Alkoxides of uranium (IV)

J. Inorg. NucL Chem., 1962, Vol. 24, pp. 863 to 867. PergamonPress Ltd. Printed in England

ALKOXIDES OF URANIUM (IV) D. C. BRADLEY*,R. N. KAPOOR'~and B. C. SM1TH Department of Chemistry, Birkbeck College (University of London), Malet Street, London, W.C~I (Received 28 November 1961)

Abstract--Lower alcohols form addition compounds with uranium tetrachloride, and uranium tetraalkoxides are prepared by the action of alkali metal alkoxides on uranium tetrachloride. The preparation of dichlorobis(diethylmethylsilyloxy)uranium(IV)is reported. The uranium tetra-alkoxides, which are oxidized extremely readily, have physicochemical properties which are typical of polymeric metal alkoxides. ANHYDROUS uranium tetrachloride reacts vigorously with methyl, ethyl, and isopropyl alcohols to give alcoholates of formula UCI4,4ROH. These compounds are nonvolatile green solids which are oxidized readily in air. They may be recrystallized from the parent alcohols, but they are insoluble in benzene and other non-polar organic solvents. Similar addition compounds of thorium tetrachloride have been reported previouslyO) and in these and other 1:4 addition compounds of uranium (IV) and thorium (IV) which are listed by COMYNS(2)it is likely that the metals exhibit the co-ordination number eight. The butyl alcohols also react with uranium tetrachloride but replacement of chloride increases in the order Bu n < Bu s < Bu t. This type of solvolysis occurs also with titanium and zirconium tetrachlorides, and the possible mechanism has been discussed by BRADLEY eta/. (1) The reaction of sodium ethoxide and sodium i-propoxide with uranium tetrachloride was investigated by ALBERS e t a / . <3) and the preparation of uranium tetramethoxide, tetraethoxide, and tetra-t-butoxide has been reported by GILMAN and hisco-workers.(4) The preparation of uranium tetra-n-propoxide and tetra-i-propoxide from uranium tetrachloride and sodium alkoxides in dimethylcellosolve is now described, and the preparation of uranium methoxide and uranium ethoxide has been repeated using reactions involving uranium tetrachloride and the appropriate lithium alkoxide. All attempts to prepare uranium tetra-t-butoxide were unsuccessful. GILMAN and his co-workers(4) reported that the compound was soluble in petroleum * Present address : Department of Chemistry, University of Western Ontario, London, Ontario, Canada. t Present address: Department of Chemistry, Gorakhpur University, Uttar Pradesh, India. (1) D. C. BRADLEY, M. A. SAADand W. WARDLAW, J. Chem. Soc. 2002 (1959). (2) A. E. COMYNS, Chem. Revs. 60, 115 0960). (3) H. ALBERS, M. DEUTSCH, W. KRASTINAT and H. YON OSTEN, Bet. Dtsch. Chem. Ges. 85, 267 (1952). (4) R. G. JONES, C. KARMAS, G. A. MARTIN a n d H. GILMAN, J. Amer. Chem. Soc. 78, 4285

0956). 863

864

D . C . BRADLEY,R. N. K~aPOORand B. C. SMITH

ether and comparatively stable to oxidation, but sublimation or oxidation in solution(S) gave a red crystalline product to which the formula UO2(OBu*)2, 4ButOH was assigned. Alkoxides of uranium(IV) are oxidized very readily, and the action of ammonia on uranium tetrachloride--alcohol addition compounds gave derivatives of uranium(V). During attempted preparations of uranium tetra-t-butoxide in the absence of air, oxidation always occurred. A grey-brown quinquevalent solid was formed from which a brown sublimate of uranium penta-t-butoxide could be obtained. Oxidation and evaporation of a petroleum ether solution of the original quinquevalent compound gave red crystals of uranium hexa-t-butoxide whose infrared spectrum showed no peaks corresponding to hydroxyl groups. It seems likely that the red compound reported by GILMAN(S)was uranium hexa-t-butoxide, which has also been prepared(e) by disproportionation of UO(OBut)4, ButOH during the attempted preparation of uranyl-t-butoxide, UO2(OBut)2. The ease of oxidation of uranium(IV) alkoxides compared with the relative stabilty of hydrated U 4+ ions is worthy of comment. In view of the precautions taken to exclude oxygen we are tempted to suggest that uranium(IV) alkoxides actually reduce alcohols with liberation of hydrogen by analogy with the reaction between an alkali metal and alcohol. It is significant that uranium(IV) alkoxides obtained in this research were all insoluble compounds and it may well be that insolubility lowers the reactivity of these compounds. Uranium is not alone in showing this behaviour since vanadium(iII) alkoxides(~) and cerium(III) alkoxides(8) are also extremely readily oxidized to quadrivalent compounds under comparable conditions, and niobium(IV) alkoxides ~9) are similarly oxidized to quinquevalent niobium alkoxides. Attempts to prepare uranium tetrakistrialkylsilyloxides were unsuccessful. Reaction of uranium tetrachloride with the lithium salt of diethylmethylsilanol gave UCI2(OSiEt2Me)2, a distillable green solid which oxidized on exposure to air. Uranium tetra-alkoxides are less volatile than corresponding titanium or zirconium alkoxides but more volatile than cerium and thorium alkoxides (see Table 1)~ Uranium tetramethoxide did not sublime even on heating to 300°C/ca. 10=4 mm and uranium tetraethoxide was regarded previously as non-volatile.(lo) Volatility in metal alkoxides is determined largely by the radius of the metal atom and the size and shape of the alkyl groups and from the atomic radii of these elements (see Table) it would be expected that uranium(IV) alkoxides would be very similar in volatility to zirconium alkoxides and more volatile than cerium(IV) or thorium alkoxides. However, it has been pointed out previously(H) that the relative volatilities of cerium(IV) and thorium tertiary alkoxides are less consistent with atomic radii than with the radii of tetrapositive ions in octahedral co-ordination. (5) R. G. JoNes, E. BXNDSCHADLER,G. A. MARTIN,J. R. TmRTLE and H. GILMAN,J. Amer. Chem. Soc. 79, 4921 (1957). (e) D. C. BRADLEY,A. K. CHAYrERJ-EEand A. K. CHATTER~EE,Proc. Chem. Soc. 260 (1957); J. Inorg. Nucl. Chem. 12, 71 (1959). (7) D. C. BRADLEYand M. L. MEHTA. Unpublished results (1960). Is) D. C. BRADLEYand A. K. CHATTERJ~. Unpublished results. (9) D. C. BRADLEYand I. M. THOMAS. Unpublished results. Ii0) R. G. JONES,W. BINDSCHADLER,G. KARMAS,F. A. YO~MANand H. GILMAN,J. Amer. Chem. Soc. 78, 4287 (1956), (11) D. C. BRADLEY,A. K. CHA'rrel~ee and W. WARDLAW,J. Chem. Soc. 2600 (1957).

0.64 0"87 1"05 1-02 1.10

1"32 1 "45 1-42 1-65 1-65

Ti Zr U Ce Th > 300/0.05fl4)

> 200/vac.~ls)

I03/0-I¢13) 180/0.1"~13) 220/0-01"

M(OEt)4

* Sublimation temperature.

170/O-l*~ts~ 280/10-s*~17~ > 300/ca.10 - 4 > 200/vac.~ts~ > 300]0.05~14)

M(OMe)4

> 200/vac.

124/0.U13) 208/0.1 ~13) 240/0.01"

M(OPrn)4

C. BRADLEY, R. C. MEHROTRA and W. WARDLAW,J. Chem. Soc. 5020 (1952). C. BRADLey, R. C. MmmOTRA, J. D. SWAW,VICKand W. WARDLAW, J. Chem. Soc. 2025 (1953). C. BRADLey, M. A. SAAD and W. WARDLAW, J. Chem. Soc. 1091 (1954). C. BRADLEY, A. K. CnAT~.ltJL~ and W. WARDLAW, J. Chem. Soc. 2260 (1956). C. BRADLL~V,A. K. C t ~ T r e n J ~ and W. WARDLAW, J. Chem. Soc. 3469 (1956). C. BRADLEYand M. M. FAKTOR, Nature, Land. 184, 55 (1959). C. HANCOCK. Personal communication. MOELU~R, Inorganic Chemistry, p. 135. J. Wiley, London (1952). MOELLER, Inorganic Chemistry, p. 141. J. Wiley, London (1952).

(A)

(A)

M

t12) D. tt3) D. ~t4) D. cts~ D. ¢16~D. ttT) D. ClS~D. ~tg) T. ~20) T.

M 4+ radius~20)

Atomic radiu#19)

TABLE I . - - V O L A T I L I T I E S ( o C / M M ) OF TETRA-ALKOXIDES

49/0.1(12) 160/0.1(12) 160/0"01" 160 -- 170/0"05 *~16) 200 --210/0-05-0.1 *(16)

M(OPrt)4

oo Os Lit

<

o

O m.

>

866

D.C. BRADLEY, R. N. KAPOOR and B. C. SMrr~

Consideration of the M 4+ radii (see Table 1) would suggest that uranium compounds should be less volatile than cerium compounds and more volatile than thorium compounds. In view of the uncertainty in the values of atomic or ionic radii and of the lack of precision in the volatility data it would be pointless to attempt a more detailed interpretation of these properties, but it appears that ionic radii afford a more reasonable correlation with volatilities than "covalent" radii. Although we could not determine the molecular weights of the uranium(IV) alkoxides, there can be little doubt from their volatilities that they are polymeric and involve uranium in a co-ordination number of six or higher. In particular, the striking increase in volatility caused by chain branching in the propoxides is characteristic of the behaviour of polymeric metal alkoxides.

EXPERIMENTAL Reactions were carried out in the absence of oxygen, and products for analysis were sealed in evacuated tubes. Preparation of alcoholates Anhydrous uranium tetrachloride reacted vigorously with methyl, ethyl, and i-propyl alcohols to give the following green solid addition compounds: uranium tetrachloridetetrakismethanol (Found: U, 46.7; C1, 27.6; MeO, 22.6. Calc. for C4H16C1404U: U, 46-8; CI, 27.9; OMe, 24.4%); uranium tetrachloride-tetrakisethanol (Found: U, 41.7; C1, 24.8; EtO, 31.7. Calc. for CsH24CI404U: U, 42.1; C1, 25.1; OEt, 31.9%); and uranium tetrachloride-tetrakisisopropanol(Found: U, 38.5; CI, 21.8; PriO, 38.2. Calc. for C12H32CI404U: U, 38.4; CI, 22.8; PrIO, 38.1%). Similar reactions with n-, s-, and t- butyl alcohols gave green solids in which C1 : U ratios were respectively 2.9, 2.5 and 0.4. Reactions of alcohOlates Ammonia was bubbled through a solution of uranium tetrachloride-tetrakisethanol in benzene-ethanol. Ammonium chloride was removed by filtration, and analysis of the brown tarry residue obtained on evaporation corresponded to triethoxyoxouranium(V)ethanol, UO(OEt)3,EtOH. (Found: U, 54.4; OEt, 39.9; valency, 5.2. Calc. for CsH21OsU: U, 54.7; OEt, 41-6%.) Distillation at 160°-180°C/0.1 mm gave uranium pentaethoxide" 0, 21) (Found: U, 51.7; OEt, 46.8; valency, 5-0. Calc. for C10H2sOsU: U, 51.4; OEt, 48'6%.) A similar reaction of ammonia with uranium tetrachloride-tetrakisisopropanol gave a light brown solid which recrystallized from i-propanol to form golden crystals of uranium penta-i-propoxide.t21, 22) (Found: U, 44.7; OPal, 55-2; valency, 5.0. Calc. for C15H35OsU: U, 44.6; OPd, 55.4%.) Preparation of alkoxides Uranium tetramethoxid¢ was prepared by filtration of the green precipitate formed on mixing methanolic solutions of lithium methoxide and uranium tetrachloride-tetrakismethanol. (Found: U, 65-2; OMe, 38.8. Calc. for C4H1204U: U, 65.7; OMe, 34.2%.) Ethanolic solutions of lithium ethoxide and uranium tetrachloride-tetrakisethanol formed a green precipitate of uranium tetraethoxide, which sublimed at 220° C/0.01 ram. (Found: U, 57.0; OEt, 42.8; valency, 4.0. Calc. for CsH2004U: U, 56.9; OEt, 43.0%.) Lithium n-propoxide in dimethylcellosolve reacted vigorously with uranium tetrachloride to form a green precipitate. The solvent was removed and uranium tetra-n-propoxidb sublimed at 240° C/0.1 ram. (Found: U, 50-6; valency, 4-1. Calc. for Ct2HzaO4U: U, 50.2%.) Lithiumi-propoxide and uranium tetrachloride reacted similarly to form uranium tetra-i-propoxide which sublimed at 160° C/0-01 ram. (Found: U, 50.5; OPt I, 49.2; valency, 4.0. Calc. for CIzHzsO4U: U, 50"2; OPr ~, 49.8%.)

Alkoxides of uranium(IV)

867

Attempted preparation of uranium tetra-t-butoxide (a) Uranium tetramethoxide was heated under reflux with benzene-t-butanol, and methanol was removed slowly as an azeotrope with benzene. The solution became red and a brown solid formed. Impure uranium penta-t-butoxide121. 221 was isolated from the reaction mixture. (b) Anhydrous uranium tetrachloride and t-butanol were added to a solution of potassium amide in liquid ammonia. Evaporation of ammonia and extraction of the residue with dry petroleum ether gave grey-brown crystals of a quinquevalent uranium compound. (Found: U, 45-8; valency, 5-0. Calc. for UO(OBut)3,ButOH: U, 43.5 ~.) A brown sublimate of uranium penta-t-butoxide was obtained at 130°C/0.1 mm. (Found: U, 40.2; valency, 5.0. Calc. for C20H45OsU: U, 39-5 ~.) The grey-brown colour of a solution of UO(OButh,ButOH in petroleum ether changed to red on standing in air for three days. Evaporation gave red crystals of uranium hexa-t-butoxide which sublimed at 100°C/0.1 mm. (Found: U, 35.6. Calc. for C24H5406U" U, 35"2~.) Silyloxide reaction Uranium tetrachloride reacted with lithium diethylmethylsilyloxide in dimethyl cellosolve-benzene to form a green precipitate. Distillation yielded dichlorobis(diethylmethylsilyloxy)uranium(IV), b.p. 260°C/0"01 mm. (Found: U, 43.0; CI, 13.05; Si, 10.4; valency 4-0. Calc. for C10H26C1202Si2U: U, 43.8; C1, 13.1; Si, 10.3 ~.) Materials Anhydrous uranium tetrachloride was prepared from uranium trioxide and hexachloropropene323) Organic solvents, of "Analytical Reagent" grade when possible, were dried carefully by conventional methods before passing through columns packed with molecular-sieve. Analysis Chloride was determined gravimetricaUy as silver chloride. Uranium was oxidized and precipitated as the trioxide, and weighed as U3Oa after ignition at 800°C in a platinum crucible. The valency of uranium was established volumetrically.t21) Alkoxy groups were estimated by the method of BRADLEYet al.t TM Acknowledgement--This work was supported by the United Kingdom Atomic Energy Research Establishment, Harwell, and the authors thank Mr. F. HUDSWELLfor his interest and encouragement. 121) D. C. BRADLEY, B. N. CHAKRAVARTIand A. K. CHATTERJEE,J. Inorg. Nucl. Chem. 3, 367 (1957). (22) R. G. JONES,G. KARMAS,G. A. MARTIN,J. R. TmRTLE,F. A. YOEMANand H. GILMAN, J. Amer. Chem. Soc. 78, 4289 (1956). 1231j. A. HERMANNand J. F. SUTTEE, Inorg. Synth. 5, 143 (1957). 124~D. C. BRADLEY, F. M. ABD-EL HAUM and W. WARDLAW,J. Chem. Soc. 3450 (1950).