COMMUNICATIONS TO THE EDITOR
Oct. 20, 1964
4497
Strong Hydrogen Bonds.
11. The Hydrogen Difluoride Ion
Sir : There has been much interest in the bond energy of the hydrogen difluoride ion. Current textbooks list values ranging from -27 to 58 kcal./mole. The first value is just slightly greater than the experimental heat of reaction of cesium fluoride with hydrogen fluoride, while the latter is due to a value corrected for lattice effects by LVaddington.' We have now determined an enthalpy of - 3 i kcal./mole for the reaction
+
( C H B ) I X F ( S ) H F ( g ) +(CHz)aNHF,(s)
This is the largest experimentally determined hydrogen bond energy reported to date. Further, lattice expansion should be slight for this reaction.* Thus the value of -37 kcal. /mole should be within 1 to 2 kcal. of the heat of the hydrogen bond in HF2-. 25
27
Fig 2 --Variation of decomposition pressure of tetramethylammonium hydrogen difluoride with temperature
30 -
pressure with temperature, for 90.5 to 122.0°, is shown in Fig. 2. The region of composition between 2 and 3 moles of H F to (CH3)4NFwas not studied owing to a failure of one of the gauges. Acknowledgment.-We are grateful to the National Science Foundation for support of this work.
20 -
-I"
2.6
I/T (OK)d03
El 10-
- 3
DEPARTMEKT OF CHEMISTRY SAMUEL ,4 HARRELL UNIVERSITY OF CINCINNATI DARLH MCDANIEL CINCINSATI, OHIO 45221 RECEIVED JULY31, 1964
0
2 Mdles HF/Me4NF
3
Fig. 1.-Pressure-composition isotherm for the system hydrogen fluoride-tetramethylammonium fluoride a t 90.5".
Tetramethylammonium fluoride was prepared under anhydrous conditions by the method of Tunder and Siege13 with slight modification. Anhydrous hydrogen fluoride was prepared by the thermal decomposition of potassium hydrogen difluoride which had been dried according to the method of K i l p a t r i ~ k . ~The interaction of hydrogen fluoride with tetramethylammonium fluoride was carried out in a vacuum line constructed of type K copper tubing with all joints silver alloy brazed. The sample tube and valves were constructed of monel. Pressure was measured by means of Wallace and Tiernan Model 145 gauges constructed of monel with actuating bellows of Ni-Span-C. The entire copper apparatus, except the gauges, was thermostated a t the desired temperatures. The temperatures used were high enough so that the hydrogen fluoride was present as a monomer. The pressure-composition isotherm for this system is shown in Fig. 1, and the variation of (1) T . C. \*'addington. J . Chem. S o c . , 1708 (1958). ( 2 ) See Fig. 7 of D. H. hIcDaniel a n d R . E. Vallee, Inovp. Chem , 2, 996 (1963). for t h e effect of cation size on t h e heat of reaction of hydrogen halides with t e t r a a l k y l a m m o n i u m halides. (3) R . T u n d e r a n d B . Siegel, J . I W O STw ~ r l .. Chem., 2 6 , 1097 (1963). (4) M E. R u n n e r . G . Balog, a n d lf.Kilpati-ick, J . A m . Chem. S o c . , 78, ,5183 ( 1 9 5 6 ) .
The Occurrence of the Re3Br9Group in Compounds Derived from Rhenium(II1) Bromide
Sir: It has been shown recently' that rhenium(II1) chloride is built up of Re3C19groups which contain a triangular Res metal atom cluster. This result had been anticipated from, and, in turn, provides an explanation for, the fact that numerous compounds prepared directly from rhenium(II1) chloride contain the ReaC19 group. '--j We now wish to report evidence that rhenium(lI1) bromide similarly gives rise to numerous compounds containing the Re3Br9 group which also contains the triangular Re3 metal atom cluster. Several new compounds,6 the spectra of their solutions, and spectra of solutions of rhenium(II1) bromide will be described in support of this statement. The presence of the Re3Brg group has been conclusively demonstrated in MzRe4Br15,where 11 represents quinolinium, pyridinium, or tetraethylammonium (1) F. A . C o t t o n and J . T. M a g u e , P r o 6 C h e m S o c . , 23Y (19641, I M W K . Cheni., 3, 1402 (1964) (2) F . A . C o t t o n . el a l . , Science, 130.5 (1964). (3) F . A . C o t t o n a n d J . T. M a g u e , I n o v g . C b e i n . , 3 , 1094 (1964) (4) J. A . B e r t r a n d , F . A . C o t t o n , a n d W .A I)ollase, ibid., 2 , 1166 (19fi3). B. H . Robinson, J E. Fergusson, and B. I< Penfold, P ~ o r C'hrrn . .So(., 116 (1963). ( 5 ) J . E Fergusson, B. K . Penfold, a n d W T Robinson, .YatiL!'r. 201, 181 (1964). (6) Elemental analyses have been carried nut o n all new c o m p w I n d ~ mentioned and agree satisfactorily with the proposed iormulas.
COMMUNICATIONS TO THE EDITOR
4498
since the two compounds are not isomorphous.? X coniplete structure determination\ of rhenium( 111) broniidc. should soon provide a conclusive answer on this point. The compound IieaBr9[(C6Hj),P]:{has also been prepared, but since no satisfactory solvent has been found, its spectrum in solution has not been recorded. However, it is isoniorphous with the corresponding chloro compound whose trinuclear structure has been suhstantiated by spectral studies,' and, in addition, these (C6HJaP compounds are isomorphous with IieZiBry[(C6Hj)aAh]3. Finally, it may be noted that the compound (CyHyN)2KeaBrll is to be compared with CsKe;jBrlo, recently rep~rted.~ The greater size of the quinoliniuni ion is perhaps responsible for stabilizing [IieaBrll]--*. complete report on these and other compounds as well as analogs in the rheniuni(II1) chloride system is in preparation.
18
'6
N
-
0 X + .-
14
C
3 rc)
; 12 I
Q CL
2 IO
.-Q 0 .'c 'c
a 0
8
0 C
'= 0
6
(8) F . .4. C o t t o n a n d S . J . L i p p a r d , unpublished work, still in progress (9) J . E . Fergiisson a n d B. H. Robinson, Proc. C h u m . Soc., 189 (1964) (10) T h i s work is being supported b y t h e U. S. Atomic Energy Crim-
0
C .-+ X
w
mission. (11) K.S.F. Predoctoral Fellow 1962-196.5; ( H o n o r a r y ) , 1962-1963.
4
L
O
W'uodroa
Wilson Felli,w
F. A . Cor.ros DEPARTMENT OF CHEMISTRY MASSACHUSETTS ISSTITUTEOF TECHNOLOGY S. J. LIP PARD'^ CAMBRIDGE, MASSACHCSETTS 02139
0
=
Vol. Mi
2
RECEIVEDA~-GLST 10, 1964
0 Frequency in
ern-'
x IOm3
Fig. l.-l-isible spectra of rheuiumiII1) bromide and some derivatives. ( R e B r j ) n in 4gC; aq. H B r (-------); (ReBrs), in acetone (- --); (CsHsS)2ReaBrlj in acetone (00000: [CsH,S)?[ReBrs(C,I-Ij),ChH,l'lain CHC1, Ke2Br,, in a q . H B r (- - - ) ; ( - - - - - - - - - ) : [ReBra(CsH,)sAs];i in acetone ( . . J.
ion, by a single crystal X-ray study,' and in ReaBry[(C2H5)2C6H5P]3, because the latter compound is isomorphous with ReaCls[ (C".Hj)2C&P]3, the structure of which has been determined.3 The compounds [ReBr3(C6H5)3X~],I and (C9H8N)2Re3Brll,which have now been prepared from rhenium(II1) bromide as well as solutions of rhenium(II1) bromide in aqueous HBr and acetone, have visible spectra which are all very similar t o one another and to the spectra of (CjH6P\')SRe4Brlj and Re3Brg[ (C2H3)?C6H3Pl3.These are shown in Fig. 1. Not only are all of the spectra shown in Fig. 1 similar to one another, but they are quite similar to the characteristic spectrum of compounds containing the Re3Cly group.'," The main difference is t h a t in the bromo compounds strong ultraviolet absorption. presumably of B r - + K e charge-transfer type, sets in a t lower frequencies, and this tends to obscure the absorption band around 19,000 c m - l . However. the characteristic pattern of absorption is clearly the same in both the KeaCIR compounds and the rheniurn(II1) bromide derivatives, and is presumably characteristic of the Re3cluster itself. The appearance of this characteristic spectrum in the HI3r and acetone solutions of rhenium(II1) bromide suggests that it consists of KeaBr9units, though these may not be connected in the same way as in the chloride
-
( 7 , F. A . C o t t o n a n d S , J . 1,ippard. iizorn ( ' h e i i z . , i n pt-esc. T h e coinp o u n d consiii- < > f rciual numbet of l
