(13) The iron is subjected to the action of dilute hydrochloric acid, which dissolves the iron, and leaves on the surface a pellicle of homogeneous graphite, which adheres well to the surface of the iron. The piece to be preserved is next treated, in a hydraulically closed receiver, by hot or cold water, or, better, by steam, in such a manner as to completely dissolve and remove the chloride of iron formed. Finally, the piece of iron is left to dry in the receiver, from which all liquid has been removed. A solution of india-rubber, gutta-percha, or gum-resin in essence of petroleum, is then injected. On the essence being evaporated, there remains a solid enamel-like coat on the surface of the iron. Instead of previously eliminating the iron salt, it may be utilised in forming a kind of vitreous enamel. For this purpose, the iron is immersed, after treatment with the acid, in a bath of silicate and borate of soda. A very pure and brilliant silico-borate of iron is formed, which closes up the pores of the metal.

As to the disengaged chlorine, it combines with the free soda, forming chloride of sodium, which remains dissolved in the liquid. (Scient. Amer.)

(14) Bower-Barff Process

Briefly, this process, as now worked, is as follows: - The iron goods, whether rusty or not, are, without preliminary treatment of any kind, placed in a suitable chamber sufficiently capacious to hold about one ton weight of contents, and this chamber is heated by the combustion therein of carbonic oxide gas, produced near at hand by several gas furnaces, an excess of air over that requisite for combustion being admitted also into the chamber, after having been heated in its passage through coils of pipes placed immediately underneath the operating chamber. A film of magnetic oxide forms upon the immediate surface of the iron articles, and this appears to be surmounted with one of ferric oxide (Fe2O3) and it is by the subsequent reduction of this substance by means of carbonic oxide that the coating of magnetic oxide is increased to the requisite extent. In brief, the excess of air present in one stage of the process seems to form ferric oxide, and when the proportion of air present is reduced (as it may be at will) so that carbonic oxide is present, then the ferric oxide becomes reduced to the lower state of oxidation, its oxygen contributing to the production of carbonic anhydride.

The time required varies from 3 to 12 hours, and the magnetic oxide as thus formed exhibits a very pleasing French-grey or leaden tint. Should the colour, however, be unsuitable to the intended use of the iron articles, the more costly metals may be deposited upon them. (Kingzett.)

Prof. Barff has at various times pub-lished further details of the results of the process. As to the action on the strength of the iron, bars treated have been tested for breaking and tensile strain, and the result is that the strength of the iron is not affected. The coating gives great hardness to the surface of iron, when the coating is sufficiently thick (even less than 1/16 in.). An ordinary flat rasp will not remove it without great labour, and it resists emery powder, will for a long time resist a rasp, and remove pieces of steel from it. Substances which adhere to iron, zinc, and enamel will not adhere to it. Saucepans in which sticky substances are cooked can be cleaned with the greatest ease, after they have been oxidised, a simple wipe removing all dirt. A urinal in constant use for months had no deposits on it. Water was evaporated in an oxidised pan for 6 weeks - common tap water; the water never boiled, but was slowly evaporated; the deposit found was removed with a duster: it did not stick to the iron. This is a matter of great importance to boilers, and for pipes through which water is to be conveyed. Articles coated can be submitted to a high temperature, even a red heat, without the coating being injured or disturbed.

At present iron wire cannot be treated successfully - the wire can be treated and will not rust, but it cannot be bent to a sharp curve without the coating coming off. Riveted iron plates can be most successfully treated; the process tightens the rivets and assists the caulking. Weights were treated for the Government, and submitted to tests, and the process is now recommended by the department for the standard weights throughout the country. Prof. Barff has not yet met with any sample of cast iron which could not be properly treated. Wrought iron requires a somewhat different treatment; a lower temperature, about 900° F. (493° C.) suits it best, and steel also. It is not well to expose articles very different in bulk at the same time; all that are put into the muffle should be pretty nearly equal in bulk. For instance, very heavy articles, such as a 56-lb. weight, should not be treated with gutter spouts. Cast and wrought iron should not be treated together. For the chamber at present in use, 2 ft. x 2 ft. 6 in., and 1 ft. 6 in. high, the outlay on fuel for subjecting the articles within the chamber to superheated steam for 10 hours is 5s. In a manufactory this cost would be greatly reduced.

A West Bromwich firm, working the patent under a royalty, inform him that the cost will not approach that of galvanising iron, and he supposes it will little exceed that of the periodical three-coat paintings. The increase caused by oxidation can be estimated, and to fit nut and screw for each other, allowance must be made for this in casting. As to the treatment of riveted articles, and the danger that the coating film would be removed, there is some difficulty in this; he supposes boilers will be treated in large chambers when made up. When the rivets were hammered or pressed into the plates, the coating is removed, and of course these spots would be attacked by rust. The remainder of the plates might be protected from abrasion; the practical difficulty is to re-oxidise the rivets in situ. To accomplish this he proposes to cover the band of rivets with a porous cap of silicate cotton, and to subsequently re-oxidise that portion. But even if this were not successful, the rusting of the rivets would be of little practical injury, as it would take so many years to rust through a rivet, and the corrosion could not spread laterally, on account of the repellent character of the black oxide.