L.EYRING,H. R. LOHRAND B. B. CUNNINGHAM
1186
rorrrsponds to A Itieitti tiistatice o f atbottt X.T, A between neighboring ions. This value of nieari distance is not unreasonable if we recall that the crystallographic ionic radii of Na+, K + and C1are 0.95 1.33 and 1.81 A., respectively, and that the average IFngth of water molecules in pure water is about 3 A. When the concentration of salt solution is so high that the ions can no longer get enough water molecules from the system to become fully hydrated, one may expect the oppositely charged ions to share their water of hydration or even to form pairs by direct contact of bare ions. Under such circumstances, one may expect the activation energy for “tracer-diffusion” to be so high that the diffusion coefficients may be very much smaller than the corresponding values a t infinite dilution. However, since in these highly concentrated solutions the ions are no longer fully hydrated, they would diffuse with a smaller mass, and hence faster inherent velocity. It is not possible to decide which of these two effects dominates by making “tracer-diff usion” measurements in potassium chloride solution in this concentration range because of it low solubility. Preliminary measure-
Vol. 74
tiient shows that the “tracer-diffusion” coefficient of Na in 12 molal lithium chloride solution is
about one-fifth as large as that in infinitely dilute solution, indicating that the effect of increase of activation energy discussed above is the dominating one. However, more experimental work is desirable before definite conclusions can be drawn for “tracer-diffusion” in concentrated electrolytic solutions. Measurements of the diffusion of various tracer-ions in different salt solutions a t various concentrations and temperatures are in progress in this Laboratory. The results will be reported in subsequent articles of this series. Acknowledgment.-The author is indebted to Professor H. S. Harned whose constant encouragement and support made the present work possible, and to Professor J. G. Kirkwood who read and commented on the manuscript. He should thank Professor H. C. Thomas foi kindly providing facilities for handling radioactivity. He is also indebted to Professor J. W. Kennedy and Mr. A. hf. Friedman who first brought to his attention the problem of Al-correction. NEWHAVES,COSNECTICUT RECEIVEDSEPTEMBER 19. 1951
__I____
[CONTRIBUTION FROM THE DEPARTMENr O F CHEMISTRY AND
RADIATION LABORATORY OF THE
UNIVERSITY OF CALIFORNIA]
Heats of Reaction of Some Oxides of Americium and Praseodymium with Nitric Acid and an Estimate of the Potentials of the Am(II1)-Am(1V) and Pr(II1)-Pr(1V) Couples1 BY L. EYRING,~ H. R. LOHRAND B. B. CUNNINGHAM The heats of reaction at 25’ of Pr203(o.c form) with 6.02 M and 1.00 M nitric acid, of Prot(,) with 6.02 M nitric acid and 6.02 &I nitric acid-0.1 M fluoboric acid, and of Am02 with 6.02 M nitric acid-0.1 M fluoboric acid have been measured From these measurements and other thermodynamic data heats of formation of Pr +&), Pr(iYOs)a(.,,, Pr(K0s)1(8q. In HNO,.MH,O), Pr(n‘Oa)a(.q. ID HNO$’I.LH~O),PT(NO&(.q, in HNOs‘O.OllHBF~’7.LH1O), Pros(,), Am and AmOz(,)are calculated. From these data and estimated entropy values the potentialsof the couples: l/zHz@), and Pr+3(.,) f H+(,) = H+(.q) = Pr+4(.,) * / r r H a , gare ) estimated to be - 2 4 volts and -2.9 volts, respectively, with an uncertainty of about +0.2 volt in each case.
+
+
It has been shown previously3 that the potential of the Am(II1)-Am(1V) couple in acid solution is more negative than -2.0 volts. The work described in this paper was undertaken to evaluate the magnitude of the potential more precisely. The chemistry of americium is of special interest in that it is the first of the transuranium elements in which the stability of the tripositive state is comparable to that observed for some of the lanthanide elements. I n the course of development of this work it was necessary to use a substitute material as a “standin” in perfecting the techniques which were intended ultimately to be applied to americium, which is available in very limited quantities. Praseodymium was chosen for this purpose because (1) Presented a t the 118th Meeting of the American Chemical Society a t Chicago, Illinois, September, 1950 (2) S o w a t the Department of Chemistry and Chemical Engineering, State University of Iowa, Iowa City, Iowa. Part of the data reported here was included in a dissertation submitted by L. Eyring to the Graduate Division of the University of California in partial fulfillment of the requirements for the degree of Doctor of Philosophy. (3) Cunningham National Xuclear Energy Series, Plutonium Project Record, Vol. 14B, “The Transuranium Elements: Research Papers.” Paper KO 19 2 (McGraw-Hill Book Co , Inc., New York, U
Y 1949).
+
of a convenient similarity in the chemical properties of its dioxide to that of americium. The work reported here thus permits an approximate evaluation of the potential of the (111)-(IV) couple of praseodymium as well as that of americium. Since i t was known that the potential of the americium couple was so negative as to make the tetrapositive state unstable in acidic aqueous solution, it did not aDpear feasible to measure the potential in the conventional manner by incorporating the couple in a reversible chemical cell. We chose, therefore, to evaluate the free energy of this reaction by the less accurate method of evaluating its heat and estimating the entropy change. Unless otherwise noted, our AH and A S values refer to a temperature of 298’ K. Results are expressed in kcal./mole for the reaction as written. Accepted values for entropies, heats of formation, etc., are those given in “Selected Values of Chemical Thermodynamic Properties” (hereafter abbreviated SVCTP), issued by the National Bureau of Standards. Our results are not corrected to unit activities, since the activity coefficients of the f3 and +4
March 5, 1952
REACTION OF
OXIDES.OFAMERICIUM AND PRASEODYMIUM WITH
ions in our solutions are not known. Such corrections probably are negligible compared with our experimental errors.
Experimental Method Preparation of Compounds. A. PrzOs(c).-The praseodymium used in the experiments described herein was obtained as “Pr~OI1”from Johnson, Matthey and Co., Ltd., of London. This material was found to contain about 4% of sodium and potassium, and one-half of one per cent. of other rare earths, principally neodymium. The relatively pure oxide was further purified by D. C. Stewart and R. C. Lilly of this Laboratory by a cation-exchange column separation procedure using Dowex-50 resin. The praseodymium from the column runs was precipitated as oxalate from 0.1 M HC1-0.25 M NH1C1-0.25 M citric acid solution by the addition of oxalic acid t o give a concentration of 0.1 M. The praseodymium oxalate in a platinum container wa: ignited in air at 650” t o produce the black oxide, ‘‘PrsOll. This oxide constituted the stock material from which the succeeding praseodymium calorimeter samples were prepared. No impurities were detected by spectrographic analysis of a 50-pg. sample of this material. Elements analvzed for. and their limits of detection in micrograms, were as follows: Al, 0.01; Ba, 0.1; Be, 0.005; Ca, 0.1; Ce, 0.1; Dy, 0.1; Er, 0.1; Eu, 0.01; Fe, 0.05; Gd, 0.1; Ho, 0.1; K, 0.1; La, 0.01; Lu, 0.01; Na, 0.1; Nd, 0.05; Sm, 0.1; Sr, 0.01; Ta, 0.5; Tb, 0.1; Yb, 0.01; Y. 0.01. A sample of oxide in a platin’um container was placed in a vacuum system and pumped down t o remove adsorbed moisture and gases. Pure hydrogen ( l / a atm. pressure) was then admitted from a tube of uranium hydride maintained a t 360’. The oxide charge was heated t o 500’ and maintained a t that temperature throughout the reduction. After two or three minutes the black oxide began t o change t o yellowgreen PrzOa. From time to time the sample container was removed from the reduction apparatus, capped and weighed on an Ainsworth FDJ microbalance. No further decrease in weight was observed after one hour of heating with hydrogen. Therefore, an apparently adequate time of one and one-half hours was adopted for the reduction t o PriOa for calorimetric measurements. X-Ray diffraction results on this material shpwed it to be the cubic “C” form with a = 11.14 =k 0.01 A. When the temperature of this product was increased to 1000“ in vacuum it changed over t o the pale green hexagonat “A” form with a = 3.859 i 0.003 A., c = 6.008 i 0.003 A. If the PrsO11 was reduced a t lOOO”, the “A” form was produced directly. Praseodymium sesquioxide (“C” form) prepared as described above was removed immediately into the dry atmosphere of a nitrogenfilled “dry box.” There it was quickly loaded into weighed sample bulbs and sealed off with Apiezon “W” wax ready for reweighing and calorimetric runs. The weighings were carried out on an Ainsworth F D J Microbalance. B. PrOt( (,,).-Praseodymium dioxide was prepared from the sesquioxide by heating the latter in a quartz bomb in a high pressure of oxygen gas. The bomb consisted of a thick-walled quartz tube of about 8 mm. 0.d. and 2-3 mm. i.d. with a rounded bottom and a constricted portion a t the other end. The constriction ensured a thick wall when the end was sealed. After the bomb was loaded with PrzOa the open end was attached to a system which was alternately evacuated and flushed with oxygen gas previously purified by passage through a copper coil immersed in an alcohol-liquid nitrogen mixture. Finally, with oxygen in the system, the end of the bomb was dipped in liquid nitrogen and oxygen was liquefied in an amount previously calculated t o produce 100 atmospheres pressure in the tube at 500”. At this point the bomb was sealed off with a gas-oxygen flame. The quartz bomb was then placed in a stainless steel jacket which served as protection in case of an explosion. The complete assembly was put into a muffle furnace and heated a t 500’ for 8-12 hours. The resulting product was a reddish-black monophasic substancp having a fluorite type structure, with a = 5.395 5 0.005 A. The oxide was transferred to a small phosphorus pentoxide desiccator inside a “drv box” readv for loadinn into wekhed calorimeter bulbs. B d b s and samples wereweighed on a
NITRICACID
1187
quartz fiber balance similat to that described by Kirk, Craig, Gullberg and Boyer.‘ C. AmzOI and Amon.-The americium used in these experiments was obtained from P--decay of Pun41produced by successive (n,y ) reactions on P U ~irradiated ~* with pile neutrons.6 The separation, concentration and purification of americium from a source of this kind has been described elsewhere.’ The results of a spectrographic analysis of our americium stock solution are given in Table 1. The symbol “
