sale. This amount varies with the coal, but is usually from eight to ten thousand cubic feet per ton of two thousand pounds. The quality of this gas is more fully described later, where the by-products of the ovens are discussed somewhat at length. THE PRODUCTS OF THE BY-PRODUCT OVEN. Coke. An investigation of the subject will immedi- ately show that the essential distinction between the operation of the retort oven and that of the ordinary beehive is that in the former the coal is coked without the admission of air, by heat applied from the outside, while in the latter the air is admitted to the oven and the combustion takes place immediately over the body of coal. The result is that in one case the hydro-car- bons are simply distilled off, with a certain breaking down and deposition of carbon on the coke, so that a yield of coke greater than the so-called ''theoretical" can be counted on, wliilu in the other case the most of the hydrocarbons are burned in the ovens, some carbon is deposited, and some of the fixed carbon of the coal is burned, resulting in a yield of coke less than the theo- ALABAMA COAL IN BY-PRODUCT OVENS. 127 retical. As an illustration of the difference jn yield resulting from this difference in method of coking, a good yield of coke from Connellsville coal in a beehive oven is 65 per cent., while in a good retort oven it is easy to get 75 per cent., an increase of about 10 per cent. Of course this increase reduces proportion- ately the percentage of ash, phosphorus, etc., remain- ing in the coke, so that the retort oven yields more coke and a purer coke than the beehive from the same coal. This increase in yield varies with the proportion of fixed .carbon, ash, etc., in the coal. The quality of the coke made in the by-product ovens has long been a subject of discussion, especially among the blast furnace men of Europe. The English authori- ty, Sir Lowthian Bell, made a series of careful tests a number of years ago and pronounced against the coke in comparison with that made in beehive ovens, and his conclusions were accepted by English ironmasters. But improved construction and practice have combined to produce a better coke, and it is reported that Sir Low- thian Bell has modified his views to such an extent that a plant of retort ovens is now being built at his own works those of Messrs. Bell Brothers. On the Conti- nent retort oven coke is now the standard, and in this country we are just beginning to realize that a coke not made in the old beehive oven and not having the famous silvery gloss of coke quenched in the oven is proving itself quite equal to it in fuel value. The essential difference between beehive and retort oven coke lies in its hardness and shape, caused by the different application of the heat in the oven. In the beehive the coal is spread out in a layer 23 or 24 inches thick over a surface some twelve feet in diameter. The bottom of the oven having been cooled by the quench- ing of the previous charge and by contact with the new 128 GEOLOGICAL SURVEY OF ALABAMA:.- one. the coking begins at the top and extends down- ward, reaching the bottom in from 32 to 34 hours. The coke has ample opportunity to swell and develop a cellu- lar structure in accordance with the composition of the coal, and entirely independent of any attempts at con- trol. The typical form of beehive coke is therefore long finger-like pieces, widening toward the bottom of the oven and with an inch or two of spongy coke at each end. The inability to control the formation of the cells makes it essential that just the right coals are used, or the requisite hard body, resistant alike to pressure and the action of hot carbonic acid in the blast furnace, -can- not be obtained. The fact that the coal from the Con- nellsville distrk-t gives just the requisite structure when coked in the beehive oven is the reason for its present pre-eminent position as a blast furnace fuel in America. In the retort oven the coal lies in a high narrow mass, about 5 feet high and from 16 to 20 inches wide. The previous charge having been pushed out rapidly by ma- chinery and quenched outside, the oven is hot when the fresh charge is introduced and the evolution of gases be- gins immediately from the coal lying in contact with the hot sides. The flow of gases being from the sides, they meet in the center and rise to the top, where they es- cape, forming a sort of cleavage plane midway between the two walls. Thus the pieces of retort coke are stouter than the long, slowly developed "fingers" of the bee- hive oven, and are a little shorter than half the width of the oven. The end of the piece next the wall is denser and the end next the cleavage plane is more spongy than the main body. The cellular structure is more compressed than bee- hive coke, principally on account of the narrow retort that permits no expansion in the direction of the flow of the gases, and also because the depth of the charge is ALABAMA COAL IN BY-PRODUCT OVENS. 129 usually about two and one-half times as great as in the beehive. The cellular structure of retort coke is depend- ent somewhat on the proportions of the ovens, the tem- perature and" the time of coking. . The ability of the retort oven to coke coals that cannot be used in the beehive is due to the more rapid applica- tion of the heat, fixing the pitchy or coke-making por- tion of the coal before it has time to escape, and the formation of a firm cellular structure by the pressure. During the past year a conclusive test has been made indicating the relative values of retort and beehive coke made from the same high grade coal, of a quality adapted to both the retort and .beehive practice. For a year or more a blast furnace has been run either entirely or largely on retort coke made from the Connellsvill 6 coal. The furnace was blown in on retort coke, and run for some months without any signs indicating anything unusual in the fuel. Subsequently a portion of beehive coke was used in the fuel charge, arid from time to time the fuel was changed from all retort coke to all beehive coke, or to a portion of each, without any indications in the working of the furnace that there was any difference in the fuel. It is probable that prolonged and accurate compari- sons would show that the hardness of the retort coke would result in a somewhat lower fuel consumption and a cooler furnace top, owing to the weaker action of the furnace gases on the harder coke; also, that the blast pressure would have to be slightly higher than with the beehive coke. It is quite within the bounds of possibility that some of our American coals, equal in chemical purity to the Connellsville, yet inferior to it in adaptability to the conditions of the beehive oven, may prove to make a coke in the retort oven that will be of equal value in 9 ISO Gfioio6t6AL StTRVEV o every respect with the Connellsville beehive coke. In- deed, experiments already made would seem to point in that direction. Objection has been made to the retort coke on the ground that it is watered outside the oven, thereby de- stroying the carbon glaze found on coke quenched within the oven and increasing the percentage of moisture in the coke. Careful tests have proved that retort coke is somewhat more resistant to the action of hot carbonic acid in the top of the furnace than is beehive coke from the same coal, which seems to show that the carbon glaze has in practice no value. The absence of a glaze on retort coke is no indication that carbon is not de- posited from the gases, for in the fiKSt place the yield of coke is always higher than the so-called "theoretical" yield, and in the second place, as the coke is leaving the oven the glaze can plainly be seen, but its brightness is destroyed by the water. A long series of tests have shown that coke properly quenched outside of the oven need not contain over i to f per cent, of moisture, but the amount of moisture in the coke after its arrival at the furnace is altogether another question, and depends more on the time it is on the road and on the humidity of the atmosphere than on the method of quenching. The effect of moisture in the upper part of a blast furnace is an open question. Experiments have heen made by leading furnacemen which indicate that its cooling action on the ascending gases saves the coke in a measure from solution in the hot carbonic acid and permits more coke to reach the zone of fusion, with the result that the fuel consumption is noticeably lowered. The Connellsville beehive coke is, perhaps, thB most perfect blast furnace fuel in the world, and it is not claimed that retort coke made from this coal is a superior fuel to the beehive product. But to the gr eet^tiitum ALABAMA COAL IN BY-PRODUCT OVENS. 131 nous coal fields of this country, to which the Connells- ville district does not bear the relation of one to the hundred, tho retort oven comes with a promise of help. Many coals that, although pure enough chemically for metallurgical use, make a soft coke in the beehive oven, when coked in the retort oven give a structure so hard- ened and strengthened that the product is an entirely acceptable metallurgical fuel. in other cases, when the impurities are too great for furnace or foundry use, oi % the structure is hopelessly weak, or when the coal is dry and lies dead in the beehive without a suggestion of coking, a coke can often be made in the retort, oven that is easily salable for domestic purposes, brewers-' and m listers' use, and for many other uses where a clean - burning fuel, free from smoke, is desired. The demand for coke for these purposes is growing rapidly, and the supply of this market should be very profitable , in a properly located and designed plant, from which the gas and other by-products would have a ready sale. The ability of the retort oven to coke coals that have hitherto been considered non-coking, brings into' promi- nence the subject of laboratory tests of coals for coking purposes and of the coke made. A chemical analysis of the coal or coke, while important, does not -fully indi- cate its value, and physical tests are quite as important. The coking qualities of a coal are hardly shown at all by an ordinary chemical analysis, and an actual test in the oven is the usu.J method for determining this point. A laboratory method for making this test has been re- cently developed by Louis Campredon in the laboratory of the Vignac Works, France. His method is similar to that used in ascertaining the binding power of cement. The principle is the mixing of the coal -with an inert body and carbonizing the mixture in a closed vessel ; the greater the binding or coking power of the coal the more 132 GEOLOGICAL SURVEY OP ALABAMA. inert matter will it bind into a solid mass. The prac- tical operation of the method is as follows : Pulverize the coal finely, passing it through a sieve -of fine mesh. A suitable inert body is a fine siliceous sand of uniform grain, but somewhat coarser than the coal. Several equal portions of coal (say of 1 gin. each) are mixed with variable weights of sand, and the mixtures are heated to a red heat in closed porcelain crucibles, so as to carbonize the coal . After cooling , either a dry powder or a more or less hard coked mass is obtained. After a few trials it is easy to determine what maximum weight of sand a coal can bind together. Taking the weight of coal as unity, the binding power will be given by the weight of the agglomerated sand. The binding power is nil for a coal giving a powdered coke, and it has been found to be 17 lor the most bind- ing coal yet tried by the experimenter, while pitch is 20. Experiments by this method show that there is no rela- tion between the proximate analysis and the binding power of coals, confirming actual even experience. THE BY-PRODUCTS. These consist primarily of ammonia, tar and gas, and in addition to the increased yield of coke are the sources of profit from the by-product oven which are wholly lost in the ordinary beehive. Some retort ovens, such as the Otto-Coppee, for example, are without the by-product apparatus, and burn the gas to heat the ovens without washing it. These recover no ammonia or tar, but use the excess gas for raising steam, evaporating about 1.5 pounds water per pound of coal coked. But the by- products are so easily saved and the profits therefrom make such an acceptable addition to the right side of the ledger that they can hardly be neglected. A brief consideration of each one rnay be of interest. ALABAMA COAL IN BY-PRODUCT OVENS. 133 Ammonia. This substance is given off from the coal in the oven very slowly at first, but as the temperature of the charge rises the quantity increases, and after some ten hours the evolution is quite rapid. As the coking approaches completion the yield becomes much less and stops altogether, although usually a quarter or more of the nitrogen originally in the coal still remains in the coke. The yield of ammonia varies very much in differ- ent coal, and depends partly on the amount of nitrogen and oxygen in the coal. It varies also with the tem- perature at which the coal is coked. Perhaps the most reliable method to determine the yield from any coal, ex- cept by an actual oven test, is by the distillation of a sample of the coal in a small retort, under the same temperature an I conditions as are present in the oven. But the results are liable to be misleading unless the operation is conducted by an experienced person, as it is hard to maintain the proper conditions. The ammonia from the ovens is collected in the hy- draulic main and condensers, along with the tar, by the cooling and scrubbing of the gas. The ammonia occurs in two forms in the liquor : *' fixed " and " volatile ; ' the former containing the sulphates, chlorides, cyanides, etc., while the latter contains the carbonates, sulphides, and, according to some, free ammonia. The bulk of the fixed salts is condensed first and the volatile later. The ammonia liquor is quite weak when it is first drawn from the tar, usually containing from to 1 per cent, of ammonia. This weak liquor may be either converted directly into sulphate, and sold as fertilizer, or by puri- fication and concentration it may be converted into aqua ammonia or anhydrous ammonia, which is used so largely through the South and elsewhere in refrigerat- ing and other apparatus. Ammonia liquor was formerly valued by the hydo- o r THE E or 104 GEOLOGICAL StTOVEt OS* ALABAMA. meter, but this method is deceptive, as the density of the liquor is affected by the condition in which the ammonia occurs. The more accurate method is the dis- tillation of the liquor with some caustic lime or soda, which drives off all the ammonia, volatile and fixed. The distilled ammonia is absorbed in standard acid, and the excess of acid is afterward titrated with a standard alkali solution. The yield of ammonia is usually reckoned as ammonium sulphate, although it may be sold as liquor or sulphate, or in a more con- centrated form, according to the market. The yield of ammonia from the coals in the vicin- ity of Pittsburg is from 16 to 22 pounds of sulphate per ton of coal. Tar. Since ttie manufacture of illuminating gas by the water-gas process has attained prominence the mar* ket for tar is very much improved. Very large quanti- ties are used for roofing, paving, etc., and in Europe much is distilled and separated into pitch and the va- rious lighter oils, which are further treated for the almost endles number of valuable substances which they contain. In this country but little of this is done as yet, and the tar is used mainly for the cruder purposes. Properly developed, its manufacture into the more valu- able products should yield very satisfactory profits. Our' chemical manufacturers are beginning to realize this fact, and plants for the distillation of tar are growing in number and in importance. The rapid increase of by- product ovens, and the consequent large amount of tar which will be put- on the market in the near future makes it necessary to find another outlet for it than the cruder uses, and it is probabje that tar distillation, will be an important industry in this country before many years. The main products of the distillation of tar are, light ALABAMA COAL tS V FRODtctf dVKKS. 136 oil, croosoto or heavy oil, naphthalene, anthracene and pitch. *i ;..-: The yield and quality of tar from retort ovens depend on the coal and also on ths temperature at which the distillation takes place. The tar from the leading re- tort ovens. is usually of excellent quality and commands the bast price. The yield of the coals in the vicinity of Pittsburg is from 70 to 80 pounds per ton of 2,000 pounds of coal. Some coals yield as much as 100 pounds or more. Gas. The gas that is obtained from retort ovens is a by-product, the value of which varies greatly with the locality in which the ovens are situated. When the ovens are at the coal mine the gas is frequently valuable only for steam raising purposes, and at the usual prices of coal at the mines would be worth but a very few cents per thousand fest. An intermediate condition would be when the ovens are adjacent to an iron or steel works, where the gas could be used for heating furnaces, soak* ing pits, etc., where it would supplant producer gas being much more conveniently applied and easily freed from all impurities. The most favorable locations for obtaining a good value for oven gas are those adjacent to large towns, where there is a demand for illuminat- ing or fuel gas. The discovery and use of natural gas in the country has caused a great demand for fuel gas, especially for domestic purposes, and many hundreds of thousands of dollars have been spent in attempt to sup- ply this demand. But while these experiments have been going on the bee-hive coke ovens of Pennsylvania alone have been quietly burning to waste nearly 1,000,- 000,000 feet a week of a very superior quality of fuel gas without exciting any special attention. Cokeoven. gas from properly managed retort oven is approximately the same article as that from the retorts 136 0EOLOGICAL SURVEY OF ALABAMA. of a gas house, the processes of manufacture being simi- lar. It usually contains rather less illuminants, how- ever. Its quantity and composition vary with the coal used and the temperature of distillation, but made from good gas coal it may be used for illuminating purposes after being passed through the ordinary lime boxes to remove the sulphur, etc. If from the nature of the coal the illuminating power of the gas is low, it can either be enriched by any of the well known methods or burn- ed with incandescent burners or used as a fuel gas ; for the lack of 1 or 2 per cent, of illuminants will not appre- ciably affect its fuel value. In arranging and oven plant for the supply of fuel or illuminating gas, it is necessary either to provide a holder of rather large dimensions or with a smaller holder, to have not less than, say, twenty-five or thirty ovens, that shall be drawn in rotation at approximately even intervals ; for in common with other substances containing hydrocarbons, when coal is distilled in an oven or elsewhere the gases given off are not at all uni- form in composition, but change constantly as the distil- lation progresses. The following are analyses of retort-ovens as from Eu- ropean and American coals : TABLE XIV. Nitrogen Methane _,* I. -rexueiJi II. jtige uy III. IV. V. acid ....... 3.0 0.90 1.4 3.27 3 oxide 8.8 4.90 6.5 7.95 7.4 n N 58.0 58.57 53.37 52.77 51.7 2 4 5.74 0.5 1.99 5.5 24.7 27.56 36.1 31.45 32.3 3.1 2.33 2.2 2.57 2.8 Totals 100.0 100.0 100,0 100.0 100.0 ALABAMA COAL IK BY-PRODtJCT OVENS. 137 It is often asked what is the difference between coke oven gas and natural gas? This is readily answered by a comparison of the above analyses with the following, which is from gas sold in Alleghany and Pittsburg by the natural gas companies : Carbonic acid. 0.3 per cent. Nitrogen 0.2 per cent. Carbonic oxide 0.0 Methan ...96.9 Hydrogen... 0.0 " Qlefines 0.8 It will be noticed that this latter gas is almost pure methane, or marsh gas, while the coke oven gas is prac- tically a mixture of methane with hydrogen. The nat- ural gas of the above analysis contains about 980 Brit-^ ish thermal units per cubic foot, while the coke oven gas usually contakis from 560 to 590 heat units. It is a familiar fact to those who have seen natural gas burned for lighting purposes that the large amount of methane causes it to burn with a smoky flame, and the light from it is therefore poor, although the proportion of olefines or illuminants, is not always as small as in the analy- sis given above. It has been stated by good authori- ties that when a gas having such a large proportion of methane is burned in the ordinary way without regener- ation of the air by the products of combustion, that the available heat units are not greater than those from a gas of similar composition to the coke oven gases given above, owing to the fact that such a very large amount of air is necessary to burn the methane, and the amount of heat absorbed in bringing the inert nitrogen up to the temperature of the combustion chamber is so great that it counterbalances the superior heating value of the gas. Of course, if the heat carried away in the products of combustion were returned to the furnaces by regenera- tion, this loss would not be nearly so great. IDfl GEOLOGICAL StfftVEtf O# ALABAMA. The principle source of luminosity in the gas is benzol. This substance is separated from the gas in some of the German by-product works, and is used for the manu ? ao- ture of the aniline colors. It is a highly volatile sub- stance, somewhat similiar to the naptha products of petroleum distillation, and is very difficult to transport. Its removal from the gas renders the latter useless for illuminating purposes, but does not materially affect its fuel value. Benzol is also obtained in the distillation of tar, but not in large quantities. To sum up briefly, then, it will be seen that the coking of coal in. the by-product retort oven differs in the re- sults obtained in the following particulars from the same operation in tho bee-hive : From tha bee-hive oven. we obtain coke. The article is of excellent quality if the coal is just adapted to the purpose, but the yield is from 5 to 20 per cent, lower than the analysis of the coal shows should be gotten. In addition to the coke there is a great deal of smoke, but those living near the ovens hardly look on this as a valuable product. From the by-product retort oven, we have coke again, and always more than the analysis of the coal indicates. It has yet to be proven that any coal which makes good bee-hive coke will not make equally good retort oven coke. Moreover an excellent metallurgical coke can be made from many coals that are worthless for the bee-hive. In fact it is largely for this reason that re- tort ovens have been so widely introduced in Conti- nental Europe. In addition to the increased yield of coke we have from a ton of coal, from 16 to 22 pounds of sulphate of ammonia, from 70 to 100 pounds of tar, and from 3,OOU to 10,000 cubic feet of gas. The manufacture of coke is about the only metallurgi- cal operation that we Americans, proud of our wonder- OOAL ttf B * ful progress in all the mechanioal arts, still conduct after the manner of our ancesters before the Revolution- ary war. Let us introduce the by-product retort oven into the chain of iron manufactir e,confi lent that it will not be unworthy to be linked with the mining and haul- age of our coal by electricity, the digging of oar ore by steam shovels, and our blast furnaces smelting 700 tons of iron in a day. CHAPTER V. Coke Furnaces. The largest furnaces in Alabama are 80 feet high, and 19 feet 6 inches wide in the bosh, or widest part. The greatest amount of pig iron ever made in a furnace in one day in this State was 265 tons,* and for its produc* tion there were required 588 tons of ore, 62 tons of limestone and 265 tons of coke, all of 2,240 Ibs. It is by no means unusual for a furnace to make 200 tons of iron a day, and for this there would be required 480 tons of ore, 280 tons of coke, and 25 tons of stone, if the proper amount of hard ore were used. The aver- age number of tons of material handled per ton of iron made is about 4.44 in coke furnaces', so that for the 835,851 tons of coke pig iron made in 1895, there were handled 3,711,178 tons of material, of which 2,089,627 tons were ore, 442,176 tons were stone (limestone and dolomite), and 1,179,375 tons were coke. These are approximate figures. The amount of ore required to make a ton of iron varies from 2.10 tons to 2.87 tons, the average being close to 2.50. The average amount *This output has been exceeded by about 50 tons since the intro- duction of 16 tuyeres. 140 GEOLOGICAL SURVEY OF ALABAMA. of coke used per ton of iron made is .1.41 tons of 2240 pounds, the range being from 1.16 to 1.60. The average amount of stone used per tou -of iron made is about 0.53 ton, the range being from 0.10 to 0.88. The amount of each material entering the furnace per day is not a matter of guess, or of indifference, but is carefully determined from the chemical analysis. It is customary to fill the* furnace and keep filling it by ' 'charges," each ' 'charge" being- composed for the most part of ore, coke and stone. Thus, for instance, a "charge "-may be composed of 5,600- IDS. of coke 10.080 pounds of hard ore, -2,749 pounds of soft ore, and 620 pounds of limestone, and the furnace will take from 80 to 90 charges per day, and should yield 200 tons of iron. The proportion between the various elements of the charge, as well as the total weight of the charge, and the num- ber of charges per day, are all subject to change, but unless there is urgent necessity the daily alterations should be very slight. Having once established the proper burden, it is not advisable to change it, nor is it- necessary to do so if the materials can be provided in sufficient quantity and with sufficient regularity, and uniformity of composition. But changes of burden are very frequently made, so frequently in fact that the ne- cessity for them constitutes the greatest obstacle in the path of successful furnace management in this State. It is the lion in the way, unchained at that. In com- paring furnace practice in Alabama with furnace prac- tice in Pennsylvania, for instance, one is impressed at the outset with the frequent and in many cases violent changes in the burden in the first place, and in the second with the large tonnage handled per ton of iron. This tonnage is referrable to the raw materials going into the furnace, and to the cinder CO&E FURNACES. 141 which, of course, has to be removed. This condition of affairs will remain as it is now until better ore can be obtained, as the ore comprises about 56 per cent. by weight of the burden, being more than the stone and the fuel together, and is subject to wider varia- tions in physical and chemical composition than cither the stone or the fuel. In discussing furnace burdens, therefore, it must be understood that we do so with some reservations. To present the matter briefly and in a general way, as be- comes the character of this -publication, and yet truth- fully as far as we shall go, is difficult. Generalizations can be accepted only with the grain of salt, and should be based on a certain set of conditions. Given these we' may derive valuable-information, but to utilize them to the best advantage one must know more than appears on the surface. It may be advisable to take up the subject first from the standpoint of the coke furnace, and then discuss, briefly, the charcoal practice. We will divide the coke practice into two main heads : 1st. Burdens composed, so far as concerns the ore, of. hard ore and soft ore, the proportion of the hard. ore rising from 48.2 per cent, to 100 per cent. 2d. Burdens composed, so far as concerns the ore, of lard ore, soft ore, and brown ore, the proportion of brown ore rising from 1 .30 to 100 per cent. 1st. Burdens composed, so far as concerns the ore, of hard ore and soft ore, the proportion of hard ore rising from 48.2 per cent, to 100 per cent. In order that the samo basis of comparison may be used, we have taken the delivery prices of the raw ma- terials as follows : Per -ton of 2,240 Ibs. Hard ore. ., 67.5 cts. per ton. Soft ore. 55.4 " Limestone. 63.4 " Coke , .$1.75 " " 142 GEOtOGICAt stBVEY OF These prices are very close to the averages for ship- ments during 1895. The table that has been prepared is based on actual furnace records, and comprises results obtained from the examination of 82,v17 charges, the amount of pig iron represented being 50,360 tons. The years selected were 1889, 1890, 1893, 1894 and 1895. The tons re- ferred to are of 2,240 Ibs. The table includes the year, the private number, the number of monthly charges, the percentage composition of the ore burden and of the total burden; the iron made, per charge, and for each month, and the percentage of foundry grades (including F. F. or 4 F., but excluding gray, forge, mottled and white); the consumption of ore, stone and coke in tons per ton of iron made ; the cost of the ore, the stone and the coke per ton of iron ; the percentage distribution of this cost ; and the pounds of coke required to make a pound of iron. The calculations have been some- what laborious but the results are extremely interest- ing and important. They do not cover as much ground as could be wished, but the pressure of other matters compelled an abridgement of the original plan. We will give a table of results from the same furnaces, consecutive months and at certain intervals. It con- tains .the results of 32,917 charges, and 50,360 tons of iron. Each horizontal line of figures represents monthly re- turns Four furnaces are represented, the ore, stone and coke being the same for any one furnace during the period, and all tons of 2,240 pounds. tftirJ ACE'S. 143 TABLE XV. ILLUSTRATIVE OF COKE FURNACE PRACTICE WITH HARD AND SOFT RED ORE. Increasing percentage of Hard Ore in Ore Burden of Hard and Soft. Delivery Prices: Hard, *& 0 1C 10 CO -* 1C ~0 1 J CO f CO 9tio!}g o ,_;__; ; os 22* 1C tO OC CD ' p.Tt? J J j oi os os CD , 2 i a o o . 1 i! CJ CO CO I CO CO ^ *QW) C5 CSJ (N o CO rr O rfi 5S* i o o c 5 - M ' IU ' x i 5^ CO ^ CM OS a^ooi- CY ? CO(:O . n o (M ! OC (M 01 9uo;g; 9.TOi <*>^ ! CSI C<] (M '"I g H -uno^ jo -i.ojaj g^S CM !M JO CM r*oj in 1 CO ^1 C 1 ^ oS.rut{Q ja c | co i>- o .33 O H 11 8!f sss 1 lOCCO 'OUOIg, lamcs 1C 5 t-r-tQO :' ^ o5 S5 c^ PH| '^s S'o^ '1JS os as r- CD OSOSCO CO 1C ! Th O OS "GC*5? TtlTT'/^ '^ t^* ^ SOXJ*toL|^|i fo QQ ^2~- II -.T9qran>T 22 ^^ ii ^-1 4 ^ 144 GEOLOGICAL SURVEY OF ALABAMA. Results from the same Furnace. Consecutive Months. 1C "*< 1C C i 1C CO*C 'l 1 T 1 ( 5 1C CO CD Q T^ 1C 1C Tfl Tf Tfl 1C ^r 1 CM 1^ C^l r- 1C TJI.IC iC J CM CM 1C CM % CM CM CM 00 t* CM CD CO O OS 00 OO 1>- CO OC CD OO OO OS OC t^OO CM 10 t^CO OS 00 OS CM si OS 00 rf( CO CO OS OOC3SIC OS r^ t C 1C !> 1 CO 1C O5 1C o i * < TP Tfl -^ -^ | CM TT" g^^ %%% 8 8 g 1C 1C OS CO O cc co co h o T CO t>- TH O5 1C 1 OS "*< C rr> rti Tfi r^- ic 01 OS OS OS CD CO I OS OO OS 2 T? CO "H 1C CM co -M' CO CO CO CO Tt< C"3 CO CO +9. 1? 01 - CM 1C CO i COr-^ . CO 1C CO "*l CD CO CM - CO 1C ^ CO CD 1C OS i * t^ OO 1 Tfl 00 OS b-i 1 OS ^ ^ -f 1C I_J5 -* T^ rti CO CO CO c3 1C CO - OC OS CO OS "^- i -H OSCOCNI i^SS^ CO S! .- OS OS OS 1C i,~ 1C CD CC CO OS g iCt- SS2 gllg CM 30 QO OO OO ^ co co co a) OS OS OS j> OO OC 00 ^ COKE FURNACES. 145 A critical examination of this table will show : 1st. The amount of ore used per ton of iron made in- creases with the per centage of hard ore in the burden, rising from 2.39 tons with ^1 per cent, to 2.52 tons with 66 per cent., and 2.78 tons with 90 per cent. 2d. The amount of limestone used per ton of iron made decreases with the increase of hard ore, falling from 0.69 ton with 51 per cent., to 0.45 ton with 66 per . cent, and 0.12 ton with 90, per cent. With 50 per cent, of hard ore in the ore burden the consumption of stone is 1545 Ibs. per ton of iron made, with 66 per cent, of hard ore it is 1008 Ibs. and with 90 per cent, of hard ore it is 269 Ibs. In one furnace for a period of three months the consumption of stone per ton of iron was 0.75 ton. 3d. The amount of coke used per ton of iron made in- creases with the increase of hard ore, rising from 1.34 ions witji 51 per cent, to 1.57 with 66 per cent, and 1.61 with 90 per cent. In the case of one furnace car- rying 50.6 per cent, hard the consumption of coke per ton of iron made for a period of three months was 1.52 tons. Coke is always the most costly ingredient of the bur- den. In the table under discussion it does not fall be- low 53 per cent, of total raw material cost per ton of iron. The tendency towards increasing consumption of coke with increasing amounts of hard ore leads, there- fore, to increase cost for raw materials to a ton of iron. The consumption of coke per ton of iron, the quality of the coke, ore and stone being the same, depends to a very great extent upon the amount of air and its pressure and temperature, which is blown into the fur- nace per unit of time. Instances are on record in Ala- bama where the consumption of coke per ton of iron 10 146 GEOLOGICAL SURVEY OF ALABAMA. with very heavy lime burdens over considerable periods did not exceed 1.25 tons, but the furnace was well equip- ped as to boilers, engines and stoves. Under such cir- cumstances it has been said by one of the best furnace- men in the Birmingham district that he could use all hard ore (of the best self-fluxing type) and make iron with 1.25 tons of coke without impairing the quality of the iron. It must, however, be said that the use of crushed hard ore tends to diminish the consumption of coke, for hard ore in large lumps is not easily penetrated by the reducing gases. When a large piaca, weighing from 50 to 75 Ibs. is exposed to the heat of the furnace in descending the outside of it is first effected. The car- bonic acid is removed, the oxide of iron begins to part with its oxygen, and processes of disintegration are set up which continue until the ore is broken into small fragments. It may be assumed that the oxide of iron is not com- pletely reduced until each piece is exposed to the deox- idizing gases. This takes place with comparative rapid- ity if the ore is porous, as with certain kinds of brown ore, or if the fragments of ore are sufficiently small. They must not be too small, else the current of gas is checked, the burden packs and the furnace "hangs." But if the size of the ore particles be small enough to allow of easy gas-penetration while not so small as to cause irregularities in the descent of the burden, we should have comparatively favorable conditions for re- duction. It would appear that the hard ore has a two- fold advantage over the soft ore, first as regards the ad- mixture of lime for making a self-fluxing ore, and second in having the lime combined with carbonic acid. The first advantage renders possible the saving of ex- traneous lime. Using 80 per cent, of hard ore and 20 COKE FURNACES. l'7 percent, of soft ore in the ore burden there are required 582 Ibs. of limestone, as against 1,680 Ibs. for 50 per cent, hard and 50 per cent, soft, a saving of 31 cents per ton of iron in favor of the heavier hard ore burden. This saving, however, may be more than counterbal- anced by the greater amount of ore and coke required in the heavier hard ore burden. It may not be possible to obtain better ore, i. e., so far as concerns its iron-con- tent, but it can be improved by crushing. Crushing does not increase the amount of iron, but it does increase the reducibility of the ore by enabling the gases from the coke to act upon a larger surface of iron-bearing ma- terial. It does more than this. It furthers the evolu- tion of the carbonic acid in the ore, and this.renders the ore more porous. Crushing and calcination have a common purpose, viz., to increase the reiucibility of the ore by increasing the amount of iron-bearing surface exposed to the reduc- ing agencies. The use of crushed hard ore is rapidly extending in Alabam i, and it will not be Ion:* before t!i3 advantages o o attending it use will force themselves upon those who seem at present to be indifferent to the matter. In a paper on "Large Furnaces on Alabama Material," (Transactions American Institute Mining Engineers, Vol. XVII, p. 141. 1839). Mr. F. W. Gordon said that the results at Ensley proved the possibility of making a pound of iron with a pound of coke. Since that time and with a better coke thai was thon used it has hap- pened for a day or so that a pound of coke mide a pound of iron, but the coke iron that has been made in the Birmingham district with a ton of coke per ton of iron is insignificent in amount, and there is no reasonable expectation that it will be increased in our day. The 148 GEOLOGICAL SURVEY of ALABAMA. present consumption for the best coke is 1.34 Ibs. per pound of iron. If any hopes were entertained as to the possibility of any one of the Ensley furnace making a pound of iron with a pound of coke even for a week at a time they must long since have been abandoned in the cold light of facts. 4th. The tendency of the percentage of foundry grades of iron is towards a decrease with the increase of hard ore. While this is not strongly accentuated still it ap- pears to be too evident to be neglected . Individual cases may be cited wherein the percentage production of foun- dry grades during a month was higher when the per- centage of hard ore rose to 80 per cent, than when it was at 52 per cent., as by numbers 34 and 20. But on the other hand when tlie ore burden was composed entirely of hard ore, as in No. 33, the percentage of foundry grades touched its lowest point, viz., 59.4. The influence of increasing amounts of hard ore 'on the quality of the iron is of the utmost importance iu the discussion of this subject. Too much stress can not be put on it, for it determines the price at which the pro- duct must be sold. The higher the percentage yield of foundry irons the more valuable is the output. Any thing, therefore, that tends to interfere with the make of foundry iron should be most carefully investigated, and conclusions drawn from authentic records must be the chief evidence. Thirteen cases have been examined, the number of charges being 82,917, and the amount of iron 50,360 tons. Three cases in which the percentagejof hard ore in the ore burden was 50.9 per cent., 50.9 per cent, and 52.3 shows the following percentages of foundry grades respectively, 99.2 per cent., 96. 2 per cent. ,'^90. 2 percent., the average being 95.2 per cent. COKE FURNACES. 149 The total number of charges was 8,853, and the total iron made 14,798 tons. Four cases in which the percentage of hard ore in the ore burden was 48.2, 50.9, 51.1, and 52.3, show percent- ages of foundry grades, respectively, 83.9, 68.3, 88.6, and 87.0, the average being 81.9. The number of charges was 11,325, aad the iron made 16,845 tons. In these cases the average percentage of frard ore in the ore burden was 50.6, as against 51.3 in the first case, while the average percentage of foundry grades was 81.9, as against 95.2. While there was a very small difference between these two cases in respect of the amount of hard ore used there was a marked dif- ference in the percentage of foundry grades made, 95.2 per cent, and 81.9 per cent. Three cases were examined in each of which the per- centage of hard ore in the ore burden was 65.9. In one of them with 1,508 charges and 2,070 tons of iron* the percentage of foundry grades was 95.7. In another with 1,343 charges and 2,61 > tons of iron the percentage of foundry grades was 87.8. In the third with 1,512 charges and 2,898 tons of iron the percentage of foun- dry grades was 93.2. The average of 4, .{63 charges and 8,483 tons of iron was, in foundry grades, 92.2 per cent. Finally, three cases were examined in which the per- centage of hard ore in the ore burden rose from 80.7 to 100. In one of these with 80.7 per cent, hard there were 1,805 charges, 3 315 tons of iron, and 93.8 per cent, of foundry grades. In another with 91.5 per cent. 3ia"d there were 1,995 charges, 3,901 tons of iron, and 83.9 per cent, foundry grades. In the third with 100 per cent, of hard there were 1,576 charges, 3,005 tons of iron , and 59.4 per cent, of foundry grades. Averaging the results from the two furnaces carrying 150 GEOLOGICAL SURVEY OF ALABAMA. about 50 per cent, of hard ore in the ore burden we find that with 20,178 charges and 31,643 tons of iron the per- centage of foundry grades was 88.5. Comparing this with the results from the furnace car- rying 65.9 per cent, of hard ore, with 4, 363 charges, 8,483 tons of iron and 92.2 per cent, foundry grades, there seems to be an advantage of 3.7 per cent, foundry grades for the higher percentage of hard ore. Taking these two together and comparing with them the results from the burden averaging 90 per cent, of hard ore there is found to be a decided falling off in the percentage of foundry grades. Perhaps ail that can now be said is that there seems to be a tendency towards inferior grades of iron when the percentage of hard ore in the ore burden passes 66. The smaller the yield of iron from the furnace the higher is the percentage of foundry grades, and this seems to be independent of the amount of hard ore carried. Out of 8 cases in which the monthly yield was between 3,900 and 5,000 tons there were 37.5 per cent, in which the yield of foundry grades fell below 87 per cent. In 5 cases in which the monthly yield was between 2,500 and 3, 500 tons there was only 1, or 20 per cent, in w^hich the per- centage of foundry grades fell below 87. Whether we may conclude from this that rapid driving on a hard ore burden tends to lower grades of iron is not quite clear. Provided that the furnace has sufficient engine power to furnish the requisite blast and stoves enough to furnish the requisite heat there does not see in to be any good reason why she should not work off on foundry grades satisfactorily, even with a very heavy hard ore burden. But to attempt to make high grade iron with hard ore (limy) burdens and insufficient blast,. or heat is apt to cause numerous disappointments. _ COKE FURNACES. 151 ORE BURDENS COMPOSED OF HARD, SOFT AND BROWN ORE, THE PRODUCTION OF BROWN RISING FROM 1.3 PER CENT. TO 100 PER CENT. The table embodies the results from 40,270 charges, and 66,653 tons of iron. The delivery prices for the raw materials are as follows, per ton of 2,240 Ibs. Hard ore 67.5 cents . Soft ore. . 55.4 " Brown ore 1.00 " Coke 1.75 " They are the same as for the table giving the results from ore burdens of hard and soft ore, except that, in addi- tion we have brown ore. They are not assumed prices but such as was actually paid in the Birmingham District during 1895. Three furnaces are represented, the ore, stone and coke being the same for any one furnace during the period. Each horizontal line of figures represents monthly returns : 152 GEOLOGICAL SURVEY OF ALABAMA. TABLE XVI ILLUSTRATIVE OF COKE FURNACE PRACTICE WITH HARD AND SOFT RED Oi(E AND BROWN ORE. Increasing Percentage of Brown Ore in Burden of Hard, Soft and Brown. Delivery Prices: Hard, 67.5 cts ; Soft, 55.4 cts. ; Brown, $1.00; Stone, 63 4 ct. ; Coke $1.75. Tons of 2, 240 Ibs. Same Furnace. Consecutive Months. iuui jo punoj J9d 95fOQ jo spunoj ?^SSfeSg oo CO CD p, +3 -^ CD CD OO o < c^i co co oo T 1 '^JOg OC 1C 1C -C 1C c: CO o o !!^::: 95100 O & 1 iC 1>- CM ' C 1C r- ( CO CO ^ti *& <*- 99- euo, g t^ GC CD CD tC CD i- 9JQ CO "^ ^ CO T 1 "^ "^T 1 cc ^^-^^^ A- Consumption : Tons per Ton of Iron. PWL CD t t^ CO (M O O CO 1C ^H 04 00 1C C o 9 M -T 1 C Tf TT 1^ CO 'X, ^ Th 01 co co co co- cc Tf t- 9UOIS co, ic -+ 1 cc' r-i t- x O4 I CD CO O "^ 'N Ci" ^3 i ! Ol O 7 ; CO 9-0 0) II C " p cr T? B CQ "a o "o -t-3 g & JO "JU90 .T9J 00 CD I-- CD 1C CM 1C CD CD CD CD CD CD CD a ' {13JOX OC )C OC 7-1 CD CS. OC CD co cc tc h- IC CD -OC O -C CD CD 1C 9S.n?i{3 ja^ Ci l-~ CD CD 1C ^+* CO CD 1C t CD CD CC5 CD CD H^OO -ti 1C CD CD C CD CO co co M co co co co o^ O5 OO CS t CD CD 9UOag CD -D. COO CD ^h- r 1 "as" o -UMoaa CD CO t- CO 1C t- t- O O O i 01 CO CO r-i t- CO CD CO 1C CO ^J^S CO O h -** "T 1 CD CD rfv '^^ ^ rV| '^l ^) '^J P-'H O * 1C ON| T 1 ~H OO t^ Tjn CD C*? >C O-l T-I -i oj 01 oi o-i c-i c-i o O! 0^ Q UMCUg CO -^ 1C ^ 00 00 OJ i i ' CO "f CO t cc Wog )C i O O CD O O CO 1C iC (C -* C 1C O-! 1C pa H ^t 1 H C CD "^ ^ T i co r- co CD oo 01 "N ^ S3 8q t^ 1C O O^l O- CO I oc -r CD CD ic co ^t< ^OCD CD CD CD oca: CO CO OC' jgqinnsyl ^^IS^^S 01 a i ^ c -f c ic >c ic ai ~ CD CD CD CD CD CD CD ^. "^ ooooaoooooacocj, iHi (rHiti IT i r-i ^ COKE FURNACES, 153 IO i-H IO rH t^ OC * CM l>- O5 O5 | i I CD CO CO IO O CM O CD O5 i O CO O O ii ITS CD 00 LO co 01 CM 10 10 10 CM CO 00 CM. Q CD i I <* O O5 OC CD XF JO 't 1 ^ T CSI CO C4 ^ o r^ o 10 a8a. A Pi o 5 S > 154 GEOLOGICAL SURVEY OF ALABAMA . TABLE XVI ILLUSTRATIVE OF COKE FURNACE PRACTICE WITH HARD AND SOFT RED ORE AND BROWN ORE. Increasing Percentage of Brown Ore in Burden of Hard, Soft and Brown, Delivery Prices: Hard, 67 5 cts. ; Soft, 55.4 cts. ; Brown, $1.00; Stone, 63.4 cts.; Coke, $1.75. Tons of 2,240 Ibs. Same Furnace. Consecutive Months. uo.ii jo punoj J9d 95JOQ jo spunoj rr rzS 10 co CQ co TP CN ^H ! p I Jj *-( 5^ j! i PH !|00 05 OC ."J CO 00 i OS OS t CD C CO CD p.">H OO &l t*- O5 CC 1C CD CD t- 10 CO lO c" hH M c o H 0) 43 02 O .^ CO ^t* h i - C^ * H I OC lO O -H OS 00 '9^[OQ 01 oo o cc rr -t' CO CO CO -*i iM O O T O O CD 1C lO 10 10 000000 i o 9.TQ CD "" CO "^ CO * OO O5 t-~ O CO CO T-J T-H r-J csi ci r i T i i 1 T-l OS OS C 1 } CO OS OS OS OS OS OS OS 05 WO! CD 1C O CD ^ CD 10 "f t^ co GC r-~ >i Ol (M T< 'M 'M OS i 9Snmod co c-5 co c^ co ib o CO Per ct. of Total Burden 8J|OQ 1C 7S CD CD t lO 00 9UO, S C4 CD O CM ^ 05 GC O 00 X O 1 -^ 7C| 4 ' Ol ^ OS UMOag CD CO CO 00 l>- 05 ^MMMi.1 05 CO 1JOS a 3 o i-H Sfe- oT O CO _o O "9< {> o Q uo.ij jo pimoj .iad a^oQ jo spunoj o aT CO "V -u en C" O O ^1 ' "'B 1 o r" - - Oi 0> P^Pn 8^00 8U01S c^ CO a.iQ EH Q OO t^ >- C t- O2 O g P- f) . ^ r 1 ^ ' to co io co ^^ iO t^ 00 CD OS 0) PH C" Q o3 O O CD C O >O O W <3J 8882SS Ofl O o C o 00000 i ft OQ PS 00 O o 0) ^L^cigci 5 fa J^ cJ ^ i.. O r ' O Q o os o o o 0) O B j>.8 o r/3 ^2 o 'ocS ' ^ i rM>- O CO CO CO CO CO CO 03 i *" o o o o o o CM "^5 r/5 (N O III 03 C IM O C rf> iO CO O O 00 OS O CC OS s I** , , S3 43 I O d> c o C d) T3 43 - CM co as co r~ r i io ex; os OC' t^ i - t~ OC t ^^* ^ sj O ! O3 r ~2 K> H o 'o cr 1 CO 1 si O CO CD OS CD 7M .5, .2 QJ f/5 "5 QQ e || CD C OS O O OS O 00 f 8 C rl ~ : CM O O CO O o> o o oc o os QO os tO CD lO tC lO QJ 43 si i Q^ /""s 1 r-, . ^ ^2 C3 'ii .CD r- oo ds i .2 00 Moquin^ -^ iO CD s- os os os f^ M oo co oo ut it by no means follows that it is the best way, or thu only way. Leaving out the content of sulphur, as not seriously affecting any of the grades above tlrajr Forge, there sliould be certain ratios established "between silicon and combined carbon for the Soft and 'Foundry irons! " The variation in "the iimoant of silicon does/ol course, influence the quality of the iron, and one might go" even farther -and allow that it influences the iron "more than any other single element. " But combined car- bon is by no means to be neglected. " In 29 complete analyse! of iron graded as No. B Fouti^ dry, I found that the silicon vailed from 1.45 to 3.83 per cent. ; , the average being 2/37 percent. Five of the sam- ples should have been graded as No. 1 Soft, as the silicon was between 3.04 and 3. 17 per cent. , and one should Have been No. 2 Soft with silicon 3.83 per cent. These ' irons were all graded on the yard "by a careful and com- petent man, .yet in 6 cases out of "29, or 20.7 per cent., the iron graded as No. 3 Foundry was really Soft. Ex- 12 GEOLOGICAL St/RVEY OF ALABAMA. eluding those six, the average silicon in the other 23 was 2 1C per cent., a result not far wrong, if at all, as No. 3 Foundry may vary from 1.90 to 2.20 per cent, of silicon. In the six cases in which the silicon was over 3 percent, the combined carbon was 1.04 por cent., and in the 23 others it was 0.82 per cent., the average of the 29 being 0.87 per cent. The combined carbon in No. 3 Foundry does not usually run as high as 0.82 ppr cent., the average being about 0.40 per. cent. In the Soft irons it should not be above 0.40 per cent., but in some cases especially when the iron resembles No. 3 Foundry, it may go to 1.00 per cent. We have then to discriminate between Soft irons with over 3 per cent, of silicon, and the normal amount of combined carbon, and irons which contain over 3 per cent, of silicon and upwards of 1 per cent, of combined carbon. Grading on fracture and appearance some of these latter irons would be put in No. 3 Foundry ; grad- ing on silicon content they would go in the Soft irons, with the understanding that the combined carbon was abnormally high. The same principle holds good in respect of the other Foundry irons, although in a less degree. It is this ten- dency of the lower grades of Foundry iron to show higher percentage of combined carbon than is usually the case that renders grading by fracture and appearance some- what uncertain. In case of doubt a silicon estimation will enaole one to decide whether or no the, iron should be put in the Soft grades, and an estimation of combined carbon will show whether or no it should be stated that this element is above the average. The multiplication of grades may go on indefinitely according as the fancied needs of consumers increase in number. ORE. 179 There was recently completed an agreement between the chief producers of Alabama coke iron whereby cer- tain uniform prices for standard grades were to be ob- served. It is a very good thing as far as it goes, but it does not go far enough, nor strike very heartily at the root of the trouble. The main point is to secure uniform grading, and this can certainly not be gained merely by establishing uni- form prices. A local trade association could take the matter in hand, but a simpler and it seems to us a more satisfactory plan would be for the companies that made the agreement as to prices to make a similar agreement as to grading, and put a competent man in charge of it. The price depends upon the grading. It is not enough for the iron-masters to meet and say what the names of the grades shall be, nor to fix the price at which the grades thus named shall be sold. Unless there is at the same time an agreement as so what kind of iron shall be classed as No. 1 Soft, or No. 3 Foundry, the agreement as to uniform prices is of little use. It is sure to happen that permission to ask a special price for a special iron will be solicited, and un- less it is known what this iron really is, what relation it bears to the grades .whose prices are already fixed and agreed upon, how can there be anything but confuv io.? One may Bay: "I an making an iron, which to all ordinary grading would be put in No. 2 Fou'iidry.' Bub ib card38 leas thai i.5) per oeiu. of silicon and is therefore not a typical No. 2 Foundry and I wish to ask a special price for' it." He has called in his chemist and knows that the iron is not No. 2 Foun- dry , although it closely resembles it in granulation, color, fracture, and face. He wishes to sell it on analysis, for this is really the gist of the whole matter. By all means let there be uniform prices, but if the 185 GEOLOGICAL SIHtVETOT ALABAMA. grading is not uniform what do the uniform prices amount to, aft^r all? They are simply grade^s-p litters, and will inevitably lead to more' confusion than -at pres- ent exists, ifjtbey are not based on the chemical analysis of $ie iron.- Some people are inclined to regard the chemical grad-- rag ! of pig if on as a sort of Panjandrum, or Mysterious Monster, lying in wait for the unwary. But no chemist who understands the situation in Alabama can declare oat and wit lor laboratory grading, as no chemist can ctoubt that the present system is out of date, illogical, aftd cumbersome . The purpose to which pig iron is put depends abso- lutely upon its composition - the color, fracture, granu- lation,, and face having nothing to do with it except in sto for as they indicate the existence of cei'tain ingredients", -vVho'se actual percentage can be deter mined only by the chemist. As regards grading the inferences t6 be drawn from data obtained on the iron yard" ar' reliable only if confirmed by laboratory tests, and are to be accepted only when they are so .confirmed. -.-.What changes are to be suggested? -First' the main- tenance of a chief grader, whose business H should be to regulate the grading under conditions imposed by the sejparate Companies. Second, th^ establishment of a oentral laboratory devoted to pig M^i analyses. Third, tfhe diminirtioTi of. the number of 'gravies and the substi- tution tferefor of aot m^tne tliaii^ix grades, differentia- ^d by the content in silicon , and combined carbon, and possibly -sulphur. These six grades might be as follows : / :;:-. - ; . >-.! '180- Silicon. Combined Carbon. Sulphur. Silvery Irons, 5 to $ 0.1^ to 0.30 0.01 to 0.04 Soft Irons, 3 to '5 0.20 to 0.60 0.01 to 0.05 Foundry Irons, 2 to 3 0.30 to 0.90 0.01 to 0.07 Gray Forge, 1 to 2 0.40 to 1.25 0.04 to 0.09 Mottled, 0.6 to 1 0.50 to 1.80 0.06 to 0.11 White. 0.1 to 033 ( 1.00 to 2.50 O.OS to 0.30 This scheme, or some ..'modification of it in line with its general provisions would retain the present nomen- clature, and bring it into closer accord with laboratory results. It would do away with five grades, which are no more than side-grades at best t and would enable the grader to exercise better discretion in the yard. The rapidity and accuracy with which the estimation of sili^ con, and combined carbon can now be made, render.it possible to have the results from the cast-house by the time the iron is ready . to break and .pile. . The estima- tion of silicon now leaves very little to be desired, and while the estimation of combined carbon in pig iro-n is not so accurate as in steel it is sufficiently so for the purpose ifiThand. If objection be made to such a rad- ical change much could be done to improve the present system without} decreasing the number of grades, or in- terfering with the nomenclature. If a systematic record of the pigs sampled were kept it would be possible to control the grading within narrower limits than now maintain. The following plan is suggested for use by graders. Have stout manila envelopes prepared, 3x6 inches in size_. On the_.fronfc have the following blank form printed, viz : 182 (jflota&iOAL fttraVair off ALABAMA, . . . .Company. .Tons. Grade No Furnace. Division Made 189 .. Sampled 189 .. (Mark out the word that does not apply.) Fracture, Regular, } Fine. Granulation, > Medium. Irregular, ; Coarse. } Smooth. Face, > Pitted. ) Blistered. Chilled edge (Signed) On the back of the envelope have the following blank form printed, viz : Charges Burden. Pounds. Hard Ore Soft Ore Brown Ore Q , n I Limestone StoQe [ Dolomite Coke Total , .. To be taken before each cast. Time. Devolutions of Engine Heat, Pressure. Average. pm i&oif. 183 Such envelopes were prepared, after consultation with Mr. Brannon, and Mr. W. J. Sleep, manager of the American Pig Iron Storage Warrant Company in this (the Birmingham) district, and were used for a consid- erable period. They answered the purpose admirably, so long as there was co-operation on the part of all the officials concerned. But while the chemist was glad to have the information, and while the grader found that it was just what he needed, in many cases the samples were either not taken at all, or the blanks were not properly filled out. Orders sent out from the general office were not obeyed and samples that should have been taken were utterly neglected. One of the annoy* ing things in connection with the study of Alabama pig iron is the curious indifference of furnace managers to the collection of systematic information in regard to their product. Most of the companies- have t)aeir own laboratories, and the chemists are alive to the impor- tance of the subject. The cost of the collection of such information as is outlined in the blanks is merely nomi- nal. The grader fills out the blanks in regard to color, etc., when the samples are taken. It is done in less than five minutes. The additional information is ob- tained from the furnace office in five minutes more. But when the general office has sufficient interest in the matter to order that the blanks should be filled out the inexplicable indifference of the furnace superintendents may block the matter. Whether it is they think they do not need the information, or whether they regard the request and the order as an unwarrantable interference with their own particular business, or both, is not in evidence. Two of the best judges of iron in the State, Mr. Bran- non and Mr. Sleep, whose daily business brings them in close contact with all kinds of iron, and upon whose 184 BOLOGUCAt StBtfEt 0# ALABAMA, judgment large suras of -money depend, are agree.d that there is urgent need of raore systematic information in the grading .of iron. The composition of heretofore well recognized grades has changed, graders need to know what these changes are .and how they may be recognized, the laboratories are well equipped and the chemists anxious to assist in every possible way the progress and success of the business. Onp of the chief officials of a large company has said : "For the enlarge- ment of .the. domestic market., the most desirable thing. to be done, in my judgment, is to secure uniformity in grading and naming iron, and selling it upon terms of uniformity. * * \ *-.*.*.. It is scarcely too much to say that the whole question of grading iron is assuming. a more complex condition." And yet the same old. absurd conglomeration continues, and graders are asked to tell at a glace the chemical composition of eleven dif- ferent kinds of iron. . Of course they can not do it,. and they should not be expected to do it. If any of these . observations apply, to the domestic. market, and in fact they all apply, with what greater force do they apply to the foreign market? Great efforts have been made to secure a foothold for Alabama iron in England or the continent during the last two years, and a gratifying degree of success has been attained.. Shipments on foreign orders for the. year 1897 approximate 220,000 tons,. and it is likely that the trade will grow, if the producers recognize the de- mands of the foreign consumers. Alabama iron going abroad has to compete with standard brands such .as Eglington, or -Clarence, or Middlesbrough No. 3, whose uniform composition has enabled the consumer to know. just what to expect. If he buys No. 3 Foundry, Alabama make, he has a right to expect that the silicon, shall not be in excess of the amount p esent in the standard brands. . The for- eign market has grown up on pretty much the same- foundation as the domestic market, viz. , cheapness. , In. spite of irregularities of composition Alabama iron has been sold in this country because it was made at a less cost and could be sold for less money than other iron.. Lack of uniformity does not distinguish all the coke iron made in Alabama, for there, are companies in the State that are very careful in grading, but there has been and is now a good deal of complaint that our irons are irregular in composition. But this has not pre- vented the development of the pig iron industry, with domestic sales, and may not prevent the further exen- sion of the foreign market. The cultivation of new markets, especially those situated at a great distance and. which have p^lv recently been compelled to look J;o us for. some of their material, can be successfully under, taken only by the exercise of the greatest care. A con sumer may for a time put up with what does not ex- actly suit him if he is buying it cheap. If forborne reason, temporary or permanent, he finds his usual supply curtailed lie must go elsewhere. The question, of cost is, of course, the main one, but the requirements, of his own market, i. e., for his own manufactured pro- ducts, must bo consulted, and he" cannot continue to buy a cheap article if he can not use it to advantage. With- in the limits of their own grades Alabama irons are known and appreciated in nearly all the States of the Union and in many foreign countries, and what is said here is not to be taken as captious criticism, for nothing could be further from the intention of the writer. But it has seemed to him that the changes which have: been slowly creeping into the grading of iron should be recognized at their full value. The names of the grades do not mean what they once meant, the names have re- GEOLOGICAL SURVEY OF ALABAMA. mained, but the composition of the irons has changed. This is no secret. It is perfectly well known to those who have given the matter even cursory attention, and the only thing to do is to act upon the common know* ledge. A system which almost every day in the year forces the yard grader to class as Soft as iron which does not carry over 2 per cent, of silicon has had its day, and should give place to a system founded on the actual chemical composition of the iron. This is true no matter whether the iron is intended for home consumption or for the foreign market, but it is particularly true for those who wish to sell their iron abroad. CHAPTER VII . THE COST OF PRODUCING PIG IRON IN ALABAMA. In January, 1894, there appeared an article in the Engineering and Mining Journal, New York, that gave the cost of making pig iron in Alabama at $6.37. The items were as follows TABLE XIX. li tons of coke @ $1.51 $1.89 2 1-5 tons of ore @ 0.67 1.48 tons of stone @ .65 . 50 Labor 1 .25 Repairs 0.50 Supplies . 50 Selling expenses . 25 Total., , $6.37 This article was unsigned and the author is at present unknpwn. The closeness with which he approximated the real cost will appear later. In June of the same year Mr. EG. Pechin, formerly editor of the Iron Trade Review, Cleveland, Ohio, pub- lished in the same Journal an article on the cost of making pig iron in Alabama, and expressed the opinion that it was then costing, at two plants, not above $6.50 per ton, and possibly less. The author of the unsigned article and Mr. Pechin were both very near the truth. It is now proposed to discuss the matter at some length and to submit figures that may be relied upon as the cost in detail of making pig iron in Alabama during 1894, 1895, and 1896. What the cost was in 1897 is an- other matter and will not be entered upon at this time. My excuse for discussing the matter must be that an erro- neous opinion seems to be current in some quarters that pig iron can be and is made here for less than $5 per ton. It is possible that some iron has been made in the State at a cpst closely approximating $5, but it is not thought that so low a cost has been possible for any length of time. During the years 1894, 1895, and 1896 the low- est cost that I am conversant with was $5.71, and I do not think that any company has maintained, for any length of time, say several months or a year, a cost account lower than this. It may be that some furnaces with exceptional conditions as to the supply of raw mate- rials may approximate this amount by the year, and even, at times, have made iron at a less cost than $5.71. In the report made by Mr. Carroll D. Wright, United States Commissioner of Labor, in 1891, as to the cost of making pig iron in this country, it is stated that, exclud- ing interest, depreciation of value of plant, and charges for freight of product to places of free delivery, the low- GEOLOGICAL SKfe^Sy^O* 1 ALA&A&A. e/sfc. cost reached in any Southern State during the year 1389-90 W.AS $9.16, At that time this was the lowest. cost reported in the entire United States. .The .details of this cost were made up. as follows. Materials : Ore . . .. .'". ...... ''.'.' :..... . .'$1 96 Limestone . . . , 324 Coke..... ... .. , .. ... 4 243 Total.. ...:../....... $6 527 Other expenditures : Labor... . ........ ..........;.... ;. . . $1 737 Officials and clerks. ..........;..... 156 Supplies and repairs . '. . 703 Taxes. . .......:..,.;.. 038 Total. ....... ... ;........... $2 634 X>rand total. .... ..... v. . ..... ... . ../.... $9 161 An Alabama furnace in operation during this period was making iron at the following cost, arranging the items as above ; Materials : Ore $2.587 Limestone 0.397 Cinder, scrap, etc 0.099 Coke 4.471 Total., . .$7.554 ~ PIG IRON. 189 'Other-expenditures: . 'r - Labor...'.../.. .... ,': $1.835 Officials and clerks. 0.178 Supplies and- -repairs. . -. . . . 0.283 Taxes ... 031 Total ....... .,,,$2.327 Grand total. .'.' 9.88 At a certain furnace plant in the State, producing in 181*0 about 140, 00.0. tons of pig iron, the cost was *as fol lows : ..;.. I Material , , $<>,&2 - . Labor... 1.86 Sundrfe* . 0.83 Total ; $9.01 .The average cost of making iron in Alabama in :i 889-90 was about $9.50, although it must be said that some furnaces made iron for about $9. During the period- of 1890-1897 the cost of making iron was -about $3 less than it was in 1890, and the low- ) esfr cost reached over any considerable period was ahowt $5.76, with a possibility that some furnaces were able to . make it fur ,ahout.J&.5j50 over a limited period. It is proposed, in, the following pages, to give detailed cost sheets of the production of a very large amount of " pig iron, and then to discpssy (briefly, the reasons for the deductions of coat within the last six or seven years. 190 GEOLOGICAL SURVEY OF ALABAMA. COST OF MAKING PIG IRON IN ALABAMA IN 1894, 1895, 1896. TABLE XX. LABOR ACCOUNT. Cast-house Cinder-yard Engines and boilers Furnace office . ... . . Iron-yard. ..... .. Laboratory ........ Lights : . .... .... . Locomotives. . . . .. Salaries Sand , ... ..:... Stables ."...' Stock-house. . , . Tracks, ... . . .. . . : ;. Water... '.-?:. ....;.. Extra, 1894. Cents. 1895. Cents. 1896. Cents. U,7 19.3 19.4 4.2 5.4 6.7 4.4 8.0 8.3 2.8 1.0 1.1 16.0 17.2 17.1 0.9 . . . . .... 0.8 08 0.8 9.9 11.0 10.2 3.3 2.6^ 2.5 0.9 0.1 ' .. ., r ;- 0.8 0.8 12 26.6 29.5 25.9 2.0 2.^ 1.2 1.1 1.4 r . : .'.. ':.' 1.0 0.4 Total 83.5 90.8 96.6 ' TABLE XXI SUPPLIES. Cast-house ........ Cinder-yard Engines and boilers Furnace office 1894. Cents. 5.5 1.4 3.1 0,8 189.5. Cents. 8.0 1.7 4.2 0.8 1896. Cents. 9.2 1.8 4.0 PIG IRON. 191 Iron-yard 0.8 0.8 0.5 Laboratory 0.2 0.4 0.5 Lights.... 0.4 0.6 0.6 Locomotives 11.0 6.0 6.8 Sand 2.5 0.3 Stables 0.3 0.4 0.3 Tracks 1.0 1.7 1.4 Stock-house 2.4 -2.4 1.3 Water.... 1.1 1.3 2.0 Extra supplies 2.4 2.3 3.0 Total 32.9 30.9 31.4 TABLE XXII. CURRENT REPAIRS. Cast house. . . 5.2 . Current repairs Cinder-yard 1.4 for 1895 and Engines and boilers 4.8 1896 taken at Iron-yard 0.6 20 cents. Locomotives 0.6 Stock-house 1.5 Tracks... 2.0 Water., , 1.0 Extra.'. '.: . '/.'.., 1.40 Total.... ..':.;.-' ...-.. 18.6 Putting these items together with the others that ap* ply to the matter we have the following : TABLE XXIII. AVERAGE COST OF PIG IRON IN 1894 1895 1896 Ore...., $1.86 $1.754 $1.716 Limestone. 0.16 0.240 0.128 Coke , . . . . 2.78 2.840 2.735 Total for materials.. , $4,800 $4.834 $4.579 GEOLOGICAL SftRVE* OF ALABAMA. Materials.'. '. ." $4.800 $4 834 $4.575 Labor . . 0.835 0.^98 0.966 Supplies . .', 0.328 0.302 31~4 Current repairs. . 0.181 0.200 0.200 G : ueral expenses. . .' 0.077 0,070 0,093 Eelining. 0.170 183 0,200 Taxes. . 0.026 0.025 0.080 Insurance . . : 0.003 0.005 0.006 Bad debts.... 0.037 033 0.030 Total. ".'.",.. ... $6.457 $6.650 $6.464 The lowest cost during 1891 was $5. 7 \', the highest was $7.8X v ancl thejwerage sslling price of No. 2 Foan* dry iron was $7.28 The lowest cost during 1895 was $5.84, the highest $7;02^ a^nd the average Celling price of No. 2 F. was $7.15. The .lowest cost during 1896 was $5.74, the highest $6:84; and 'the "average selling price of 'No. 2 F. was $7.22. The percentage distribution of the various items of cost is as follows : - " TABLE XX IV. Ore 1894. Per Cent. --..- 28.8 1895, Per Cent. 26.3 1B96, Per -Cent. 2^.6 Limestone Coke... ....>,.,; 2;5 43.1 4,0 42.6 .. 3.0 42.3 Materials Labor Supplies . Current repairs General expenses ..... Kflining Taxes. . 'Bad "debts. Total . , 100 .00 100 .0 TOO .00 PIG IRON. 19& The average cost of the raw materials during these three years was as follows, per ton, stock-house de- livery : Hard Ore. Soft Ore. Brown Ore. Limestone. Coke. 1894. . .$0.753 $0.566 $1.01 $0.605 $1.875 1895... 0675 0.535 1.09 0.634 1.758 1896... 0.672 0.572 1.07 0.647 1.727 There was also used some mill cinder at an average cost, per ton, of 75 cents, and a little blue billy at an. average cost p^r ton, of $1.71. But the proportion of these two materials was small, and the items may be neglected. We are now in a -position to compare the co^ts of these years, one with another, so as to be able to observe the course of the industry at a time when it- is likely that the costs were as low as they will be for some time to come. Unless large expenditures are made for improvements it is likely that these costs will stand, as the lowest for quite a while. The first thing that attracts our attention is the close agreement in the costs for the three years, the greatest difference being only 21 cents as between 1894 and 1895, while as between 1895 and 1896 there is a difference of only 1 cent, This close agreement may, in part, be due to the system of book-keeping employed, not, of course,, with any intention of misleading but merely to harmo- nize the costs of one year with those of another in a gen- eral way. For instance, take the years 1894 and 1895, where there was a practical identity of cost In 1896 the cost of raw materials was 24 cents less than J >94 while the labor costs were 13 cents more. The cos!} in 1894 which were in excess of those in 1896 are as follows : 394 GEOLOGICAL SURVEY OF ALABAMA. Cents. Ore 14 Stone 4 Coke 5 Supplies 1 Total 24 While those that were less in 1894 than in 1896 are as follows : Cents. Labor .. 13 Repairs 2 General Expenses 2 Relining . . 3 Taxes 5 Total 25 There is certainly a very judicious balancing of ac- counts as between these two years that at the close of 1896 it should be found that there was a difference of but one cent. It leads to the supposition that arbitrary charges have been made, based, it may be, on the ex- perience at that particular plant but liable to excessive variations. The cheapness with which the raw materials are mined and delivered in the stock-house has conditioned the building up of the industry of iron making more than any other single circumstance, perhaps more than all other circumstances combined. It is this feature of the matter that has made progress possible, for the labor costs and other expenses are not as low as they are among the chief competitors of the State in the iron market. The furnace yield of iron from the ores, taking a general average, is 41%, and it takes 2.47 tons of ore to make a ton of iron. Handicapped with such low grade ore it PIG IRON. 195 has yet been possible to assemble this ore, with the lime- stone and coke, and make iron at an expense, for raw . materials, of $4.57 over a period representing about 160,- 000 tons. 1 his would not have been possible except for the topographical and geological features of the district. If one inquires as to the future of the iron industry in the State he. can be best answered not by referring to what has been done, but by judicious investigations into the possibilities of continued cheap raw materials. On the average the percent of the total cost of making iron in Alabama borne by the raw materials during the three years we have selected was 72.7. During the census year 1889-90 it was about 74%, so that there has been of recent years a reduction in this most important item of 1.3%, and as much as 4% if we take the year 1896 as the criterion. In labor costs there has been a percent- age reduction of 5%, as the labor cost, 7 to 8 years ago, was over 19% of the total cost, while during the period 1894 to and including 1896 it was 14.3%. The saving in labor has been nearly four times as much as the sav- ing in raw materials. Realizing that the great advant- age given by cheap raw materials was not the only fac- tor in maintaining a position in the iron market, the producers of iron in Alabama set themselves to reduce the cost of converting these materials into pig iron, and -as the labor cost was and is the most important after the materials strenuous efforts were made to diminish it. It may be possible, by introducing mechanical appliances around the furnaces, not only in the stock-house but also in the cast-house, to bring down the labor cost by twenty-five cents per ton of iron, so that it would not exceed, let us say, 60 cents. But this implies the ex- penditure of large sums of money and more care in the preparation of the stock before it reaches the furnace. The advantage of cheap raw material, we must remem- 196 GEOLOGICAL SURVEY OF ALABAMA. ber, is one that is apt to create a false sense of security. Relying too much upon what nature has done leads one to neglect doing what he should do. Then too, cheap materials with an enormous drain upon them all the while, putting nothing back while taking a vast deal out, after so long a time and the time may not be so very long, after all fail to respond to the demands made upon them. Their inevitable tendency is to be- come dearer as they become scarcer. To counterbalance this tendency there must be economies in other direc- tions, such for instance, as a more scientific and less wasteful system of mining, reductions in freight charges on raw materials, ownership and direct working of the mines and quarries, and, particularly, improve- ments at the furnaces for handling stock and products. Whether or no we have already seen the cheapest pig iron in Alabama is a question for the future to decide. I do not propose to enter upon it at present except to say that there has been too much reliance placed upon the (Cheapness with which 'materials*have been assembled and a great deal too little upon improved methods. When cheap materials become dearer and no improvements have been made in other directions then the cost of mak- ing iron in Alabama will begin to increase, and many of the advantages she now enjoys will be lost. The constant drain that has been made upon tne so- called soft red ores, i. e. the ores that carry from 46 to 48 per cent, of iron with less than 1 p^r cent of lime and that can be delivered in the stock-house for 55 cents per ton, has already made itself felt. There is very little of such ore now left within easy reach of Birmingham and the furnace practice is in a state of transition. From this time on the ore mixture will be made up more large- ly of the limy ores and the brown ores (limonites.) PIG IRON. 197 "There are some furnaces of exceptional situation, as for instance at Ironaton and Shelby and possibly at Shef- field that can secure brown ore in the stock-house for 60 to 80 cents per ton, but this is by no means the general situation. The average co-t of brown ore at the stock- house is close to $1, if indeed it be not nearer to $ i .10. With hard ore at 70 cents and brown ore at $1 a mix- ture of 20 per cent brown and 80 per cent hard would cost per ton of iron, $1.91, taking the iron in the hard ore at 37 per cent, and in the brown ore at 50 per cent. The average cost of the ore mixture, with varying pro- portions of hard, soft and brown ore, during the y<-ars 1894-96, was $1.77, a difference of 14 cents per toa against the hard-brown mixture. This disadvantage may, of course, be counterbalanced by using less limestone, but it may well be that more coke will have to be used, so that the difference is not likely to be less than 14 cents per ton and m~y be more. But the furnace practice, with progressive exclusion of the soft ore, has not been sufficiently extended as yet to permit a positive opinion, and the matter must await further developments. 1 he acquirement of limestone and dolomite quarries by the furnace companies and the direct working of them, without royalties, or profits to the contractors, has al- ready resulted in notable economies in respect of fluxes. The development of the by-product system of coking, with the result of giving cheaper cok , is also a most promising outcome of recent months* in the Birmingham, district. In connection with the blast furnaces of the Tennes- see Coal, Iron and R.-iilway Company, at Ensley, near Birmingham, the Soivay Process Company is erecting 120 Semet-Solvay ovens, and expect to have them in operation by the close of 1898. The ordinary bee- 198 GEOLOGICAL SURVEY OF ALABAMA. hive coke is being improved, and we may, I think, expect that its quality will be still further insist- ed upon by those who buy in the open market. What ever the future may hold for the State in respect of the cost of making iron, one thing appears to be certain, viz. that the most rigid economies and the very best practice will be required to maintain as low a cost account as has been reached during the last few years. The development of the home market for pig iron, while not a factor of its cost, is yet of no little import- ance as affecting the future of the industry. The capac- ity of the rolling mills now built in the State is 183,300 tons per annum. For this amount must be substracted 19,200 tons representing the capacity of mills which, in all likelihood, will not be in operation again. This leaves 164,100 tons as the total annual capacity of the mills that may be counted upon as consumers of pig iron. The pipe works making gas, water and soil pipe have a total annual capacity of 21,000 tons. If we allow 25,000 tons a year for axles, mine and car wheels, and iron used in the construction of railroad cars, &c. &c., we shall have 210, 100 tons as the total annual capacity of the mills, pipe works, &c. To this may be added 23,000 tons as the annual capacity of the steel works now built. The grand total, therefore, is 233.000 tons per annum, and represents the amount of pig iron that can be work- ed up in the establishments in the State. But it is not likely that the amount of domestic pig iron so used is above 175,000 tons annually, or a little over 18 per cent, of the annual production of pig iron, and I am inclined to take it at not over 15 per cent., or about 142,000 tons. In the State there are 7 rolling mills, 2 steel works, 2 bridge works, 7 pipe works, 2 car axle works, and 4 car wheel works to use up nearly a million tons of pig iron. This statement does not include the foundries, but even PIG IRON. 199 with these included the capacity for finished goods does not reach 20 per cent, of the production of crude iron. The ^tate needs more extensive and better equipped foundries, machine shops, and other establishments for using what is made at home. From the Birmingham district alone there were shipped in 1897 749,065 tons of pig iron. During the year 2 8,633 tons were exported^ as against 65,000 tons in 1896. The State exported 1.5 times as much pig iron as was used within her own bor- ders. The home consumption has not kept pace will the home production, and the developments of the last ten or fifteen years have been in the direction of crude iron and not in that of finished goods. With respect to pig iron and its products the State is pretty much in the condition in which the Southern States were a few years ago with respect to cotton and cotton mills. There has been a great awakening with respect to cotton, why not with respect to | ig iron? These two products, the one natural and thf other manufactured, represent the crud- est of crude mat* ri ils, for neither can bo utilized in the economic arts until it is transformed into something else, the cotton into c< tton goods, the pig iron into castings, wrought iron, and steel. Unless there is a great change in the consumption of pig iron we shall continue to be hewers of wood and drawers of water for those whose in- telligence is no greater but whose f jrosight is keener than our own. 200 GEOLOGICAL SURVEY OF ALABAMA, CHAPTER VIII. COAL AND COAL WASHING. According to Dr. Eugene A. Smith, State Geologist, the area of the several coal fields of the State is as fol- lows ; in square miles : Oahaba 400 Coosa 150 Warrior ' 7800 Total 8,350 By far the greater amount of coal is mined in the Warrior Field, the chief operations being in the coun- ties of Jefferson, Walker and Tuscaloosa, in the order of prominence. In Bibb county the mines in and around Blocton furnished last year (1897) 671,077 tons. Adding to this amount the 84,673 tons mined in Shelby 'County, \ve have a total of 755,850 tons to be credited to the Cahaba field, or about 13 per cent, of the total pro- duction. The Coosa field produced 67,-584 tons, or about 1 per cent, while the Warrior field produced '5 ; 0'24,031 tons, or more than 85 per cent, of the total output. At present, and it may be for many years to come, the Warrior field is and will be the great source of the coal mined in the State. Its area is very much greater than that of the other two combined, the coal is certainly as good and the facilities for mining and trans- porting it are better than in the other fields. The best and largest seams of coking coal are in the Warrior field, but for steam and domestic purposes the War- rior coals are no better than those from the Cahaba field. Some of the Coosa coals are also well adapted GOAL AND COAL WASHING. 201 for coking, steam, and domestic use, but they have not as yet come much into market. The following tables, taken from the reports of Mr. E. W. Parker, Statistician of the Department of the In- terior, will exhibit the condition of the coal industry in Alabama, during recent years. 202 GEOLOGICAL SURVEY OP ALABAMA. X M PQ oo 00 TH e 2 4-4 CO .2 ** 3 - o s -u> o D xp i -j uc CD UC 'oo" CO CO CO CO co i-T -H t^ T^ !>.-* ^-co^ coco CO t woosooccocMO ID - i i Q O '-^'^'OO O- r- 1 GO 1 C Q 55 l^gS^^sS : C5^iOCD 00 . O T-" co' O5 Tf O 1 'J^ O5 7^ co" * 05"-^^ oo co 05 . 05 CN CN CD O h-^ . "^t 1 '"^ r ~^ . Tf 05 CCOO . w o _ _ 1.44 1 73 1 75 1.50 Tuscaloosa .... Walker 1.05 1.00 1.03 1.03 1.07 1 02 1.05 98 1.06 0.8" 97 0.90 1.13 0.85 93 0.79 Gen average. . . 1.03 1.07 1 05 0.99 9o 90 0.90 0.88 204 GEOLOGICAL SURVEY OP ALABAMA. s 0, I P4 o O cS r O fi ce ^ PH a s- O ,0 ce Stat pa^o[di9 ueui jo paA'o[duid uetu jo SA'Bp uauijo pa^o^diue uaui jo 9SKJ9AY 9SRJ9AV iO < i CC CO CD tO CR C^ CC 0 CD CX3 ^1 ' )O CO (M c CO CM CO i CM O ^ O l>- iO CM l>- O O l C CD lO O tC CM CM CM T-I c>4 COAL AND COAL WASHING. 205 The number of mines reported in the coal producing counties in 1896 and 1897 was as follows : TABLE XXVIII Giving the Number of Mines in the Coal Producing Counties, in 1896 and 1897. COUNTIES. Number c f Mines. 1896 1897 Bibb 5 6 Blount 1 1 Cullman . . . . 1 Etowah 1 1 Jefferson 32 40 St. Glair 2 2 Shelby 5 7 Tuscaloosa . . 6 6 Walker 26 23 Winston . .... 1 2 Total . 80 86 206 GEOLOGICAL SURVEY OF ALABAMA. XI 05 .0 tf! I 0, p 00 ) O CD t ilTi CM CM CD 00 ** 'N OO CO t^ "M 'f I-- CO CO C-J CO O CM CO O5 t^ CO O CO ^ 10 o'jo os oT 1 otTod ifli "* o od'oTo j^co io~ co icTco'i-TVcocifoo c-fcM o Oo 1C Co C^ CO CD lO CD ^** CO CO ^ i CO C"! OC IQ CO O5 CO T-I CO CM CO Tf (N rH C ; ; ; ; ; il ! : : : : COAL AND COAL WASHING, 207 1 "~ I GC O^ Ci C^J O ^ T-l ?5 -lOC"Jr-l O5 -it 1 '^ l OO I'' *ft C^Ot-rHl-GClOT-iICOCOO I CO CD CO^-< C-l t C^OC" OS ^o^ 1 o > .0 nd a c3 1 1 OQ ;2 : J ^o^^^ t-- oo^ 1 ! ^ i c^ co o CSJ . . *tf 00 CO . . CD 00 ^H CO H-i CD T3 X '$ S x g Ix! o K? 3 CD O O i -(MO CD Oi -OOl * 00 CO C^ T i C^l Oi CO iH i^oo : of ^'3~ r 1 il * 6 W >> 00 51 < X Q ft g -< O ~ oo i i r as rf co 00 OrH t-O CS COCO CD .:_ (M OO C^ O_ '^C^OS o"cb~ 1 co" GcTaTirf CD i . rH CC ^O 10 CO 1 CD _ O EH 00 -g r-i : '.'.'.'..'. CD ^ O O :::,:: r cj jd ce r I r/2 C3 O *-! H S JZ5 b o -4_3 3 a, -j > 3 O .... ^ 5 -S WD g '? s ^Q^J^JJ-J I "3<*'"<3.S IS ^-< 3 -^ D4J^ "^ t>- t>- PQ PQ O W *~J OQ OQ E^ r* 1^ COAL AND COAL WASHING. 218 The total number of mines was 80. The 70 mines report- ing in the State in 1896 worked 14,814 days, an average of 211.6 days per mine. The highest number of days re- corded was 312 in Jefferson County, and the lowest 42 in St. Clair County. Five (71 per cent.,) mines worked more than 300 days, 19 (27.1 per cent.) worked between 250 and 300 days, 23 (33 per oent.) worked between 200 and 250 days, 14 (20 cent.) worked between 150 and 200 days, 5 (7.1 per cent.) worked between 100 and 150 days, while 4 (5.7 pei' cent.) worked lea* thai 100 days. The number of days worked, by counties, as given in the above table, is obtained by dividing the total num- ber of days reported from each county by the total num- ber of mines making the returns. It is not altogether fair to the mines working a considerable number of days to group them with mines working irregularly, or with small mines. Thus, by the table, Blount County has to its credit 273 days, but produced only 32,760 tons, while Jefferson County, producing 3,729,719 tons has 238 days to its credit. Perhaps a better insight into the business would be gained by dividing the total amount of coal credited to each county by the number of days worked . Proceeding in this manner we have the following table, giving the amount of coal produced per day in the va- rious counties during the year 1896. TABLE XXXTI. Giving the amount of coal mined per working day per county in 1896 : 214 GEOLOGICAL SURVEY OF ALABAMA. TONS. Bibb ...... 3,025 Blount. . ... ... , , . 320 Etowah 13 Jefferson. . . . ,..,;.. 15,671 St. Glair ? ...... 211 Shelby.. -... . . 358- Tuscaloosa . 929 Walker... .. . . 5,382 Winston 13 Total. . , . , 25,722 Dividing these figures, in turn, by the number of mines reported will give a general average of the ton- nage output per day per miner per county in 1896. : TABLE XXXIII. Giving a general average of the tonnage per day per 1t1i* T"\\T f\ f\ 11 t*\ 4~ TT T f% "I Qdi miner per county in, 1896. Bibb... ., ,. 3.56 Blount 3.00 Etowah 1.00 Jefferson . . 4.14 St. Glair . . . . . . 2.24 Shelby 1.66 Tuscaloosa 2.25 Walker 4.02 Winston 0.30 Working the 8 ft. seam at the Blue Creek Mines, Jef- ferson, Co., 504 miners working 275 days produced, in COAL AND COAL WASHING. 215 1896, 662,295 tons of coal, an average of 4.77 tons per day per miner. On the 4 ft. seam at Pratt mines, Jefferson County, 338 miners secured 354,084 tons in 270 days, an average of 3.88 tons per day per miner. On the thinner seams in the northern part of Jeffer- son Co., averaging 2i ft., 273 men secured 93,343 tons in 196 days, an average of 1.74 tons per day per miner. The amount of coal obtained per day per miner does not altogether depend upon the thickness of the seam. There are other circumstances as well, for instance the quality of the coal itself, its surroundings as regards ease of mining, whether it has to be blasted down, or can be under cut and wedged down, etc., etc. It does not follow because of the thickness of the seam that the miners make better wages, for the thicker the seam, other things being equal, the less is the rate paid per ton for mining. According to the report of Mr. James D. Hillhouse, State Mine Inspector, the following was the output of coal and coke, and the number of coke ovens in Alabama in* 1897, by counties and by classification. TABLE XXXIV. Giving the output of coal, and coke, and the number of coke ovens in 1897, by counties and by classification ; also the number of days worked. The total number of m^n employed was 7,743 miners; 2,270 inside day men, and 1,088 outside day men, a total of 11,101, as against a total of 9,894 in 1896, and 9,766 in 1895. The total number of mines was 86. GEOLOGICAL SURVEY OF ALABAMA. > X XI SUO lonpojj 8UOAO suox jo CD : :S 5S t CO > I CO GO CO CO CO CO t^- CN O^ I OO co" i-T I 10" C5 CO CO O > aj *S a8 j^ *>""" 'JS'j^tcZS ** *3Jl^i O I CO Oi O CO OOC5O1CO(M^1OCO O5OiO5O5 CCCOCX3OO 222 GEOLOGICAL SURVEY OF ALABAMA. In 1891, of the 2,144,277 tons of coal made into coke only 0.4 per cent, was washed slack, i. e., of every 100 tons of coal sent to the ovens less than one-half a ton was washed slack. In 1893 there was fifty times as much washed slack used for coke as in 1891, and in 1895 more than 140 times as much as in 1891. There was a remarkable increase as between 1892 and 1893, viz. : from 4.3 per cent, to 21.1 per cent., as also be- tween 1893 and 1894, viz. : from 21.1 per cent, to 43.1. From 1894 on the increase in the use of washed slack has not been so marked as in the previous years. The use of washed slack enables the mine owners to avail themselves of what would otherwise be of little value, and to make a better coke of this material than is made of run of mines coal. Results of washing slack coal from the Pratt seam. Amount represented about 5,000 tons. COAL WASHING. 223 d CO sC O& ,rH ^ gs -Cr-H paqSBM CO CD ; ^jg 9 *"* 1 = O O O O* CO CD rH GO M ^ 1? d tO CD CO OS ^^ paqs^un t^ QO C-1 OO < I CN 1C r- 1 o a, 00 % S ' '. d ^^ ^^ ^^ OS CD CN CN CN cc a; OOCN iO r-i -dv^i "P8USBM t^ *M rH CO rH Q ^j a; CO h- &fo CO CO H c ^ Ti CN a W (M O O O t^ OO CD CD t- CO .-j ^ d ^i IO CO CN - H .s paqsimun 35 O t^ i i -H CO iO "- ^ ^ i 3 lO CD ' r N 1 c O O O OS M . SCO to 3 paqgBAV CO ^H CN CD rH i CO CD ,-J 2 O OO a ge M M CO 2 o po^un O OO tO CN rH t--CD lO OS CO t^rH ~* CN O rH S r2 d O rH 00 iO O CO QC rH JH c O O O O rH COOO CN 02 O CO CO CO rH Tfi r- ( O 4 ,! co paqsA\ CO -H OS CD i 1 -. CO 10 tf ^ C 1 Cn ^ H 0- I^ r^ ' 2 o 00^ ffl k>l H QJ c kxl fc W o H ^ paqsuMiin CO CN O OS rH rH CO CO 1 g g^ CO 'CN 2 3 w^ paq 8 AV CN -CDr^ W O j; i ; |oos CH -5 3 - <{ _i 0) H 2 o J3 paqsBAMifj 00 CN OS.5 O OJ ICO r-J -H (M CD ^ d p O CN O CO 00 O o i M u C 8OO OS -H O5 CN PH : i^ScN M &^ 'P81{8B^ H CN O t- -H n co co CO c o COOSCN iQ 3 _ t> 1 ' ^ 2 o T-H ocor- S- OQ H.S "paqstj v\ufj OS CN OO -H rH D c '5 ; OS r^\ O ^ " f t- OS CD . CN CN "no M CO CO 3 pa M88AV 00 O OSOS rH o ** aj o *^ s M|^| 224 GEOLOGICAL SURVEY OF ALABAMA. Disregarding the changes in the volatile matter and fixed carbon as not affecting the efficiency of the wash- ing as much as the reduction of the ash and the sulphur r some important deductions may be derived from an ex- amination of these tables. The Robinson washer does not size its materials ; everything through a If inch screen, for instance, goes direct to the washer, and no attempt at sizing is made. The above sizes of coal were obtained by using hand-screens, but they were not sent to the washer by separate sizes. Of the material going into the washer 27 per cent, passed a % inch screen. 16 " " K inch " but was retained by a % in. screen 10 " " %inch " " " Kin. " 16 " " V 2 ' inch " " " %in. " 15 " " %inch " " " Kin. " 9 " 1-inch " " " %in. 7 " " l^inch " " " lin. " Calculating the average ash from the ash in each separate size we find it to be 11.90 per cent. This was the ash in the slack going into the washer. Of the ma- terial coming from the washer, excluding the refuse slate, sludge, etc. 28 per cent, passed a % inch screen. 21 " -K inch " but was retained by a % in. screen . 10 " " %inch " " Min. " 13 " " ^inch "' " " %'m. " 13 " " % inch " " " Y z in. " 7 " " linch. " " " Kin. " 5 " .-" l^inch " " " lin, " Calculating the average as before we find it to be 7.4/ per cent., and the reduction of the ash is 37.23 per cent. That is, this slack lost 37.23 per cent, of its ash by being washed, a result somewhat lower than is ob- tained by considering the slack as a whole without re- COAL WASTING. 225 gard to the ash in the separate sizes. Using the same method for calculating the sulphur in the unwashed slack we find it to be 1.65 per cent., and in the washed slack 1.35 per cent. The sulphur, therefore, was reduced by 18.18 per cent. In other words 100 parts of ash in the unwashed slack become 62.77 parts in the washed slack, and 100 parts of sulphur in the unwashed slack become 81.82 parts in the washed slack. 15 226 GEOLOGICAL SURVEY OF ALABAMA. rt o C C - o g 3 0} r/3 ^ & ~~xT o > a 02 CD C/3 3 H.5 do CC I A !o ob _JL_|I. O 03 :- ^d a u C3 I >O ~" a2 COAL WASHING. 227 The analysis of th.6 sludge, corresponding to. these re- sults, was Per cent. Volatile matter. . . : . 24.73 Fixed carbon 44.14 Ash 31.13 100.00 Sulphur 4.12 The pmvst coal that could be picked out, by hand, from the coal here in discussion, had the following com- position : Per cent. Volatile matter 33.00 Fixed carbon 64.60 Ash.. 2.40 100.00 Sulphur 1 .25 In washing operations it is, however, impracticable, if not impossible, to obtain coal of this degree o*f purity. Owing to loss of coal, there is a point beyond which it is impracticable, to reduce the ash. This point varies with each coal, and to some extent also with the purpose for which the washed coal is intended. In this State only the slack coal is washed, and practically all the washed slack is made into coke. Reverting to the state tnent already made that all of the material through a If-inch screen is called slack, and is sent to the Robinsod-Ramsey washer without further sizing, the question is : to what point shall the ash in the washed coal be brought in order that the washing may be considered satisfactory? There are three elements entering into this question : 1. The amount of ash in the original slack. 228 GEOLOGICAL SURVEY OF ALABAMA. 2. The waste of coal in the operation. 3. The demand of the furnaces for a superior coke. The maximum amount of ash to be left in the washed slack depends to a great extent upon the demands of the blast furnaces and foundries for coke, for if the demand is active and prices good the waste in the washing is not of so much importance. It is always important, and should be carefully looked after, but there are times* when its importance is greater than at others. Consider- ing all the elements entering into the question, the- amount of ash to be left in the washed slack, whatever it may be, is to be termed "fixed" ash, and the differ- ence between this and the total ash in the unwashed slack is removable ash. For instance, if the ash in the unwashed slack is 11.90 per cent., and the ash in the washed slack is 7.47 per cent., we may regard this lat- ter as the fixed ash, and 4.43 per cent, is the removable ash . But in this particular case the reduction of the ash from 11.90 percent, to 7.47 per cent, was not as good work as should have been done. With coal of this nature the ash should be reduced to 6.75 per cent, instead of 7.47 per cent., for the coke should not carry over 10 per cent, of ash. The best results with this particular coal were to re- duce the average ash by 43 per cent., and the sulphur by 26 per cent. , taking the records over considerable periods. The four following analyses represent about the best prac- tice on the large scale, using unwashed slack, and the- Robinson-Ramsay washer. For convenience of compar- ison the average composition of the unwashed slack is> also given : COAL WASHING. 229 UNWASHED SLACK DRY. Per cent. Volatile matter 30.06 Fixed Carbon 58.04 Ash 11.90 100.00 Sulphur 2.40 WASHED SLACK DRY. 123 Per cent. Per cent. Per cent. Volatile matter.. . 32.43 32.46 32.55 Fixed carbon 60.91 60.86 60.64 Ash.. 6.66 6.68 6.81 100.00 100.00 100.00 Sulphur 1.91 1.89 1.93 The reduction of the ash was 43.5 per cent., and of the sulphur 20.4 per cent. The yield of 48-hour coke, over a H-inch fork, from this washed slack was 58.78 per cent., or from 5 tons of coal 2.94 tons of coke. There may be instances in which the Robinson-Ramsay washer, on coal of the kind herein described, has done, perhaps, somewhat better work than this, but it is not thought that under average conditions the results are any better than these. 230 GEOLOGICAL SURVEY OF ALABAMA. c CD O g bc^ CO CO O CO CO 3 S ~ _1 gj 6 o 4-3 O 6 00 \ \.Tf (D Th ^ *o M 00 C*T\ cu CD H O PH j^ CO 10 CO 1 1 o C a* 3 C 2 1 TH O co" X o -t- CO rH d a g to _j CO H o c o FH CO *o O \" co 8 M O __ PH d o o rH CD PH a JH rH 43 c ^ ^ i rH 'o 'bC^' N o o CO 06 O CO O cc - 'o CO /} -jj * s f c 5 -1- 1 6 30 o 6 8 ^ Js ^ OH CO P^ -2 PH w ^ ^j CD CO CQ ^3 a* 1 C3S 1 co o CD OJ GO ,- H OH 3 C a ^ | ^ 5 !H CD PH O CO s p O CD P^ X (C g 6 oc TJ g CD O " 3 CD CD t> "co id CD IL, CD t> O 1 C !at T ~^ O O /v* c c O ^J s o o CD tf PH o c3 T 6 \00 (N 6 I O O CD PH co\ O ^ 00 i CO ^ co i I? ^ CC ^ CO < -i* t 3 J3 jg 4A o o +a O -u O ^ . X .^ 00 JJJ 1) CD CD \^ PH PH COAL WASHING. 231 The Robinson-Ramsay washer does very well on slack in which there is little or no bone coal, and where the dif- ference bt-tween the specific gravity of tne coal and the slate is considerable. For instance, the average specific gravity of the refuse slate, as from the above tables, is 1.71, the highest being 2.2 in the material through i- inch and left on -J-inch screen, the low-^r. bein^ 1.42 in the material through -J inch screen. Th sp -ific gravity of the pure slate, without intermixture of coal, may be* taken at 2.40, but there is very little such material in the unwashed slack, for the refuse slate of highest spe cific gravity. 2.21, had with it 4 per cent, of coal which carried 8.9 per cent of ash. In washing coal it is not so much a question of remov- ing the pure sl.-ite from p,ire coal, because- this can al- ways be done, a? of separating si * to-carrying coal from coal of a lire, iter or iess degree of puriry. The question as to wh it is coal, is nor g'-ivral, but special, and has to be answered in the light of each individual caso. With the coal urn er < iscussion, and with this washer, the writer is inclined t') think that material above 1.35 sp. gr. cannot \v 11 be considered coal, for the lowest ash in coal recovere 1 from refuse ^late by a solution of this specific grav:ty was 8.70 per cent. Perhaps the limit should not be above 1.30. Taking it at 1.30, and the specific gravity of the refuse slate, with coal attached to it, at 1.71, the difference in specific gravity is 0.41. With this difference, and with this particular coal, this washer may be depended upon to handle a large amount of slack ev ry day, and to do this work very well. But it is not designed to treat coal in which the specific gravity of the impurities, such as bone coal, etc., ap- proaches that of the coal itself. Possibly if the slack were properly sized, and each size sent to its own washer, better results could be obtained- 232 GEOLOGICAL SURVEY OFALABAMA. What was said to be the Luhrig system was introduced into the Birmingham district in 1890-91 , but the machines were neither properly constructed nor properly managed, and the washing operations soon came to nothing. It is much to be regretted that this was the case, for when the Luhrig system is designed after a study of the coa itself, there is no better coal washing system. In Ala- bama, however, only two systems are in use, on a large scale, the Robinson-Ramsay and the Stein. The Camp- bell tables have been introduced to work some of the Walker county coals, and have given fair results. Until the fall of 1897 the Standard Coal & Coke Co., Brook- wood, Tuscaloosa county, Alabama, was the only estab- lishment using the Stein washer, but the Jefferson Coal & Railway Co., Lewisburg, Jefferson county, has recently had built, under the personal directon of Mr. Stein him- self, a very complete washing plant of a capacity of 40 tons an hour. At Brook wood the Stein washer has given excellent results. It is to be regretted that no detailed investigations of the washing operations there are ac- cessible. In the proceedings of the Alabama Industrial and Scientific Society, Volume VI., Part I., Mr. F. M. Jackson said in regard to THE STEIN WASHER *. "It has enabled the Standard Coal Company to pro- duce a coke of uniform quality and of extraordinary structure, the average analysis of which invariably runs below 10 per cent, of ash and 1 per cent of sulphur. The analysis for the last six months shows the average ash to be 8.80 per cent., and. sulphur 0.74 per cent., whereas the. coke formerly carried as much as 18 per cent., and never under 13 to J 5 per cent., with sulphur 1.50 per cent, to 1.75 per cent. The loss in washing is from 6i per cent, to 9 per cent, of the weight of the slack, and the loss in coal is never over 3 per cent, under COAL WSAHING. 233 ordinary conditions, and often is as low as 2 per cent." Mr. Jackson refers to Mr. John Fulton's book on coke for further information in regard to the Stein washer at Brookwood. From this authority we learn that it was the first of its kind erected in the United States, having been built in 1890, and has a daily capacity (10 hours) of 500 tons. The following analyses, taken from Ful- ton, show the reduction in a&h : Unwashed Washed r? -L Cl \Jflt -H iC~ duction of ash Coal. Coal. Coke. in coal. 15.32 8.15 10.10 46.9 14.10 7.50 9.50 46.9 15 07 6.50 56.8 20.83 8.10 10.50 61.3 17 . 18 7.60 10^50 55.5 16. -8 6.50 9.27 60.2 20.90 5.50 73.5 17.37 5.40 69.0 18.63 7 15 61.7 21.12 4.81 6.10 77.5 Average ... 17 . 69 6 . 72 9 . 33 60 . 93 This is certainly an excellent record. So far as con- cerns the making of coke from this coal the information is satisfactory, but inasmuch as the Stein system is based on the sizing of the coal before it goes to the jigs, it would have been more complete had the efficiency of the washing as referred to each size been given. We must, therefore, in the absence of specific data infer that these results are averaged from the separate results, and yet the variation in the efficiency of the washing forbids this assumption, for the removal of the ash va- ries from 46.9 per cent, to 77.5 percent. The Robinson- Ramsay washer, on some coals, removes from certain sizes as high as 72.65 per cent, of the ash, but does not reach anything like so high a result, considering all the <;oal that goes into it. One may be permitted to doubt if the results at Brookwood represent all the sizes of ooal. For instance, a certain coal had 21.12 per cent. 234 GEOLOGICAL SURVEY OF ALABAMA. of ash before washing and 4.81 per cent, after washing,. a reduction of the ash of 77 .^ per cent., and the ash in the coke was 6.10 per cent. But one would like to know what size this was, and what proportion this particular size bore, |n weight, to the total amount of slack sent to the jigs. Looked at from the standpoint of actual results, certainly these figures leave but little to be de- sired, and this, after all, is the main consideration. To remove 77.5 per cent, of ash from coal carrying 21.12 p^r cent, is certainly good work, but one cannot refrain from asking why this result was not reached with coal of 14.10 per cent, ash? If 77.5 per cent, of ash were removed from this kind of coal the resulting coal would carry only 3.18 per cent., instead of 7.50 per cent., which it did carry under a removal of 46.9 per cent. Two facts standout prominently from these analyses, viz.: the best results were from the dirtiest coal, and that from a coal practically useless for coke-making there was obtained a coal that makes excellent coke. There are two points of view in coal washing opera- tions practical and scientific and to some it might appear that if the practical results are satisfactory the scientific aspect of the matter may be left to take care- of itself. But it will generally be found that the best practical results are reached by the aid of the best scien- tific information, and that there is a very real and a very vital connection between good practice and good theory. A careful study of what is going on very often leads to improvements ; and from an examination of what is- done we come to a decision as to what should be done.. At the Florida meeting of the American Institute Mining Engineers, 1895, Mr. J. J. Ormsbee had a paper entitled "Notes on a Southern Coal-Washing Plant." The analyses and tables in that paper were made by myself, but it is not necessary to repeat them here, as. COAL WASHING. 235 the foregoing analyses and tables were made (also by myself) in 1896, and show all that is required in a dis- cussion of this kind. Almost the whole of Mr. Orms- bee's paper is taken up by material which was furnished by myself. In 1890, when this State was visited by the British and German iron-masters Mr. Jeremiah Head, of Eng- land, was in the Birmingham district, and again in 1894, with his son, Mr. A. P. Head. In 1897 the elder Mr. Head published in the Transactions of The Feder- ated Institution of Mining Engineers, Newcastle- Upon-Tyne, the results of his observations in the prin- cipal coal districts of the Southern States. ' What was said in regard to Alabama is reproduced here. The analyses he quoted are omitted, for the reason that those already given in these pages are sufficient to enable one to judge of the quality of the coal and coke in the State. Mr. Head's long familiarity with such matters, his openness of mind, and frank way of speaking render his remarks extremely interesting and important. His inference that the labor troubles of 1894 were in any wise connected witli the employment of negroes in the mines is a mistake. It was not a question of negro labor but of the recognition of the Labor Unions. The effort was made to prevent not only the negroes but non-union white miners as well from working at the wages offered. Mr. Head's remarks are as follows: Birmingham District.- We now come to the important coal-fields in the State of Alabama, of which the city of Birmingham is the focus, and to which, to a great ex- tent it owes its existence ; as also does the neighbouring 23G GEOLOGICAL SURVEY OF ALABAMA. city of Bessemer, and several others. The principal Alabama coal-fields are : Square Miles. Warrior, estimated to extend over.. . 7,8.00 Cahaba " " ... 400 Coosa " " ... 345 Total 8,545 These coal-fields differ essentially from those already described, in that they do not exist as a succession of flat beds in mountains at a considerable elevation above the sea ; but as a series of parallel elliptical synclinal basins below the ground-level, with their outcrops rising to it all around. The Forest of Dean coal-field is of the same nature, and in South Wales there are coal deposits of both kinds. The general str ke of these coal basins is from northeast to southwest. The dip is naturally greatest at the outcrop, then gradually lessens and dis- appears ; and finally rises in the same way on the op- posite side. The Warrior coal-field contains no less than fifty seams, of which twenty-five are thought to be workable, but only three are actually worked. The thickness of the coal in these varies from 3 to 14 feet. The Cahaba deposit contains twenty coal-seams, of which three, from 2 to 6 feet in thickness, are worktd. The total production of the Alabama coal-field was : In 1870 11,000 tons. In 1880 340,00 " In 1886 1,800,000 " In 1889 2,903,350 " Since 1890, the coal and iron trades have been suffer- ing from a terrible depression, from whi^h they are only just recovering, and therefore recent statistics do not in- dicate the productive powers of the district. COAL WASHING. 237 The Alabama coal is mostly of a coking quality. In 1890, there were 4,647 coke ovens built, and 270 under construction; but, by the end of 1891 the number had increased to 6,000. With the exception of 64 Thomas coke-ovens, all the ovens are of the ordinary beehive type 101 to 12 feet in diameter, and 5 to 7 feet high in- side. The charge is usually 5 tons of small coal, which produces 3 tons of coke. In the principal, or Great Warrior, coal-field there are numerous mines, for the most part with coke-burning plants attached. The following are typical ones viz., Bine Creek, Pratt, Adger, Blocton, and Johns mines. Average analyses of the coal of the Warrior coal-field, and of the coke made from washed and unwashed coal, are given below : Coal. Coke. ^ From unwashed Coal. From washed Coal. Fixed carbon 61.51 87.02 90.48 Volatile matter. .. 31.48 1.02 1.11 Ash 5.42 10.12 750 Sulphur 0.92 1.77 0.83 Moisture 0.67 0.07 0.08 With one exception, these mines are all worked from the outcrop, the winding shafts being at such an angle with the horizon as will admit of entrance and egress on foot. The tubs are hauled in and out by engines and wire ropes running on rollers. There are three entries at the Blue Creek mine, which are together capable of yielding 2500 tons of coal per day. The one exception referred to is the Pratt mine, which has a shaft 200 feet deep, worked by a winding engine and head-gear in the usual way. Pumping is effected by a force-pump below, driven by air compressed at the surface. This mine alone produced over a million tons of coal in 1889. 238 GEOLOGICAL SURVEY OF ^ALABAMA. Under the system of working prevalent in the Ala- bama districts, galleries are driven off the main slope at intervals of 300 feet. The intervening body of coal is worked out by driving stalls 40 feet wide, and 60 feet from centre to centre, for a distance of about 275 feet. This leaves a pillar of 20 feet between the stalls, which is worked back to the heading, as soon as the stalls are finished. In this way all the coal is taken out between the galleries, leaving pillars to protect the entries. When the galleries have been driven about 3,000 feet even these pillars are removed. By this means not more than 5 per cent, of the available coal is lost. Ventila- tion is usually effected with one continuous current, but sometimes a split-current is adopted. In that case a Guibal fan is placed at the air shaft, on each side of the main haulage slope, so that each side is independent of the other, and each gallery takes its supply of fresh air direct from the main-slope. The hewing is generally done by hand; but at the time of the writer's visit 13 Harrison pick machines were in use; they are able to uhderctit to a depth of 4i feet along 90 feet of face, with one attendant per shift. At the time of the writer's first visit to Alabama, in .1890, the coal-slack was nowhere submitted to a wash- ing-process before being charged into the coke-ovens. The disadvantages arising from the comparatively poor calorific value of the resulting coke was felt to be a seri- ous drawback to the development of the iron trade. Consequently great attention was given to coal- washing plant, and it was not long before the Robinson and Ram- say coal-washer was introduced and adapted to Ameri- can requirements. The writer believed that all coke now used at Alabama blast-furnaces was produced from washed coal- slack. The beneficial results were shown, in the comparative analyses which he had given, and COAL WASHING. 289 had contributed materially to make possible the extra- ordinary development which was at present in progress in the Southern pig iron trade. At the Pratt mines coking-plant, flues, built in the walls between the ovens, and communicating with them, draw off a portion of the gnst's and convey them to the boilers, where they . are burnt. It was claimed that no less than 375 tons of coal per week, was saved by this arrangement. If we take the area of the three principal Alabama coal-fields, at* the estimate already given, viz., 8,545 square miles, which is equivalent to 5,468.800 acres, and reckon the workable coal at 8 feet 4 inches thick in the aggregate ; and as yielding 10.0 tons per inch per acre, we shall find that the total quantity of coal is 54,688,000.000 tons, which consumed at the rate of the present total production of Great Britain (viz , 180 mil- lion tons per annum) fixed the duration of the Alabama coal-field at 303 years. At the time of the writer's last visit to Alabama, viz., in the autumn of 1894, the price of coke delivered at the blast furnaces was about 6s. 9d. per 2240 pound 2 , or half the current price in England. Negro labor is mainly employed, the latitude being about the same as that of Morocco, and the climate b ing, therefore, almost tropical, the population requires less food and protection than in colder regions. Great efforts were made in the spring of 1894 by the leaders of the local trades unions to force the negroes into their organization. When they found that impossible, they endeavored to frighten them out of the trade. Several were shot, others maltreated, and. for a time mine-managers went about armed with re- volvers. , By aid, however, of a loyal militia, headed by a, capable and courageous governor, the trade union- ists were eventually beaten ; and thenceforth the south, having the benefit of chea x p negro labor, has been able 840 GEOLOGICAL SURVEY OP ALABAMA. to compete advantageously in all parts of the United States.. The distance from Birmingham to the Gulf ports is 258 miles to Pensacola, and 276 miles to Mobile. The railway rate to the former port, including shipping charges, is 4s. 6d. per ton, or say 0.20d. .per ton per mile. This rate enables coal-producers to put coal free- on-board at these ports, for, say, 8s. per ton, or as low as bunker coal at the northeastern ports of Great Brit- ain. The railway facilities enjoyed by .the Americans are in striking contrast with their absence here, British heavy products having to pay from three to five times the above rates per ton per mile. It is not, however, in direct coal exportation that British producers need fear American competition. It is more in heavy goods, such as pig-iron, steel rails, and billets, which absorb in their manufacture from 1| to 2i times their own weight in fuel. In the case of such ex- ports, only one railway and sea freight is paid on all material used in producing 1 ton of product carried. Such goods are also practically undamageable, and are not much affected by delays of transit ; and being use- ful as ballast they are taben at low sea-freights along with cotton cargoes. Alabama pig iron so favored, is already arriving in considerable quantities in European markets ; and for the time being at all events, coals, or the products into the manufacture of which they enter are being literally "carried to Newcastle," or right into several of the coal-producing districts of Europe. Calorific Power of Coal. The subject of the calorific power of Alabama coals and cokes has not received the attention its importance demands. It is very rarely that any interest is mani- fested in the matter. COAL WASHING. 241 So far as is known, Prof. 0. H. Landreth of Vander- bilt University, Nashville, Tenn., was the first to make tests of the heating value of Alabama coals. This he did in the spring of 1885, and the following table, taken from Mineral Resources of the United States, 1886, p. 289, embodies his results. 16 24 2 GEOLOGICAL sftfcVictf '&F ALABAMA. f ' * f ! 1 ' / I,, rft-n rtR T TJ ( ) 1o ; ' C " 1 *i *'j M , i( '(> < i i "A "H '\i\ '! i . I OG >}Jj:f,i r.Ji'/r ..IH1':>T f 'i [ i V f i K ! '. '/. , V j I H'l 07 i i i Tl } i i c 1 O r^^H ot! eh 'g ) 3^l'.t |l '> .fifiir,(.n;IA lo > h?7 jinihiyfl erlt lo ^)>i'.),t a o/l nt ; |l:: : "SS5SS&2 ; 3 1881 lo ;2uruja oil? ni Lib or, c W'gj'TZJ g ' QC G55 OS O5 OQ. O5 O^ 9^ o H'v'vuuunU \n'v;Hv'sU\. aioil 1 ' ^ 1 .ei[jj80i gfd B&ibodmd ||| O c .2 H 1(2 S3 , .d o?^ 2221^2252 +3 >H & S ^ * 0!5 "^ c cS C * "SI b ^ I- 3 coco^S^o^^ ^ aog t^^t^.t-It-acoit- v C3 rV| TO > <^ o j_q ^2 W c ^ pq ^ H 13 : : : : bC a |H '. '. OP M j O . . \* \i \ ' 8i CO S3 83^ ' 5^-g c P7^,- % Os3 ; ^'~~' i= "s3c3'3 SSoWKfi^^ . A M A H A OOAUx > WASaiKO w l A ' ) I OCX! Ofl t > 243 During the last few years,: as opportunity offered,,, the The writer has made ultimate Unil/ies, a|ad cil trifle tests of some of the; principal coa/ljs0f ! fkwJpfcate. : * cS 9 "q S o^a COOOCVlOCOlCt CO-f T^-CO o c3 Biuouimy r-iaCOCOCOCOOOCOt^l^ OOiOSOOOCDOCCO^OOCD a o3 jD jni[d[ngi CD 1C C O O O OC CO GO O t^ C5 *-^ O^ ^O c3 ^ qsy CD lO O ' O OCO'Cidt^-UO OO61O --OOr- iiOiCCDOO O a U890J^I]^ ||^||^^2 o CE cNcooocociosos^Mas ' 10 *4 1 o uaS^xQ T I "' * -2 aaSojpiH <*< * t- - a p uoqj^Q 'T-i CD Oi t CO CO 1C lO CD OO 00 ' / -s . 9 . . . ... X LJ a CO ^ .& :::.:: : : : X Q j-H rCj -r-t Hrt W jr 1 1 PH H-) PQ s '.'..'.'.'.'..'.. ^ c^ 49 C^ "fj PQ M . t>.-^CC i- t- co co t-- QO ' O' ? iibio'O"^ O 'rr CD -i CM ^-i < - o^J ?CM t^. CM CD t- CO r-i ^ ' >->-. d_^_g lO O CO 1C O O iC t- O O5 CO GO 1C CM 00 OO O ^t 1 Tfi CD l> O ^ OC 10 10 CM 1C ^ CM CD CD 00 TJI 00 !c g CO t~ h-l>- CO O CM OO 00 i O "f ^ >C 1C rf< TJH 'J CO t^ t>- C GO CC t |^ t^ t~_ OO t^ OO rt! iiW f ' >I(X/I If 1 oil . AM Afl AJ/: ii ..)'!"> '!<{ <.!..' <>1 ^' I rno'fl r'rnr.v r.^iTir.'i It As was stated in chapter 1. two processes for , concen- n'>l)'i1ri "!'?/' > ' r. >/iM-!'tF oT vtunh i- 1 !! > ,1io^ Ollp , , trat^ng the low grade ores of the .^tate have , beep , tried , t .9^ a ^rge scaje .. ^he , experiments^ for the, most ^rt, were, .confined, to, the soft, ; yr lime-free ?( red ( ores, and, to the 'hard', or limv orep. They were oased on two dif- .. >, i rr '/'n; '/ '>'M!i:i.-fr> 101 M* IM> IvHHjn.)^ HI nyi', ferent principles, first the, artificial magnetization of the .i,'i'j.>{i( rTii','> ; - > *jf\ ] n> !l!' f)(l I , < H >'! ') ' > f i * ITiO'J I i'3''M tH'o '*j of the ore, and subsequent separation over a special machine , and second .the trejatr^ent' of tlie { ore , merely dried and crushed, in a saturated magnetic fiel4, this ;r . t \ v ii.Mu>.1r-ir> :HK^ i-n I. :T- -m.1 ni. process not requiring that the ore should be magnetic. lofljn i-MMmif fiTj OT mi h T>ji-M f>,i l>n.i; -no -i! I !'!' -i r P. 1 1 09 y/on /ji,ffr^ r )*?*fil1. lo jintnou .if-iv/ql HiflJ . v :<> a description of the .experiments which he carried on < fj j -T[ fninoO <>! JJ/>jlfH ifl ll""'!/! oa.t h(f,i: thru ill, 7/ Tn Tr I rP' H IK ROIrJilt'i )I .y r >BUTUl *)nl.in 'trj; MJfJi'iil /!fr;i;< '" r. [ M i fff '.H) wol-J J i'/f I'ftv Ofli HUv/ OTlf&nOTO during the following year it was found that it was not v/on ' & > 1 :-- ;;; !'i-/''"P'irfh lo .-!ioM)n(J.n-." !> SM')> JslJlO^QJ T^insr -MJ! f l> .ire j rrt^ .. was concentrated magnetically without being at all ''"t "ififf) >. r>r,m;M Q xi I o1-">nofj l)OP,.r^V0lTI'T1l rtO*0 magnetic in the ordinary acceptation of the term. v/n & t[i t ojtnij f>,i hu.n >T').iri') 'Mlrilffr,- of iirMir, ii.f.vi; inr>'H Atlanta paper is here given, and the results from the < , tiioTlr, -^irn yi'ii', 1 -. l>fi.c /I'M (1 i J^VM i , nt.p.')H )l c. jxi^'t I oii.J Wetherill machines, also. Mfi 'M! } no fttroq oji ,).r, , no'tr to ti'-) TMJ The deposit '6f red 1 fossilif erouis ! ore 1 '(Clinton-) ' attains maiimtirn! thi^knds^ l iw ( the 1 f iiamediate vicinity of fiKf ,',tn .) 'jfH 8 ) <> (noil e^iltJBO ti ; niB')H. ^ boo 5! HP, f 248 GEOLOGICAL SURVEY OF ALABAMA. Birmingham, where the Eureka seam (now termed Ish- kooda) i^ from 18 to 24 feet thick. The upper portion of this seam, near the outcrop, is what we term soft ore, inasmuch as the lime has been removed by leaching. Under cover the ore becomes hard and the amount of lime it carries varies from 12 to 25 percent. In mining the soft ore it is customary to remove the over-burden and to take the ore from open cut, the tracks being at different levels to facilitate the handling. The over- burden varies in thickness from a few feet to 40 feet, and is stripped on the dip for distances varying from 50 to 300 feet from the crop. The dip of the seam increas- es as one goes towards the southwest, the average being close to 20 degrees. In the early years of iron-making iii the district it was customary to remove from 15 to 20 feet of the ore and to send it all to the furnace, but of late the mining has been restricted to 10 or 12 feet and there has been left in the ground from 8 to 10 feet of ore. This lower portion of the seam is now considered too low in iron and too high in silica to permit its pro- fitable use in the furnace. It carries about 40 per cent, of iron and about 35 per cent, of silica, the silica in- creasing with the vertical depth below the mining mark. Not less than 500,000 tons of this low-grade ore is now stripped, the upper 10 or 12 feet of workable ore having been removed and sent to the furnaces. Nothing re- mains now but to shift the tracks and to mine the lower portion also, thus making the entire thickness of the seam available for the furnace. With the exception of the Irondale seam, 5 feet thick and carrying about 53 per cent, of iron, at no point on the mountain can the entire seam be mined for furnace purposes unless the lower portions be subjected to some process of concen- tration. The Irondale seam is distinct from the big, or Ishkooda seam ; it carries from 6 to 8 per cent, more of LOW GRADE ORES. 249 iron, also more alumina, and can be profitably mined from wall to wall. This, however, is not the case with the Ishk'ooda seam. Ic is not likely that, on the aver- age, more than one-half of it can be used now for the manufacture of iron, and unless the remaining portion can be concentrated, it is practically of no use whatever. The stripping that has been done i-^ chargeable to the ore mined and sold, so that the lower portion of the seam can be mined at a very slight expense. It can be loaded into the railroad-cars and laid down in any stock house in the vicinity of Birmingham for 40 cents per ton. This statement applies to such ore as has been already stripped, and from which the upper portion has been removed for use in the furnace, leaving the ques- tion of tracks and loading-appliances already provided for. It applies, therefore, to what may be considered a limited amount of ore, and it is so in a certain sense and as compared with the enormous deposit of such low- grade material along the mountain. A concentrating- plant taking 500 tons of ore per day and wo. king stead- ily 365 days in the year would require nea iy 3 years to use up what is now ready for mining; ar.ri when we consider that the mining of the upper portion is going on all the while, thus increasing the amount of the low- grade ore left, it is not likely that such a plant would be able to uso the uncovered portion in five years, if the removal of the better ore should cease on the first day of October, 1895. So much for the Ishkooda opening; but there are several other mines along the mountain, within 2 and 3 miles of Ishkooda, that exhibit the same conditions, and I feel warranted in saying that the available supply of 25-cent. ore will not fall much short of 1,000,000 tons. When one considers the immense amount of low- grade ore that has not been touched,, from Grace's Gap GEOLOGICAL ^pRVE^ OF ALABAMA. fl] satisfied that for many years tOj pomp, the supply or pre t for a cpncentpating-plapt will % am ,>J e - ( '^m I* fr ^Ipnpwledge of the subject an4 fro in an acquaintance of f several years, with the ore-situation in - } the 'Birmingham s district, I have no hesitation in ^saying .that a concen- tf tratingrylant of the capacity menticme.^ above v would ( no^ experience anv ^serious! ( difficulty ^in/^ obtaining low- jgrade pres suitable for Goncentration and ( at ; a .price that rf^A^WW?! * Jft f rftHft^ felTSR^i ?P?ffRr out i' i -h,;< ,i i *'j .-;lfi')' nj- -t')l rif/;r!vftin)'riM !o A^'ii'M/ 'rli IM '*vnoH CRUSHING THE ORE. U c >') rf')nH ni'>t.;t^ RJflT .no.) 1 The size of the 1 '6re ! racist suitable 'for^magnetizatioh- is that ot'ah'en's egg, ^With^uch pieced the -magnetiza- tidn is' through -e veil 1 td the ce'nter^ahd when th ! e prop er heat hafe been used there is no sign of lou'ping, -or incip- ; ient fusion. -'The'ore is" df a deep 'Velvety bladk' : color. -At iittes the grains of ; satfd are s'dniewhat' Whitened ! but "fdi-th'e most part they 'are 1 boated Witli 'a film one of black magnetic oxide: P| The 'grains ( df ^sand ' are- Tdiarided ' ! in 1 'the 'original ore, 1 atid in 1 'the ^m'aign^ti^ed drfef 'tftey 'are ' ' v bf f th4 ; same -'ph-V sl'cJall ' ''nature . ' [ ; If tH^ ; heat j b ; e" tb(6 : high - ' 'tfh& sand adheres closely ! td the' ' W&; ! ' incipient fusion having 'se ( 15 ! ih , ! 'and ! the 1 subsequent ! ; separatidn ; i j s l 'i^o* ' so 'as 'large as a' 'cbcdah'trt uniformly t6 ! th'e cdh'ter^but there is danger is using ore of this 'sizes ; for 1 ' whefl : the 'interior is at ia 'suitable '"ter9'pei*ature the'extefidr ! '*y ! apt H t6 be' td6 hdt , and 1 there" niay a;rise j 'mdr'e or les's' tendency 'towards' Iduping, We^ ; h'ave ! fotmd' ! the iriost ' 'stiitable ' heat 'is a'fullred and ' it ! is' 'difficult to 1 maintain a large terior may just right while the interoir of the lu'mp 'not be red-hot /and 'When ! brttken will still be unreel oiced. A M AM A.I / IjO'W.GRAtKK Q&EiS.K MOM l s a : large* netized costing; o>n the outside extending ,a ^hir-d, ^, a halfiof ,thq distance to .the, Center. :Thte, Qoat,Uag );i >fQyld be of ^^uHWaiplc colpr, , while,,, $he : , center shpm-tt'be, .original iie^. jcplqr, ; qf a very large nunaber of : piece^ s fro^ n t;J> t e t kilqjUn4er ing .cpu.ditlops of wpr tlaat ^as .m^gn^.fci^ at outsicle. , , ,,,,, ,(, ,, ,, |; ,, ( |^-i,, HI ni H A It was our practice to , charge. *tke, ; kilpL^;ith (; pf e^gs, as , ^early^ifpr^a jn t( fi^e ^.pos^i^le, SQ.^Uat ,., fc^ gas ,^r- / pqeu^s, .should; meet wife abouf '^ ne : .^ftra^ , r W^WWn as i they, tr-aver.se, , tjie sp.a^e -. and ,.:, , orq,, the, cabs beJLag loaded witli of fine ore is .necessarily .made , in thq, , k,Un itselfi t>y Jbhe gtf. t}ae,qre,; against itself;, and aA^i at its best, are herewith given. The sample^ w$re,, drawn frora^the p^odupei; iDftmediateily iLp.faant.pf.ith^pip? con- 252 GEOLOGICAL SURVEY OF ALABAMA. veying the gas into the kiln, care being taken that no air was drawn into the sampler. They were analyzed at once for carbonic acid, oxygen, carbonic oxide and hydrogen, acetylene not being determined, nor marsh gas, although this latter compound exists in producer- gas to the amount of some 3 per cent. Acety- lene seldom occurs in producer -gas beyond a few tenths of one per cent., and may be neg- lected. As to marsh gas, it does not seem probable that this gas, in and for itself, can be used to magnetize ore, as the reactions that occur when it is passed over red-hot ore, no other gas being present, are theoretically not such as would lead to a magnetization of the ore. I am unable to speak with confidence on this point, how- ever, as it is a matter of great difficulty to prepare this gas in a state of purity. The ordinary reactions by which it is prepared from sodium acetate yield a gas which is seriously contaminated with hydrogen, render- ing it useless for magnetizing experiments, as the hydro- gen is itself a powerful reducing agent, and the magnetic oxide produced by passing marsh-gas from this source over ore would be due to the hydrogen primarily. The question of the effect of pure marsh-gas on red-hot ore is one of scientific rather than of practical moment, as the producer-gas usually employed contains it only to the maximum extent of some 3 per cent. It can, of course, be prepared pure from zinc methyl ; but none of this substance could be procured from deal- ers in this country, and the question has been dropped for the present. The effective agents in magnetizing ore are carbonic oxide and hydrog9n and if the producer is operated under the best conditions there will be enough of these to do the work. The average content of carbonic oxide while magnet- LOW GRADE ORES . 253 izing was 25 per cent. ; of hydrogen, 13 per cent. ; of carbonic acid, 6 per cent. ; and of oxygen, 0.40 per cent. MAGNETIZATION AND CONCENTRATION OF IRON-ORE. TABLE XL. Analyses of the Producer- (las Used. Carbon- ic Acid. c 0> 00 > H Carbon- ic Oxide T3 o> Remarks. 14.00 None. 8.46 5.93 Bed 4 inches ; color dark gray ; 1 after starting fire. hour 13.20 None. j 6.00 5.60 Bed 6 inches ; color dark gray ; 2 after starting fire. hours 5.00 0.40 30.80 12.90 Bed 2% feet; color grayish-red ; burns well. 6.80 0.40 25.10 11.90 Bed 3 feet; color orange-yellow lent gas. , ex- 8.00 None 24.09 13.84 Bed 4 feet ; color orange-red ; good gas 4.30 1.83 22.18 12.63 6.00 0.60 25.40 14.60 10.05 9.00 0.40 21.40 Average. The analyses of the waste gases showed that all the carbonic oxide and the hydrogen were consumed in the kiln. After the gas been burning all around for 10 hours, the di charging of the kiln can begin. It will be under- stood that the first 10 or 15 tons, lying at the bottom of the kiln and thus beyond the limit of the heat, are not 254 :< '"' GEOLOGICAL Ja1l"aM must be' seat bae"k a/s atv - tion of ore has been removed so as to give place to ore that Has beea su&clently heated aaa magnetized^ ail or the ore coming to the shutes has traversed the zone of highly-heated gas and has been exposed to it&- influence. As the ore is withdrawn from the shutes fresh ore is charged into the ,kiln",,and the operation is continuous. When the ore comes down to the shutes red-hot, the cur- rent of gas is changed, and instead of passing into the combu'^tion-charalDe^, 1 |i/t , is' .passed into ; the' magnetizing- chamber, from which it passes over the ore unmixed with ajr and therefore capable of reducing tU.^,, ferric, oxide in the ore into tire magnetic oxide. In experiment- ing with the kiln, we found that even when the gas-valve leading into the combustion-chamber was Closed and the valve leading into the magnetizing chamber opened, there was still too much air going into the kiln, and we ,y = , wull lY-M^nirm 'ft>i;<> ; .' V P ' .,-> [ji <}[ ri^ Of {! luted the shute-doors with clay. It was extremely diffi- cult to prevent the gas from burning in the ore and thus wasting its reducing power; but by constant attention and keeping the shute-doors well luted, we succeeded in preventing this to a great extent. The, reducing-gas was passed over the ore for an hour, when one or two shutes were opened and a cab of ore withdrawn. It was at a full red heat when drawn, and was spread out on the ground to cool. It retained its heat for several hours, but when finally cool enough to handle was of a dull : black color, arid tvhen coarsely powdered, highly magnetic." '1%$ -temperatifire -of 'the kiln pas measured bj^. an Uehling-Steinbart pneumatic pyrometer of the latest* ( , tinted* ffto ! 900 de:fcd '1350-^guiiFailt^ feita^llOO'de 5 ^.* 1 ' ' n^'- of "the most trying difficulties w ; e experienced was in getting the 6re ! j doWn to the shutes thoroughly and uniformly 'magnetized, Sometimes the greate'r* JHKff a cab' would be' well magnetized Svtiite a ^oi'tioii * of it taken 'from the same stiute afc the j: same i time' 'ivfrtild not ' be magnetic at all. This was found to be due to the" fact ! tiiat s H nacl not'been exposed to 'th^' gas for a suffi- cient length of time . A!S probi of t'h is 1 ,' ' t took some 1 ' 6^ ^ the gas that was going into the kiln 'an $ ' 'some of tnd non-magnetized ore frbin a cafe,' heated ttie'ore'to a f ul'l ! redness in a glass tube and pa&syd ; tlib >i a;s%vW'it! U) Ii; J{? became' magnetic iii J : a: : fe'w" moments^ nj After this, ^w^*" allowed the' gas' to paBs ' ovei* the' ' ^'re for ; a ( longer tith'e ',' ' ' 7y !I; 1 ' 1 ( air excluded, we obtained better results. One thing 'Wa'sl ''' proved to ( Q\ir entire .satisfaction, viz., that when the ore was exposed for a sufficient length of time at a full red heat t0 a jpur^ei^tjOf producer-gas, it beca^ne.;higl)ly n^ag- netd<3,[and tji8ft,,this e^eqt.w.as to /a considerable extent, u . independent , &$ , the >,&!%$ ,iOf , , the lumps,.: The, idifficulty already alluded to, the tendency of the larger Jumps; $$,,. loup, was hard to overcome. The outside of these pieces would be magnetic while the interior would ; not b^ changed at all, or at 1 best would exhibit 'very feeble mag- netism Now and then a lump as' larjgfe as 1 *a ' cdcba'-ftut 1 Would' come down in a very satisfactory conditioh, biit : M the whole it was found desirable to exclude these large lumps froiii the kiln 'and tb ( ubie ore 1 'thatT'wus 1 6f the 1 size of a hen's egg. Another serious difficulty 'Was in the irregular manner in which' 'thd ! ! ote came down to the ' shute's ' ' M'a kiln of this 'cons true tlb A' it Wye'ry 'difficult? M > to get a uniform heat all round. At times the kil ; n f!l and too cool somewhere else. When it became too h*6'l! ! ! on one side there was nothing to do but to draw ore from the shutes on that side and let the ore descend until 256 GEOLOGICAL SURVEY OF ALABAMA. the normal heat was restored. This naturally disturbed the course of the operation elsewhere in the kilns, and had a tendency towards allowing insufficiently-magnet- ized ore to come down to the shutes and in a measure to occupy a space outside of the area of magnetization. When the operation was proceeding satisfactorily, we got from the kiln 110 tons of ore per day of 24 hours, and worked in this way for several weeks. A part of the ore was magnetic, and a part was not. It was culled for separation. The separating machine could not treat half the ore that was magnetized every day ; and the remainder was sent direct to the furnace without separation. CONCENTRATION OF THE MAGNETIZED ORE. This was effected over a Hoffman separator, at first, and afterwards over a Payne machine, which proved to be an excellent separator. The magnetized ore was first sent to a No. 3. Gates crusher, screened over a revolving screen of 8 meshes per linear inch, the heads from the screen going into a pair of rolls and thence into the conveyor with the tails from the screen, and so on to the bin above the separator. Between the end of the conveyor and the bin there was another screen of quarter-mesh size to re- move the small lumps that jumped the rolls or passed down between the ends of the rolls and the housing. All the material going to the separator passed this screen and nearly all passed a screen of 8 meshes per linear inch . The fineness of this material is given in the following table. LOW GRADE ORES. 257 TABLE XLI. Fineness of Material Going to the Separator. Per cent. Left on 8-mesh screen 3.00 Through 8- " " and on 10-mesh 6.50 I 10- and o n 20-mesh 28 50 ( 9Q- " 30- " 31 50 t 30- " 40- u ... 6.50 t 40- ' ' 50- " 9 50 I 50- 60- " 60- " 70- " 2.50 3 50 ; 70- " 80- " None 4 70- 100- " ... 3 50 t 100-mesh. 5.00 This represents the average fineness of tho material sent to the separator during the course of the experi- ments, as several determinations were made from time to time. It was not found practicable to run the separator at a greater speed than would give about 700 pounds of heads, per hour, as we had difficulty in disposing of our tail- ings in greater quantity than this, owing to the confined space in which we had to work. The separation was attended by a good deal of dust until we regulated the feed to this point, and even then it was far from pleas- ant. Special care has been taken of this in the plans for the alteration of the plant ; and we shall remove the dust by means of an air-blast. The average content of iron in the ore sent to the separator was 45 per cent, and of silica 30 per cent The average content of iron in the heads was 58.86 per cent, and of silica 11.51 per cent. ; in the middlings 51.12 per cent, of iron and 21 per cent, of silica. At the very start we found that some portions of the ore were more highly magnetic than others, and that 17 258 GEOLOGICAL SURVEY OF ALABAMA. the less magnetic material manifested a strong tendency to go into the tails and not into the middlings. In other words, the tails contained magnetic ore that should have gone either into the heads or at any rate into the mid- dlings. Adjustment of the machine and changes of the amperage enabled us to correct this to some extent ; but we did not succeed in doing away with it entirely, and throughout the entire course of the work we were troubled with incomplete separation. Repassing the tails over the machine always resulted in obtaining more heads and middlings th-m in the first pass, and we finally concluded that it was practically impossible to get even tolerable tails by one pass. To this conclusion it seems that all have come who have tried magnetic separation, even of highly magnetic natural magnetite, viz., that it is in all cases advisable to use two machines or, better still, two drums, and to pass the middlings and tails from the first to the second, increasing it may be the amperage on the second machine or drum, arid, perhaps, also regrinding the material from the first ma- chine before sending it to the second. As by far the greater part of the expense in the magnetic separation of ore is incurred before the ore is sent to the separator, the additional expense of sending it to another machine, even should it be reground, is comparatively slight. It may be of interest to some to know the distribution of the iron and the silica in the heads according to the fineness. I give, therefore, in the following table some analyses covering this point. Numerous analyses have been made to show just where the best ore was, and if -finer grinding would enable us to improve the quality of the heads. From these I select the following : LOW GRADE ORES. 259 TABLE XLII. Analysis of Heads According to Fineness. Original Ore: Insoluble, 28 per cent. ; Iron,. 44 per cent. Percent. Insoluble. Iron. Left on 8-mesh screen 3.00 12.76 63.20 Through 8- on 10-mesh 3creen. . 6.50 12.50 62.70 10- " 20- " .. 28.50 13.00 61.30 20-" 30- " .. 31.50 13.40 60.00 30- " 40- " .. 6.50 13.70 60.30 40- " 50- " .. 9.50 15.40 58.25 50- " 60- " . . 2.50 13.90 60.80 " 60- " 70- " .. 3.50 14.00 60.70 70- " 80- " .. None. 70- " 100- " .. 3.50 14.70 60.00 100-mesh.. 5.00 16.10 57.00 Average 13.94 60.42 It might be inferred from these analyses that the amount of iron decreased with the fineness ; but that this is not always tV.e case will be apparent from the following analyses representing the heads at another period of the work : TABLE XLIII. Analysis of Heads According to Fineness. Original Ore : Insoluble, 32 per cent. ; Iron, 40 per cent. Per cent. Insoluble. Iron. Left on 10-mesh screen .... 2.90 12.65 59.15 Through 10- on 20-mesh .... 18.30 12.58 59.09 k< 20- " 30- " ...... .... 21.70 12.72 59.25 80- " 40- " .... 10.00 12.65 59.20 40- " 50- " . ... 10.00 12.40 59.48 50- " 60- " .,.. 13.30 11.05 61.73 60- " 70- " .... 8.50 11.08 61.80 70- " 80- " . . . . None. 70- " 100- " .... 10.00 11.45 61.40 100-mesh . 5.30 10.80 62.00 Average 11.93 60.33 260 GEOLOGICAL SURVEY OF ALABAMA. There does not seem to be any fixed rule as to this matter ; sometimes the percentage of iron increases with the fineness and sometimes it does not. It may be chargeable to the nature of the ore, if easily pulverized or not, the degree of magnetism in the ore (about which very little is known, whether the ore be natural or arti- ficial magnetite) ; the intensity of the current ; the speed: of the machine ; or a combination of these causes. So far, nothing has been said as to the removal of phosphorus. This element is present in the ore to about 0.30 per cent., but it is not removed in the separation. It seems to be present as phosphate of lime, entirely amorphous, and most intimately mixed with the iron,. We have not been able to remove it, or even to diminish it to any considerable extent. No matter how finely the ore is ground, the heads still carry more phosphorus than is allowed in Bessemer ore. It can be entirely re- moved by chemical means, and brought from 0.30 to 0.008 per cent, at one operation. It has been found that dilute sulphuric acid will dissolve out the phosphorus from the heads without affecting the content of iron se- riously, and in this manner heads carrying from 58 per cent, to 60 per cent, of iron and 0.008 per cent, of phos- phorus have been prepared. A word now as to the cost of carrying out this process on scale, let us say, of 100 tons of raw ore per day of twenty- four hours. We will assume that the plant is erected on the mountain in immediate proximity to the ore, and that the gravity system is employed for conveying the ore from the mine to the kiln and from the kiln through the various operations until the concentrates are loaded on the cars. We will allow, also that it requires 3 tons of raw ore to 1 ton of concentrates carrying 55 per cent, of iron, and. that the yield of such concentrates from one kiln is 27 LOW GRADE ORES. 261 tons per day of 24 hours. In other words, we allow that from a kiln holding 100 tons of raw ore we obtain daily 1 tons of magnetized ore fit for separation. The cost of producing 1 ton of concentrates of 55 per cent, iron will be about as follows : 3 tons of raw ore, at 25 cents, $0. 75 Crushing, including labor, . 05 Discharging kiln 006 Crushing, rolling and screening 0.05 Separating and disposing of tailings 0.05 Superintendence, 0.04 Night foreman 0.02 Engineers, 04 3 tons of coal for producer, at $1.25,. 04 3 tons of coal for boilers, 0.04 Oil, supplies, etc., 01 $1.15 These are the estimates that have been made from our experience with the process at Bessemer, where we had to work under unfavorable conditions, and wh re the cost per ton of 55 per cent, concentrates was 40 cents higher than the above figures. If we are able to in- crease the percentage of iron in the concentrates, as we expect to do, 'he cost per ton will be lessened accord- ingly. On the other hand, should we not be able to do this, but have to allow for 3 tons of raw ore per ton of 55 per cent, concentrates, as above, the cost will not vary much from that given, viz., $1.15. We come now to the question, is a ton of 55 per cent, ore of the fineness already given, worth $1.15 at the works, or $1.30 at the furnace? In valuing an ore for furnace practice, two methods may be used, the one based on the nature of the iron desired to be made from it, whether special high-grade Bessemer or basic open- hearth : the other, disregarding this feature of the ques- 262 GEOLOGICAL SURVEY OF ALABAMA. tion, as based on ordinary grades of foundry-, forge- and mill-iron made in this district. Both methods are in common use, and both are independent of the reducibil- Hy of the ore, this factor of the question rot being gen- erally considered. The matter, then, narrows down to the question as to- whether this ore, under the conditions now maintaining in the Birmingham district, is worth to the furnace $1.30 per ton delivered. This may, perhaps, be answered to the best advantags if we inquire as to its value if it alone were to be used iin the furnace. As a matter of fact, unless it be made nto briquettes, eggettes, or other suitable shape, by means of some binding material , it can not be thus used ; but for the purpose of this calculation we may assume that it can. We will assume that the limestone to be employed as flux contains 3 per cent, of silica, that the coke used as fuel contains 10 per cent, of ash, or 5 percent, of silica, and that the ore contains 55 per cent, of iron and 13 per cent, of silica. What will it cost to make to make a ton of iron with these ingredients, allowing 2400 pounds of ckke per ton of iron? 1.82 tons of ore at $1.30 $2.36 1.20 tons of coke at $1.75 2.10 . 66 tons of stone at . 60 . . . . 39 $4.85 This cost is, of course, to be taken as representing the cost of the materials entering into a ton of iron, and does not include labor costs, repairs and interest, and is based on ordinary fouadry-irons with slag carrying 35 per cent, of silica. Aside, however, fron considerations affecting the cost of making iron, with or without these concentrates, in LOW GRADE ORES. 263 the Birmingham district, the success of the process will bring into use very large deposits of soft ore now prac- tically worthless, and enable the owners of such ore-lands to realize more on their investment than they could otherwise hope to do. The supply of the better grades of soft ore is not indefinitely great, and even where the qual \>y of the seams justifies mining, with the exceptio of some narrow seams of high- grade ore, very little more than half the seam is now being taken. It follows that the original cost of the ore-lands must be doubled if the lower part of the ore is not used, and in charging off the cost of the land this fact must be considered . If this process- w^ill enable us to utilize the whole seam, t jp, middle and bottom, all al- ng rhe Red Mountain, the supply of soft ore is very greatly increased and the cost of making iron will continue lower than it we had to mine ore un- der ground. To this paper may be added the following observa- tions. The difficulty of effecting a uniform and regular magnetization in the Davis-Colby kiln, was to a great extent obviated hy reconstructing the kiln , so as to provide 4 chambers each with its own in-rake pipe from the pro- ducer and its own draft-pipe into the main connecting with the central stack b'lilt alongside the kiln. Each of these compartments hold about 22 tons of. raw ore. The advantage of thus dividing the kiln was at once ap- parent. Each compartment was a separate kiln, inde- pendent of all the others, and such reducing action as was desired could be carried on at will. If any com- partment became too hot the amount of gas going into it was decreased by closing the valve, if not hot enough additional gas was let in. Each compartment being provided with its own discharging door it could be em- plied without interfering with the others. There is no better kiln for calcining ore than the Davis-Colby, but 264 GEOLOGICAL SURVEY OF ALABAMA. when it came to magnetizing ore it was found necessary to reconstruct it. In regard to the magnetization of the fossil ores. It bas occurred 10 more than one person to endeavour to take advantage of the fact that they (in common with non-magnetic iron ores generally) become magnetic when exposed, at a sufficient temperature to the action of reducing gases. But so far as the writer is aware these were the first experiments on a large scale to im- prove the quality of the low grade ores of the Clinton formation, the so-called red fossil ores. Iti March 1897 I receive 1 a very interesting conm mi- cation fro n Mr. Jio. T. Hinlect, Wythivilli, Vi., de- tailing s) n3 ex :>9rim.3fifcs ii } mil) 10 y^a^s a^o with the fossil ores of that part of Virginia, the S. W. portion. Mr. Hamlett wrote : "About 10 years ago I examined these ores and con- cluded I would make an experiment with them, simply for my amusement. I t^ok several pieces as large as my fist, and put them, on an ordinary wood fire and left them th *re to roast all night. Next morning I pounded them up in an iron mortar to the size of ordinary blasting p w ler. I t'l^n took a small, cheap pocket m ignet of the usual horse-shoe typi j , and was some what sur- prised at the ready way in which I could pick out the ore and leave rhe grains of silica. My little magnet would draw up every particle of the ore. I then sent about 100 Ibs. of the, ore to Mr. Clemens Jones, of Pen na, informing him of ray experiment. I was aware of the f act that his attention had been turn- ed to this subject of concentrating fossil ores by roast- ing them and u-ing elect? icity as an agent in his work. In due course or' time I received a letter from him stacing that he had male a very successful and en>our- LOW GRDE ORES. 205 aging test with the ores sent, him that they averaged 24.30 per cent, of iron as received, but that lie had no difficulty whatever in concentrating them up to 48 per cent without crushing them too fine, etc., etc. There the matter dropped, so far as I was concerned, and there it has remained until this hour so far as these ores are concerned. Mr. Clemens Jones piper on the Magnetization of Iron ore" was read at the New York meeting of the American Institute of Mining Engineers, September, 1890, but there is no mention in it of the Virginia ore, and he seems to have confined his experiments almost entirely to limonite (brown ore) . So far as the writer is aware Mr. Hamlett was the first to experiment even in a small way with the mag- netization and concentration of the red fossil ores, and this fact would certainly have been mentioned in the Atlanta article had he known of it. It was stated in that article that we used the Hoff- man separator. We did so at first but afterwards used the Payne Separator, and obtained from it excellent re- sults, making about 100 tons of concentrates. It was in every way superior to the Hoffman machine, and is certainly well adopted for concentrating magnetic ore. Taking every thing into consideration it was thought that the experiments conducted on so large a scale pr.nnisfd to develop into a valuable adjunct to the Birm- ingham iron industry. But hearing of the Wetherill process it was decided to try this also, as it held out hopes of our being able to dispense with the magnetiz- ing of the ore, and this would be a great desideratum. CONCENTRATION BY THE WETHERILL PROCESS. J3o the concentrating plant was remodelled, and two furl 266 GEOLOGICAL SURVEY OF ALABAMA. size Wetherill machines were put in. It is not our pur- pose to describe the Wetherill process. Briefly, it is based on the fact that when iron, bearing minerals, properly prepared as to size, etc., are brought into a saturated magnetic field they are attracted in propro- tion to the strength of the current, and the amount of iron in the material. Non-magnetic ore is attracted just as if it were magnetic, and for all practical pur- poses these machines, whose magnets are actuated by a current of electricity, act on red fossil ore as if it were magnetic. A report was made to the Wetherill Con- centrating Company on the resu.ts of various trials lasting over several weeks, and formed a part of a pa- per read before the Pittsburg meeting of the American Institute Mining Engineers, February, 1896, on ''the magnetic Separation of Non-Magnetic material" by Messrs H. A. J. Wilkens and H. B. C. Nitze. Mr. Wil- kens was present when the experiments were being conducted, representing the Wetherill Company as its general manager, and with the writer had charge of the work. Messrs Wilkens and Nitze prepared a most exce]]ent piper on the Wetherill process generally and from it is taken the following description of what was accomplish- ed in concentra f ing the fossil ores of the Birmingham: district. "Clinton Fossil Ores Of more general interest on account of the greater application of the process and the large extent of the field, are, perhaps, the results obtained on the red fossil hematite ores of the Birming- ham district in Alab ama. The richer, soft ores of this district, such as are used in the furnaces , average from 45 to 48 per cent, in iron, and from 30 to 24 per cent, in insoluble matter. Such res occur, however, only in a few localities, which are LOW GRADE ORER. 267 limited in extent, and are now almost exhausted. By far the greater portion of the leached ore-beds consists of material running from 35 to 45 per cent, in iron and from 45 to 30 per cent, in insoluble matter. This latter class of ore cannot be used in the furnaces to advantage, and is therefore practically worthless, unless the per- centage of iron be raised by concentration ; and at the same time the insoluble matter be proportionately de- creased. Structurally, the ores as a rule fine-grained, the aver- age size of the distinct particles being such as would pass through a 10 mesh screen. On examining the product of separation it is seen that the ore consists of : 1. Rounded silica grains, which, owing to a coating of iron oxide, are found by analysss to contain from 10 to 15 per cent, of iron. 2. Rounded grains of more highly ferruginous Fma- terial; Tunning, perhaps, 30 per cent in iron. 3. A binding material of hematite, which in itself carries a varying amount of insoluble matter, depend- ing upon the locality of the ore, fineness of grain, etc. Various working tests were made on material from a great number of localities, and the results were verified by some 500 analysis. Space will not permit of a detailed account and dis~ cussion of the results ; it is merely intended here to pre- sent a general idea of what was accomplished. The previous magnetization experiments had been made entirely on the richer soft ores, such as are now being used directly in the furnace, and of the composi- tion given above. Concentration tests on this material by the Wetherill process gave the following results (Calcu- lated on a basis of 100 tons of raw ore) : 268 GEOLOGICAL SURYEY OF ALABAMA. Iron. Insoluble. Original ore gave 48.03 . 25.20 57 tons of heads with 57.10 13.10 28. " " middling with 46.20 25.40 15. " " tails with 10.00 70.80 It was further found that about 20 per cent in weight of this ore could be brought up to : Iron 59 .15 % Insoluble 10.45 The above results compare most favorably with those previously obtained by the magnitizing roasting process, particularly in the proportional amount of heads that were produced and the comparatively small percentage of iron carried in the tails. For the purpose of compari- son, the following results of the process are given, (cal- culated on a basis of 100 tons of raw ore) : Iron. Insoluble. Original magnetized ore gave 49.05 22.05 15 tons of heads with 59 .00 1 1 .06 35 " middlings with 52.00 20.00 50 " tails ' 44.00 28.00 Only the more perfectly magnetized material was used on the concentrating machines. In the magnetizing process is to be considered not only the cost of roasting, but also the imperfections attending it, such as the incomplete magnetization, the louping of the ore-lumps, and the inability to use a large percent- age of fines in the kiln. There is no doubt, moreover, that the raw material is better ^adapted for concentration, on account of the uni- formity in the magnetic properties and physical struc- ture of the several ingredients. The tests by direct concentration on the lower-grade LOW GRADE ORES. 269 ores showed a proportionately greater increase in the percentage of iron than those on the higher-grade mate- rial. The quality of the heads was, however, not as good, which shows that the hematite matrix in the low- grade ores shows a larger percentage of inherent insolu- ble matter than that of the richer ores. Among others the following results were obtained : (Calculated on the basis of 100 tons of raw ore) . Iron. Insoluble. Original ore gave 41 .H8 37.51 69 tons of heads with 52.00 23.00 31 " tails " 18.40 70.00 About 25 % in weight of this original ore was raised to: iron, 56.40 % ; insoluble, 17 %. Tests were also made on the so-called " hard ore,'* which represents that portion of the ore-bed from which the lime has not been leached. The raw ore of this character, as ussd at ihe furnaces, averages : iron, 35.50 ; insoluble, 17.50; lime, 16 %. From this were obtained from 50 to 60 % in weight of heads, containing : iron, 48 ; insoluble, 10.50 ; lime, 10 %r In preparing this paper Messrs. Wilkins & Nitze had in view an account of the Wetherill process as applied to various ores, not only of iron, but of zinc, and man- ganese, and to monazite sands, etc. It was not the pur- pose to speak particularly of the results reached in the Birmingham district on the low-grade Clinton ores. Their paper, therefore, while fully indicating the lines along which work was carried on here could not deal in detail with every feature of it. As the writer is con- vinced that some such method of concentration will eventually be used here it may not be out of place to 270 GEOLOGICAL SURVEY OF ALABAMA. give other results reached in experimenting, on a com- mercial scale, with the Wetherill process. The question was discussed by- the writer in the En- gineering and Mining Journal, New York, Vol. LXII, pp. 75, 105, 124, 151, and the description given here is taken partly from that publication, and in addition from his own note-books. The Wetherill Inclined Magnet Machine, and the Flat Magnet Machine were used, sometimes one and sometimes the other. The soft red ore was passed through a 15-mesh screen, and fed to the machine run- ning at 8 amperes arid 100 volts. Iron. Insoluble. 100 tons original ore gave , . .39.20 40.16 52.4 tons heads with 56.40 17.10 6.9 " middlings with 38.85 41.35 40.7 " tails with 16.70 74.10 The gain of the heads in iron was 43.8 % over the original ore, and the redaction of the insoluble siliceous matter was 57.4 % ; number of tons of raw ore for 1 ton of 56.40 % concentrates, 1.91. That is to say, from 1.91 tons of raw ore carrying 39.20 % of iron there was obtained 1 ton of ore with 56.40 % of iron. This result given here were not obtained at a single operation, and the course of treatment was as follows : 1st pass, amperes, 10 ; volts, 100. Iron. Insoluble. 100 tons original ore gave 39.20 40.16 59.3 " heads and middlings with.. 54. 10 18.80 40.7 " tails " 16.70 74.10 The heads and middlings from the 1st pass were re- passed at 8 amperes and 100 volts, and we obtained^ wo LOW GRADE ORES. 271 products, viz. : middlings, 4 % of the original ore, with 31.40 % of iron, and 52.20 % of insoluble matter; and heads and middlings, 55% of the original ore, with 54.10 per cent, iron, and 18.70 per cent, insoluble matter. Finally these second heads and middlings were repassed at r> amperes, 100 volts, and two products obtained, vi;; : middlings, 2.9 per cent, of the original ore, with 46.30 percent, of iron, and 30.50 per cent, of insoluble, and heads (final heads) 52.4 per cent, of the original ore, with 56.40 per cent, of iron, and 17.10 per cent, of insoluble. We could have stopped with the first heads and middlings, and have had 59 3 per cent, by weight of the original ore, with f 4.10 per cent, iron, and 18.80 percent, of insoluble matter. We may say, then, that from an ore carrying 3c).20 per cent, of iron, and 40.16 per cent, of siliceous m itter we obtained at the first pass 59 per cent, by weight of concentrates with 54.10 per cent, iron, and 18.80 per cent of siliceous matter. The gain in the percentage of iron was 38 per cent, above the original ore, the* reduction of the siliceous matter was 53 per cent., and for one ton of concentrates there was required 1.69 tons of raw ore. One hundred tons of this raw ore would yield 59 tons of concentrates with 54 per cent, of iron, and 41 tons of tails with 16.70 percent, of iron. In our operations \the amount of raw ore passing a 40-mesh screen was 33 per cent, of the ore, and this con- tained 49.4 per cent, of iron, and 28.5 per cent, of siliceous matter. The fines from this low-grade ore are much richer in iron than the coarse stuff. They carry from 49 per cent. to^54 per cent, of iron even when the original ore carries only 37 per cent, of iron. The ferruginous portion of the ore is softer than the more sandy portions, and it is possible to effect a very considerable concentration merely by crushing the dry 272 GEOLOGICAL SURVEY OP ALABAMA. ore and screening over a 40-mesh screen. The amount of material passing through a screen of this fineness varies from 25 per cent, to 35 per cent., so that we might expect to get to 54 per cent, of iron in one- fourth to one-third of the raw ore simply by crush- ing and screening. There is an increase of iron in the material finer than 40-mesh, but hardly enough to merit attention. The material through a 40-mesh screen was, therefore, called fines, and can be concentrated somewhat. In working on the fines we used the inclined-magnet ma- chine, and obtained results as follows : Iron. Insoluble. Fines through 40-mesh 49.40 26.50 10 amperes, 100 volts, gave 12.6 per cent, of heads with 55.30 17.12 22.8 per cent, of middlings with. . . .51.75 21.10 64.6 per cent, of tails with 45.80 30.35 The gain of the heads in iron was 11.9 per cent., and the loss of insoluble matter was 35.4 per cent. Numerous experiments with this and similar material satisfied us that it would not be profitable to attempt its concentration. It should be briquetted at once without further treatment, or mixed with 'heads' and briquetted. Material through an 8-mesh and over a 15-rnesh screen was tried on the inclined-magnet machine, with the fol- lowing results : Iron. Insoluble. Raw ore through 8 over 15 mesh, 24 percent 35.40 46.34 6 amperes, 100 volts, heads, 45.5 per cent. 50.20 24.34 Middlings 19.0 per cent. 43.00 34.95 Tails 55.5 per cent 15.40 75.35 LOW GRADE ORES. 273 By repassing the middlings, the yield of 'heads' could be increased perhaps to 50 per cent., so that there would be 50 per cent, of heads, instead of 35.40 per cent. But it would not be advisable to use ore of this degree of ooarseness, as the mechanical separation of the ore into ferruginous portion plus matrix is more perfect in ma- terial through a 15 or 20-mesh screen than in coarser stuff. Crushing the ore merely separates it into two por- tions, the one carrying iron, the other carrying silica, and the object of the separation is to divide the one from the other. The following results from concentrating low grade soft red ore 'by the Wetherill process are taken from the writer's note-books. Iron. Insoluble. Original ore 34.90 47.12 Gave. 52 per cent, of heads with 49.20 25.84 20 per cent, of middlings with 39.20 41.00 28 per cent, of tails with 14.00 78.14 Original ore 36.80 45.56 46 per cent, of heads with 52.90 21.24 15 per cent, of middlings with 37.45 43.62 39 per cent, of tails with 17.20 74.68 Iron. Insoluble. Original ore 39.20 40.16 Gave. 51.6 per cent, of heads with 52.50 22.60 11.4 per cent, of middlings with 32.05 51.89 37.0 per cent, of tails with. 16.10 74.76 Another trial of this ore under somewhat different con- ditions resulted as follows : 18 274 GEOLOGICAL SURVEY OF ALABAMA. Iron. Insoluble. Original ore 39.20 40.16 Gave. 66.4 per cent, of heads with 53.80 19.02 43.6 per cent, of tails with. 24.70 62.20 And a third trial, varying the treatment : Iron. Insoluble. Original ore 39.20 40.16 Gave . 52.4 per cent, of heads with 55.40 17.10 6.9 percent, of middlings with 38.85 41.35 40.7 percent, of tails with 16.70 74.10 These last results having been already quoted. Iron. Insoluble, Original ore 34.82 47.60 Gave. 42 per cent, of heads with 55.60 17.00 18 per cent, of middlings with 37.95 43.17 40 per cent of tails with 13.50 79.88 Iron. Insoluble, Original ore 42.00 36.42 Gave. 59 per cent, of heads with 51.00 25.20 23.7 per cent, of middlings with.'. . .45.70 31.76 17.3 per cent, of tails with 12.90 79.80 Original ore 37.30 42.90 Gave. 47 per cent, of heads with 53.25 19.05 24 per cent, middlings with 30.26 51.94 29 per cent, tails with 13.70 78.70 Original ore 37.36 42.73 Gave. 46 per cent, of heads with. . . 50.50 22.12 15 per cent, of middlings with 36.80 42.73 LOW GRADE ORES. 275 39 per cent, of tails with 15.80 74.20 A great many more analysis could be given, all bearing on this question, as the writer has devoted much time to the study of the matter. But these will suffice to show what was done, and to indicate the lines along which future investigations will doubtless be conducted. So far as concerns the low grade soft red ore of the Birm- ingham district ic may be said that it far exceeds in quantity the richer ores, and it can be mined more chap- ly than these. The vast deposit of low-grade ore carrying from 33 per cent, to 40 per cent, of iron can be utilized. Now they are practically worthies, and the exhaustion of the richer ores is proceeding very rapidly. There will come a time, and that soon, when the soft red ore as now used will become so scarce as to forco the iron companies to discontinue its use, or pay more for it. The careful experiments that were made demonstrated beyond any question that an ore of 35 per cent, of iron could be concentrated to 52 per cent., and that 2 tons of raw ore would yield 1 ton of such concentrates. This means that ore now worthless can be made into concen- trates richer than any soft red ore now used in the Birm- ingham district, with the possible exception of the Iron- dale seam. There is not in the entire State a more in- viting field for cultivation by the far-seeing iron-master. The enormous expense incurred by Mr. Edison in concentrating the low-grade magnetites of Sussex Co. New Jersey, would not be required here. It is true that he takes an ore of about 17 per cent, of iron and concentrates it to about 63 per cent., and it is also true that his concentrates are Bessemer ore, and worth four or five times as much as the Alabama product would be. But the market for the Alabama concen- trates would be at the very door of the works, and the 276 GEOLOGICAL SURVEY OF ALABAMA. cost of production would be far below the cost in New" Jersey. In urging this matter upon the attention of the pro- gressive iron makers in Alabama, it is hoped that steps will be taken to put to profitable use what is now use- less, and yet is capable of being made of the highest use. We can never avail ourselves of the resources that nature has so bountifully supplied unless we overcome the obstacles that nature herself has placed in our path. The utilization of ,the low-grade soft ores if not now a necessity of the situation will speedily become so, for the other ore is disappearing ; there is not enough cheap brown ore to take its place, and to replace it with limy ore means an increase of the cost account. But tLe low-grade soft red ores are not the only ores- that lend themselves readily to concentration. There- are very large deposits of 'hard' red ore (limy ore) that can not be used because of the low percentage of iron and the high percentage of siliceous matter. In view of the results obtained with the Wetherill. process one is forced to the conclusion that concentra- tion based on previous artificial magnetization cannot be recommended. It is true that the final heads from magnetized ore carry more iron than the final heads from the Wetherill process, but on the average this dif- ference is not above 5 or 6 per cent, and can Dot coun- terbalance the difference in the cost of the two schemes. Furthermore, the waste of iron in the tails from mag- netized ore is very much greater than from the Wetherill machines. Unless all of the ore is thoroughly mag- netized this loss will be constant, and unavoidable. The cost of thorough and uniform magnetization would be very great, even if possible at all. The writer may l>e pardoned for having taken an encouraging view of concentration based on magnetization in 1895, because LOW GRADE ORES. 277 it seemed then to be the only solution of the problem. To concentrate three tons of ore into one would have paid then as'it will pay now. It is probable, from addi- tional study of the subject, that in the magnetization process there would have been required three tons of raw ore forgone ton of concentrates carrying 55 per cent, of iron, but by using the Wetherill process two tons of raw ore will make one ton of 53 per cent, concentrates. To mine and treat one ton of ore for two per cent, of iron does not present many attractive features. The Wetherill process is carried on at so much less expense throughout that if it gives approximately the same results, this feature alone would commend it. The sole advantage that the magnetizing process possesses over the other is in the higher percentage of iron in the final heads, and this advantage disappears entirely when we consider the cost at which it is gained. These two processes have been described because they are the only processes that seem to merit attention, and of these the magnetizing process must now be excluded. If it is asked why either one is to be considered we re- ply because the supply of cheap soft red ore carrying from 4'5 to 48 per cent, of iron is being rapidly depleted, and in a few years will be practically exhausted. This may not be a welcome truth to some, and others will deny it, but it remains, in spite of surprise and denial. So far as concerns the soft red ore the time is not dis- tant when a much higher price will be paid for it than now maintains. The great bulk of the ore on the Red Mountain, near Birmingham, which uninitiated visit- ors regard with wondering eye, is too poor in iron to be used in the furnaces. If used at all it will have to be improved by concentration, or the furnace practice will foe confined to hard (limy) ore and brown ore. We may keep the great out-crops of ore for a sort of 278 GEOLOGICAL SURVEY OF ALABAMA. show-place, as they are to some extent now, and con- tinue to publish photographs showing 15, 20, and 25 feet of ore as evidence of the prodigality of nature, But there is not a single place on Red Mountain, from Irondale to Raymond, where even 12 feet of ore is mined, and the huge seams taken as a whole are worth- less. It is all very well to take visitors to some great cut in the seam, and ask them what they think of that for ore. What they will think depends entirely upon how much they know about the ore. If they do not know much their astonishment will be all that the most accomplished 'boomer' could wish, but if they know the ore they will be apt to ask how it is proposed to utilize such low-grade stuff. This low-grade material, which exists in very large masses, can be utilized by concentration, but until this is done it is commercially of no importance. Concentration of the 'Hard' (Limy) Red Ore. The following experiments were made with the Weth- erill process on the ordinary 'hard' (limy) ores of the Birmingham District. Concentration of Hard (Limy) Ore. Two experiments on the ordinary limy ore are first given . Iron. Lime. Insoluble. Original ore 37.60 15.00 16.20 gave 55 per cent, heads with 48.70 9.76 10.26 15 per cent, middlings with 29.00 21.40 18.20 30 per cent, tails " 18.20 25.12 27.00 With an ore not so good but still passable : Original ore 34.50 17.10 18.04 gave LOW GRADE ORES. 279 64 per cent, of heads with 45.40 11.45 12.25 7 " " middlings " 25.80 24.02 17.95 29 " " tails " 13.55 27.10 30.34 To bring the iron up from 37.6 per cent, to 48.70 per. cent., and at the same time preserve the self-fluxing na- ture of the ore is very encouraging. The second re- sults are still better. The low-grade limy ore was then tried with the follow- ing results : Iron. Lime. Insoluble. Original ore.. 31.80 10.79 33.10 Gave. 44 per cent, of heads with. . .43.15 8.80 19.66 6 " middlings with. 29 .45 12.40 32.90 50 " tails " .22.80 12.52 43.82 Original ore 32.80 9.90 33.70 Gave. 58 per cent, of heads with. . . .44.50 9.00 17.30 10 " middlings with. 35. 90 13.20 23.28 32 " tails " .21.60 8.80 42.70 Other experiments on similar limy ore showed similar results. In its original condition this low-grade limy ore is not self-fluxing, i. e., it does not carry enough lime to flux the siliceous matter, and by concentration it does not become so. But it is greatly improved. In the one case the ratio in the raw ore between the lime and the siliceous matter is 1 : 3, but in the heads it was reduced to 1 : 2.2. In the other case the ratio fell from 1 : 3. 4 in the original ore to 1 : 1.9 in the heads. The original ore is worthless, the. concentrates, while not self-fluxing, are still very good semi-hard ore. The re- lation of the low-grade * hard ' ore to the * hard ' ore mined is approximately the same as that of the low- 280 GEOLOGICAL SURVEY OF ALABAMA. grade soft ore to the soft ore mined. Take for instance, the big seam on Red Mountain. In places it is 22 -feet thick, but will average about 20 feet. Where the lime has been leached out the whole of the seam is soft ore, but only the upper 10 feet is mined, the lower 10 feet being too low in iron and too high in silica to allow of its profitable use in- the furnace. As the seam goes under cover the lime increases and the ore becomes ' hard, ' or limy, and when the lime and the silica are in equal proportions the ore is said to be self-fluxing, as has been fully explained in the chapter on ores. The 8 or 10 feet of the ' hard ' ore next to the roof of the seam is the better portion, just as this part of the leached, or soft ore is the best. The 8 or 10 feet of the ' hard ' ore next to the floor of the seam is too low in iron and lime, and too high in silica to be used. It must be concentrated, just as the corresponding part of the seam towards the outcrop must be concentrated. The following sketch will explain the relative posi- tions of the usable soft and hard ores, and the unusable. LOW GRADE ORES 281 282 GEOLOGICAL SURVEY of ALABAMA. In this sketch the distance along the dip to which the 1 soft,' or lime-free ore goes is taken at 300 feet. This is not always the case. Sometimes the * hard ' or lime- ore, begins much nearer the crop, and at places the soft ore extends further than 300 feet. But no matter whether the distance is more or less than 300 feet even- tually the lime-ore replaces the other and extends from wall to wall. The sketch shows that about one-half of the soft is mined and used, the remainder being unfit for use. It also shows that about one-half of the ' hard * ore is mined and used, the remainder being unfit for use, It is the lower half in each case that must be concen- trated. It is not proposed, at present, to attempt the concen- tration of the upper half, either of the soft, or of hard ore, inasmuch as the prices at which they are delivered render the competition even of better ore very severe. But taking the best case in which one-half of the big seam can be mined, as the sketch shows, the other half is practically worihless as it is. This is the big seam at its best, and there is not much of the minable portion of it left. But in many places, as between Red Gap and Lone Pine Gap, on Rtd Mountain, near Birmingham,, the entire thickness is of low-grade, none of it is fit to use, and the 20 feet would be available for concen- tration. Reasoning from analogy we can expect the entire- seam under cover, and when it becomes limy to be also of low-grade. The question of concentrating the low- grade limy ore, is, therefore, of no small moment. Allow- ing for the sake of the argument that the furnace prac- tice in the Birmingham district will be based more and more on the use of limy ore, and that there will be less and less ' soft ' ore used, where is the limy ore to come from? The estimates as to the amount of limy ore LOW GRADE ORES. 283 available will have to be greatly reduced, and when larger and larger demands are made upon it, as will cer- tainly be the case if the use of sof tjore is lessened or dis- continued entirely, it is doubtful if they can be met, except at an increased cost. Regarded from any stand- point, whether that of soft ore, or of hard ore, concen- tration becomes a ver> live question, and one to which no prudent manager can refuse to give earnest heed. The self-fluxing limy ores of the Clinton formation are highly esteemed, and justly so, for while not rich in iron they carry the lime necessary for fluxing their own silica. This is a great advantage, and any plan that promises to increase the available supply of these ores certainly merits the most careful consideration. Another suggestion that has been made in respect of im- proving the quality of the limy ores is to calcine them and send the hot ore to the furnace. Taking an ordi- nary limy ore, i. e. With iron ,37 %, silica 16%, lime carbonate 28%, if the carbonic acid were entirely re- moved the analysis would show iron 42%, silica 18%, lime 17.9%. One hundred tons would weigh 87.7 tons, and in respect of weight to be handled there would be a positive advantage. Of the raw ore there would be re- quired 2.7 tons per ton of iron, of the calcined ore 2.38 tons, a saving of 716 Ibs., of ^ore per ton of iron. In other words, a 150 ton furnace running on all hard ore requires 405 tons and would*require 357 tons of calcined ore. It it was charged with as much calcined ore as raw ore the output would be 170 tons instead of 150 tons, a gain of 13 % . Some experiments were tried here, but were not con- ducted long enough to warrant one in giving an opin- ion as to the results. There is no difficulty in removing the carbonic acid in a gas-fired Davis Colby kiln, as we found that the ore from the shutes contained only a few 284 GEOLOGICAL SURVEY OF ALABAMA. tenths of a per cent, of carbonic acid, whereas it carried nerly 1 7 % as charged into the kiln. The ore would, of course, still be self-fluxing, and the question would be whether the removal of the carbonic acid outside of the furnace, with the consequent trans- formation of the carbonate of. lime into caustic lime, would benefit the ore more than it would cost. Without entering upon any lengthy discussion, as the matter has not yet passed the experimental stage, we may regard the question briefly, from a physical and a chemical standpoint. Physically the ore would become more porous as the expulsion of the carbonic acid would, to a great extent, destroy its compactness. It would lose in weight, but this would be more than counter- balanced by the gain in the per centage of iron. Its increased porosity would allow easier penetration for the reducing gases of the furnace. Against this may be placed its increased fria- bility, and the consequent production of a greater quan- tity of the fine material in the furnace. Chemically, we should have to consider the effect upon the combustible gases of the introduction of caustic lime instead of car- bonate of lime. The carbonic acid has to be removed and the question narrows down to a single consideration, viz : Is there any advantage in removing it outside of the furnace? The heat within the furnace removes it quite as effec- tively as the heat of a kiln, but then we would have to weigh the effect of large volumes of hot carbonic acid on the coke, with solution of carbon, &c. Cokes differ markedly in this respect, and each one has to be exam- ined in and for itself. If the calcined ore is charged direct it would carry a considerable amount of heat into the upper part of the furnace and it would be more diffi- cult to maintain a cool top. This, however, need hardly LOW GRADE ORES. 285 be considered, as the additional temperature, due to charging hot material, would be derived, not from reac- tions within the furnace, but from extraneous sources. A cool top under ordinary conditions means that the heat within the furnace is used in melting the stock, and is not escaping in the gases. But if a hot top is due to extrane- ous heat, such, for instance, as hot material charged, there would be no injurious effect upon the zone of fusion . It might be advantageous to have a hot top if the heat was not derived from the reactions within the furnace, as the gases to be consumed under the boilers and in the stoves would arrive at the burners at. a higher tempera- ture. Aside from such considerations, however, it seems advisable to use the calcined ore direct. Where it is stocked, or allowed to remain even for twenty-four hours in the air, it rapidly takes up water and becomes pasty. When the slacking of the caustic lime is completed the material appears dry but in reality contains not only water of hydration but carbonic acid also. When the water of hydration is expelled the lime becomes pulver- ulent and dusty, blows about in every breeze and is troublesome to both bottom and top fillers. It can be dampened with water from a hose-pipe, of course, but in that case the mass becomes pasty, and the stockhouse uncomfortable. If the ore is not used direct, (the kiln being in immediate proximity to the furnace), the ad- vantages to be obtained from calcining begin to disappear at once, and continue to become less and less the longer the interval between calcination and charging. CONCENTRATION OF BROWN ORES. Some experiments on concentrating brown ores were made with the Wetherill process, but we did not proceed far enough to obtain any very positive results. We 286 GEOLOGICAL SURVEY OF ALABAMA. found that an ore carrying, on dry basis, 45% of iron, and 18% of silica could be improved so that about 55% of it carried 52% of iron. In the paper by Messrs. Wilkens and Nitze, already quoted, are given results from the trial of some Virginia brown ores. Thus a brown ore from Iron Gate, Alleghany county, gave the following results : Iron. Silica. Original ore 43.08 31.29 Gave Concentrates, 63.4% with 51.04 11.24 Tails 36. 6% with 31.74 Washer tailings from Barren Springs, Va. Iron. Silica. Original ore 32.03 29.93 Gave Concentrates, 30% with.... 53.14 7.43 Tails 70 % with 22.98 39.58 It may be that some such process will be found to be applicable to low-grade brown ores, especially to wash- er-tailings and kiln screenings, but for the ^most part calcination will be used on brown ores for improving their quality. There are doubtless many brown ores whose initial content of iron is so low as to forbid the expense of cal- cining, and some magnetic process may eventually be applied to them. But for brown ores that carry from 40 to 45% of iron, dry, calcining is to be preferred. Calcining is not commonly practiced in Alabama. Some of the charcoal furnace calcine their brown ore, but by far the largest users of the brown LOW GRADE ORES. 287 ore, the Woodstock furnaces at Anniston, and the furnaces at Sheffield and Birmingham do not use cal- cined ore. When calcining is practiced one of two methods are used, the old fashioned open air pile fired with charcoal breeze ; or the new fashioned gas-fired kiln. The former method needs no description. When properly managed it gives fair results, but can not be depended on to give uniformly calcined ore. Even with careful attention, which it seldom gets, a part of the ore will not be calcin- ed at all. a part will be proper]y calcined, and a part will be 'louped'. Attention is being drawn more and more to calcining in gas-fired kilns, and of the various kinds the Davis- Colby is preferred. In this kiln the current of heated gas and flame is drawn across the ore as it descends be- tween the outer walls of the combustion chamber and a central space connected with the stack. The kiln is built of any convenient size, from 100 to 150 tons capacity, &nd is fired with producer gas. Allowing 7 per cent, of hygrocopic water, removable at 212 deg. F, and 7 per cent, of combined water, remov- able only at red heat, a kiln holding 125-140 tons of raw ore will deliver from 107 to 120 tons of thoroughly and uniformly calcined ore per 24 hours, with a consumption of 2i to 3 tons of coal. To calcine one ton of raw ore (2240 Ibs.) requires about 52 . Ibs, of coal. The advantages of the gas-fired kiln are economy of labor, and uniformity of product. These advantages maintain under all conditions, except where the price of coal is prohibitory, and even there the wood-fired or charcoal-fired producer may be used. The use of all brown ore in coke furnaces may be ren- dered necessary by contracts specifying that the iron shall be made from brown ore, or by proximity to de- 288 GEOLOGICAL SURVEY OF ALABAMA. posits known to be very considerable. A determination on the part of furnace owners to make a special high grade charcoal iron would also entail the exclusive use of brown ore. A kiln to treat 140 tons of raw ore per day, with pro- ducer and all necessary fittings, will cost about $7,000, and will yield ordinarily about 120 tons of calcined ore. This amount would contain from 60 to 65 tons of iron, and would be equivalent to 20 per cent, of the ore bur- den for two 150 ton furnaces. The freight on a ton of raw ore from the washer to the furnace may be taken at 25 cts. in the Birmingham dis- trict, and if the ore averages 47 per cent, of iron we would have 1052.8 Ibs. of iron costing for freight 25 cts. The freight on a ton of calcined ore would also be 25 cents, bat it would contain 54 per cent, of iron, or in the ton 1209.6 Ibs. of iron. So far, therefore, as concerns the transportation charges we would get 1209.6 Ibs. of iron in the calcined ore at the same price paid for 1052.8 Ibs. in the raw ore. Each ton of calcined ore delivered at the furnace would contain 156.8 Ibs. of iron'more than a ton of raw ore. If it requires 4 men in the stockhouse r as bottom-fillers, to handle 140 tons of raw ore per day y containing 65.8 tons of iron, 3 men could handle the 121.7 tons of calcined ore required for the same amount of metal. So far as concerns the handling of the ore in the stockhouse there would be a saving of one man at each furnace by substituting calcined ore for raw ore. The economy becomes even more striking if we con- sider the kiln as situated at the furnace, so that the bot- tom-fillers could draw the ore from the shutes. At one well managed plant this has been the practice for several years. The trams come in from the washer and dis- charge into the kiln. The bottom-fillers draw from the shutes into the buggies, and the hot ore goes at once to LOW GRADE ORES. 289 the furnace. At this establishment it has been shown that there is great advantage in the use of calcined ore, irrespective of the easy way of handling it in use, and it fortunately happens that it is able to compare, for a term of years, the practice on raw ore, pile-calcined, and kiln-calcined ore. It is not going too far to say that it would be profitable to erect kilns at the furnaces, even when the ore has to be hauled at a freight cost of 25 cts. per ton, or even more. Excessive freight charges on ore would, of course, militate against this proposition, but until they rise be- yond 40 cts. per ton calcining would be advantageous. The erection of kilns at the mines, except under unus- ual conditions, can not be recommended, for the reason that the life of a brown ore deposit is uncertain. But at the furnace, and especially where coke is made on the spot and it is possible to calcine with waste gases from the ovens, this objection is removed. The furnace operator would be able to buy ore from the smaller mines which can hot incur the expense of building kilns, the entire process would be under one management, and the utilization of gases now going to waste would, of it- self, show a profit. It is a truth of general application that it pays to cal- cine brown ore, for it has been shown to be beneficial wherever it has been carefully and faithfully carried out. 19 290 GEOLOGICAL SURVEY OF ALABAMA. CHAPTER X. BASIC STEEL AND BASIC IRON. The manufacture of basic open-hearth steel in Ala- bama began on the 8th of March 1888 at North Birmingham. It was the first attempt at steel making in the State, and this furnace was among the first basic open furnaces built in the United States, if not the first. The enterprising character of the men composing the Henderson Steel and Manufacturing Company in under- taking at this early date to enter upon the production of basic steel when there was but one other establishment in the country is deserving of the highest praise. There was very little known about basic steel then, for the development of the industry has been rendered possible during the last 10 years. The Henderson Steel nnd Manufacturing company may, therefore claim, to have been the pioneers in an industry which has grown to very large proportions elsewhere in this country and which now promises to be of increasing-importance here. While the operations at North Birmingham did not attain the commercial success so well deserved by the faith and progressiveness of the promoters, technically the process even then was successful. In its essential construction and operation the furnace did not differ from those now used, for although what was known as the Henderson process was employed yet there was no real difference between it and the more recent modifica- tions of the basic open-hearth. To the kindness of Mr. H. F. Wilson, the secretary of the company, the writer is indebted for some data con- cerning this furnace. It was of 13 tons capacity, and made 200 heats before it was closed down. The. maxi- mum out put in any one day of 24 hours was 25 tons, BASIC STEEL AND BASIC IRON. 291 and about 1600 tons of steel were made. The steel was sold, as ingots, to the Bessemer Rolling Mill Company, Bessemer, Ala., for about $22.00 a ton and they made most excellent boiler plate of it. Crellin and Nails, Birmingham, manufactured boilers of it, and some of their work may now be seen in the grain mill of Mr. B . B. Comer, Birmingham. The pig iron used was mottled and white of local pro- duction. Mr. E. E. Robinson was melter. The follow- ing table gives the composition of the heats and the analyses of the steel from heat No. 93 to 105, inclusive 292 GEOLOGICAL SURVEY OF ALABAMA, I I 4 o o to be a n o *~^ '. rj CT^ rl W P i f Cu c Q 5 cc i ! < i IT'COC 1C 1O id iQ tO 1O lO lO 1C C C iC lO C <-C 1C 1C 1C 1C 1C tO cocot-cot-cooocot-cocot-t- CM o !> O O CM ^ co co oo oo o I co ^ t> T"^ CO OC r-^ CD *~4 CO CD GO "^ I "^ CD_CD CD_OQOO' tOi IT iO5O O5 05' O1O1O5O5O5O5O5OOOCDOO O I. 05 O . BASIC STEEL AND BASIC IRON. 293 Additional information in regard to the early history of steel-making in Alabama is contained in a pamphlet entitled ''Basic Steel. Report of committee on its suc- cessful and economical manufacture by the Henderson Steel and Manufacturing Company, North Birmingham, Ala., August 27th, 1890." This committee was composed of A. B. Johnston, president Birmingham Chamber of Commerce; W. H. Hassinger, manager Alabama Rolling Mill, Gate City ; G. L. Leutscher, chemist Tennessee Coal, Iron and Rail- way .Co. ; P. Leeds, superintendent machinery Louisville and Nashville Railway Company ; -and H. R. Johnston. Mr. Gogin was at that time manager of the steel 'Company. Tliis committee reported that on August 19th, 1890, there was charged into the furnace White Pig Iron from Pounds. DeBardeleben furnaces Bessemer, Ala 15,000 Pit scrap 5 ,525 Miscellaneous scrap 4,514 Brown ore, 55 per ct. iron 742 Spiegel 200 Ferro-manganese 200 Total metal 26,181 The quantity of fluorspar and limestone was not given. The yield of metal was Pounds. 24 steel ingots 22,250 Pit scrap 1,510 The yield then was 85 per ct. of ingots and 6 per ct. pit scrap, and the loss of metal about 9 per ct. The committee, further reported that basic billets and slabs .could be made for $22.00 a ton. 294 GEOLOGICAL SURVEY OF ALABAMA. The analyses quoted were as follows : WHITE PIG IRON. Silicon . . - .43 per ct. Sulphur 0.149 ' Phosphorus 0.68 ' ' Manganese 0.10 " BROWN ORE. Metallic iron 56.12 per ct. Phosphorus , 0.34 ' Insoluble residue 4.99 " LIMESTONE. Carbonate of Lime 95.71 per ct. Alumina and Oxide of Iron 1,04 - " Silica 1.33 " STEEL. Silicon Trace . Sulphur 0.06 per ct. Phosphorus 0.018 " Manganese 0.29 " Carbon 0.08 '" The writer made an analysis, in 1890, of a sample of the first heat of basic open-hearth steel March 8th, 1888. which had been drawn out, under a hammer and found its composition as follows : Analysis of the first heat of basic open-hearth steel made- in Alabama, at North Birmingham, March 8, 1888 : Silicon 0.023 per ct. Sulphur 0.014 " Phosphorus 0.038 " Manganese .144 ' ' Combined Carbon , . . . . .0.484 " Graphitic Carbon ', . , ..0.095 " BASIC STEEL AND BASIC IRON. 295 The report of the committee also stated that the phys- ical tests of the steel they examined were as follows- plate f xl. 4-in. sect. 8-in. sect. Lbs. Lbs. Ultimate tensile strength per sq. in . . 48,110 48,460 Elastic limit per sq. inch 32,030 32,275 Reduction of area 54.7 perct. 57.4 perct. Elongation 32.0 " 28.0 perct. A sprue of the first group of ingots was forged into a bar 1 inch square, and was bent when cold, with a sledge until perfectly folded. Not . the slightest flaw could be detected at the fold. Excellent razors and knives were also made of this steel, and some of them are still in use in Birmingham, It is, therefore, to be concluded that the first basic open hearth furnace in Alabama, and one of the first in the United States, beginning operations in March, 1888, made excellent steel of native materials. The process was handicapped with white pig iron high in sulphur and of irregular composition, as also by lack of experi- ence on the part of the operators, and many other ob- stacles besetting a new enterprise, but the promoters had the courage of conviction, and went as far as their means would permit. They are entitled to and should receive the highest commendations for what they did, for they laid the foundations of the steel industry in this State. The times were not ripe for the commercial success of the enterprise then, and it was not until the middle of 1897 that they seemed to hold out promise of fruition. The Jefferson Steel Company succeeded the Hender- son Company, and operated the North Birmingham furnace in 1892 and 1893, making, perhaps, 1600 tons of steel, under the management of Ernst Prochaska. 296 GEOLOGICAL SURVEY OF ALABAMA. The operations were suspended during the summer of 1893. Here the matter rested as to Birmingham until 1897, for the crude experiments carried on under the Hawkins process at North Birmingham in 1895 can not fairly be included in a historical sketch of the rise of the steel industry here. The amount of basic open hearth steel made at Birm- ingham, all of native materials, except as to spiegel, ferro-manganese and fluorspar, up to July 22nd, 1897, would not exceed 3500 tons, if indeed it is above 3000 tons. Basic Open-hearth Steel at Fort Payne. Steel was next made at Fort Payne, but in spite of repeated inquiries no definite information could be se- cured. BIRMINGHAM ROLLING MILL COMPANY. In 1897 the Birmingham Rolling Mill Company, which had been in successful operation for a number of years, and which of late had been buying steel billets in Pennsylvania and rolling them into shape here, took up the matter. The citizens of Birmingham subscribed to the undertaking to the amount of $40,000 and the first basic open hearth furnace went in July 22nd, 1897, being followed by the second on October 25th. Both furnaces were designed and built by S. R. Smythe & Co., Pittsburg, Pa., with a capacity of 35 tons each to the charge. The iron used was the basic iron made at the Alice furnace, within 200 yards of the mill. The quality of the metal has been and is now of an excellent quality, as the following analyses of the first 245 heats will show, in respect of chemical composition. The chemical composition of the metal is given in the following tables : BASIC STEEL AND BASIC IRON. 297 Analyses of the first 245 heats of basic open-hearth steel made by the Birmingham Rolling Mill Company, Birmingham, Alabama, from July 22nd to December 31st, inclusive, 1897. SULPHUR. 0.015 to 0.020 ieais. . ..31= 70 12.7 0.020 " 0.025 . . . 69= 28.1 0.025 " 0.030 . . .81 = 33.1 0.030 " 0.035 ...33= 13.5 0.035 " 0.040 ...17= 7.0 0.040 " 0.045 .-. . 8 3.2 0.045 0.050 ...2= 0.8 0.050 " 0.055.. ... 1 0.4 0.055 " 060 ] = 0.4 0.060 " 0.065 2 0.8 245 Average sulphur 0.028% PHOSPHORUS. Heats. % 0.001 to 0.005 . 100= 40.8 0.005 " 0.010 . 49 20.0 0.010 " 0.015 . 15 6.1 0.015 " 0.020 . 16= 6.5 0.020 " 0.025 . 18= 7.3 0.025 " 0.030.. . 13 5.3 0.030 " 0.035 . 8 3.3 0..035 " 0.040 5 2 0.040 " 0.045 . 6 2.4 0.045 " 0.050 3= 1.2 0.050 " 0.055 1= 0.4 0.055 " 0.060. . 4= 1.6 2&8 GEOLOGICAL SURVEY OF ALABAMA 0.060 0.065 0.065 " 0.070 0090 " 0.095 0.095 " 0.100 0.100 0.150 0.150 " 0.200.. 2= 0.8 1= 0.4 l^= 0.4 1= 0.4 1= 0.4 1= 0.4 245 Average phosphorus 0.012 % Average manganese, 0.45. carbon,. . . 0.18. silicon.. ..0.008. It will be seen that in 181 heats out of 245, or 73.9 per cent., the sulphur reached a maximum of 0.030 per cent., while in 64 heats, or 26.1 per cent., it was above 0.030 per cent. In only 14 heats out of 245, or 5.6 per cent., was it above 0.040 per cent. In a list of sulphur estimations in basic open hearth steel, given by H. H. Campbell (Manufacture and Prop- erties of Structural Steel, 1896, pp. 321 and 322), the number of heats examined was 973. Of these, 255 heats, or 26.2 per cent., showed a maximum sulphur of 0.030 per cent., while 618, or 63.5 per cent., gave sul- phur above 0.030 per cent. The conditions as to sulphur are then seen to be in the case of the Birmingham steel almost the reverse of those maintaining in the basic steel quoted by Mr. Campbell. In the Birmingham steel 73.9 per cent, of the heats showed a maximum sulphur of 0.030 per cent., while in the steels quoted by Mr. Campbell, and pre- sumably of northern make, there were 63.5 per cent. above 0.030 per cent, in sulphur. In the Birmingham steel there were 26.1 per cent, of the heats with sulphur above 0.030 per cent., as against BASIC STEEL AND BASIC IRON. 299 63.5 per cent, in the other steels. Furthermore, in Mr. Campbell's steels there were 143 heats out of 973, or 14.7 percent., in which the sulphur was above 0.040 per cent, as against 14 heats out of 245, or 5.6 per cent., of Birmingham steel, and in Mr. Campbell's steels there were 87 heats out of 973, or 8.9 per cent., in which the sulphur was above 0.050 per cent., as against 4 out of 245, or 1.6 per cent., in the Birmingham steel. It is, however, in respect of phosphorus that the chief obstacles were encountered and successfully overcome. The sulphur may be considered an element whose maximum in the steel may be more easily controlled than that of phosphorus, especially when the pig iron used is low in sulphur. If the maximum sulphur in the pig iron is 0.050 per cent, the removal of 50 pe r cent, would cause the steel to carry from this source, 0.025 per cent. But with phosphorus at 0.75 per cent, in the pig iron 86.6 per cent, must be removed to bring the steel down to 0.10 per cent, the maximum allowable under most circumstances, while 93.3'per cent, must be removed to bring it to 0.05 per cent. Basic open hearth steel has been made in Birming- ham of pig iron, pit scrap and ore, in which the phos- phorus was below 0.050 per cent, and in some cases be- low 0.010 per cent. The phosphorus estimations given in the preceding lists are of steel made with various mixtures of pig iron and scrap and ore, and there is practically no difference between them. An examina- tion of the list shows that 149 heats out of 245, or 60.8 percent, gave a maximum phosphorus of 0.010 per cent, while 180 heats out of 245, or 73.4 per cent, gave a max- imum phosphorus of 0.020 per cent. Putting the phos- phorus limit in the very highest grade of basic open- hearth steel at 0.030 per cent, we find that 86 per cent. 300 GEOLOGICAL SURVEY OF ALABAMA. of the heats showed a maximum of this amount, and in 40.8 per cent, of the heats the maximum phosphorus ivas 0.005 per cent. In the results given by Mr. Campbell (ut supra) we find that in 157 heats out of 973, or 16.1 per cent the maximum phosphorus was 0.010 per cent, as against 60.8 percent in the Birmingham steel with a maximum of 0.010 per cent. In the northern steels there were 770 heats out of 973, or 79.1 per cent, in which the maxi- mum phosphorus was 0.020 per cent, as against 73.4 per cent, in the Birmingham steel with a maximum of 0.020 per cent. The percentage of heats in the north- ern steels with maximum phosphorus 0.020 per cent, is somewhat higher than in the Birmingham steel. In the northern metal there were no heats in which the phosphorous was below 0.005 per cent. while, as before stated of the Birmingham steel 40.8 per cent, of the heats had maximum phosphorous 0.005 per cent. Of the northern steels there were 898 heats out of 973, or 92.3 per cent, with maximum phosphorus 0.030 per cent, as against 86 per cent in the Birmingham steel. But when one considers the number of the heats of north- steel in which the phosphorus is above 0.030 percent it is found that they are 75 out of 973, or 7.7 percent, while the corresponding percentage in the Birmingham steel is 13.7, nearly twice as many. Taking everything into consideration, however, with due regard to the newness of the conditions surround- ing the production of steel in Birmingham, and the fact that the results here given are from many different mixtures in the furnace we conclude that in chemical composition the steel compares very favorably with stan- dard makes of ngrthern steel, and that the severest specifications could be successfully met. The following table gives the results of the examina- BASIC STEEL AND BASIC IRON. 301 tion of some basic open hearth steel plates made by the Birmingham Rolling mill, for elastic limit, tensile strength, elongation and reduction. All the chemical analyses, as well as the physical tests were made by Mr. David Hancock and the writer in the Phillips Testing Laboratory, Birmingham. TABLE XLV. Giving Physical Tests of Basic Open Hearth Steel Plates made by the Birmingham Kolling Company, 1897 1898. Specimen of Plate. Size. Elas. Limit Lbs. Per sq. Inch. Ten. Str. Lbs Per sq. Inch Elongation in 8 Inch per Cent. Reduct of area Per Cent. 5-16 inch. 85.360 65,600 25.7 49.6 5-16 inch. 34.720 62,440 27.2 52.6 5-16 inch. 35,200 63,720 27.5 51.5 5-16 inch. 33,300 58,290 26 49.6 5-8 inch. 33,930 57.900 ;5.0 53.0 5-8 inch. 28 900 53,680 32.5 51.0 5-8 inch. 31.040 52,510 27.0 52.8 5-8 inch. 32,360 53,390 31.7 56.5 7-16 inch. 31,400 50,520 32.0 64.0 7-16 inch. 32,360 50,650 ?0.7 61.6 7-16 inch. 29,960 51.130 30.0 60.8 7-16 inch. 32,790 53,960 27.2 57.4 7-16 inch. 32.760 53,360 26.5 55.7 7-16 inch. 32,260 53,420 30.5 58.0 1-4 inch. 39,560 58,420 27.8 53.1 1-4 inch. 41,450 57,260 25.0 54.9 1-4 inch. 43,040 64,380 25.0 55.1 1-4 inch. 43,470 63,310 25.0 50.6 1-4 inch. 44,280 58,480 26.7 55.8 1-4 inch. 44,850 57,490 26.0 54.9 1-4 inch. 43,590 56,680 26.0 54.9 l%round. 32,680 50.520 32.5 63.5 l^round. 37.560 58,940 30.0 53.9 The plates tested were 16 inches long over all, Siaches long and 2 inches wide between fillets, with a fillet ra- dius of H inches. They were pulled on a 200,000 Blehle Testing Machine, with automatic extensometer and electric registration, the elongation being after- 302 GEOLOGICAL SURVEY OF ALABAMA. wards checked by measurements. Numerous other tests might be given but is is thought that these will be sufficient to show the quality of the material made from the basic iron of the Birmingham district. Up to the 1st. of May 1898, 500 heats had been made and the two furnaces are now in active operations. The material is made into boiler and tank plates, fire-box sheets, rounds, flats and squares, and is sold under specifications as to chemical composition and physical tests. It is certainly excellent work even for an old estab- lished steel works to make basic open-hearth steel of such quality that in 245 heats practically 74 per cent, contained a maximum amount of sulphur of 0.030 per cent, and 86 per cent, a maximum of 0.030 per cent, of phosphorus. These results have been reached in Birm- ingham by the first open-hearth furnaces on regular run, and have been extended over nearly six months. Can they be continued indefinitely? Are these results typical of what may reasonably be expected in the fu- ture? Were there any favorable conditions surrounding these 245 heats from July 22d to December 31st, that would not maintain in any number ? These are vital ques- tions, and upon the answers to them depend the future of the manufacture of basic steel in Alabama, as, indeed, in the entire South, for if this steel cannot be made in Alabama, it cannot be made anywhere south of the Potomac river. In the Birmingham district, as, indeed, everywhere else, there are two aspects of the steel industry techni- cal and commercial. While the metal produced may be of the best quality so far as concerns chemical and phys- ical tests, and while assurance may be given that the raw materials, of which the pig iron is made, exist in very large quantities, yet, after all, the main question is, whether the steel can secure and hold a profitable market. BASIC STEEL AND BASIC IRON. &Q3 Technically the basic open-hearth steel made at Birm- ingham -is of a superior quality. The pig iron, which is the chief constituent, can be made here at a less cost than anywhere in the United States. These are facts beyond dispute. But they are not the only considera- tions which affect the establishment and development of the steel industry in Alabama. It is comparatively easy to convince even the most skeptical that excellent- steel can be, has been, and is today, made here in quan- tities that fully warrant the assertion that the matter has long since passed the experimental stage. What is to be done with the metal after it is made? Can steel- makers in Alabama enter the. steel market and obtain for their product the footing now enjoyed by Alabama pig iron, for instance? These are questions which only the lapse of time can fully answer. An industry may be established technically, and that within a compara- tively short time, while its establishment commercially may be protracted through a number of years. This is is a matter which in some of its aspects is disconnected from the quality of the metal, and depends not only upon the management, but also and particularly upon the especial kind of competition which the metal has to meet. In rectangular shapes, in rounds, in tank and boiler plate, in sheets, in structural material and agricultural steel, the competition varies according to circumstances, and a fully equipped plant must be able to enter the market offering the best inducements for each class of goods. These are matters, however, which may be left to take care of themselves. Once established, the two facts that excellent steel is made here, arid that the chief materials of its production are obtained in the district, and the 304 . GEOLOGICAL SURVEY OF ALABAMA. growth of the industry follows in accordance with the usual laws of industrial development. With the exception of the magnesite for the lining of the furnaces and fluorspar, there is not a single material which cannot be furnished either in the Birmingham district or within easy reach of it. Manganese ore for ferro-manganese and spiegel, iron ore for basic pig, ferro-silicon and "fix," limestone for flux, can all be obtained here as cheaply as at any point in the United States. With large works there might be some difficulty in securing wrought and steel scrap to supplement the scrap produced at the plant itself, but excellent steel has been made here without the use of outside scrap. It is not necessary to use the pig and scrap process, for the pig and ore process has been used with very satisfactory results. Speaking from a full knowl- edge of the subject and with due regard to the emergen- cies that may arise, it is asserted that there is not a sin- gle thing required in the manufacture of steel that can- not be produced here with the exception of magnesite and fluorspar. This statement may cause some surprise, for while it is known that basic iron, which is the chief raw material for the steel-maker, is made here, yet it is not known that ore for ferro-manganese, ferro-silicon, spiegel and "fix" can be obtained in Alabama. It has been sup- posed that the resources of the State were limited to the pig iron and the limestone , but this is not true. There is no special ore needed for ferro-silicon, and it can be made of Red Mt. ores quite as readily as from the ore now used elsewhere. Ten years ago, without any spe- cial effort to make high-silicon iron, it was made here with 7 per cent, of silicon, and this amount can be in- creased to 10 per cent, if a sufficient demand should arise. As to ferro-manganese and spiegel, manganese ores of 44 BASIC STEEL AND BASIC IRON. 305 to 48 per cent, of manganese can be delivered in Birm- ingham for $8 a ton, while the deposits of magnetic ore not yet utilized can be drawn upon for material carrying 60 per cent, of iron to be used in the pig and ore process. But failing this, brown ore has already been used with good results. As to basic iron, the industry has been established here two and a-half years. The iron has been shipped to the following steel makers : Aliquippa Steel Co Pittsburg, Pa- American Steel F. Co St. Louis, Mo. Apollo I. & S. Co Pittsburg, Pa. A. & P. Roberts Co Pencoyd, Pa. Birmingham Rolling Mill Co .Birmingham, Ala. Builders' Iron Foundry Boston, Mass, Burgess S. & I. Co Portsmouth, Ohio. Carnegie Steel Co. Ltd . .Pittsburg, Pa. Cleveland Rolling Mill Co Cleveland, Ohio. DeFour & Bruzzo Italy. DeKalb Company Fort Payne, Ala. Elmira I. & S. R. M. Co Elmira, N.Y. Granite City Steel Co E. St. Louis, Mo. Illinois Steel Co Chicago, 111. Jefferson Steel & Mfg. Co Birmingham, Ala. Jones & Laughlins, Ltd Pittsburg, Pa. Kellogg Weldless Tube Co Findlay, Ohio. Kirkpatrick & Co Pittsburg, Pa. Midland Steel Co Muncie, Ind. Mt. Vernon Car Mfg. Co Mt. Vernon, 111. Nashua I. & S. Co Nashua, New Hampshire. Naylor & Co Pitts.! .hard and brown, or from soft and brown. There are results from the use of hard, soft and brown, and from hard ore and soft ore, with no brown. There is but little good in discussing the adaptability of brown ore alone from this purpose, as it is already known to be suitable, as also that the cost of production would tte considerably higher, even if brown ore could be obtained in sufficient quantities. The best practice will, therefore, be to continue the use of the three ores already tried, while striving to in- crease the proportion of limy ore. The low cost of basic iron in the Birmingham district is certainly a strong argument for its production here. So long as ore suitable for producing this kind of iron can be laid down in the stockhouse for 1-J- to 2-J- cents per unit of iron, the manufacture of basic iron mav be com- mercially profitable. Whether the manufacture of basic steel will follow upon the manufacture of basic iron is another question. BROWN ORE, OR LIMONITE. An average analysis of the brown ore used in the pro- duction of basic iron is as follows : Per cent. Hygroscopic water 7.00 Combined water 6.00 Metallic iron 48.54 Silica 11.22 Lime .84 Alumina 3 .61 Phosphorus 0.38 Surphur 0.09 320 GEOLOGICAL SURVEY OF ALABAMA. If carefully mined and washed the brown ore is of fairly uniform composition . No calcined brown ore has- been used in the production of basic iron. Some good basic iron was made in 1892-93 from brown ore exclu- sively, but of late it has been used to the extent of about 20 per cent. only. Good basic iron has been and can be made without using brown ore, but if it be omitted there is an increas- ed risk of an excess of both silicon and sulphur. For instance, it was found that the best results were ob- tained by using an ore burden containing 20 per cent, of brown ore, irrespective of the percentages of hard and soft ore, which may vary within wide limits. So far as concerns the ore burdens the records cover a consider- able range, from 36.10 per cent hard, 42.0 per cent. soft r and 21.3 per cent, brown, to 64 per cent, hard, 36 per cent, soft, and no brown. Of 30,222 tons of iron spec- cially examined with reference to the ore burdens on w^hich it was made, the brown ore showed the following percentages, viz : ; 8.9 ; 10.6 ; 14.5 ; 19.1 ; 20.0 ; 20.1 ; 20.3 ; 21.1 ; and 21.3. When running exclusively on hard and soft ore the average silicon was 0.68 per cent., the av- erage sulphur 0.043 per cent., and the average phos- phorus 0.70 per cent. With an ore burden of 52.3 per cent, hard, 27.5 per cent, soft, and 20.3 per cent, brown ths average silicon was 0.47 per cent., the average sul- phur 0.033 per cent., the phosphorus remaining the same. Important as are these differences between the silicon and the sulphur, they become even more so when it is stated that the chances of exceeding the 1 per cent, of silicon with an ore burden containing no brown ore are nearly four times greater than when 20 per cent of brown ore is used, and the chances of exceeding 0.050 per cent. BASIC STEKL AND BASIC IRON. S2T sulphur are more than twice as great. Furthermore^ the range of both silicon and sulphur is wider when brown ore is omitted than when 20 per cent, of it is used. Lastly, the average consumption of coke per ton of iron with no brown ore was 1.53 tons, and with 20* per cent, of brown ore it was 1.19 tons. The saving of flux with increase of hard ore is a par- tial offset to the advantages arising from the admixture of brown ore, but after deducting this the balance is de- cidedly in favor of the use of brown ore. THE FLUXES. The basic iron of 1892-93 was made with limestone as flux, but during the last 12 months dolomite has been exclusively employed in the production of the best quality of basic iron, The experience of the last year was not favorable to the use of limestone. The basic iron fell off in quality, and varied widely in composition, when limestone was used. It carried 4 per cent, of silica and 53 per cent of lime, with 0.40 per cent of oxide of iron and 0.60 per cent, of alumina, on the average, but var- ied widely in composition. The dolomite that was used had the following average composition : Per cent. Silica 1.50 Oxide of iron 0.60 Alumina, 0.40 Carbonate of line 54.00 Lime 30.31 Carbonate of magnesia. 43 .00 Magnesia. 20.71 The value of magnesia as a desiliconizer and desul- phurizer in the blast furnace is still somewhat in dis- pute, but the experience with dolomite here has proved, beyond question, that it can be used with great advan- 21 322 GEOLOGICAL SURVEY OF ALABAMA. tage. Dolomite has to a large extent supplanted lime- stone in the Birmingham district within the last year as a flux on ordinary grades of iron, and is exclusively used on basic iron. The amount of dolomite used, per ton of iron, varies, of course, with the amount of hard limy ore used. For basic iron the variation was from 0.12 ton with 81.2 per cent, limy ore and 38.8 per cent, soft ore to 1.08 tons with 36.2 per cent, limy ore, 53.2 percent, soft, and 10.6 percent, brown ore. In pounds, per ton of iron, the variation, then, was from 260.8 to 2419.2, certainly a wide range, and one that shows the fluxing power of the limy ore to great advantage. When 268.8 Ibs. were used the consumption of other ingredients, per ton of iron, was 2.64 tons of ore and 1.55 tons of coke, and the make of iron, under these con- ditions was 520 tons. When 2419.2 Ibs. were used the consumption of other materials, per ton of iron, was 2.25 tons of ore, and 1.32 tons of coke, the make of iron be- ing 2068 tons. In making 7424 tons of basic iron the consumption in tons per ton of iron was : Ore 2.10 Dolomite 0.92 Coke 1.23 The ore burden being composed, in percent, as follows : Limy ore 36.1 Soft ore 42.6 Brow T n ore . 21.3 And the total burden Limy ore 17.8 Soft ore . ..21.0 BASIC STEEL AND BASIC IRON. 323 Brown ore . 10.5 Dolomite '22.0 Coke .28.7 This matter will be discussed more fully under the heading 'Furnace Burdens.' FUEL. All of the basic iron is coke iron, the coke used being the ordinary 48 hour ' 'bee-hive," made from washed slack-coal. The average analysis is as follows : Coke. Ash of Coke. Per Cent. Per Cent. Moisture 0.75 Silica 45.10 Volatile matter 0.75 Oxide of iron. . . . .12.32 Fixed Carbon. . 89.00 Alumina 31.60 Ash.. . 9.50 Lime. . 1.50 100.00 Magnesia trace. Sulphur 1.00 Sulphur 0.10 Phosphorus 0.02 The ultimate strength of the coke is about 2000 Ibs. for a 1-inch cube, and the compressive strain about 500 Ibs. The apparent specific gravity is 0.89, the true specific gravity 1.80, the percentage of cells by volume is about 45, and the volume of the cells in 100 parts by weight is about 47. In structure the coke is generally fine grained and close, and breaks into lumps rather than fingers. It is a small celled coke with strong walls, and carries a good burden, 1 Ib. carrying as much as 2.54 Ibs. The consumption of coke, in tons per ton of iron, varies from 1.56 when using 64. 6 per cent, limy ore, 35. 4 percent, soft, and no brown to 1.05 when using 60 per cent, limy ore, 20 per cent, soft, and 20 per cents. 324 GEOLOGICAL SURVEY OF ALABAMA. brown, the respective returns being based on 1221 and' 2984 tons of iron. It is much better to state the matter in this way thanj to give the average over a long period during which the burden is changing constantly. When no brown ore is used the consumption of coke is high. For instance with 64% limy ore and 36% soft,, the make was 3521 tons, and the consumption of coke per ton of iron, was 1.52 tons. With 81.2 per cent, limy ore, and 18.8 per cent, soft, the make was 520 tons, and the consumption of coke 1.55 tons. With 64.6 per cent, limy ore, 26.5 per cent, soft, and! 8.9 per cent, brown the make was 1140 tons, and the consumption of coke 1.24 tons. Lastly, with 36.1 per cent, limy ore, 42.6 per cent, soft, and 21.3 per cent, brown, the make was 7424 tons, and the consumption of coke 1.23 tons. Per pound of iron, then, the consump- tion of coke varies, according to the burden, from 1.36- Ibs. down to 1.18 Ibs. Many other instances could be given but these are sufficient for the present purpose. Coke of the kind described above can be secured here- in large and regular shipments. During the last few years great improvements have been made in the Birm- ingham district in the manufacture of coke, especially in utilizing slack-coal and the best coke now made here will compare favorably with the best coke made any where else in the United States, as has been abundantly substantiated not only by chemical and physical tests, but also and particularly by furnace records. As re- gards basic iron there are records of the production of more than 22,000 tons showing the average consumption of coke, per ton of iron, as 1.26 tons. Considering the physical and chemical irregularities of the ore, points which have always to be borne in mind when discussing BASIC STEEL AND BASIC IRON. 325 rthe blast furnace practice in the Birmingham district it is a good result to obtain a pound of iron with 1.18 Ibs. & g o d * 2 PQ d oT o w g 43 > S i-3 o co H ^ M <^ n3 d S S s M O - CQ O> 02 A ^ ^ OH g ^ 3 l- W M 00 d s *H 3 PQ tH uap.inq 'sqj seia.T^o 9>[oo - q{ y cs oo oi O5 T 1 T 1 T i ( Phosphorus. 6 fc 1 1 ^ rr Oi QO T-H t^ CD CO t^ CO Tfi O t^ ^J 0^ CO O5 t^ rH t^ cr> CD t~ rH CO O5 1^ O5 t-co coco Sulphur. 6 , ft -4-3 1 * Tj CD OS ^ rtl *& "F ^t" O O O O CO 1 O t-- i lO SS OJ *O (M CD CO ^ ^ O O O T^ CC O CD Tfl Tf< O 1O Silicon. J 4d cc C3 Q ^ CO 05 o co -ti CD CD rfi Oi 1C CO OC O CO OO * OJ 05 CD 00 t- - QO ^ O5 CD lO 1C * OO Consumption. "o B O ^ c JH O a^ 02 I l^^ox (M CO Thi T 1 10 CO CO CO ^H -tl !f TfH 9^ C4 CD OC 1C 1C 1C r^l 1C _- a^iiu -oioa CO OS rt< (M CD CO lO T ( d d o o 9.TQ ^f cc O -* 1 CO CO CO CD o.i oi 01 oj ap:i UOJI JO SUO ^./M o Ol Ol O5 OJ lO CM i i 1C CO'T-H" " 00 O) U) * C I c JOO 9CJIUI ^opn ws 00 CO GO Tf CO CO CO T}H CO CO CO CO 1C O T CO ^ 10 03 lO " CD ^ t>- ^^ GOOD OT-H < - OI J pjH WOS pjB H ,1 CO - OI _iisi O "* tO OO CD O GO OO CO W CO ' CD 04 Ttl -f - 1 1 CO CD CD OC saS.iBi{Q jo -o^ GC Oi Oi C5 Tt< >o co r GC^CD lO OI i-H 90U9.T8J9^[ i-H OICO -* 328 GEOLOGICAL SURVEY OF ALABAMA. jo -sq| 9^00 qn TABLE XLVIII. Burdens of Limy (Hard) Ore, So Tons of 2240 Ibs. I 8 K 3u 5i O> .5 OB 9310 'UIO^OQ UOJI JO SUOJ, UIOfOQ Of ore burden jo CDCOOOOi^COOO CO CO C^-l OC JO * ^ CO CO CO C^l CC ^J 8ggo88 ooooocooct^ocoio COTf(' inia ' \ - - 73.604 97,562 Alabama ) New England ) New York, [ .................. 22,692 79,141 New Jersey, ) I . . 20,244 29,720 Wisconsin \ 336,403 556,491 Virginia and Alabama are grouped together, but cer- tainly the production of Alabama alone would not fall much short of 50,000 tons; and about 40,000 tons for 1897. The returns for 1897 under New York and New Jer- sey include also New England. Virginia and Alabama include Maryland. Ohio and Wisconsin include Illinois and Missouri. CHAPTER XI. FURNACES, ROLLING MILLS, &c. Coke Furnaces in Alabama. (From the Directory of the Iron and Steel Works in. the United States, Amer. Iron and Steel Assoc.. Phil a., 1898. Jas. M. Swank, Manager.)* Clifton Furnaces, Clifton Iron Company, Ironaton, Talladega county; two stacks; No. 1, 55x13, changing to 70x16, built in 1884, blown in April 16, 1885; No. 2, 60x14, built in 1889-90, and blown in during 1891; built to use charcoal for fuel, but changed to coke in 1895; six Co wper stoves ; fuel, Alabama coke ; ore, local brown hematite ; product, foundry pig iron; total an- nual capacity, 72,000 gross tons. Brand, "Clifton." T. G. Bush, President, Anniston ; Augustus Lowell, Vice-Preside at, Boston, Mass.; C.. L. Pierson, Treas- urer, Boston, Mass. ; Paul Roberts, Secretary and As- sistant Treasurer, Ironaton. Selling agents, Matthew Addy and Co., Cincinnati ; C. L. Pierson & Co., Boston and New York. Fort Payne Furnace, DeKalb Furnace Company, Fort Payne, DeKalb county. One stack, 65x14, built in 1889-90 and blown in September 3, 1890; three Sie- mens-Cowper-Cochrane stoves ; fuel, coke ; ores, red and brown hematite , product, forge and foundry pig iron; annual capacity, 27,000 gross tons. (Formerly operated by the Fort Payne Furnace Company). A. L, Tayles, President; E. Dudley Freeman, Treasurer. Idle and for sale. 348 GEOLOGICAL SURVEY OF ALABAMA. Gadsden-Alabama Furnace, Gadsden, Etowah county; one stack, 75x16, built in 1887-88, and first blown in October 14, 1888 ; three Wliitwell stoves ; fuel, coke ; ores, local red and brown hematite; product, foundry and basic pig iron ; annual capacity, 35,000 gross tons. Brand, "Etowah." Owned by Thomas T. Hillman, George L. Morris and Mrs. Aileen Ligon, of Birming- ham. Idle, and for sale or lease. Hattie Ensley Furnace, Colbert Iron Company, les- see, Sheffield, Colbert county; one stack, 75x17, built in 1887 and blown in December 31st, 1887 ; three Whit- well stoves; fuel, coke; ore, local brown hematite; product, foundry pi'g iron, annual capacity 48,ooo gross tons. Brand, "Lady Ensley." A. A. Berger, Presi- dent: "Wade Allen, Vice-President ; J. V. Allen, Secre- tary and Treasurer ; A. J. McGarry, Manager. Selling agents. Rogers, Brown & Co., Cinti., N. Y., &c. Mary Pratt Furnace, W.T. Underwood, Birmingham, Jefferson county. One stack, 65x14, built in 1882, and first put in blast in April, 1883 : rebuilt in 1889 ; three Whitwell stoves; fuel, coke; ores, local brown and red fossiliferous ; annual capacity 30, 000 gross tons. Brand, "Mary Pratt." Idle for several years. Philadelphia Furnace, Florence Cotton andiron Com- pany, Florence, Lauderdale county. Main office, 330 Walnut St., Philadelphia, One stack, 75x17, com- menced by the W. B. Wood Furnace Company in 1887, and completed by the present company in 1890-1 ; three Whitwell stoves, each 70x20; fuel, coke; ore, brown hematite from Lawrence county, Tenn.; product, foun- dry pig iron ; annual capacity 45,000 gross tons. Brand, ''Philadelphia." Robert Dornan, Vice-President ; James Pollock and William H. Arrott, committee for bond- holders ; E. Cooper Shapley, attorney, Girard Building, Phila. For sale. Idle since 1893. FURNACES, ROLLING MILLS, ETC. 349 Pioneer Furnaces, Pioneer Mining and Manufacturing Company, Thomas, Jefferson county ; two stacks, each 75x16.5 ; No. 1 built in 1886-88, and blown in May 15, 1888; No. 2 built in 1889-90, and blown in February 22nd, 1890; eight Siemens-Cowper-Cochrane stoves; fuel, Alabama coke; ores, red and brown hematite from the company's mines near the furnaces; product, foundry pig iron; total annual capacity 95,000 gross tons. Brand, "Pioneer." Edwin Thomas, President, and Samuel Thomas, Vice-President, Catasaqua, Penna.; George H. Myers, Secretary and Treasurer, Bethlehem, Penna. Selling agents, Matthew Addy and Co., Cincin- nati ; W. R. Thomas, 50 Wall St., N. Y. , Dallett & Co., 201 Walnut Place, Phila. Sheffield Furnaces, Sheffield Coal, Iron and Steel Company, Sheffield, Colbert County. Three stacks, each 75x18, built in 1887-88; No. 1 blown in during Sept., 1888, and No. 2 blown in during Oct., 1889 ; No. 3 not yet blown in ; Nos. 1 and 2 rebuilt in 1891 ; nine Whit- well-Cowper stoves ; fuel, Alabama and Virginia coke ; ores, Alabama and Tennessee brown hematite ; product, foundry pig iron ; total annual capacity. 150,000 gross tons. Brand, " Sheffield." A. W. Willis President, E. W. Cole, Vice-President, T. D. Radcliffe, Secretary, Sheffield; S. B. McTyer, Treasurer, J. J. Gray, Jr., Superintendent, Sheffield. Selling agents, Rogers, Brown and Co., N. Y., Miller, Wagoner, Feiser & Co. r Columbus, Ohio.; Hickman, Williams & Co., Louis- ville, Ky. Sloss Furnace, Sloss Iron and Steel Company, Bir- mingham, Jefferson County. Four stacks: No. 1, 82.25x18, built in 1881-82, put in blast April 12th, 1882, and rebuilt in 1895 ; No. 2, 68x18, built in 1882 : No. 3, 73x16.5, built in 1887-88, and blown in during Oct., 1888; No. 4, 73x16.5, built in 1887-89, and blown in 350 GEOLOGICAL SURVEY OF ALABAMA. during Feb., 1889> five Whitwell, eight Gordon-Whit- well-Co wper, and three two- pass 18x70 stoves ; fuel, coke; ores, red fossiliferous, hard and soft, and brown hematite; ores and coal mined on the company's prop- erty within ten to fifteen miles of furnaces ; product, foundry and mill pig iron ; totol annual capacity, 200,- 000 gross tons, Brand, " Sloss." Sol Haas, Presi- dent; E. W. Rucker, Vice-President ; J. W. McQueen, Secretary, A. H. McCormick, Treasurer. Selling agents, D. L. Cobb, Louisville and Chicago : Rogers, Brown and Warner, Phila. ; Hugh W. Adams and Co., 15 Beek- man St., N. Y. Spathite Furnace, The Spathite Iron Company, Flor- ence, Lauderdale County. One stack, 75x14, completed in December, 1888, and blown in during in during Oct., 1889 ; rebuilt in 1893; three improved Pollock 'stoves ; fuel, coke ; ores, Spathite and brown hematite from Iron City, Term.; product, spathite pig iron ; annual capac- ity, 30,000 gross tons. Brand, "Spathite." (For- merly called North Alabama Furnace.) J. Overton Ewin, Receiver; J. H. Short, Superintendent. Selling agents, Rogers, Brown & Co., Cincinnati. Sold Nov. 25th, 1895, to Louisville Banking Company. Louisville, Kentucky. Idle and for sale. Spathite Furnace, No. 1, Spathite Iron Company, Nashville, Tenn. Furnace at Birmingham. One stack, 65xl5i ; commenced building February 9,^1890 ; blown in August 23, 1890; remodeled in 1897; three Massicks and Crooke stoves ; fuel, Alabama coke ; ores, spathite and brown ; product, spathite pig iron ; anuual capac- ity, 40,000 gross ton. (Formerly called Clara Furnace) , Thomas Sharp, President (died 1898) ; William M. Dun- can. Vice-President; John P. Helms, Secretary and Treasurer. Talladega Furnace, Talladega Furnace Company, Tal- FURNACES, ROLLING MILLS, ETC. 351 ladega, Talladega County. One stack, 72x18, built in 1889, and blown in October 5th, 1S89 ; three Ford and Moncur stoves, each 62x26 ; fuel, Alabama and West Virginia coke ; ore, local brown hematite ; product, Bes- semer, foundry and forge pig iron; annual capacity, 40,000 gross tons. Brand, ' ' Talladega." Rudolph Gut- raann, President; William P. Parrish, Secretary. Idle for several years. Tennessee Coal, Iron and Railroad Company, Bir- mingham, Jefferson County. Thirteen stacks in Jeffer- son County. Five stacks at Bessemer: Nos. 1 and 2, each 75x17, built in 1886-87 ; No. 1 put in blast in 1888, .and No. 2 in 1889 ; seven Whitwell- stoves ; Nos. 3 and 4, each 75x17, built in 1889-90 ; eight Whitwell stoves; No. 5, or Little Belle, 60x12, built in 1889-90, three Whitwell stoves. Oxmoor Furnaces, at Oxmoor, (formerly called Eu- reka Furnaces) two stacks : No. 1 75x17, completed in July 1877, and rebuilt and blown in during Dec. 1885 ; No. 2, 75x17, first blown in in March, 1876, and rebuilt and blown in during Aug., 18S6 ; seven Whitwell stoves. Fuel, Pratt and Blue Creek coke, made in Company's ovens ; ores, local brown hematite and red 1'ossiliferous from tlio company's mines ; product, 'foundry, mill and basic open-hearth pig iron ; total annual capacity, 126,000 gross tons. Brand, " DeBardeleben." Alice Furnaces, at Birmingham, two stacks : No 1, 75x15, built in 1879-80, and put in blast November 23d, 1880 ; raised to present height in 1890 ; three Gordon- Whitwell-Cowper stoves ; No. 2, 75x18, built in 1883, and put in blast July 24th, 1883 ; three Whitwell stoves ; brand, "Alice," product, basic and foundry pig ; an- nual capacity, 113,000 tons. Ensley Furnaces, at Ensley. Four stacks, each 80x20, built in 1887, 1888, and 1889; No. 1 blown in 352 GEOLOGICAL SURVEY OF ALABAMA. March 19. 1889; No. 2, December 1st, 1888; No. 3,. June 5th, 1888, and No. 4 April 9th, 1888 ; four Gordon- Whitwell-Cowper stoves to each furnace. Brand, " En- sley." Fuel, Pratt coke made in the company's ovens ; ores, red and brown hematite from the company's mines product, foundry, and forge pig iron ; annual capacity of Alice Furnaces 113,000 gross tons ; of Ensley furnaces, 292,000 tons. Total annual capacity of the thirteen stacks, 823,000 tons. N. Baxter, Jr., President ; James Bowron, 1st Vice-President and Treasurer; A. M. Shook,. 2d Vice-President; George B. McCormack, General Manager; T. F. Fletcher, Jr., Secretary and Assistant Treasurer; H. D. Cooper, Auditor; Erskine Ramsay, Chief Engineer ; John Dowling, Superintendent of Bes- semer Division A. E. Barton, Superintendent of En- sley Division. Selling agents, Rogers, Brown & Co. y Cincinnati, and branch houses ; Matthew Addy & Co. r Cincinnati and St. Louis. Trussville Furnace, Trussville, Jefferson County. One stack, 65x18, built in 1887-89, and blown in in April, 1889 , three Whitwell stoves : fuel, Alabama coke ; ore, local red hematite ; product, foundry pig iron ; annual capacity, 30,000 gross tons. Brand, "Trussville." Owned by Messrs. Hogsett, Ewing and Thompson, Un- iontown, Pa. Williamson Furnace, Williamson Iron Compay, Birm- ingham, Jefferson county. One stack, 65x13.66, built in 1886, and first blown in in October, 1886; three Mas- sicks and Crooke stoves ; fuel, coke made at Coalburg ; ores, red fossil and brown hematite ; product, foundry and mill pig iron; annual capacity 18,000 gross tons. Brand, "Williamson." C. P. Williamson, President and General Manager ; H. D. Williamson, Vice-Presi- dent; J. B. Simpson, Secretary and Treasurer. Idle since 1892. FURNACES, ROLLING MILLS, ETC. Woodstock Furnaces, The Woodstock Iron Works,. Anniston, Calhoun county. Two stacks, each 75x16,. built in 1887-89, and one blown in October 10th, 1889 ;. seven Whitwell stoves ; fuel, Alabama coke; ore, local- brown hematite; product, foundry pig iron; annual capacity of No. 4, 60,000 gross tons. Brand "Wood-- stock." John D. Probst, President, and George Glover,, Secretary, New York ; H. Atkinson, Vice-President and Treasurer, and A. H. Quinn, Assistant Treasurer, Annis- ton. Woodward Iron Company, Woodward, Jefferson- county. Two stacks, each 75x17, one built in 1882-83,. and put in blast in August, 1883, and t.he other built in 1886 ; eight Whitwell stoves ; fuel, coke made from the company's coal ; ore, red fossiliferous, mined within three miles of the furnace ; specialty, foundry pig iron ;;, total annual capacity, 100,000 gross tons. Brand,. "Woodward." J. H. Woodward, President; Frank M. Eaton, Secretary; Silas Hine, Treasurer; J. H. Me Cune, General Superintendent. Number of coke furnaces in Alabama, 37 completed stacks, and 1 stack partly erected. Annual capacity of coke furnaces in Alabama, 1,965,- 000 gross tons. Number of coke and bituminous furnaces in the Uni- ted States, 247 ; annual capacity 15,114,700 gross tons. Alabama has 15.0 per cent, of the total number of coke furnaces, 10.8 per cent, of the total annual capaci- ty, and produces 16.0 per cent, of the total amount of coke iron. Dividing the period 1876-1895 into 4 sub- periods of 5 years each we have the following comparisons : 1876-1880, coke furnaces built 4 ; production in 1876 r 1,262 tons; in 1880, 35,232 tons; increase 33,9^0 tons r or 28 times. 23 354 GEOLOGICAL SURVEY OF ALABAMA. 1881-1885, coke furnaces built 6 ; production in 48,107 tons; in 1885, 133,808 tons; increase, 85,701 tons, or 2.78 times. 1886-1890, coke furnaces built 29 ; production in 1886, 180,133 tons; in 1890, 718,383 ions; increase 538,250 tons, or 3.99 times. 1891-1895, no coke furnaces built. The greatest activity was displayed in the period 1886- 1890, as of the 33 completed stacks in 1895, 20 or 74.5 per cent, were built during these years. It was not un- til 1888 that the production of coke iron passed the 200,- 000 ton mark, and not until 1889 did it rise above 500,- 000 tons, and assume respectable proportions. Until 1897 the year 1895 witnessed the largest production of coke iron ever recorded in the State, 835,851 tons, ex- celling the output of 1892 by 11 tons. Of the 835,851 tons 387, 793 tons, (46.4 per c?nt.) were made during the first half of the year, 18 furnaces being in blast June 30 fch, and 448,058 tons (53.6 per cent.) in tlie second half, 20 furnaces being in blast December 31st. The 60,265 tons made in the second half of the year in excess of the output during the first half may be taken as representing the increase due to the upward tendency of prices .which seemed to be genuine about that time. The production of coke iron since 1876 is given in the following table : FURNACES, ROLLING MILLS, ETC. 355 TABLE XLIX. Production of Coke Iron in Alabama. Tons of 2240 pounds. YEAR. TONS. YEAR. TONS. YEAR TONS. YEAR. TONS. 1876 1877 1878 1879 1880 1881 1,262 14,643 15,615 J5,937 35,23-.' 48.107 1882 51,093 1883 102,750 1884 116,264 1885 133,808 1886 180,133 1887 176.374 1888 1889 1890 1891 1892 1893 5L/,289 608,034 718,383 717,687 835.840 659.725 1H94 1895 1896 1897 556,314 835,851 892.383 932,918 Charcoal Furnaces in Alabama. i [From the Directory to the Iron and Steel Works in the United States, American Iron and Steel Association, Phila. Jas. M. Swank, Man- ager. ] Attalla Furnace, Buffalo Iron Company, Nashville* Tenn. Furnace at Attalla, Etowali county. One stack, 55x11, built in 1888-S9, and blown in June 15th, 1889; iron stoves ; ores, red and brown hematite from Etowali and Cherokee counties ; product, car-wheel pig iron ; annual capacity, 18,000 gross tons. Brand, ''Attalla." Robt. Ewing, President; J. A. Cooper, Secretary and Treasurer. Idle since 1892. Bibb Furnace, Alabama Iron and Steel Company, Brierfield. Bibb county. One stack, 55x12, built in 1864 to use charcoal; rebuilt in 1881, and remodeled in 1886 to use coke ; returned to the use of charcoal in 1890; re-built in 1892 ; warm blast ; ore, brown hema- tite, mined in the vicinity ; product, car-wheel pig iron ; annual capacity, 14,500 gross tons. Brand, "Bibb." T. J. Peter, President. Selling agents, C. R. Baird & Co., Phil., De Camp & Yule, St. Louis ; Forster, Hawes & Co., Chicago. Idle since 1894. Clifton Furnace, Clifton Iron Company, Ironaton, Talladega county. One stack, No. 2, 60x14; built in 356 GEOLOGICAL StTRVEY OF ALABAMA. 1889-90, and blown in in 1-891; hot blast; ore, local' brown hematite; product, car wheel and malleable pig: iron; annual capacity, 22,000 gross tons. Brand, "Clifton." (See Coke Furnaces) . Jenifer Furnace, Jenifer Furnace Company, Jenifer, Talladega county. Central office, Anniston. One stack r 56x11, built in 1892, and blown in December oth, 1892 r taking the place of the old stone stack built in 1863 ; two Hugh Kennedy stoves, each 45x16 ; ore, local brown hematite ; product, car- wheel pig iron ; annual capacity 12,000 gross tons. Brand, ' 'Jenifer." (One stack, built in 1863, abandoned and dismantled in 1872.) John H. Noble, President, and John E. Ware, Secretary and Treasurer, Anniston. Selling agents, Rogers, Brown ............. 3.40 Galveston, Texas . . . ................... , 5.97 5.10 Hamilton, Canada ................... . 4.30 Kansas City, Mo ..... . .............. '. . 4.40 2.50 Louisville, Ky ........................ 2.00 Minneapolis, Minn .................... 4.95 2.50 Mobile Ala., export .................... 1.00 Montreal, Canada ..................... 5.60 *Nashville, Term ....................... 1.00 2.50 New Orleans, La., export .............. 1.60 Newport News, Va .................... 2.6 "K ^ v i XT v - ............ 5.13 o./5 New York, N. Y., ., , oc , ' ( rail and water ...... 3.25 Norfolk, Va ...... ................. .... 2.35 Omaha, Neb ......................... 4.50 4.75 Philadelphia, Pa., all rail .............. 4 .02 " rail and water ....... . 3.25 *From Birmingham. Pensacola, Fla., export .............. ... 1.00 368 GEOLOGICAL SURVEY OF ALABAMA. 4.40 Pittsburg, Pa 3.70 14.47 Portland, Oregon 12.84 San Francisco, Gal 12.84 2.90 Savannah, Ga 2.90 3.25 St. Louis, Mo 2.75 5,10 Toronto, Canada 4.30 .Youngstown, Ohio 3.30 From the Sheffield District the all-rail differential is 40 cents under the Birmingham rate. The distances from Birmingham to these points is about as follows : From Birmingham to Distance in Miles. Atlanta 167 Baltimore . - . 1,050 Boston 1,450 Buffalo 950 Chattanooga 143 Cincinnati 504 Cleveland 767 Chicago 650 Columbus 630 Denver 1,400 Detroit 766 Galveston 800 Hamilton 975 Kansas City 850 Louisville 394 Minneapolis ...... 1,050 Mobile 276 Montreal 1,600 Nashville 209 New Orleans 417 Newport News .... 800 New York 1,225 FURNACES, ROLLING MILLS, E^C. 369 Norfolk 775 Omaha * 1,000 Pensacola 260 Philadelphia 1,150 Pittsburg 817 Portland 3,675 San Francisco 3,000 Savannah 448 St. Louis 528 Toronto 996 Youngstown ...... 875 The pig iron produced in Alabama goes into almost every State of the Union, and into many foreign coun- tries. The transportation rates, therefore, are most im- portant to the stability of the industry. Taking the figures given in the preceding statements as to the rates and the distances, it will be found that the highest rate per ton-mile from the Birmingham district is to Atlanta, a distance of 167 miles, to which point the rate is $1.30, or 7.78 mills per ton-mile. The lowest rate is to Louis- ville, a distance of 394 miles, to which point the rate is $2.00, or 1.97 mills per ton-mile. As it might be of some interest to know what the rates per ton-mile are for pig iron, the following table has been constructed, based on the above rates and dis- tances, and all rail freights. TABLE LIII. Giving the freight rates per ton-mile on pig iron from the Birmingham district to points as below, in mills. Atlanta 7.78 Baltimore 3.58 Boston 3.68 Buffalo 4,05 34 370 GEOLOGICAL SURVEY OF ALABAMA. Chattanooga 5 .24 Cincinnati 2.24 Cleveland 4.30 Columbus 4.60 Denver ... 6.56 Detroit 4.44 Galveston . . 7.46 Hamilton 4.51 Kansas City . . 5.18 Louisville 1.97 Minneapolis. . 4.71 Mobile 3.62 Montreal 3.50 Nashville 4.78 New Orleans 3.83 Newport News 2.94 New York 4.19 Norfolk 3.03 Omaha 4.50 Pensacola 3.85 Philadelphia 3.50 Pittsburg 4.53 Portland 3.49 San Francisco 4.28 Savannah 6.47 St. Louis 5.21 Toronto 4.32 Youngstown 3.77 Freight tariff for coal and coke in effect in the Spring of 1898, from Birmingham to Atlanta, Ga $1 .05 Augusta, Ga 1 .80 Charleston, S. C 2.05 Columbia, S. C 2.20 FURNACES, ROLLING MILLS, ETC. 371 Columbus, Miss 1.05 Dallas, Texas 4.75 steam coal, $4.95 coke. El Paso, Texas 6.44 Greenville, Miss 1 .15 Houston, Texas 2.90 coal, $3.55 coke. Macon, Ga 1.50 Meridian, Miss 1.15 Mobile, Ala 1.50 Montgomery, Ala 1.10 New Orleans, La 1.40 steam coal, $1.60 coke, Pensacola, Fla Savannah, Ga 1.80 Selma, 1.00 Shreveport, La 2.15 Vicksburg, Miss 1.55 Bunker rate to Mobile $1.10 " " New Orleans. . 1.40 " " Pensacola 1.10 The same rates hold for export. These rates are per ton for carload of not less than 23 tons of 2,000 pounds. INDEX. Page. Alabama Coal in By-product Ovens 115 Axle Works 364 Basic Iron, Burdens 320-345 Basic Iron, Composition of 315 Basic Iron, Cost of : 314 Basic Iron, Firms using ; 305 Basic Iron, Manufacture of 305-345 Basic Iron, Production of 346 Basic Iron, Specifications for 311-314 Basic Steel 290 et seq. Basic Steel, Chemical and Physical Tests. .292, 297-302 Basic Steel, First Production of 290 Basic Steel, Materials for 304 Basic Steel, Report of Committee on, in 1890 293 Bessemer Ore, not found 17 Bessemer Rolling Mill : 291 Bessemer Steel, Statistics of 309 Big Stone Gap Coke, Analysis of 84 Birkinbine, John, Statistics of Iron Ore 30 Birmingham Rolling Mill Company, Basic Steel. . . .296 Black Creek Coal, Calories of 242 Black Creek Coke, Analysis of 80 Blair, A. A., Analysis of Iron Ore 44 Blast Furnace Burdens 67, 141-165 Blast Furnace, First in Alabama 8 Blast Furnace, List of 345 et seq. Blauvelt, W. H., Semet-Solvay Ovens 123-139 Blocton Coal, Calories of 242 Blocton Coke, Analysis of 84 Blue Billy Iron Ore 61 Blue Creek Coal, Calories of 244 374 Page. Blue Creek Coke, Analysis 78-81 Brannon, W. H., Grading Pig Iron 169-173 Bridge Works 363-365 Brookside Coke, Analysis of , .85 Cahaba Coal, Calories of 242 Campbell, H. H., on Basic Steel 298 Campbell Coal. Washer 218 Carnpredon's Method of Testing Coking Coals 131 Carbon, deposited in Coking 101-104, 130 Carbonic Acid, Removal of, from Limy Ore 2^3-285 Carbuilding Works 364 Charcoal Furnace Practice 164 Coal, Analysis of 244-246 Coal, Area 201 Coal, Beaver Creek, Pa., Analysis of 246 Coal, Blue Creek, Ala., Analysis of .244 Coal, Carnegia, Pa., Analysis of 246 Coal, Clinton, Pa., Analysis of 246 Coal, Henry Ellen, Ala., Analysis of 244 Coal, Hoytdale, Pa., Analysis of 246 Coal, Mary Lee, Ala., Analysis of 244 Coal, Pocahontas, Va., Analysis of 246 Coal, Pratt, Ala., Analysis of 244 Coal, Pratt, Ala., in By-product ovens 115-120 Coal, Pratt, Ala., in Bee-hive ovens. 96, 100-110 Coal, Thacker, Pa., Analysis of 246 Coal, West Va., Analysis of 246 Coal, Colorific Power of 240-246 Coal, Changes of, in Coking 109 Coal, Coking, Campredon's Method of Testing 131 Coal, Freight Tariff on .-. 371 Coal, Mines, Statistics of 202-21?' Goal, Prices of . . . . , . . .203 375' Page. Coal, Production of 202, 366 Coal, Ultimate. Analysis of 100, 109, 244 Coal, Used in Coking 221 Coal, Washing Plants 218 Coal Washing, Results of 223-234 Coke, Analysis of 78-88 Coke, Ash, Analysisof 78, 87-88 Coke, Bee-hive 93-111 Coke, By-product 115-139 Coke, By-product, Structure of . . . 128 Coke, By-product, Use of in Blast Furnace 129 Coke, Changes of Coal in Making 109 Coke, Classification of 76 Coke, Connellsville, Analysis of 83 Coke Consumption 89, 141-163 Coke Furnaces 139, 345 Coke Furnace Practice 141-163 Coke from Lump Coal, Analysis of 87 Coke from Run-of-Mines Coal, Analysis of 86 Coke from Washed Slack, Analysis of 87 Coke Oven Gas, Analysis of 107, 1 17-119, 136 Coke Ovens, Statistics of 92 Coke, Otto-Hoffman 1 15-120 Coke, Physical Structure of 78-^8, 101-107 Coke, Production of 92, 366 Coke, Semet-Solvay . , 123-139 Coke Yield of Pratt, in Bee-hive Oven 96-100 , 110 Coke Yield of Pratt, in By-product Oven 117-119 Concentration of Low-grade Ores 247, et seq. Counellsville Coke, Analysis of 83 Counellsville Coke compared with By-product ) 10Q lQn Coke f IZ *' Crellin & Nails 391 Davis-Colbv Ore Kiln.. . .283 '376 Page. DeBardeleben, H. F 11 Dewejr, F. P. on Coke 106 D'Invilliers, E. V. Comparison of some Southern ) ^ Coke and Iron Ores \ Dolcito Dolomite Quarry 69 Dolomite , Analysis of 64 Dolomite, First Use of 69 Dolomite, North Birmingham Quarry 70 Dolomite, Use of, as Flux 70-75 East No. 2 Ore Mine 37 Ebelmen, on Coke Oven Gas 107 Fleming, H. S., General Description of the Ores \ ^ Used in the Chattanooga District i Forges and Bloomaries . 362 Fort Payne, Basic Steel at , 296 Fossil Red Ore Mines 43 Freight Tariff 367-36^ Fulton, John, On Coke , . . . . : 3 Furnace Burdens 67, 143-165, 320-345 Furnaces, Charcoal 355-357 Furnaces, Charcoal, When Built. . 357 Furnaces, Coke 347-355 Furnaces, Coke, When Built 353-354 Furnaces, Directory of 347-35S Gas Carbon , Analysis of 80 "Gouge," The . 37 Gogin, Mr.. 293 Grace's Gap 3? Hancock, David, Analysis and Tests of Basic Steel. .301 Hard Red Ore, Analysis of 52 Hassinger, W. H., Member of Committee to Re- ) 293 port on Basic Steel } Hawkins' Process of Steel Making 2^6 Head, Jeremiah, On Birmingham District 235-240 377 Page. Helena Coal, Calories of 242 Hematite Ores 35-54 Henderson Basic Open Hearth Furnace. 313 Henderson Steel and Manufacturing Co 2v0-293 Henry Ellen Coal, Analysis of 242, 244 Hillhouse, Jas. D., Statistics of Coal and Coke ^2 Hillman, T,T ..11 Hoffman Concentrator : . . . . 256 Iron Ore, Analysis of Brown 57 Iron Ore, Analysis of Hard Red (Limy) 52 Iron Ore , Analysis of Soft Red 44 Iron Ore, Analysis of Blue Billy. 61 Iron Ore, Analysis of Mill Cinder 61 Iron Ore, for Basic Iron 315, 317, 319 Iron Ore, Basis of Purchase 21-24, 58 Iron Ore, Black-band 16 Iron Ore, B.rown, (Limonite) 54, 57 Iron Ore, Classification of 35 Iron Ore, Concentration of Brown 285-289 Iron Ore, Concentration of Hard Red (Limy) . .278 et seq Iron Ore, Concentration 24 , -289 Iron Ore, Concentration by Wetherill Process. 265 et seq Iron Ore, Geology of 35 Iron Ore, Hard Red, (Limy) Nature of 50 Iron Ore, High Phosphorus 312 Iron Ore, Phosphorus in .. .< . . 17-18 Iron Ore, Prices of 143 Iron Ore, Production of 19, 20, 30, 366 Iron Ore, Relation Between Hard and Soft 281 Iron Ore, Section of Deposit of 38, 42, 281 Iron Ore, Screening of Brown 60 Iron Ore, Valuation of 33-34 Iron Ore, Washing of Brown .... 378 Page. Iron Trade Review, Statistics from 32 Jefferson Coke, Analysis of 80 Jefferson Mining & Quarrying Co. (Dolomite) 69 Jefferson Steel Co 313 Johnston, A. B. Member of Committee to report ) on Basic Steel ] Johnston, H. R. Member of Committee to report ) on Basic Steel } Landreth, 0. H. Calorific Power of Fuels 242 Leeds, P , Member of Committee to Report on ) Basic Steel \ Lentscher, G. L. Member of Committee to report on Basic Steel Limestone, Analysis of * 62 Limonite Ores 54-60 Littlehales. Thos. G 117 Luthy, Dr., Analysis of Pratt Washed Slack Coal . .116 McCreath, A. S., Comparison of some Southern ) -. Cokes and Iron Ores \ Magnetization of Ores 247 et seq Manganese Ore 305 Mary Lee Coal, Calories of 24*4 Mason, Dr. Frank, Analysis of Pratt Washed Slack Coal Meissner, C. A., First to Use Dolomite 69 Mill Cinder 61 Morris, Geo. L 11 Netze, H. B. C., on Concentrating Ore 266 North Birmingham Dolomite Quarry 70 Open Hearth Steel, Statistics of 309 Parker, E. W. Statistics of Coal and Coke 202-217 Parsons, W. P 117 Payne Concentrator 256 Pechm? B. 6., Articles on Alabama . . . < . . . .1 379 Page. Pechin, E. C. Cost of Making Iron in Alabama 187 Pig Iron 166-199 Pig Iron, Basic, Composition of 315 Pig Iron, Bessemer 167 Pig Iron, Charcoal, Production of ' 35s Pig Iron, Coke, Production of 355 Pig Iron, Cost of Making . 187-199 Pig Iron, Exports of 184 Pig Iron. First Used in Making Basic Steel 294 Pig Iron, Freight Tariff 366-370 Pig Iron, Grades of 168 Pig Iron, Grading, Agreement of 1888 176 Pig Iron, Grading, New System Suggested 181 Pig Iron, Prices of 173 Pig Iron, Production of 21, 366 Pig Iron, Variation in Composition of 177 Pipe Works 363 Pittsburg Gas & Coke Co 115 Pocahontas Coke, Analysis of r4 Poole, Calorific Power of Fuels 245 Porter, Jno. B. Iron Ores and Coals of Georgia, ) ..' Alabama and Tennessee ) Pottstoun Iron Co,, Thomas Steel 308 Pratt Coal, Analysis and Calories - 242-244 Pratt Coke, Analysis 78-81 Producer Gas Analysis 253 Ramsay, Erskine, Pratt Mines of the Tenn. C. I. & Ry. Co Robertson, Kenneth, Analysis of Pig Iron 74 Robertson, Kenneth, Grades of Pig Iron 175 Robinson-Ramsay Coal Washer. 218, 223, 226 Rolling Mills 359-362 St. Bernard Coke, Analysis of 85 Schniewind, F.. 115-117 380 Page. Sleep, W. J 183 Sloss, J. W -...11 Sloss Iron and Steel Company, Dolomite. ..-..' 70 Sloss Iron and Steel Company, Coal tests for. . . . 115-120 Smith, Dr. Eugene A., On Coal Area 201 Soft Red Ore, Analysis of 44 Soft Eed Ore, Section of Seam 38, 42, 281 Speakman, Wm 117 Standard Coke, Analysis of 80 Steel Works 359-361 Stein Coal Washer 218-232 Stonega Coke, Analysis of . 85 Stoves, Hot Blast 358 Swank, Jas. M., Statistics from 21 Thomas Iron, Manufacture of 308 Troy Steel & Iron Co .308 Uehling, E. A. Use of Dolomite .70 United States Geological Survey, ) OA 01 Q n QQ CM QO Statistics, ^U, 21, dU, dd, d4, U Warrior Coal, Calories of 242 Washing Brown Ore 55 Washing Coal 218 et seq Weeks, Jos. D., Death of. 91 Wetherill Concentrator 247 et seq Wilkens, H. A. J., Concentrating Ore . .266 Wilson H. F. Secretary Henderson Steel & Mfg. Co. 290 Worthington, J. W. & Co 69 RETURN CIRCULATION DEPARTMENT 202 Main Library LOAN PERIOD 1 HOME USE 2 1* *@ H U, B iMji B '-*""' 4 5 6 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS Renewals and Recharges may be made 4 days prior to the due date. Books may be Renewed by calling 642-3405 DUE AS STAMPED BELOW SENtONILL JUN 2 7 1996 U. C. BERKELEY FORM NO. DD6 UNIVERSITY OF CALIFORNIA, BERKELEY BERKELEY, CA 94720 YC 68509