Page:Encyclopædia Britannica, Ninth Edition, v. 13.djvu/304

This page needs to be proofread.

288 IKON phalian coalfields and in Ohio. "When carbonaceous matter is not present in any considerable quantity, clay ironstone forms a dark bluish-grey or greyish-yellow mass, sometimes forming layers of nodules, sometimes distinctly bedded deposits of large area, e.g., the Cleveland beds of North Yorkshire, the ores of Glamorganshire, Shropshire, Derbyshire, France, Westphalia, and various parts of the United States, notably Alabama, Tennessee, Kentucky, Ohio, and Pennsylvania. Curiously, although an ordinary con comitant of coalfields, clay ironstone is almost entirely absent from the Coal Measures of Durham and Northumberland. In many localities clay ironstone has become so changed by oxidizing and hydrating influences as to contain but little ferrous carbonate, the iron being converted into a hydrate, giving to the ore the character of a brown hematite ; this is specially noticeable in the Northamp tonshire deposits, most of which are usually classed as brown haematite, although containing some amount of carbonate, whilst occasionally ferrous carbonate is found in them having undergone but little alteration, and forming a clay ironstone closely resembling that of Cleveland. As a rule Coal-Measure ironstones are somewhat highly phosphorized ; this is especially noticeable with the Cleve land ore, which usually yields on smelting a pig iron containing between 1 and 2 parts of phosphorus per 00 of iron. The following table illustrates the composition of some of the more important clay ironstone deposits : Character of ( Ore ami < Locality.... ( Average Scotch Blackband. Dudley Ore, Stafford shire. Cleveland Ore, North Yorkshire. Dowlais, South Wales. Aber- c-oni Black- band. Blue Ore, Hanging Rock, Ohio. Dick Pattinson. Rilcy. Ratcliffe. Wormley. Ferric oxide Ferrous ,, Manganese oxide 1-0 40 5 0-13 46-30 1-44 4-80 2-60 38 06 0-74 5-92 44-29 1-13 0-45 4-10 43-37 1-50 6-05 13-51 42-48 0-13 0-59 Lime 4-0 0-76 7-77 3-06 300 3-43 4-0 0-94 4-16 3-73 0-25 i-no Silica 8 10-29 10-36 13-01 2-80 7 -52 Phosphoric ) anhydride f " Carbonic do Sulphur 32-0 5 074 30-44 0-07 1-07 22-00 0-14 0-42 32-48 trace 3050 1-88 0-35 30-76 0-15 Water 5 1-38 4-45 1-45 0-31 Organic mutter... Potash and soda 5-0 1-14 trace 0-35 0-14 6-25 0-32 100-0 98-43 97-27 100-51 100-28 99-92 Total metallic^ iron ) 32-0 36-14 31-42 34-72 36-40 41-89 Pyrites. As already stated, pyrites is never used directly by the smelter as a source of iron ; but the residue left after burning pyrites to make vitriol and extracting copper from the residue by Henderson s process consists almost entirely of ferric oxide, and from its physical characters is valuable as fettling for puddling furnaces ; so that the iron contained in the pyrites ultimately becomes largely reduced to the metallic state, either in the puddling furnace itself, or subsequently from the tap cinder produced therein on its being smelted in combination with other ores. The cupreous pyrites of Spain and Portugal (Huelva and Tharsis ores), and certain other analogous substances from other countries, containing but little silicious matter or other ingredients besides iron, sulphur, and copper, are in consequence largely used by vitriol makers. The fol lowing table illustrates the average composition of Huelva and Tharsis ores before burning and subsequently, and also of the "purple ore " or " blue billy " left when the copper has been almost entirely extracted (together with quantities of silver and gold, relatively small, but absolutely sufficiently great to be a distinct source of profit) by conversion into chloride by heating in contact with air with sodium chloride and lixiviation of the product, the "purple ore " remaining undissolved : Raw Ores. After pass ing through the pyrites kilns. Dry Purple Ore. Analyst Clapham. Alder Wright, Alder Wright. J. A. Philips. Snelus. 41-92 47-50 4-21 0-33 0-22 1-52 ]- 3-40 Average. 44-28 49-07 2-75 0-38 traces 2-34 i -is Average. 3-0 3-0

<

3-0 90-0 i -b 0-36 0-20 o -75 as sulphate 2-11 96-00 absent 0-50 0-32 0-30 0-75 4-02 94-51 absent o-io Copper Zinc Lead Silica, quartz, and substances in soluble in acids.... Lime, soda, <fec 100-00 100-00 100-0 99-92 100-00 A valuable report on the character of various British iron ores is to be found in the Journal of the Iron and Steel Institute, 1871, whilst numerous analyses and descriptions of ores from almost all parts of the world are given in the volumes published during the last ten years or so. 6. Analysis of Iron Ores and of Metallic Iron and Steel. The analysis of iron or<5s by the " dry method " (fusing with reducing agents, such as powdered charcoal, and suit able fluxes, and weighing the button of cast iron produced) has the advantage of giving in a comparatively short time a notion both of the amount of iron contained in the ore and of the presence or absence of phosphorus, manganese, &c. (judged of by the physical characters of the button), but has little to recommend it on the score of minute accuracy. The results are usually in excess of the iron actually present by ^ to T -J 7 , on account of the button containing carbon, &c. ; whilst it does not by any means necessarily follow that the reduction of a given ore on the small scale in a crucible and on the large scale in a furnace will produce a metal of the same characters in each case. Accordingly the " wet method " of analysis (solution in appropriate solvents and separation of the various constitu ents from one another or other treatment equivalent there to) is ordinarily preferred. The ore, after being finely pulverized and sifted, &c., and repre senting a fair average sample of the material operated on, is dissolved in hydrochloric acid (if this fails by itself to produce ready solu tion, the ore may be heated to a low red heat in hydrogen so as to reduce to the metallic state, and then dissolved in hydrochloric acid) ; the ferric salt present is then reduced to the ferrous state by nascent hydrogen (evolved by adding fragments of pure zinc or other reduc ing agents, such as sulphur dioxide), and the amount of iron con tained in the fluid determined either by adding a standard solution of potassium permanganate to the diluted fluid until a pink tint just appears (Marguerite), or by adding standard potassium dichromate solution until a drop of fluid just ceases to form a blue precipitate or greenish colour with potassium feriieyanide (Penny) ; the oxygen communicated by the test fluid being known from the volume of liquid consumed, the amount of iron peroxidized is known. When the ores contain iron in both the ferrous and ferric states, and the amount of each is required to be determined, the ore is boiled with hydrochloric acid, and the ferrous salt determined in one part of the solution, and the total iron in another portion. If the iron exist wholly or partially as carbonate, the amount of carbon dioxide may be determined by treating the finely pulverized ore with sulphuric acid in a suitably constructed apparatus, and weighing the appar atus after the (completely dried) gas has been wholly removed, or by absorbing the evolved gas in ammonia, boiling with calcium chloride, and weighing the precipitated calcium carbonate, a correction being made by means of a blank experiment for any ammonium carbonate originally present in the ammonia solution, or formed by absorption of carbonic acid from the air during the operation. Manganese is conveniently determined by dissolving the ore, peroxidizing if necessary, rendering nearly neutral, and boiling with sodium or ammonium acetate, whereby all iron and alumina present are thrown down as basic acetates, carrying with them all the phosphoric acid which is in solution ; to this filtrate bromine is added (or it is saturated with chlorine), and the whole allowed to stand in not too cold a place for some hours, when the manganese is precipitated as a hydrated dioxide, or oxide approaching in composition thereto, which is collected, washed, ignited, and weighed as Mn.,0 4 ; when more than a trace of manganese is present it may be determined volumetrically by several methods, e.g. , Pattinson s, consisting of addition of ferric chloride if the iron present is not already present in larger quantity than the manganese, of bromine water or calcium hypochlorite, and finally of freshly precipitated calcium carbonate, the liquid being at a temperature of 60 - 70 C. ; the precipitate thrown down contains all the manganese as Mn0. 2 (Pattinson, Chcm. Soc. Journal, 1879 [Transactions], p. 365), which may be estimated by dissolving with dilute sulphuric acid and a known amount of stand ard ferrous sulphate solution, and determining the iron which remains unoxidized by the Mn0 2 . Kessler (Zcitscli. Anal. Chemie, 1879, 18, part i.) employs an analogous method for manganese determination, adding zinc chloride and bromine, boiling for a long time to ensure that all the manganese is precipitated as Mn0 2 , and finally dissolving in solution of antimonious chloride in hydrochloric acid, and titrating the non-perchlorinated antimony by perman ganate. Alder AV right and Menke (Cliem. Soc. Journal, 1880 [ Transactions ], p. 22) find that Pattinson s process gives more satis factory results if zinc is present as well as iron in the precipitation of the manganese as Mn0 2 , the formation of oxides of manganese lower than Mn0 2 and of permanganate (which may sometimes otherwise