Somewhat similar and exceedingly danger ous are the explosions of finely divided ma terials which in their ordinary form are con sidered anything but explosive. Finely divided coal dust mixed with air has been the cause of many disastrous mine explosions and fires. In 1878 a tremendous explosion in the Washburn Flour Mills in Minneapolis, Minn., demon strated the fact that any finely divided car bonaceous material constitutes an explosive substance when mixed with certain proportions of air. Indeed, many finely divided metals which ordinarily do not burn in air or oxygen will burn readily in an ordinary gas flame if the material is in the form of a fine powder. All of these are instances of a combination of oxygen with other materials.
It is not generally appreciated that in our various forms of explosives we have combus tion or reconrbustion of oxygen with other constituents and that in such materials we com monly distinguish between rapid combustion, explosion and detonation, all being different de grees of speed of oxidation. Explosive pow ders, for example, are so regulated as to give any desired rate of combustion to meet the projectile or gun requirements.
It has been shown that even at high tempera tures the presence of a trace of moisture is necessary for free oxidation, and in absokitely dry oxygen even sulphur and phosphorus can be distilled and carbon made red hot without any combustion taking place. A jet of burn ing carbon monoxide is entirely extinguished when introduced into pure and absolutely dry oxygen, but the presence of the minutest trace of moisture is sufficient to restore the oxygen to its activity.
Manufacture.-- The following methods have at various times been tried out commercially for the production of oxygen in quantity: 1. The ignition of nitre. The oxygen from this source was contaminated with oxides of nitrogen.
2. Heating manganese dioxide. (See p. 489). This was one of the cheapest of the early methods.
3. Heating a potassium chlorate-manganese dioxide mixture as described (p. 489). Until about 1885 this was the almost exclusive source of the considerable quantities of oxygen used for lime-light purposes.
A number of processes such as the follow ing are now obsolete: extraction from air by utilizing solubility of oxygen in water; the alternate oxidation of cuprous chloride by air, the absorbed oxygen being recovered at dull red heat; alternate formation and decomposition (with liberation of oxygen) of alkaline manganates; heating a concentrated solution of bleaching powder at 70 to C. with a cobalt
catalyst, etc.
4. The alternate formation and decomposi tion of barium peroxide. Thorpe's Diction ary of Applied Chemistry) (1916) gives the following account of this process: "In 1851 Boussingault found that when barium oxide (baryta) is heated to a dull-red heat in a current of air it is converted into barium peroxide, and that at a higher tempera ture this peroxide is again resolved into barium oxide and oxygen. But his attempts to utilize this reaction as a practical and economical source of oxygen failed owing to the fact that after a few oxidations and deoxidations the baryta lost its power of reabsorbing oxygen. Many other attempts were made to overcome mining the respective phases of oxidation and deoxidation. Neither the oxidation nor the deoxidation is as complete as when two tem peratures are used, and the yield per opera tion is much less. But the duration of the operation was reduced from about 4 hours to 8-15 minutes, and the total daily yield therefore largely increased. At the same time the opera tion was much simplified, the wear and tear of furnace, retorts, etc., very greatly reduced, and the fuel required lessened. Labor was also economized, the multiplied reversals of cocks, etc., necessitated by the single-temperature method of working being effected automatically by reversing gear designed by K. S. Murray. The labor required was therefore little more than that needed for stoking the furnace, and oiling and supervising the pumps, etc. The oxygen obtained had a purity of about 93-96 per cent.
"For a producer capable of delivering 10,000 cubic feet of oxygen per 24 hours the con sumption of coke in the furnace is about 12-15 hundredweight per day, and for plant of that or smaller size the pump power required is this difficulty, but for long without success. In 1879, however, the MM. Brin freres were more successful, and took out a patent for the process (Eng. Pat. 1416 of 1880). Further improve ments were made under the auspices of the company formed to develop and work the patents, and the process was made practical and economical and was worked on a large scale at various places.