was added to water. A%ttemperatures of 200" and above, the decomposition of UO4.2H2Oleads to C 0 3without the formation of U20i. The rate of decomposition of U 0 4 . 2 H 2 0 to U,07 was found to give surprisingly good agreement with first-order kinetics, the rate of decomposition being proportional to the quantity of undecomposed peroxide. Table I shows the firstorder rate constant calculated a t 1:30", the agreement being as good as could be expected from the accuracy of the measurements. In a similar manner the rate constant was found to be 4.4 X hr.-I a t loo", 4.0 X hr.-I a t 120", and 1.3 x 10-l hr.-I a t l.iO". Xn -1rrhenius plot of these data gives a very good fit to a straight line, from which the activation energy can be calculated to be 3.5 kcal. mole. I t is questionable whether such good results would have been obtained if preparations of radically different crystal size had been used.
Discussion The results obtained in this investigation are in
striking disagreement with those reported by Duvalj and in the earlier work by Hiittig and von Schroeder.? The explanation may lie in the extremely long time required for reaction 1 to reach completion. In the thermogravimetric method used by Duval, sufficient time for reaction 1 to reach completion could not possibly have been allowed, and Hiittig and von Schroeder do not state the length of time they heated their samples. It niay also be significant that both Hiittig and von Schroeder and Duval started with moist U04, containing an amount of water in excess of that required by the formula C04.2H20,while in this investigation dried U04.2H20was employed. On the basis of our experiments it may be concluded that U 0 4 . 2 H L 0decomposes to LTL',O;in the temperature range between 90 and 1%j3and that the decomposition reaction does not give any evidence t o support the formulation of uranium peroxide as C03.H,0?.H,0. .LL-STIS
12,
TEXAS
...
[CONTRIBUTION FROM
THE
LINCOLN LABORATORY, XASSACHUSETTS IssrIrvx
OF
TECIIXO~A):;Y~
Hydrolysis of A"'BV Intermetallic Compounds1 BY J. -1.KAFALAS, H. C. GATOSAXD 31. J. BUTTOX RECEIVED APRIL 3 , 1957 AIIIBv intermetallic conipounds hydrolyze to a varying degree. Phosphides hydrolyze readily in acid media a t room temperature with formation of phosphine. . h e n i d e s and antimonides hydrolyze t o a lesser extent.
Intermetallic compounds of the type --Irl'B\' (InSb, GaXs, etc.) have become of considerable importance in recent years because of their semi-conducting properties.? During and after their preparation these compounds are generally brought into contact with aqueous media. They are usually prepared, for example, using an excess of constituent X ; A is subsequently leached by an acid which does not appreciably attack the compound itself. Furthermore, for certain applications the crystal surfaces are treated with acid or basic solutions. T h e purpose of this study was to determine whether the above intermetallic compounds hydrolyze in aqueous media to form highly toxic hydrides of the type BVHa(oiz.,PHtI,=\sH:~, SbH,;) according to the reaction .41"IBv f 3H20
--+ ;\"'(OH)X
f B'-H$
Preparation of Samples.-The representative intermetallic compounds used were I n P , Ga.\s and GaSb, containing radioactive P, As and S h , respectively. They were prepared as follows. 1nP.-Elementary P , containing P32,and In were placed in a quartz tube which was subsequentlv evacuated and sealed off. The mixture was then heated for one hour :it 1050°.3 ( I ) T h e research reported in this document was supported jointly b y t h e Army, N a v y and Air Foice under contract mith Massachusetts Institute of Technology. ( 2 ) I,, Pincherle and J. SI. Iiadcliffe, Adanptres i x P h y r . , 5 , 271 il!?jii). (3) Because the vapor prescure of free P is r e l y high a t this temperature one end of the quartz t u b e was kept a t a lowel temperature. I n this way explosions were prevented. Similar prrcaution- were taken in t h e preparation of arsenides.
T h e resulting sample was treated briefly with cold concentrated HNO:i in order t o remove the excess indium. .in S ray powder pattern confirmed the formation of I n P . T h e radioactive P3* was obtained from Ca3(P04):, containing P3*,by reduction with charcoal and S i 0 2 a t 1300'. Ga -4s .-Elementary arsenic containing radioactive Asy3 rogethcr with non-radioactive Ga.\s WV:ISplaced in it quartz tube which wits then evacuated and senlect off. The end o f thc tube containing the Gd;\s w a s heated for severdl hours above the melting point ( 1:'80°) of Ga.is. This treatment ensured hor~iogeneous distribution of radioactive arsenic throughout the Ga.1~. The radioactive arsenic was obtnineti frmn radioactive ;\sCI.j by reduction with H:$PO?and SnCL. Gash.-Radioactive Sb126 was lioniogeneously diqtributed in GaSb in the mariner described for GaAs (1n.p. of GaSb 725')). The radioactive antimony \Yas obtained from SbCl, HCI solution. 131- reduction with iron and SnC12 in Experimental Technique.-The above compounds were crushed t o coarse powders and were employed in this form. hleasurenients of the radioactive content of weighed fractions showed t h a t the radioactive species were distributed IiiJrnogenecJusly throughout the samples. The apparatus employed in this stud)- is shown in Fig. 1. The various coarse powders were placed rm a fritted disk F in the reaction chamber S. .I solution of known pH was added through the funnel P and a moderate stream of nitrogen was bubbled through the solution and through the adsorption traps TI and T:: which contained an oxidizing solution of iodine, K I and S a H C 0 3 . The stream of nitrogen was employed for the purpose of stirring the reaction mixture and for carrying any SbH7,-ISH$,PH3 or other volatilc hydrides formed into the adsorption traps a s rapidly as possible. T h e reaction chamber S was designed so t l u t no rarliivactivit>-was carried into the traps other than t h a t :issociatetl with the gaseous products. Several blank tests irere performed in which chamber S contained a radioactive solutiou, b u t not radioactive gases. In all cases, the nitre)gen stream carried no detectable radioactivity into the traps. Each sample of the intermetallic compounds was exposed t o t h e solution for 16 hours. ITpon completion of each r u n ,
:5-
liquid samples were withdrawn from the adsorption traps and counted bl- means of a cylindrical, well-type scintillation counter. X11 experiments were performed a t room temperature.
Results.-The results obtained are summarized in Table I. I t can be seen t h a t I n P reacts readily in acid solutions t o form volatile hydrides of phosphorus. This result is t o be expected in view TABLE I FORMATIOS O F VOLATILEHYDRIDESBY HYDROLYSIS OF A'"B' Intermetallic compd.
InP
GaAs
GaSb
a
426 1
HYDROLYSIS OF ATT'B INTERNETALLIC COMPOUNDS
Xug. PO, 19.57
INTERMETALLIC COMPOUSDS I N CONTACT AQUEOCSMEDIAFOR 16 HOURS Soh. used
N 2 in
JVITII
A m t . o f hydride recovered, expressed in y per cm.2 of AIIIBY
Hydride
10 NaOH S.S." PHa 1 N SaOH N.S. 0 . 1 -V' S a O H S.S. Dist. H20 S.S. 0.1SHC1 0.35 1 N HC1 45.0 10 "V HC1 Gross amounts (samples dissolve readily)
1 iL"a0H 0.1 V . KaOH 0 . 0 1 N KaOH Dist. HzO 0.1 NHC1 1 . 2 N HC1 12.0 NHC1
N.S.
AsHs
N.S. K.S. S.S. 0.35 0.35 0.35
Dist. H 2 0
S.S.
0.06 HC1 1 . 8 N HC1 6.1 NHC1 12 N HCl
0.01 0.01
SbH,
N.S. N.S.
Xot significant,
