In the Bessemer process the conditions are worse still, as these elements are all blown through the whole mass of the steel.
We know the effect of oxygen and how to eliminate it practically.
Percy gives the effects of nitrogen as causing hardness and extreme brittleness, and giving to iron or steel a brassy lustre. Such a brassy lustre may be seen frequently in open-hearth or Bessemer steel, and occasionally in crucible-steel. When seen in crucible-steel it is known to be due to the fact that the cap of the crucible became displaced, exposing the contents to the direct action of the flame. Of the effect of hydrogen we know less; there is no reason apparent why it may not be as potent as the others.
Ammonia in sufficient quantity to be detected by the nose has often been observed in open-hearth and Bessemer steel.
To settle the nitrogen question Prof. John W. Langley developed some years ago a very delicate and accurate process for the determination of nitrogen even in minute quantities; the process was tedious and expensive, so that it was not adapted for daily use; it involved the careful elimination of nitrogen from all of the reagents to be used, requiring several days’ work, in each case to prepare for only a few nitrogen determinations.
By this process it was found, in every one of many trials, that crucible-steel contained the least amount of nitrogen, open-hearth steel the next greater quantity, and Bessemer steel the greatest amount. He found no exceptions to this. For many years great efforts had been made both in Europe and in the United States to make by the Bessemer or the open-hearth process a cheap melting-product to be used in the crucible instead of the expensive irons which so far have proved to be necessary to give the best results.
There appeared to be no difficulty in making a material as pure chemically, or purer, than the most famous irons in the world, and this material was urged upon the crucible-steel makers. Careful tests of such material failed to produce the required article; in fact it was demonstrated over and over again that an inferior wrought iron would produce a stronger steel than this very pure steel melting-material, and crucible-steel makers were compelled to adhere to the more costly irons to produce their finer grades.
Prof. Langley determined the nitrogen in a given quantity of open-hearth and Bessemer steel; this same material was then melted in a crucible, and it was found that the resulting ingots contained nearly as much nitrogen as the original charge. The quantity was reduced slightly; still this steel contained more nitrogen than any other sample of crucible-steel that he had tested. The physical test of this trial steel showed the usual weakness of the Bessemer or open-hearth steel, as compared to crucible-steel.
The next step was to try to get rid of nitrogen by the use of some affinity, as oxygen is removed by manganese. Boron and titanium seemed to be the most feasible elements; boron appeared to offer less chance of success, and titanium was selected. A ferro-titanium containing six per cent of titanium was imported from Europe at some expense. As the most careful and exacting analyses of this material failed to reveal a trace of titanium, it was not used.
After many futile efforts Langley succeeded, by means of electric heat, in reducing rutile and producing a small quantity of an alloy of iron and titanium. A trial of this alloy, although not conclusive, led to the belief that such an alloy could be used successfully to eliminate nitrogen; but as its cost, about two dollars a pound, was prohibitory of any commercial use, the subject was not pursued farther.