of the collars and "biled" shirt of the bachelor until he gets a wife. The Laundry Fellowship is an Association fellowship, with about 1,800 members behind it.

Other Fellowships bear the names of Synthetic Resins, Bread, Zirconium, Fish Products, Fuel, Plastics, Soap, Enameling, Synthetic Acids, Food Container, Protected Metals, Stove, Sulphur, Oil Shale, Nickel, Flotation, Glass, Oil, Quartz, Gas, Tar Products, Emulsion Flavors, Inks, Cements, Fiber, Yeast, Silicate, Magnesia Insulation, Coke, Organic Syntheses, Insecticides, Glue, Fertilizer, Dental Products, Cleaning, Refractories, Asbestos, Fruit Beverages, and Magnesia Products.

The Bread Fellowship is the oldest, probably, at the Institute, and one of the most successful. It would be hard to overstate the importance of the work of this Fellowship, and of similar work done elsewhere, upon the quality and cost of commercial bread. The processes developed save, probably, half the yeast and half the sugar used in bread-making. One has only to compare the commercial bread of today with that of a few years ago to realize the enormous importance of scientific methods applied to this ancient art of the home.

The Fellowship labeled "Fiber" looks ordinary and uninteresting; but when one understands its ramifications and import, it has a different significance. One of the problems of this fellowship is the testing and development of fiber shipping containers. If one goes through a freight warehouse these days, he is struck with the lack of wooden boxes, and the way in which fiber boxes and cartons have taken their place. The development of containers includes not only the study of the fiber boards, multiple and corrugated, but of the adhesive, which must be cheap and at the same time proof against storage in damp warehouses and exposure to weather.

One of the tools of this fellowship is a miniature Ferris Wheel, operated by a motor and containing a series of baffles, so that a loaded container may, in a few minutes, be subjected to all the drops and bumps of a thousand, two thousand, or three thousand miles. Other tools give the actual strength of the fiber and tape employed.

The Sulphur Fellowship has a number of most interesting problems. You will realize how enormously the production of this element was stimulated by the war, as the starting point in the manufacture of sulphuric acid, which in its turn lay at the foundation of the manufacture of explosives. Now that the war is

over, the companies producing sulphur in the Texas fields have an enormous excess of this element over what the markets can possibly absorb. The question is, how to use the vast sulphur deposits. Perhaps some of those present will have some ideas on the subject. A large scale use which seems possible is as a material for large acid proof containers. Curiously enough, while the sulphur obtained from the deposits by the Frasch process is very pure (often 99.9%), the presence of a trace of oil in the sulphur makes its continued combustion in a sulphur burner difficult, because the oil forms a film which extinquishes the flame. A special burner had to be devised for the purpose.

The Yeast Fellowship, the Flotation Fellowships, and the Coke Fellowship have problems of most far reaching character and are almost research institutes in themselves. Such fellowships are of the type called Multiple Fellowships, in which the Senior Fellow is a man of unusually high research ability, in charge of a group of investigators for the solving of a group of problems.

Fellowships like those on Synthetic Resins, Synthetic Acids, Organic Synthesis, and the Pratt Memorial Fellowship are doing work of a pure-science research character, but often on a scale of which the organic chemist rarely dreams.

The stipend carried by the fellowships at Mellon Institute is by far beyond that allowed in the usual college or university. Since it is paid by manufacturers, accustomed to a business man's scale of compensation for service performed, instead of by Boards of Trustees doling out very limited funds to needy students, there is a possibility of attracting and holding men of university research character to the work of investigation. The advantage of these fellowships is further increased by the fact that by special arrangement with the Donor, the Fellow may spend a limited amount of his time in graduate study or in teaching at the University of Pittsburgh. A number of Fellows have received higher degrees in this way.

The writer will close this paper with the inscription he often pondered upon during his year in Pittsburgh. It is: "This building is dedicated to the service of American Industry and to young men who destine their life work to the industries; the goal being ideal industry, which will give to all broader opportunities for purposeful lives."

Read before the Chemistry and Physics section of the Illinois State Academy of Science, at Rockford, April 28, 1922.

CONCENTRATION OF RADIUM FROM CARNOTITE ORES.1 BY B. S. HOPKINS and G. C. RUHLE,

University of Illinois, Urbana.

Carnotite is a potassium uranyl vanadate of essentially the composition represented by the formula K2O. 2UO,. V.O,. 8 H.O. All the mineral substances present are valuable but the most valuable component, radium, is present in such a small proportion that it cannot readily be shown in the formula. It requires a ton of relatively rich carnotite ore to produce 10 milligrams of radium. It is very evident, then, that any method which is efficient in the removal of radium must be capable of effecting practically 100 per cent extraction. It is, also, clear that a very important part in the process will consist in the concentration of the minute quantities of radium after they have been removed from the great bulk of the ore.

For the first step in the process, the United States Bureau of Mines recommends the employment of nitric acid by the use of which practically all the radium, together with most of the other valuable mineral constituents present, is converted to the soluble form. The solution obtained in this manner is nearly neutralized, barium chloride is added, and the radium and barium are precipitated by adding sulfuric acid. The precipitated radium-barium sulfate is filtered off and from the clean solution, uranium is precipitated, usually as sodium uranate, and the vanadium either as ferrous vanadate or calcium vanadate.

The main advantage claimed for this process is the high recovery of radium. The disadvantages arise from the cost of nitric acid and the fact that there is only partial extraction of vanadium. In large measure the cost difficulty is overcome by the fact that a very considerable portion of the nitric acid may be recovered and used again. If the main object in view is the extraction of radium, this method is said to be especially efficient.

The radium-barium sulfate, which contains only a very small per cent of radium must now be subjected to treatment for the concentration of radium. The usual procedure is as follows: The mixed sulfates are reduced to sulfides by heating with charcoal or are converted to the carbonates by boiling with sodium. carbonate solution. The sulfides or carbonates so obtained are dissolved in hydrochloric acid and the resulting solutions subjected to fractional crystallization. This method of concentra

Read before the Chemistry and Physics section of the Illinois State Academy of Science, at Rockford, April 28, 1922.

tion depends upon the fact that when a saturated solution of radium-barium chloride is cooled from 100° to 0° the crystals formed are much richer in radium than the original solution. Accordingly if a solution of the mixed chloride is evaporated until there remains not quite enough solvent to keep all the salts in solution there will be a tendency for the radium chloride to crystallize out, while the mother liquor will become correspondingly richer in barium chloride. After this process has been repeated many times, it is found that the radium is nearly all concentrated in the crystal fractions, while the solutions at the "soluble end" of the series contain no radium.

It has been found that the concentration of radium takes place more rapidly if this process of fractional crystallization is carried out by the use of bromides in place of chlorides. This is explained by the fact that the bromides are 'more soluble than the chlorides. If a saturated solution of the

chlorides is cooled from 100° to 0°, about 50 per cent of the solute crystallizes out; but under the same conditions the bromide solution will give up only about 34 per cent of the salts. Hence, there is a more rapid concentration of the radium if this salt is used.

The concentration of the radium in any fraction may be calculated from the equation:

in which n is the number of crystallizations, A is either the actual or assumed concentration of some dish to start with and K called the enrichment factor is a constant, when the crystallizations are carried out under exactly similar conditions. It represents the relative concentration of the radium in the crystals to that in the original material. It has been shown that the enrichment factor is practically independent of the degree of acidity of the mother liquor; likewise that this factor for a bromide system is 2.6 while for a chloride system it is about 1.6. Reasoning from the familiar relationships shown in the periodic table, it might readily be concluded that if a bromide system was more efficient than a chloride system the fractional crystallization of the iodides would be considerably more efficient than either the chloride or the bromide. With this view point in mind, a series of experiments are now being conducted to determine the practicability of an iodide system of radium concentration.

Before this experiment could be carried out successfully, it was necessary to determine the best method of preparing the

iodide solution. Several methods were tried such as: (1) Superheated steam was passed over the sulfide converting it to the hydroxide, which was then heated to a dull red in a stream of hydriodic acid gas; (2) the sulfide was transformed to the iodide by boiling with an alcoholic solution of iodine; (3) the sulfide was added to a boiling solution which contained slightly more than the calculated amount of ferrous iodide; (5) the sulfide was boiled with hydriodic acid solution and a small amount of iodine in hydriodic acid was added. Of these methods the last proved to be the most satisfactory so it was employed. To test out the efficiency of the halide fractionation systems, three samples of radium-barium sulfate, each weighing 100 grams, were reduced with charcoal and the resulting sulfides were treated with hydrochloric, hydrobromic and hydriodic acids respectively. After the action had ceased they were boiled to expel hydrogen sulfide, filtered and the residues washed thoroly. The filtrates were evaporated to dryness, taken up with a small quantity of water to which was added a little of the free acid and the fractional crystallization begun, by evaporating on a steam bath until the solution were completely saturated. Then the dishes were cooled in ice water, the crystals filtered out, redissolved in pure water and recrystallized.

To test the efficiency of the solvent action of the three halogen acids the residues from the acid extractions were analyzed for their barium content and were found to contain practically the same per cent of that element. Hence, it was concluded that the acid extraction was the same in all cases and that the three solutions presumably contained the same amount of radium.

After several crystallizations of the three halide systems, equivalent amounts of the richest fraction of each were placed in the case of a charged electroscope and the time of discharge noted. The iodide discharged the electroscope more quickly than either the chloride or the bromide, but the work has not yet progressed to the point which will permit a definite statement concerning the value of its enrichment factor. Another decided advantage in the use of the iodide comes from the greater solubility of this salt over the others, hence a given amount of radium in the iodide solution occupies a much smaller volume than it does in the chloride or bromide solution. This permits the use of smaller crystallizing dishes, a material saving on a large scale operation.

John L. Niermann. Jour. Phys. Chem. 24 192 (1920). 2C. E. Scholl. Jour. Am. Chem. Soc. 42, 889 (1920).