J. I?. STARIPFER, JR., C. E. HOLLEY, JR.,
3504
nal ingredients tends t o fix the arsenic to some extent. Hence, the mixture, when introduced into the firing chamber, contains magnesium arsenates of various compositions (depending upon initial ratios added) plus magnesium carbonate and stannic oxide. The new compound may be considered as having the composition 2(2Mg0.SnOz).(3Mg0. As206). When these preformed compounds are
AND
J. F. SUTTLE
VOl. 82
added in the proper ratio and fired, a good pattern of the compound is obtained, although with slight traces of impurity because of some loss of Asz05. Once formed, however, the compound is as stable as 6MgO.As205; that is, the pattern line intensities show no change after firing for 4 hr. a t 1200". Acknowledgments.-The writer acknowledges the assistance of hlr. D. D. Shaffer and Miss Joann Schmidt in preparation and X-ray analysis.
[COXTRIBUTION FROM T H E UNIVERSITP OF CALIFORNIA, L O S .%LAMOS SCIENTIFIC LABORATORY, L O S ALAMOS,N E W bIEXICO, OF CHEMISTRY, UNIVERSITY OF N E W MEXICO, ALBUQUERQCE, NEW M E X I C O ] AXD T H E DEPARTMENT
The Magnesium-Hydrogen BY J. F. STAMPFER, JK., C. E. HOLLEY, JR., AND J. F. SCTTLE RECEIVED NOVEMBER 6 , 1959 The decomposition pressure of magnesium hydride was measured under conditions such that the precision and accuracy of the temperature and pressure measurements could be estimated and the composition of the solid phases determined. T h e diffusion of hydrogen through the walls of the apparatus and the solubility of hydrogen in the walls was considered. The apparatus and the method of making the measurements are described. The decomposition pressures were converted P t o fugacities by use of the Beattie-Bridgeman equation of state for hydrogen and fitted t o straight lines to give R In ~ H = (-17,785 f 7 6 ) / T 32.28 f 0.45 and R In ~ D =P (-17,480 i 110)/T 32.48 i 0.17. hfagnrsium hydride, MgH2. was found t o be a stoichiometric compound within the limits of error of the measurements The solubility of hydrogen in magnesium was found t o increase from 2 atom yoat 440' t o 10 atom 70 a t 560". From these measurements thermodynamic functions fcr the formation of magnesium hydride and magnesium deuteride were estimated.
+
+
sure tubing and pressure valves and fittings were used throughout the measuring system. Pressures were measured by means of Heise gauges, either 8 or 16 in. in diameter, graduated up to 5000 psi. with 5 lb. graduation marks. The 16 in. gauge was calibrated with a dead-weight tester and the 8 in. gauge was calibrated by comparison with the larger gauge. The principal uncertainty in the pressure readings was the 5 psi. uncertainty in the reproducibility of the gauge readings themselves. The temperatures of the reactors were maintained by means of a surrounding furnace which could be held constant within 1' at any temperature from rooni temperature up to GOO". With the smaller reactor a 5,18 in. thick copper liner was used in the furnace to even out temperature fluctuation. With the larger reactor there was no room for the copper liner, but the larger mass of stainless steel effectively served the same purpose. Control of the furnace was by means of Experimental a Wheelco temperature controller operating from a chromelApparatus.-Since previous work had shown the necessity alumel thermocouple in the furnace, which controlled a portion of the heating current. of making measurements of pressures up to several hundred The volumes of the reactor and the other pertinent parts atmospheres and temperatures up to several hundred deof the measuring system were measured, using helium, by grees, the apparatus was designed for a maximum pressure of 350 a t m . and a maximum temperature of GOO". The reactor comparing with a known volume which had been Calibrated by Tveighing it with and without water in it. vessels were made from 347 stainless steel bar stock. T h e Materials.-The magnesium was supplied by the Dour one used for the pressure measurements was a cyliiider 3 ' / ~ Chemical Co., Midland, Mich., and was doubly distilled in. long X 2 in. 0.d. with a 23/4 X '/2 in. hole bored into it. metal which had a purity of 99.98Cjo, the chief impurities Diffusion from this reactor was too fast for accurate measurebeing iron, 100 p .p .m., copper 60 p .p .m. and zinc 50 p.p.m. ment of quantities of hydrogen present iii the gas phase at Turnings were prepared from a n ingot of this material immethe higher temperatures, so the compnsition measurements were made in a heavier walled reactor 7 in. long X 5 in. o .d. diately before they were loaded into the reactor; thus the oxygen content could be assumed to be low. No analysis with a 2 X in. hole. Since no mechanical closure was W L S made for non-metallic elements. The magnesium hyfound to be satisfactory through n temperature cl-cle, the dride was supplied by Metal Hydrides, Inc., Beverly, Mass. vessels were closed by melding t o pressure tubing which It contained 88.855 hlgHP, 7.5'3 Mg metal and 3.774 MgO. could then be attached to the measuring apparatus. PresThe MgO may have formed during handling and shipment since the material mas reported to be 93% MgH2. The Iiy(1) This work was done, in p a r t , under t h e auspices of t h e U. S. drogen was from the National Cylinder Gas Co., Denver, Atomic Energy Commission. Colorado, and was >99.94y0 H P ,
