
Dee. 1, 1870] 
NATURE 

steel may be made, that will be entirely desiliconised, de- 
phosphorised, and desulphurised, and contain but the re- 
quired amount of carbon to form steel of any particular 
degree of hardness, by arresting the decarbonisation of 
the metal whilst undergoing conversion, 
Experiments were made in May last on a large scale at 
Messrs. Park, Brothers and Co.’s Steel Works, in Pitts- 
burg, in boiling puddling furnaces fettled with such oxides, 
but without puddling or labour of stirring the iron during 
conversion ; the only labour was that of “ balling” and re- 
moving it from the furnace after the conversion was com- 
pleted. The mode of application in the above-named 
establishment was 125lb. of titaniferous iron ore (similar 
to the Norwegian) and 5olb. of fluorspar, both in a 
powdered state, and then charging them evenly over the 
sole of the furnace. 475lb. of No. 3 charcoal pig-iron, 
similar in quality to English hematite of Barrow, was then 
charged upon them, and, when melted, was allowed to re- 
main without stirring or puddling. As soon as the pig- 
iron melted, reactions began between the fluorspar and 
the titaniferous iron ore, and the silicon and phosphorus 
contained in the pig-iron. To ascertain exactly the con- 
ditions upon which the changes are made in the cast-iron 
whilst under treatment, samples were taken from the bath 
of liquid iron. The analysis of the first sample shows that 
the operation of the new process is entirely different from 
any other process, inasmuch as the silicon is entirely re- 
moved at the early stages of the process ; and, with the 
silicon, phosphorus is also taken from the iron, and the 
carbon is changed from the graphitic to the combined 
form. Other samples were taken from the bath at inter- 
vals of ten minutes. The analysis of the first sample made 
it evident that most of the later ones were steel. This 
has since been confirmed by analysing a specimen taken 
thirty minutes later from the bath, and by the treatment 
of them as steel, it having been found that they possess 
the properties of steel, forging well, and tempering and 
~ hardening according to the various degrees of carbon con- 
tained in them. At the end of the operation, the charge 
became wrought-iron, by removal of a// the carbon. This 
iron forges, welds, and is neither red-short nor cold-short. 
The analyses of the first and fourth samples taken from 
the bath have been made by Mr. W. M. Habirshaw, 
analytical chemist, of 36, New Street, New York, and are 
annexed. Also, analyses of Messrs. Sanderson’s and 
Krupp’s cast-steel, and Hoop L Dannemura-Swedish bar 
iron are annexed for comparison, taken from Dr. Percy’s 
“Tron Metallurgy.” 
First Sample. 
Refined Cast-iron, 
taken 40 minutes after fusion. 
Carbon, combined 2°7144 
Fourth Sample. 
Steel, 
taken 30 minutes later. 
Carbon, combined 0°2172 
. do graphite traces do graphite none 
Slags (silicates) none Slags (silicates) none 
: Silicon 070046 Silicon none 
- Phosphorus 00349 Phosphorus none 
Sulphur 0°1073 Sulphur very minute trace 
Titanium trace Titanium none 
Fluorine none Fluorine none 
The presence of sulphur in the refined cast-iron is mostly 
due to sulphurets mixed with the fluorspar, which was 
used in the condition in which it was taken from the 
mine, there being no convenience at hand for dressing it. 
Later experiments with other kinds of pig-iron, with 
: impure or fluorspar not dressed, show that at the stage of 
. the process where it becomes refined cast-iron, the in- 
crease of sulphur from this cause amounts to o*1051 per 
. cent. ; which with “dressed” or pure fluorspar, would be 
o’co22 per cent. of sulphur for the refined cast-iron of the 
foregoing analysis, instead of 9°1073 per cent, 
: It will be evident to the practical metallurgist that the 
refined cast-iron, when treated with pure fluorspar, be- 
comes steel of superior quality when decarbonised below 
1°90 per cent of carbon. 

95 
Hoop L Bar-iron. | Sanderson’s Cast- Krupp’s Cast-steel. 
steel, 
Carbon 0'087. Carbon not deter. Carbon 118 
Silicon O115 Silicon 0°24 ~~ Silicon 0°33 
Phosphorus 0°034 Phosphorus 0'02 Phosphorus 0'02 
Sulphur 0220 = Sulphur 005 =©Svulphur none 
Manganese none Manganese 0'03 Manganese trace 
Arsenic trace Cobalt, nickel none Cobalt, nickel o-12 
Cobalt, nickel none Copper none Copper 0°30 
Copper none Aluminium o12 Aluminium none 
Aluminium none 
From the foregoing analysis it will be seen that the 
refined cast-ivon of the patent process contains but .& part 
of the silicon, less than one half the sulphur, and about 
the same amount of phosphorus, as compared with Hoop 
L bar-iron. This is the most celebrated wrought-iron 
made in the world, and is used exclusively for making 
steel, and sells at 24/. per ton at Sheffield. 
The analysis of the steel of the patent process shows, 
as compared with Sanderson’s and Krupp’s, that while the 
latter are alloyed with carbon, silicon, sulphur, phos- 
phorus, &c., the steed of the new process is practically 
ure iron and carbon. 
In the articles before referred to, written by Mr. Williams 
(and from his intimate practical and scientific knowledge 
no better authority can be had), he says :—* To make 
perfect steel they take out a// these latter, and leave 
nothing but pure iron and carbon. Absolute perfection is 
not, of course, practically attainable in steel making, but 
it is approximated in the same degree as the purification 
of iron from everything except carbon is effected.” 
Persons wishing to satisfy themselves of the value of 
the process, by a test, should use the ordinary boiling 
puddling furnace, fettled with the purest red hematite or 
specular ores, containing the least silica, ground and 
applied wet, with the ordinary “bottom” made in the 
usual way : and when perfect steel is required, that con- 
tains neither silicon, phosphorus, nor sulphur, good pig- 
iron should be taken, such as Nos. I and 2 Barrow, or 
West Cumberland hzematite ; and treated with pure fluor- 
spar, and titaniferous iron ore from the Bay of St. Paul, 
in Canada, containing over 43 per cent. of titanic acid, 
or from Norway, containing over 40 per cent. of titanic 
acid ; using 48lb. of fluorspar to 118lb. of the ore. They 
should be ground to fine powder, and mixed, and used 
dry, and charged evenly over the sole or bottom of 
the furnace; and 475lb. of pig-iron should be charged 
upon them, and the furnace closed tight so as to exclude 
all air, for about 70 minutes. The fire should be kept to 
the highest temperature. After the metal has been in the 
furnace this period, samples should be taken from the 
bath at intervals of five minutes. It will be found that 
a little experience will soon determine the proper time to 
stop the process. 
It is best zo¢ to allow the workman attending the fur- 
nace to stir or puddle the metal during the conversion, 
as the fluorspar and titaniferous ore become viscid by the 
heat of the furnace by the time the pig-iron melts, and, if 
left alone, will remain on the bottom of the furnace until 
decomposed by the reactions of the process, when they 
pass through the iron as vapour and slag, and purify it 
more effectively than can be done by stirring or puddhng ; 
and the slag serves to protect the surface of the metal 
from the effects of the sulphur in the fuel. The only 
labour that should be ad/owed is that of removing the 
steel from the furnace at the required stage of conversion. 
The time in conversion of the steel, from the charging 
of the pig-iron, is one hour and 30 to 40 minutes ; this 
time may be shortened 30 to 40 minutes, by previously re- 
fining the cast-iron of all its silicon and most of its 
phosphorus, by a shorter and more economical process 
(Patent, No. 1,051) with hematites and specular ores with 
| fluorspar, which will become the subject of a future 
article. 
New York 
JAMES HENDERSON 
