Fig. 135.

Fig. 136.

Reverberatory Furnaces

The employment of reverberatory furnaces for smelting galena depends upon the double decomposition of the lead sulphate obtained by the roasting and of the non - oxidized sulphur when these two react on each other. It can only be applied to ores containing 55 Per cent. of lead and at most 4 to 5 Per cent. of quartz or clay, otherwise there results a formation of lead silicate difficult to reduce. The presence of blende or pyrites produces oxysulphides which render the slag more fluid; this mechanically holds some grains of lead, whence the necessity for liquation as practised at some works. Gruner prefers to (treat these rich residues in a shaft furnace, and to avoid the liquation. The rapidity of the reverberatory method differs much in different localities: it is slow and gradual in Carinthia, more rapid in Brittany, and much quicker still in British, Spanish, and Belgian works.

In Carinthia, the process is as follows: A charge is made of an ore containing a little blende and pyrites, as gangue, to a depth of about 2 in. on the inclined floor of the furnace; the roasting is conducted slowly for 5 hours, followed by stirring up for another 4 or 5 hours. By reason of the moderate heat of the roasting, the lead sulphide is only partially reduced; in the presence of the lead sulphate formed, a certain quantity of lead oxide remains in the bath, and sulphuric acid is disengaged. At the moment of stirring, the temperature rises; there is then complete reduction and decomposition of the lead oxide, with formation of metallic lead, a result which may be expressed by the following equations:

2 PbS+2 PbO. S03+30=Pb2S + 2 PbO+3 S03 and PbS + 3PbO=4Pb+3SO2. This last equation points to the proportions of oxide and sulphide desirable to maintain after the first period. There remains on the bed, after the reaction which follows the stirring, an appreciable proportion of rich cinders, containing about 40 Per cent; these are set apart for ultimate tretment. by: liquation. In Carinthia, the heating is done with wood; in some experiments in Belgium with this furnace and with a view to avoid the extreme slowness which characterises the Carinthian method, coal was employed; in this case, the air necessary for oxidation, instead of passing through the grate, is admitted by holes immediately into the furnace.

The Breton process being no longer in use, since the closing of the Albert - ville and Poullaouen works, it requires no description. Suffice it to say that it differs from the preceding in the roasting being performed more rapidly and at a higher temperature, resulting in a smaller yield and a less pure product. At Albertville, use was made of a furnace with a double hearth, one half for the roasting and the other for the reaction; this principle was also tried in Carinthia, because it gives a certain economy of fuel. An attempt was made to superpose the second hearth, but this was not successful.

The English process, properly so called, employs furnaces with an internal basin furnished with 3 working doors on each face. The object is not to raise the temperature too high at first, so as to produce lead oxide, as in the Carinthian process, followed by gentle heat to obtain the metallic lead. There is thus a succession of distinct periods of roasting and reaction, and is finished by adding a little slaked lime to cement the mattes; there remain rich cinders, which are feebly liquated and treated in a blast furnace. The fault of the English method is a too rapid - action; some lead is liberated before the roasting is finished, and consequently there is a premature softening of the mass, which impedes the subsequent roasting; this entails a loss of lead which more than compensates for the gain in economy and rapidity of working. (The English process, deals with a ton in 7 hours, while the Carinthian method only treats a charge of 4 cwt, in 9 or 10 hours.)

- The English method has long been applied on the Continent, In Belgium, for instance, they use long furnaces provided with a fire at each end, and in which the operation is conducted as if there were 2 distinct furnaces joined together. They experience great waste of lead by the rapid heating and imperfect roasting, and afford another proof of the necessity for roasting slowly and in one operation at a low temperature. Applying this principle to the English furnace, very good results have been secured at the Tarnovitz works, Upper Silesia. The arrangement of the furnace is shown in Fig. 137.

Fig. 137.

Owing to the nature of the ore, which contains about 40 Per cent. of carbonate and sulphate of lead, the first roasting period is somewhat reduced, but is nevertheless much more gradual and prolonged than in England; it is indeed pushed to that point at which the correct proportions of sulphide and oxide are obtained for the final reaction. In the figure, a is the fire; b, the bed of the furnace; c, the well; d, hopper; e, working doors.

Prof. Huntington describes the Eng - lish furnace as having a hearth about 8 ft. by 6 ft., separated, from the fire by a bridge rising to within 18 in. of the arch, the latter gradually descending towards the chimney till it is only 6 in. above the hearth. The coaling door and ashpit are on opposite sides of the fireplace, known respectively as the " labourer's " and the "working" side. The hearth is lined with slags from a previous operation, spread while still pasty, and in a manner to leave a depression where the molten lead can collect, and whence it may be tapped into a receptacle placed outside. The operations are divided into 4 consecutive stages or "fires." (1) After lead has been tapped, and while the furnace is still glowing, about a ton of fresh ore is charged to through the hop-per,a little coat being meantime put on the fire to maintain a moderate heat without sufficing to fuse the galena. After considerable roasting, the skimmings from the outer receptacle, consisting mainly of a mixture of sulphur and lead, are thrown in upon the charge, and are rapidly attacked by the oxide and sulphate,occasioning separation of metallic lead. The charge is stirred at intervals, and some small coal is occasionally scattered on it to regulate the oxidation.