Thermochemical properties of lithium

2144

T. B. DOUGLAS, F. EPSTEIN, L. J.
be approximated using the equation Pe P,= 1 cf,/cu), derived from the steady-state condition, where Pe is the equilibrium pressure, P, the observed steady state pressure, and f the ratio of the area of the orifice to the area of the sample. Thc latter has been taken as the cross-sectional area of the effusion cell, assumed t o be the effective areainvolved in exchange of molecules between vapor and solid. For cell 2, f = 0.0074 and Pe P, = 2 (average value); hence LY = 0.007. In view of the assumption concerning the sample area and the rather large scatter of our results from cell 2, we regard this as only a rough estimate of the accom modation coefficient. However, this value does predict that essentially equilibrium pressures should be observedwith cell 1, where f = 5.73 X giving PeIPs= 1.08, a deviation within the limit of our experimental error. In this comparison, allowance has not been made for a possible variation of alpha with temperature. We are unable to explain the very low values observed by Ditmars and Johnston. The line representing their least squares equation is shown on Fig. 2 (marked ref. 3 ) . Even though theaccommodation coefficient appears rather small, our estimate is much larger than would be required to give values of Pe'Ps of the order of 100. The orifice diameter of their cell was intermediate between the two used in this work. This research was supported in part by the Ofice of Ordnance Research, U. S..irmy.

flected in the free energy, is well within experimen tal error. Johnston and Bauer's values'were derived from equation7 2

+

which they developed using the data of J. Johnstons obtained from equilibrium studies a t higher temperatures. Pressures calculated from this equation are shown as the dotted line in Fig. 2. The close approach of the two curves (and the 23' values) shows general agreement of our results with those of J. Johnston, considering that the latter's measurementswere made above the melting point of LiOH. Pressures obtained using effusion cell 2 are observed to fall materially below those calculated from equation I , indicating that true equilibrium was not established. If one attributes the deviation to the accommodation coefficient, alpha may
(fi)

NOTE D D E D A

IN

PRooF.-Recently

a paper has appeared by C.

11. Shomate and

4. Cohen (THISJ O U R K A L , 77, 28; (195.5)) reportJ.

ing high temperature heat capacities for I.iO and 1,iOH. Using their results with similar d a t a for water vapor ( D . D. R'agman, et ai., J . Rcrearch, ,\-all. Bur. S t d s . , 3 4 , 143 (19451) and t h e third law d a t a of Johnston and Bauer (ref. I ) , ASo is calculated t o he 30.30 e.u. a t 600°K. for t h e reaction PLiOH(s) = I,i?O(s) HiOigI.LYith our equilibrium constant a t 60OoK;.,this entropy change leads to a calculated heat of reaction of 31.22 kcal., 0.42 kcal. larger than t h e value taken above (at 5 8 7 ° K . ) . T h e deviation is easily within probable experimental error. ( 7 ) This expression has been corrected for a n error in t h e sign of the fourth term on t h e right in t h e reference cited. (8) J. Johnston, Z.physik.Chcm., 62, 339 (1908).

SEATTLE, ~VASHISGTOS
THE

[CONTRIBUTION FROM THE NATIONAL

BUREAU F STANDARDS AND O

KNOLLS ATOMIC POWER LABORATORY]

Lithium: Heat Content from 0 to 900°, Triple Point and Heat of Fusion, and Thermodynamic Properties of the Solid and Liquid'
BY THOMASDOUGLAS,'LEO F. EPSTEIN,' B.
JAMES L.

DEXTER* AND

m T I L L I A ~H. l

HOWLAND3

RECEIVED XOVEMBER1934 11,
Lithium was distilled at 650-700° i l l z'ucuo and sealed in stainless steel type 317. From chemical analysis the purity of the first sample was 99 98 atomic % and that of the second sample, obtained with the still in uucuo, was approximately 99 99 atomic %. T h e melting curves, which are consistent with these analyses, gave a triple point of 180.54". Using a Bunsen ice calorimeter and...