I E O N 309 Amongst the alkaline salts thus deposited are considerable amounts (under certain circumstances) of potassium and sodium cyanides ( 19) ; this circumstance appears to have misled Pro fessors Bunsen and Playfair into the belief that cyanogen gas is occasionally one of the normal constituents of blast furnace gases as they escape at the top, the cyanides being more or less deposited in the collecting tube employed, and decomposed by the carbon dioxide and moisture present, with the production of hydrocyanic vapour, which on. analysis gives the same numbers as the same bulk of a mixture of hydrogen and cyanogen in equal volumes ; it is noteworthy that cyanogen has never been found by any other analyst. The following analyses indicate the general character of the waste gases escaping at the top of blast furnaces smelting various ores: Analyst 1. Bunsen 2. Ebel- 3. Tunner 4. 5. Lowtluan 6. Lowthian Playfair. men. Bell. Bell 55-35 57-06 55-1 54-51 60-93 47-2 Carbon oxide Carbon dioxide... 25-97 7-77 C 73 28 Gl 11-39 2-74 23-8 13-6 7 5 34-97 8-36 2-16 26-62 11-75 070 28-9 13-6 10-3 Marsh gas 3-75 20 defines (express ed as C 2 H 4 ) j- 0-43 100-00 10000 100-0 100-00 100-00 100-0 l.Alfreton furnaces using raw coal, calcined clay ironstone, and limestone: blast at 330 C. 2. Seraing furnace using coke, brown hteinatite with a little mill cinder, and limestone: blast about 180 C. 3. Wrbna furnace using spathic ore and charcoal: blast at 400 C. 4. Askam-in-Furness furnace using Askam haematite, Fisher s Red Buy aluminous ore, and coke: blast at 500 C.; exit gases at 380 C. 5. Average of a number of analyses of gas from an SO foot furnace using cal cined Cleveland ore, limestone, and hard coke: blast at 485 (average), and exit gases at 332 C. 6. Coltness furnace using raw coal. The precise relative amounts of carbon oxide and dioxide that accompany the proper working of any particular furnace are governed by a variety of circumstances, many of which have been carefully investigated by Percy, Bunsen and Playfair, Vathaire, Tiinner, Ebelmen, Scheerer, and others, and more especially by Lowthian Bell, with the present writer s cooperation (Chemical Phenomena of Iron Smelting), with the general result that the amount of heat produced in the furnace by the conversion of the fuel (coke or charcoal) into carbon oxide and dioxide, together with that introduced by the blast, is equal to the sensible heat carried out by the escaping gases and the molten pig and cinder, together with an amount constant for given conditions, but otherwise variable within certain limits, representing the sum of the heat absorptions during the various physical and chemical changes going on in the furnace, and the loss of heat from the furnace walls by radiation and conduction ; so that for given sizes and shapes of furnace, given ores and fluxes, and production or given kinds of pig iron and cinder, the amount of carbon oxide relatively to the carbon dioxide in the escaping gases is regulated only by the proportion of fuel burnt, and the temperature of the issuing gases ; this latter being also constant, the greater the amount of coke burnt per ton of iron run the more carbon oxide exists in the waste gases and the less carbon dioxide, and vice versa. There is always a natural limit, however, to the extent to which the quantity of carbon monoxide can be reduced and that of carbon dioxide increased in any given furnace under any given conditions (as to nature of ore, &c.) by diminishing the amount of fuel relatively to the burden ; as this limit is being approached and passed, the pig iron begins to deteriorate in quality, first being less graphitoidal or "grey" in character, then becoming entirely white iron of a less carbonized character than good pig of the kind ; finally a large fraction of the iron is wholly unreduced, and passes into the cinder as ferrous oxide (silicate), producing a strongly marked "scouring cinder," and greatly diminishing the yield. The reason for this is simply the natural character of the complex chemical changes and reac tions involved in the working of the blast furnace (dealt with in detail in 19). Poisonous Effects of Blast-Furnace Gases. Carbon oxide being, as is well known, a poisonous gas, rapidly producing death when inhaled even in small quantity (as when badly ventilated rooms are warmed by charcoal braziers, &c., or when a considerable escape of coal gas containing usually a few parts per cent, of carbon oxide takes place into the air of the room), it results that unless care be taken serious effects may be produced by the inhala tion of the waste gases from the blast furnace. Several fatal occurrences due to this cause have taken place, one of the most remarkable of which was the death of Mr Truran, manager of the Dowlais iron-works, through the escape into his office of the gases from the gas main, which was of brickwork, and newly constructed underground. Besides carbon oxide, the waste gases often contain perceptible quantities of potassium cyanide disseminated through them as dust ; the effect of moisture and carbon dioxide upon such air is to impregnate it with vapour of hydrocyanic acid (prussic acid) ; in some of the cases of poisoning by waste gases the cyanide was believed to be the chief deleterious agent. 19. Chemical CJianges taking place in the Blast Furnace. At the level of the tuyeres, the entering blast comes in contact with a mass of incandescent coke through and over which molten cinder and pig iron are dropping and running ; the almost instantaneous effect upon the air con sequently is to transform the oxygen into carbon oxide either at once or through the two well-known reactions : C0 2 + C =,2CO (2). If a hole be drilled through the walls of a furnace at the tuyere level, and the issuing gases collected (or, what is much the same thing, if the blast be shut off from one goose neck and the plug taken out so that the pressure of gases inside the furnace forces gas out at the orifice)," it is invariably found that the amount of carbon dioxide present in the gases is inconsiderable or nil, the composition of the gases being essentially a mixture of carbon and nitrogen with a little hydrogen (either derived from the moisture in the blast being converted into hydrogen and carbon oxide, or from the hydrogen of the coke first burnt to water vapour and then immediately reconverted into hydrogen and carbon oxide). Besides the carbon oxide due to these causes, there is also a small amount of that gas arising from the molten iron and cinder accu mulated in the hearth, owing to the reaction of the dis solved carbon on the last traces of iron oxide disseminated through the pig and dissolved in the cinder ; so that at the tuyere level there naturally is a little more oxygen relatively to the nitrogen than that corresponding to the oxygen of the original air and moisture in the blast, viz., a mixture of about 35 volumes of carbon oxide and 65 of nitrogen. In passing through the mass of materials in the furnace, the carbon oxide becomes more or less con verted into carbon dioxide, reducing the iron ore in virtue of the change expressed in general terms by the equation zCO + Fe x O !/ = zCO z + Fe x Oy- i . . . . (3) ; so that, were this the sole action taking place, at successive levels upwards the amount of carbon in the gases would remain constant relatively to the nitrogen, whilst the oxygen therein would increase. The actual changes, however, are far more complex than this. Thus, commencing with the top of the furnace, and proceeding downwards, when raw limestone is used as flux, it gives off carbon dioxide as it gets heated, thereby increasing both carbon and oxygen in the gases ; during the passage downwards of the ore in the blast furnace it finds itself continually exposed to a heated atmosphere containing carbon oxide and dioxide; the first effect of the gases upon the newly introduced ore is simply to heat it up, but as soon as the outer portions of the lumps have attained a temperature of something like 200 C. (dependent on the physical character of the ore), which practically is almost immediately after intro duction, reduction of the ferric oxide present commences, the carbon oxide of the gases becoming converted into car bon dioxide in accordance with equation 3. Simultane ously, however, the fuel introduced is more or less acted upon : if raw coal be not used, but coke or charcoal, as is most frequently the case, the effect of exposing this to an atmosphere containing carbon dioxide is to cause (when the temperature is sufficiently high) the occurrence of the reaction between the carbon dioxide and the carbon of the fuel expressed by equation 2. The temperature at which this change begins to take place to any considerable extent depends on the physical condition of the carbon, as does also the rate at which it goes on, which is also modified by the amount of carbon dioxide present in the gases relatively
