General Chemical Company, very little information is at present available. It is claimed, however, that it does not require the high pressures necessary in the Haber process, and that ammonia can be produced by it considerably cheaper than by the Haber process.

The Bucher process for the production of NaCN, though still in the experimental stage, is so simple of operation, and the raw material is so cheap and easy to obtain, that it promises to give fixed nitrogen at a cost lower than that of any other known process. Dr. Parsons, chief chemist, Bureau of Mines, says of it, "From the chemist's standpoint the process is today a success. There is no difficulty whatever in the chemical reactions. When Na,CO,, ground coke or carbon in any form in contact with finely divided iron are heated to redness, and nitrogen or even air passed through the mass, nitrogen in quantity is fixed as NaCN. The reaction takes place readily. No power factor of any consequence is involved, and it appears certain that if the mechanical difficulties are solved, nitrogen will be fixed in this form cheaper than by any other known synthetic process. The process has the further advantage that it would also make cheaply available cyanide, which is so greatly needed by our mining industries."

.

In discussing his process Professor Bucher says the idea is not new but that the fixation of atmospheric nitrogen to form alkali cyanides is by far the oldest of all such methods. In 1839 Lewis Thompson, while attempting to improve the method then in use for the manufacture of Prussian blue, found that KCN is formed when nitrogen is "allowed to act on a mixture of carbon and potash under favorable circumstances." In describing his method Thompson states that he found it necessary to use iron. "When iron is not employed," he says, "a much higher temperature is required." He ground together into a coarse powder two parts potash, two parts coke, and one part iron turnings, and heated the mixture in an open crucible in an open fire for about half an hour, and obtained an abundant yield of KCN.

The announcement of Thompson's results soon led to very active discussions and investigations by some of the most noted investigators of that time, including Berzelius and Bunsen. For some reason unexplained, however, all of these investigators omitted the iron from their mixture, and as a result were either not able to obtain cyanide or were able to obtain it only at very high temperatures.

The early work of Professor Bucher on this subject consisted in the investigation of the following reactions:

2Na+2C+N,→2NaCN+46,200 cal.

Na2CO3+2C→2Na+3CO-184,700 cal.

(1)

(2)

Reaction (1) was found to proceed slowly, and the time required for completion was always a matter of hours. When iron was added to the mixture, however, the reaction was almost instantaneous at a low red heat. The catalytic effect of iron was therefore established. Its commercial application would depend on the possibility of obtaining cheap sodium. The applicability of reaction (2) was investigated for this purpose. A number of experiments combining the two reactions showed that cyanide could be prepared very readily at temperatures about 860° to 950° when iron was used as a catalyzer, but that appreciable quantities could not be obtained below 1000° in the absence of iron. This established the principle of the reaction. The next problem was to transform the laboratory experiment into a commercial possibility.

To get rid of the inconvenience of passing nitrogen over a powder, such as would be formed by the intimate mixture of Na,CO,, carbon and iron, briquettes were made. Of the methods tried out, that which gave most satisfactory results was as follows: The iron or hematite and carbon were ground separately. until they would pass through a 100-mesh sieve. The two were then mixed and the grinding was continued for about an hour. The Na,CO, was then added, and the mixture ground for about five minutes longer. The dry product was then placed in a steam jacketed mixer, and enough hot water added to make a thick paste. The briquettes were made by forcing this pasty mass through an ordinary meat grinder provided with a plate containing holes one-eighth inch in diameter. The knife was adjusted so that pieces about an inch long were cut off. These were dried in an oven. They formed a hard, dry briquette, quite free from dust.

It was found necessary to use hot water in the mixing in order to assure the formation of the monohydrated sodium carbonate since the formation of any of the decahydrate caused the briquettes to crumble on drying. The transition point from the deca- to the monohydrate is at 35°. A mixture of two parts iron, two parts coke, and one part soda was found to give the best results. It was also shown that air, producer gas, or even gas from combustion chambers could be used in place of nitrogen in the reaction.

Both a batch and a continuous furnace had been tried out on an experimental scale and found to work quite satisfactorily. Certain mechanical difficulties, however, have prevented placing it immediately on a commercial scale. The chief object of the Government's investigation will be to ties.

solve those difficul

The extraction of the NaCN from the reaction product is accomplished in either of two ways, by distillation or by leaching. By the distillation process the cyanized briquettes are heated in a closed vessel to about 1000° and at 2 mm. pressure. The cyanide distills over and collects in a pool which solidifies to a clear glass. The method is claimed to effect an almost complete separation. It is necessary to observe certain precautions in the leaching process, owing to hydrolytic action on the NaCN, the formation of ferrocyanide, and the tendency of the dihydrate (NaCN, 2H,O) to "set" like plaster of Paris. To avoid these dilemmas the lixiviation temperature should be slightly above 35°, the transition temperature of the dihydrate, and the leaching carried out as quickly as possible. The solution when evaporated to dryness gives NaCN sufficiently pure for many purposes.

An idea of the cost of fixed nitrogen by this process may be had from the reactions involved. The reaction

Na2CO2+4C+N,→2NaCN+3CO-138,500 cal. would theoretically require 35,000 h. p. to produce 180,000 tons of HNO, allowing 85 per cent efficiency in oxidizing the ammonia. The three molecules would give on burning

3C0+120, 3CO2+200,000 cal.

equivalent to about 50,000 h. p. to help make up the unavoidable heat losses. Combining the two reactions it is seen that the total reaction is exothermic, showing at least that the power factor is of little consequence. The great simplicity of the process and the low cost of the materials used also indicate that the method should produce fixed nitrogen at a very low cost as compared to other processes now in use.

THE SOLAR ECLIPSE OE JUNE 8

BY WILLIAM B. THOMAS,

Jamestown College, Jamestown, N. Dak.

Since eclipses recur in a cycle of eighteen years, eleven and one-third days, nearly, the solar eclipse of June 8 is regarded as a repetition of that of May 28, 1900. Since, however, this eighteen-year period, the Saros of the Chaldeans, is not an exact dividend of the days in the year combined with the moon's motion, each succeeding appearance of the corresponding eclipse comes some eleven days later in the year; but since the earth has completed .32 of a new rotation, in other words has spun off a third of an additional day before the conditions making the eclipse arrive, at the end of the Saros, each recurrence is seen some 120° of longitude farther west than the one preceding. When it is added that every third Saros brings a specific eclipse approximately to the same longitude, it will be seen that anyone who will apply the sense of the facts just given, as far as they go, is predicting an eclipse, in a rough way.

The general conditions of the event being now given, the eclipse of June 8 may be examined. If the rotational motion of the earth were to cease for the afternoon of that date, at about 3:35, counted as Chicago time (or 4:35 counted as "daylight-saving-time" at Chicago), from a very high aeroplane, properly situated, a shadow, moving from thirty to forty miles a minute in the general direction, southeast, might be seen starting to flit over the continent. It would enter off Washington and leave the continent off Florida. We are supposing the earth to stop, but the usual phenomena of time and daylight to go on. This could not be, and, in reality, the minute of arrival, anywhere, of this unusual shadow is complicated by factors, such as the latitude of the observer, his nearness to the exact center of totality, and, of course, the rotational motion of the earth. The shadow moves forward like that of a cloud seen over a hilly countryside, but in a manner calculable mathematically.

The general data of eclipses are well enough known. For example, Oppolzer's lists trace all eclipses occurring for over eight centuries. The following years will bring total eclipses visible in the United States, namely: 1918, 1923, 1925, 1945, 1954, 1979, 1984, and 1994. Obviously, all these are not that of 1918 repeated. The same definite locality will not observe a total eclipse oftener than once in 360 years. The Naval Observatory publishes in the Nautical Almanac exact circumstances

of eclipses each year. Owing to the line of totality, full calculations, this year, are given for Denver, Col., with local circumstances for seventy-seven other points. These circumstances, as they are called, are concerned with the time, magnitude, and angle of the shadow at the beginning, middle, and end. The region of totality must be considered favorable this year. The eclipse may be observed at places easily accessible, and the uncalculable meteorological factor stands a good chance of being auspicious. The corresponding eclipse of 1900 appearing about 8 o'clock in the morning, in the Rocky Mountain district, was seen as a partial eclipse in a perfectly clear sky. The various matters of addition to theoretical knowledge which await the coming of a total eclipse render it sure that even war conditions will not prevent a careful scrutiny of this phenomenon, the present June, and the leaving of records for the future. Quite obviously, since science is progressive, equipment for study was never before quite so perfect for the making of an observation. The minute or two of totality shown in the Almanac on page 561 as occurring on the central line will be a time of close watching and careful comparison.

The area of totality for a total solar eclipse being narrow and variable, each observer who cares to do so may calculate for himself such items as (1) position of point of contact, (2) time of maximum eclipse, and (3) magnitude of maximum eclipse. A glance about page 729 of the Almanac will show the laboriousness of this work. It will readily appear that it is done largely by division of labor and the sharing of results. The reader, however, cannot inform himself of the processes employed better than by a following of the descriptions given in standard texts on the subject, and especially in the Nautical Almanac. What are known as the Besselian elements are used; in other words, the problem is stated geometrically, these elements of the problem are supplied by the Almanac, and calculations are then made. The formulas employed, respectively, for the items enumerated above will illustrate the method, though they will not be fully intelligible without reference to page 728 of the Almanac. They are: