Popular Science Monthly/Volume 29/July 1886/Rustless Iron
By JAMES S. C. WELLS, Ph. D.
OF the many methods in use for the protection of iron from rust, the one of most scientific interest is the so-called Bower-Barff process. By this treatment the iron is coated with a layer of the black or magnetic oxide of iron (Fe3O4), and, as is well known, this oxide does not undergo any further oxidation on exposure to air or water—proved by the magnetic iron-ores and sands, which withstand any amount of weathering. The liability of iron to rust is a great drawback to its use for many purposes, and the practical value of a process which will protect it, at a slight expense, is self-evident. That the process is successful in accomplishing this object seems no longer a matter of doubt, and at less cost than galvanizing or tinning. The color on cast and wrought iron is a bluish-gray, which to some may be objectionable, but, as the coating takes paint far better than untreated iron, this objection is easily overcome, and with the assurance that the paint will remain, and not soon be thrown off as it is generally. For polished work the color is a lustrous blue-black, adding greatly to the beauty of the article treated. This process seems peculiarly well adapted for gas and water pipes. Any one who has had occasion to use water which has passed through a new iron pipe, or one that has not been used for some time, knows how full of rust it is, and that only after months of constant use does it become clear again. With pipe coated with the magnetic oxide by the Bower-Barff process, no trouble of the kind can occur. The water runs pure from the first day, and if for any reason the pipes are emptied, and left so, there is no danger of their becoming coated with rust. Another important fact is, that the water coming through one of these rustless pipes is just as pure as when it entered, for the water can dissolve none of the coating of oxide, as it always does with lead or galvanized pipes. It is a well-known fact that water running through lead pipes is very apt to contain lead in solution, and the continued use of such water causes lead-poisoning, for, although the amount (of lead) dissolved may be very small, still it accumulates in the system, and finally causes sickness and disease.
Professor Venable has lately shown that water passed through galvanized pipe dissolves quite an appreciable quantity of the zinc coating, thus making it unfit for drinking purposes. Tin-lined pipes are also used, and until the introduction of the "rustless" pipe were considered the best, but were far from being all that could be desired; in many instances, after using for a time, the coating was completely destroyed. Then, again, if the pipe is to be better than lead, the tin used for the lining must be pure, because if it contains lead, which is often mixed with tin, it would be worse than the common lead pipe, the alloy dissolving much more readily than either would alone. Considerable architectural iron-work protected by the rustless process is being used with very satisfactory results. It is needless to multiply examples of its usefulness, for numberless ones will occur to the reader. Up to the present time only four furnaces have been built in this country—two in Brooklyn, one at Little Ferry, New Jersey, and one in Philadelphia. The processes by which this coating of magnetic oxide is formed differ accordingly as the iron is cast, wrought, or polished. The Bower process is the better for cast-iron, and consists in oxidizing it by means of carbonic acid and air. In the Barff method, which is the one used for wrought-iron and polished work, the oxidation is produced by means of superheated steam. This method will also give a coating of the magnetic oxide on cast-iron, but the action is very much slower than with the Bower treatment, and consequently more costly. The difference is probably due to the large amount of carbon contained in cast-iron, and which has to be oxidized as well as the iron, i. e., the carbon contained in the film of iron which is changed to oxide. It may be asked then, why, if air does the work so much quicker than steam, it can not be used for wrought-iron and polished work, as well as for cast-iron? It has been found by experience that the coating produced on the former, when air is used, is liable to scale off, which is not the case when it is treated with steam. Cast-iron after treatment seems tougher than before. I have frequently noticed, when present at the unloading of a charge of hollow-ware that had been treated, a kettle or pot fall off, and, although falling against heavy iron, it would bound off and reach the floor uninjured. The same accident happening to any such article before treatment is almost sure to break it. Whether this toughening is caused by a kind of annealing due to the slow cooling of the charge after coming out of the furnace, or whether it is that the surface of the iron becomes malleable owing to the oxidation of its contained carbon, I can not say, but think it probable that both contribute to the result. Mr. Bower, in his first experiments, treated the articles in a muffle-furnace—that is, a furnace in which only the oxidizing gas, either air or carbonic acid, was allowed to come in contact with the iron; but this was found to be costly as well as unnecessary. Now the furnaces are built so that the products of combustion of the coal, used to heat the furnace, come directly in contact with the iron undergoing treatment, and by a suitable arrangement of dampers, etc., the same furnace can be used for the Barff or steam process. In his experiments the inventor found that, if he admitted a large excess of air, the article came out covered with the red oxide of iron (Fe2O3), and that below this red coating was a thin film of the magnetic oxide. For some time he was at a loss how to regulate the quantity of air added, so as to prevent the formation of this red oxide, but finally hit upon the following plan: During a certain time the iron was oxidized in excess, that is, to sesquioxide (Fe2O3), and then it was subjected to a reducing action for a definite shorter period. In this way it was obtained covered with the magnetic oxide only. The chemical reactions that take place are given by Mr. Bower as follows: the excess of air during the first part of the reaction causes the formation of sesquioxide of iron (Fe2O3),
2 Fe O3 Fe2O3;
but, this being in contact with red-hot iron, its lower surface is reduced to magnetic oxide, as shown in the following reaction:
4 Fe2O3 Fe 3 Fe3O4.
It would seem from this reaction that no reduction period would be necessary: theoretically it would not, but practically it is required, as there is always an excess of the red oxide (Fe2O3).
This excess of the red oxide is then reduced to magnetic oxide by the reducing gases, consisting chiefly of carbonic oxide (CO), which is converted into carbonic acid as follows:
3 Fe2O3 CO 2 Fe3O4 CO2.
When steam is used instead of air, the steam coming in contact with red-hot iron is decomposed, giving up its oxygen to the iron, and forming the black oxide, its hydrogen being at the same time set free as gas. The following shows the reaction:
3 Fe 4 H2O Fe3O4 H8.
We will now pass to a description of the different processes, as carried out on the large scale. The furnaces in use vary somewhat in construction, but the principle is the same in all. The iron to be treated is placed in a large fire-brick chamber, known as the oxidizing chamber, into which the gases from the producers pass after having been through a combustion-chamber. In the latter the gases can be mixed with air, and burned, when necessary, the amount of air being regulated as required. The producers are simply very deep fireplaces, the bed of coal being three to four feet deep, so arranged that only sufficient air is admitted to burn the coal partially; that is, instead of being converted into carbonic acid (CO2), it is only allowed to take up one atom of oxygen, and thus forms carbonic oxide (CO). The latter, either alone or mixed with hydrocarbons, formed by the decomposition of naphtha, used during the reduction periods, is the gas that passes into the combustion-chamber to be there mixed with air or not, according to the result desired to be produced.
The fuel used in this country is anthracite coal, with the addition, during the reduction periods, of a little naphtha or crude petroleum. A small stream of the oil is led into the top of one of the gas-producers, where it is vaporized, and passes along with the other gases on their way to the oxidizing chamber.
Cast-iron before it can be treated requires to be "pickled," to remove any sand adhering to the casting. The "pickling" is done as follows: The iron is first placed in a bath of dilute sulphuric acid, and allowed to remain there for from ten to fifteen minutes, or even longer, if the casting is a very rough one. It is then removed and washed with boiling water, and when dry the sand which is now loosened is taken off with steel brushes. Now the article is ready for treatment in the furnace. The pieces to be treated are placed on the drag, a heavy iron plate, which is run into the furnace through a large door in one end. Before making the charge, the furnace must be heated to a white heat; having attained this temperature, the door is opened and the charge run in as quickly as possible, then closed and tightly luted, so as to exclude all air. The charge is heated with a slightly reducing flame, that is, one in which there is an excess of carbon, until it has reached a bright cherry red. During this heating of the charge only enough air is admitted to the combustion-chamber to partially burn the gases. When it has reached the proper temperature, the "gas" is turned on, that is, the supply of air is entirely cut off, a small stream of oil is allowed to flow into the producer, and the chimney-damper nearly closed. The flame in the furnace now becomes smoky. During this reduction or "gas" period, which usually lasts twenty minutes, the gases rich in carbon come in contact with the red-hot charge, and reduce any sesquioxide to magnetic oxide. At the end of this time the oil is shut off, the gas partially so, and the chimney-damper opened. When the furnace has become free from smoke, the air-valve is opened so as to let air enter the combustion-chamber in quantity only slightly in excess of that needed to burn the gases completely. This excess (of air), in conjunction probably with the carbonic acid formed by the combustion of the gases, oxidizes the iron to sesquioxide, which will be reduced by the next "gas" to magnetic oxide. When the oxidizing period, which lasts forty minutes, is over, the air-valve is closed, the gas turned full on, and the oil run into the producer as before, and the second "gas" or reduction period of twenty minutes begins. These alternate periods of reduction and oxidation are continued for from eight to ten hours, and the charge is drawn at the end of the last "gas," or, what seems to assure better results is, instead of withdrawing the charge now, to complete the treatment by admitting steam for one hour, as in the Barff process, to be presently described. By thus finishing the operation with steam a more uniform color is obtained, due probably to the oxidation of any protoxide of iron that might possibly have been produced during the last reduction, thus insuring a coating consisting only of magnetic oxide. Wrought-iron or steel, of course, requires no "pickling," as there is no sand to be removed. The furnace is heated to the same temperature as for cast-iron, and then the charge is run in and heated up with a strongly reducing flame until it reaches a bright red. The "gas" is then turned on for twenty minutes, and, when this time has expired, the chimney-damper and gas-valve are both closed tightly, and the steam-valve is opened into the combustion-chamber. The steam, in passing through this chamber, which is at a white heat, becomes highly superheated before reaching the charge in the oxidizing chamber. It is known that sufficient steam is being admitted, by the amount condensed on a cold iron bar held at one of the openings, through which the excess of steam and the hydrogen set free in the reaction escape from the furnace. The steam is kept on for from eight to ten hours, and then the charge is withdrawn. When polished work is to be treated, the furnace is not heated so highly as for wrought-iron, and, just as soon as the charge has been made, the gas is turned on for one hour, then steam is admitted, and the operation goes on the same as for the latter, with this difference, that the temperature in the furnace is kept very low—so low, in fact, that on looking into the furnace the charge is scarcely visible. If too high a heat is used, it causes the coating to scale. The steam is kept on for from eight to ten hours, and the charge is taken out. At first the articles treated are completely covered with soot when they come out, and do not look attractive, but, on rubbing with oil, which is the next step, the soot is removed, and leaves the articles a beautiful, lustrous blue-black.