The second method of making coke, and one which is still largely employed, is in the bee hire oven, many improvements in its construc tion having been made from time to time. As late as 1893, all of the 44.201 coke-ovens in the United States were constructed on the beehive plan. The chief advantage claimed for it is that it produces from prime coking coals the best quality of metallurgical fuel. A minor ad vantage is that water is applied to cool it, while the coke is still in the oven, after which the oven heat reduces the amount of moisture in the coke. The great disadvantage of the beehive oven is that, as ordinarily constructed, the valuable by-products—ammonia and tar—con tamed in the volatile matter are entirely lost. This oven is not adapted to inferior grades of coal. It derives its name from the dome shape of its interior. It is usually built of stone masonry, on a firm foundation, with its interior lined with specially designed fire-brick. The bee hive oven is usually about 12 feet in diameter and C to 7 feet high in the centre. The coal is a charged through hole in the centre of the roof, and is leveled off to an even depth of about 23 inches. The fresh charge is fired by the heat remaining in the walls from the previous charge, and the combustion is supported by air admitted through the front door, over the top of the charge. The volatile matter in the coal is driven off by the heat and burned in the top of the oven, along with a portion of the fixed carbon. The source of heat being at the top, the coking proc ess proceeds downward, and is effected by the partial combustion of the charge itself. In Eng land, and in a few American plants. the volatile matter is gathered into a conduit and carried under boilers, to raise steam for pumping water and other purposes; but usually the gases escape directly into the open air and are wasted.
As early as 1766 attempts were made in Ger many to save the by-products from coke-ovens. It is now accomplished in the retort orris, which was devised in Europe for the double purpose of saving the by-products and for utilizing for coke making inferior grades of coal. Very slow progress was made in developing the . process, and it was not until 1883 that it was put upon a paying basis. This was due to the unsatis factory design of the early coke-ovens and to the low price of the by-products, on account of the supply from gas-wo•ks. In 1856 Knab of France built a group of retort coke-ovens which had for their object, in addition to the making of coke, the double purpose of separating the tar and ammonia from the gases generated, and of then returning these gases to be burned in the flues to heat the ovens. The principal defect of these ovens was the failure to proportion the several parts to the quality of coal to he coked. This mistake was corrected by Carves of France in the Knah-Carves oven of 1873, which has proved a model for later ovens. Subsequent im provements were made by Albert Huessner in Germany and G. Siebel in France, in 1881. Coke produced in by-product ovens was regarded with disfavor till Dr. Otto introduced improvements which were patented in 1883 and are embodied in the Otto-Hoffmann coke-oven, to which the Siemens regenerator is applied. In 1887 the Semet-Solvay oven for coking dry coals, or a mixture of pitchy and dry coals, came into no tice. The Belgian ovens, designed for coking poor grades of coal, are also widely used in Europe.
The retort oven used in the Semet-Solvay process is a long, narrow chamber from 30 to 33 feet long, about 6 feet high, and from 15 to the last traces of tar and ammonia are removed. The gas is now returned to be burned in the flues of the coke-oven: but as this consumes little more than half of that generated, the rest is available for other purposes. The tar is col lected into tanks, and the ammonia is concen trated info a strong crude liquor or into sulphate of ammonia. A ton of coal will yield from 15 to 25 pounds of sulphate and 5 to 14 gallons of tar. The demand for these by-products is rap idly becoming greater, and in some of the most recent coke-ovens their manufacture is consid ered of more importance than the coke itself.
In Germany, 12.000 Otto-Hoffmann coke-ovens are in operation, of which 400 save the by-prod ucts and the remainder do not. There were, in 1901, in America. either completed or in course of construction, about 2000 such ovens. Of these only those in three plants attempt to save il luminating gas.
The accompanying table gives the yield in coke and other products of certain standard coals: 20 inches wide, depending on the quantity of the coal to be coked. The ovens are built in blocks of from 25 to 34, separated by flues in which gas is burned, the heat from which cokes the coal. The charge is introduced through sev eral openings in the top, the ovens are nearly filled, and then tightly sealed. As the heat in these ovens comes from the sides instead of the top. as in the beehive oven, the flow of gases generated is from the sides to the centre, while the free expansion of the coke is somewhat checked. As a result, some coals that in a bee hive oven make a coke that is too soft and spongy for blast-furnace use, are hardened and strengthened in the retort oven so that they are able to bear the furnace burden. The ovens being so much narrower, the process of coking requires only half as much time as in the bee hive furnace. When the charge is coked it is pushed out by means of rams through doors at each end. the doors are immediately closed, and the oven is ready for a recharge with almost no loss of heat. The coke is quenched as soon as it leaves the oven. The gases are conducted from the oven, through an opening in the top, into a collecting main. This is a hydraulic main, like that used in illuminating-gas works. The gas bubbles through the water, and a part of the tar and ammonia is condensed and separated as the gas cools, and then collected. The gas next passes through tubular condensers, where it is cooled by contact with a series of tubes through which cold water is flowing. During this process more ammonia and tar are con densed. The gas now goes through an exhauster, and, last of all, to a scrubbing apparatus, where For a complete discussion of the coke indus try, reference should be made to the "Mineral Resources of the United States," which form a part of the .Innual Reports of the United States Geological Survey: and to Platt, "Special Re port on Coke Manufacture," Penasy/ritain ±cc oar/ Geological Surrey Report of Progress (Harrisburg, 1876) : Fulton, Coke: A Treatise On the Manufacture of Coke and the Saving of By-Produrts (Scranton, 1S95) : Parker, Manu facture of Coke in 1896 (Washington, 1S97) ; Weeks, "Coke," in United States Census Office Eleventh Report on Manufacturing Industries (Washington, 1895). See AMMONIA : GAS; and TAR.