STEEL.— Bessemer Process. Spectrum.
872
TABLE X.
Carbonic acid,.
Oxygen,.
Carbonic oxide,.
Hydrogen,.:.
Nitrogen,.
Hydrocarbons,.
No. I.
2 minutesfrom start.
No. 2.
4 minutes.
No. 3.
6 minutes.
No. 4.
10 minutes.
No. 5.
12 minutes.
No. 6.
14 minutes.
10-71
*92
none.
88-37
8-57
3-95
•88
86-58
8-2
absent.
4-52
2-00
85-28
3-58
none.
19-59
2*00 prob.74-83
2-3
none. .29*3
2-16
66*24
absent.
1-34
none.
31-11
2*0 prob.65-55
100-00
100-00
100-00
100*00
100-00
100-00
Observations with spec-troscope and naked ^eye,.
Faint continu-ous spectrum.No actual flame.Abundance ofsparks thrownout.
Sodium linejust appearing,but in flashes.Abundance ofsparks.
Sodium,potassium, andlithium lines.Quantity ofsparksdecreasing.
. Full spectrum,including thecarbon lines.Flame dense andbright. Very fewsparks.
Full spectrum.Dense flame.Still fewersparks.
Full spectrum.Dense flame.Scarcely anysparks.
TABLE XI.
Series.
Minutesfrom com-mencementof blow.
Oxygen.
Carbonic
Oxide.
Hydrogen
Nitrogen.
A,.
3 to 5
4*762
9*127
_
_
86-111
A,.
9 to 10
1-699
5*999
17-555
•908
73-840
A,.
21 to 23
•967
4-856
19*322
1-120
73-735
A,.
26 to 27
•550
1-853
14-311
1-699
81-587
B,.
2 to 3
7-256
6-608
—
—
86-136
13,.
8 to 10
1-296
5-613
15-579
M12
76-400
B,.
12 to 15
•980
4-144
25-580
1-040
68-256
B,.
17 to 19
1-318
2-995
25-606
1-120
68*961
Solid Material Ejected from the Convertor. —In theearly stages of the blow large quantities of sparksare emitted with the gases. The bulk of these comeout as bright white specks, but quite a large numberscintillate beautifully on leaving the vessel. Theselatter are due to particles of metal which are caughtup as spray by the blast, but only burn when theyreach the atmosphere outside. The remainder ofthe sparks are minute globules, which on analysisgave 52-82 per cent, iron, and 27-90 silica. Theyare very remarkable on close examination, consist-ing largely of spherical boom-shaped particles, orhollow spheres, with a small opening in each.
When manganiferous iron is being blown volumesof white smoke accompany the flame from an earlystage, but with English pig iron this only makes itsappearance towards the end of the blow. It consistsof the oxide of manganese, which is carried out ofthe vessel with the blast. As soon as the carbon,silicon, and manganese have been all burnt out theiron begins to burn, and the colour of the smokechanges from a white or grey to ruddy-brown. Thischange is one of the best indications the blower canhave when treating very siliceous pig iron. If ironof low grade be used, an extremely rapid evolution
of gas takes place at a particular portion of theprocess, and large quantities of metal and slag arethrown out of the vessel. A similar eruption takesplace even with iron of fair grade if it is allowed toget too cold before the blow is commenced; but withrich graphitic pig, highly siliceous iron, or metalcontaining from 2 to 4 per cent, manganese, theprocess goes on evenly and with little or no ejectionof slag.
Spectrum of the Bessemer Flame .—The flame pro-duced by the Bessemer process is so intense, thatvery early attempts were made to utilize the spectro-scope in directing the operations. There is, how-ever, so little difficulty in blowing English pig ironby the unaided eyesight, that the use of the spectro-scope has not been followed up as it ought to havebeen. With the highly manganiferous pig iron oftenused on the Continent, there is difficulty in applyingthe spectroscope; but in special cases, and particu-larly when it is in constant use, it affords valuableindications of what changes are taking place. Thegeneral indications of the spectroscope have beenmentioned above ; but subsequently to writing thepaper, Snelus discovered that there exists some rela-tion between the time of the first appearance of cer-tain bright lines and the ultimate length of the blow.Thus, the two bright lines in the red portion of thespectrum marked c and /'appear to come into full viewwhen the mid point of the blow has been reached,if iron containing little manganese and about 2 percent, silicon is used. These lines are believed to bedue to carbon. Most of the lines in the green por-tion are due either to carbon or manganese, but thesubject has not been studied sufficiently to accountfor all the lines. The spectrum is very brilliantand characteristic. The principal lines producedby English hematite are shown in the followingdiagram:—
Fi«. III.
YELLOW GREEN
Li a b cclef g Na li ik lmn 0 p *irs t n
il I I I I I i I 1 II M I I I I I I 1 I I I I I
2 3 4 Q C
M I I I I 1 I II I I I I I I I I i I I I