No abstract

No abstract

A method is described for the measurement of the active oxygen present in a mixture of ignited rare. earth oxides. The procedure is superior to the classical Bunsen method in that no difficulty is encountered in the analysis of high ceria mixtures. Data are given for the analysis of pure cerium dioxide, of some cerium dioxidelanthanum sesquioxide mixtures, and of pure praseodymium oxide.Pi GESERAL, gravimetric methods of analysis are based upon I a separation followed by a final weighing of the desired constituent in a pure compound of known composition. However, in the analysis of rare earth materials, separation of the individual earths is impractical and one is faced with the necessity of weighing a mixture which may contain elements of the entire group. Certainly the oxalates of cerium, praseodymium, and terbium yield upon ignition in air oxides containing more oxygen than can be accounted for by sesquioxide formation ( I ) , and there is some indication of a higher oxide of neodymium (6). The situation is furt,lier complicated by the possible influence of certain other rare earths upon the higher oxides formed by the elements mentioned (2,S, 5 ) .The evaluat,ion of the true rare earth metal content of a mixture of such oxides would. be greatly simplified by a method n-hich permitted direct measurement of the so-called "active oxygen", the oxygen present in excess of that in the sesquioxide. Correction of the weight of the oxide mixture would then permit calculation of an accurate metal content.Such an analytical method would be useful also in obtaining data regarding the composition of rare earth oxides formed under various conditions, both in the pure state and in mixtures. Such studies have been made in this laboratory using the method described below and will form the subject. of a later paper.The method of Bunsen (4) has found wide application for the determination of higher-valent cations in compounds similar to the rare earth metal oxides. Hydrochloric acid is added to the compound under investigation and the chlorine liberated by the reaction is absorbed in a potassiumiodide solution. The liberated iodine is then titrated with thiosulfate solution in the usual manner.Such a procedure is not feasible, however, if an appreciable amount of cerium is present in the mixture under investigation. Cerium dioxide, once formed, is extremely inert to all but the strongest reducing agents and even extended contact with concent,rated hydrochloric acid results in very little attack. EXPERIMENTALSeveral modifications of the Bunsen method were attempted, using both arsenious oxide and sodium oxalate as reducing agents. The reaction rate between the oxides proved to be too slow to be useful, while the conditions necessary to complete the reaction with sodium oxalate were so vigorous that appreciable decomposition of the oxalate in the sulfuric acid medium occurred.A simple procedure using potassium iodide and a modified Bunsen apparatus was finally developed which gave satisfactory results.The apparatus consisted of a...

As a preliminary to a study of certain regions of the four-component system Na20-Si02-Al20g-H20 at 25°C. some confirmatory work on the three-component systems Na20-Al20g-H20 and Naü0-Si02-H20 seemed advisable. The system Na20-Al20g-H20 has been investigated by Goudriaan (3) and by Fricke and Jucaitis (2) at 30°C. The present work on this system was carried only to the point necessary to confirm the results of the latter investigators, allowing for the 5°difference in temperature.The system Na20-Si02-H20 has been investigated by Harman (4). The results presented at this time differ in several important respects from those reported by that investigator. MATERIALS Three materials were used for the preparation of the solutions: AlüOg'3 H20, Na20-Si02-9 H20, and a solution (ca. 50 per cent) of sodium hydroxide.The alumina trihydrate was prepared by slow hydrolysis of c.p. sodium alumínate solutions, filtered off, rapidly washed with cold water on a suction filter, and dried in a steam chest. One such preparation gave 35.3 per cent loss on ignition (theoretical, 34.65 per cent H20) and contained less than 0.2 per cent sodium oxide.The hydrated sodium metasilicate was prepared by water recrystallization of the c.p. salt, was filtered with exclusion of carbon dioxide under a ' This article is part of a thesis presented by J. W. Sprauer to the Graduate School of Purdue University in partial fulfillment of the requirements for the degree of Master of Science.

Recently, however, certain earths have been found to possess very well developed "anomalous" valences. A stimulus, supplied by such important discoveries, appears to have been given to the work of prediction, verifica-121 CHEMICAL REVIEWS, VOL. 16, NO. 2 pressed the belief that lanthanum oxide does not absorb oxygen unless cerium is present. Marc (67) and Meyer and Koss (78) also commented on the catalytic effect of cerium on the absorption of oxygen by lanthanum oxides. On the other hand, early work by Cleve ( 22)-probably with impure materials-tended to show that La203 absorbed oxygen at red heat.Higher sulfides have been prepared by Biltz (5), and by Klemm, Meisel, and von Vogel (55). The first investigator prepared LaS2 by heating the sulfate in a stream of hydrogen sulfide. At temperatures above 600°C. the compound breaks down, giving La2S3 and S. It yields hydrogen persulfide with acids and apparently has the structure La2S3 S and is a "polysulfide." These results would seem to be confirmed by the magnetic susceptibility measurements of Klemm, Meisel, and von Vogel, whose data show that no increase in valence occurred during the reaction.The work of Andrieux (1) has shown the existence of borides of lanthanum and other earths having the well-defined formulas MBs. These were prepared by the electrolysis of fused borate and fluoride mixtures in graphite crucibles. These compounds, like the various aluminides, stan-