STEEL.—Bessemer Process. Heat Generated. 873
These appeared in the following order in a “ blow ”lasting eighteen and a half minutes:—5'57, turnedon blast; 6T, Ha; 6-4, Li; 6 - 5, A; 6'5±, K; 6-6,cand,/; 6-61, t; 6-7, K; 6T4J, lines towards blueend gone; 6T5£, all lines gone except Ha, Li, and k.
If the gas from the convertor be collected andburnt at a jet, we get the ordinary spectrum of car-bonic oxide.
Generation of Heat by the Bessemer Process .—Thissubject has been investigated by Akekman, Jordan,Kupelwuser, and others. But in making the cal-culations it has been generally assumed that thecarbon is entirely converted into carbonic oxide,whereas it will be seen, from the foregoing analyses,that of the total carbon is converted into car-
bonic acid, and only g-J into carbonic oxide. Thisintroduces a considerable error into previous esti-mates. There are also some factors that are neces-sary to an accurate result, which must be assumedfor the present. Therefore, the following figurescan only be approximate.
We will take the case of a Bessemer pig ironcontaining 2 per cent, silicon and 3'5 per cent, carbon,which is about the average composition of the molteniron run into an English convertor. A considerableportion of the iron itself is oxidised during the blow,possibly less than Akerman assumes; but for con-venience we will accept his estimate of 6 per cent.There is a good deal of doubt about the calorificpower of the iron so burnt; and as to the exactspecific heats of some substances under consideration,probably the following figures will be sufficientlyaccurate:—
Calorific power of C burnt to C0 2 = 8080“ C “ CO = 2403
“ Si “ Si0 2 — 7830
“ Fe “ FeO = 1201
Specific heats of CO,.— '21
CO.— -24
“ N.— -24
“ O.— -21
Molten iron.— *16
We will also assume with Akerman that thetemperature of the molten pig iron is 1400° C., andthat the gaseous products leaving the convertor gooff at 1400° C. (This latter figure must be belowthe actual).
Taking cwts. as the unit, 100 cwts. of iron con-tain 3-5 cwts. of carbon; ^jj- of this, or 1 - 51 cwt.,are burnt to C0 2 , and or 1’98 cwt., to CO.
. •. Carbon burnt to C0 2 , produces 8080 X l’Ol= 12,200 units of heat; but 1'51 cwt. C requires
or 4-02 cwts. oxygen, and forms 1-51 +
6
4'02 = 5'53 C0 2 . As the oxygen is obtained from
the atmosphere, there will be ^ ~ 13T5 cwts.
nitrogen mixed with it. How both the C0 2 and Hleave the convertor in a gaseous state, and thereforethey carry with them (5'53 X '21 X 1400) + (13'45X "24 X 1400) = 6145 units of heat, leaving 12,200— 6155 = 6055 units to increase the temperatureof the bath. Similarly the carbon burnt to CO pro-duces the following results:—2403 X 1'98 — 4757
VOL. II.
units produced ; but 1-98 C. requires - 1 X ® _
2’64 cwts. O, and forms 4'62 CO.,. With the O
V 77
there will be---= 8-83 H, and the CO and
23
H will carry away (4'62 X -24 X 1400) 4- (8-83 X•24 X 1400) — 4419 units, leaving 4757 — 4519 =
238 to add to the heat of the bath.
Again, 2 cwts. silicon burnt to Si0 2 produces 2
X 7830 = 15,660 units. Silica being a solid body,mostly remains in the vessel as slag. As it is im-possible to ascertain with accuracy the differencebetween the specific heats of the silicon and silica,and of the iron and oxide of iron, it will probablybe sufficient in each case to deduct the heat requiredto raise the oxygen consumed from 0° to the tem-perature of the bath.
2 X 16
Two parts silicon require _2-28 oxygen.
2*28 X 77
and with this will be associated-—- = 7 63
nitrogen ; (2-28 X -21 X 1400) + (7-62 X -24 X1400) = 3230 units have to be deducted from 15,660,leaving 12,430 for heating the bath.
Six cwts. Fe burnt to FeO, produce 6 X 1201 —6X8
7206 units, and require —-- = 1-71 oxygen, with
28
which then will be 1-71 X 77 — 5.72 H. (1'71 X -21
23 V
X 1400) + (5-72 X -24 X 1400) = 2424. Therefore,from the iron burnt, there will be 7206 — 2424 —4782 units for heating the bath.
Summing these results, we have for adding to theheat of the bath :—
From C converted to C0 2 ,. 6055
“ O “ CO. 238
“ Si “ Si0 2 ,. 12,430
23,505
If there were no loss of heat by radiation andabsorption by the sides of the convertor, the tem-perature of the bath would be (without taking into
23 505
account the loss in weight of the metal) ’
' ' 100 X TG
= 1469° C. higher than at the commencement; sothat the final temperature would be 1400 + 1469 =2869° C. A considerable reduction must be madefor the heat taken up by the lining of the vesseland for loss by radiation, but at present no exactfigures for these can be given. It is probable thatnot more than three-fourths of the heat developedis really taken up by the metal, and thus the finaltemperature would be about 2000° C. This isthe figure given by Akerman ; but the amountof silicon has been taken at 2 per cent, insteadof 1 per cent., which he assumes for Swedish iron,while the heat developed by the combustion ofcarbon is really greater than he makes it. It will beseen, however, that more than half the heat developedis due to the silicon. This agrees with practicalexperience. It will also be evident, that since the
110