THE PRESENCE of paramagnetism in the uranium chalcogenides makes a study of their thermal properties seem of special interest for exploration of the mechanisms

Heat capacities have been measured and derived thermodynamic properties calculated for the structurally related USI.9 and USe2 phases from 5 to 350°K. Values of the heat capacity (Cp), entropy (S°-S~) and Gibbs energy function [ (G° -H~) IT] at 298.15°K in cal mole -t °K -t are 17.68, 25.91 and 13.82 for US~.9 and 18-92, 31.98 and 17.86 for USe2, respectively. Schottky-type anomalies are present in US1.9; a ~-type thermal anomaly with ASt ~ 0-19cal mole -* °K -~ occurs at 13.1 °K in USe2. THE PRESENCE of paramagnetism in the uranium chalcogenides makes a study of their thermal properties seem of special interest for exploration of the mechanisms by which spin order is approached. For US1.9 and USe2, both with tetragonal structures, magnetic susceptibility measurements[I-3] indicate the presence of 2 unpaired 5f electrons with comparatively weak atomic interactions between neighbors since the Curie-Weiss law is fulfilled at higher temperatures with a Weiss constant of about 50°K. A structural study [4] shows that some of the uranium atoms in the tetragonal US~_~ structure are distributed at random; this might cause further anomalies in the behavior of the heat capacity or in the zero point entropy. E X P E R I M E N T A L The preparation of the USt., and USe2 samples was carried out by reacting elements of previously described[l] purity in evacuated and sealed quartz tubes in the right proportions. In the case of the sulfide, the reaction was performed by heating the mixture to 970°C over a period of 3 weeks and keeping it at this temperature for 2 days. After cooling, the sample was then crushed and homogenized at 800~C for 1 week. After the heat capacity determinations were completed, chemical analyses were made on the USt.9 sample by K. Jensen of the Argonne National Laboratory through the cooperation *Supported in part by the U.S. Atomic Energy Commission. 1. F. Gr0nvoid, H. Haraldsen, T. Thurmann-Moe and T. Tufte, J. inorg, nucl. Chem. 30, 2117 (1968). 2. W. Trzebiatowski and W. Suski, Bull. Acad. poi. Sci. Sdr. Sci. chim. 9, 277 (1961 ). 3. V. L Cbechernikov, A. V. Pechennikov, E. I. Yarembash, L. Martynova and V. K. Slavyansldkh, Soviet Phys. JETP 26, 328 (1968). 4. R .C .L . Mooney Slater, Z. Kristallogr. 120, 278 (1964). 2169 2170 E . F . WESTRUM, Jr. and F. GR~JNVOLD of H. E. Ftotow. The results indicated 79.78+_0-10 per cent by weight of uranium, 20.15+0.07 per cent by weight of sulfur by a distillation process and 20-42-40-09 per cent by weight by gravimetry. (Theoretical: U = 79-62, S = 20.38). The reported results have been corrected for high recovery on standards. Hydrogen, oxygen, and nitrogen were 1"3 ---0-4 ppm, 470 + 30 ppm, and 25 -+6 ppm respectively accordhag to L. F. Krout of the same laboratory. Spectrochemical analys/s by J. Lech indicated in (ppm by weight, factor of two precision): A! 10, As < 20, Cr ~ 4, Fe 20, K < 100, Mg ~ 5, Mo < 20, P < 100, Si ~ 60, Ti < 200 and Zn < 50. Spark spectral analyses from the Chemical Engineec/ng Department of ANL indicated in (ppm atomic): Cr 24, Fe ~ 220, K ~ 21, Mn 20, P ~ 67 and Sc ~ 20. Mass spectrographic resuRs showed A! ~ 13, Mg ~ 33, Si ~ 1500 in the same units. The selenide was prepared in the same way, but the temperatttre of homogenization was 840°C. Analyses similarly performed at the Argonne National Laboratory on the diselenide sample indicated Uranium: 59-92.4.0-10 per cent by weight, 3 determinations (theoretical U = 60.12) Selenium: 39.79 -40-08 per cent by weight, 6 determinations (theoretical Se -39.88) Hydrogen: 5.6-4-0-3 ppm Oxygen: 780--20 ppm Nitrogen: 12--I ppm. By spectrochemical analysis (in ppm by weight, factor of 2 precision): Al 10, As < 20, B 2, Cr ~ 4, Ca 10, Fe ~ 30, K ~ 100, Mg 30, Mo < 20, P < 100, Si ~ 40, Ti < 200 and Zn < 50. By spark spectrometry (in ppm atomic): Cr = 12, Fe = 220, Mn = 20, P = 67, Sc = 7. By mass spectrometry (in ppm atomic): AI ~ 20, Mg 25, Na ~ 67, Si 1100. Measaren~nts were Lade in the Mark 111 vacuum cryostat[5] provided with an electronic adiabat/c-shield control system consisting of three separate channels of recording cirCuitry with proportional, rate, and reset action. These kept the temperature differences smaller than a millidegree and thereby reduced heat exchange with the calorimeter to a magnitude negligible in comparison to other sources of error. The gold-plated copper calorimeters used (laboratory designation W-29 for USI.9 and W-28 for USe2) had capacities of 42 and 92 cmL The heat capacities of the empty calorimeters were determined separately and small corrections were applied for differences in the amounts of helium gas, indium-tin solder, and ApiezowT grease for the loaded and empty calorimeters. The heat capacity of the USI.g sample 046-551 g), represented about 92 per cent of the total below 10°K and decreased to 66 per cent above 90°K. For the USe~ sample (196-420g), the corresponding values are 92 per cent and 60 per cent, respectively. All measurements of mass, temperature, resistance, voltage, and time were based upon calibrations or standardizations made by the U.S. National Bureau of Standards. R E S U L T S A N D D I S C U S S I O N Experimental heat capacity values for the USI.o and USes samples are presented in Table 1 in chronological order for the mean temperatures of the determinations after correction for curvature of the heat capacity. The data are given in terms of the defined thermochemical calorie, equal to 4-1840 J, and an ice point of 273.15°K. They are considered to have probable errors decreasing from about 3 per cent at 5°K to 0-5 per cent at 10°K and to less than 0.1 per cent above 20°K. For US1.9, the shape of the heat capacity curve (see the comparison with US~[6] in Fig. 1) is nearly normal, while for USes a k-shaped anomaly is present, having its maximum at 13. t °K (see Fig. 2). The smoothed heat capacities and the thermodynamic functions derived from them by means of a digital computer using a previously described program[7] 5. E. F. Westrum, Jr., J. chem. Educ. 39,443 (1962). 6. E. F. Westrum, Jr. and F. Grclnvold, J, inorg, nucl. Chem. 3e, 2127 (1968). 7. B. H. Justice, Doctoral Dissertation, University of Michigan (1961); U.S.4.E.C. Reg. TID-12722 (1961). Uranium ehalcogenidesIII Table 1. Heat capacities of US~.9 and USe* 2171