3164
COMMUNICATIONS TO THE EDITOR
Vol. 81
However, it served as a basis for study
proton acceptor group an element of small atomic radius and strong electronegativity (i.e., N, 0, F). diene structure plus one hetero atom. The basic Although these conclusions may be valid when very structure is strong hydrogen bonding to anions is involvcct (e.g., HFz-), their extension to covalently bonded Rj, ,Ro negative atoms is not possible. On the contrary, M R1 we wish to present spectroscopic evidence which R4Y Y demonstrates that the relative electron donor abiliRd*R2 ties of the alkyl halides decreases in the order I > I Br > C1 > F. Furthermore, the hydrogen bonds The compounds reported here all contain phenyl formed by compounds of these elements with suitRI, R,, RB, Rq. The nature of Rg and Re substituents able proton donors are weak relative to those of the may vary with the possibility of one or both being other electronegative elements. absent, e.g., with As, R5 = C6H5 and R6 is not The order given above has long been known to present. Also, R5 and R6 may represent a second hold for the intramolecular hydrogen bonding ring as in a spiro compound. Preparation of spiro exhibited by the o-halophen~ls,~The criteria for compounds imposes several limitations on the gen- strength, or more precisely the enthalpy, of the eral reaction, i e . , the need for (1) a hetero atom hydrogen bond is taken to be related to Av, the with a valence of four or greater and ( 2 ) use of 1 spectral shift in cm-l from the “free” to the mole of a tetrahalide of the hetero atom plus 2 moles “bonded” peaks4 However, the interpretation of LTPBD. of the o-halophenol data is ambiguous because, in Referring to structure I, compounds have been the rigidly held aromatic systems, the small covaIent prepared with hl = S, P, Sn, As, Sb and Ge (also radii of the lower atomic weight halogen atoms may l l= Sn and Ge). These spiro compounds with I TABLE I5 are all high melting (>160”) crystalline solids. AV VALUESOF B-SUBSTITUTED ETHANOLS Molecular weight determinations and elemental analysis verified the ring structure. A typical Substituent F C1 Br I E t 0 EtS E t ? N member of the series is pentaphenylphosphole A v , cm.-l 12 32 38 46 30 92 166 (M = P, Rs = C6H5) m.p. 261-262’; mol. wt. TABLE I15 calcd. 464, found 463 (mass spec.); calcd: C, SHIFTSDUE T O INTERMOLECULAR HYDROGEN 88.1; H, 5.4; P, 6.65; found: C, 88.2; H, 5.5; FREQUENCY BOSDINC(an.-’) P, 6.55. AV A” Work is in progress to prepare heterocycles of Compound CH30Hb Phenolb this type with R groups I-IV other than phenyl. n-Butyl fluoride 19 40 Other synthetic routes to these compounds and the Cyclohexyl fluoride .. 53 preparation of other type compounds having this n-Butyl chloride 37 50 basic unit are being investigated. Concurrent Cyclohexyl chloride .. 66 with the synthetic studies, we are attempting to n-Butyl bromide 41 69 determine the chemical nature of these new ring Cyclohexyl bromide .. 82 systems with regard to the possible aromaticity of r%-Butyliodide 47 ii the five-membered ring, ability to form ?r electron Isopropyl iodide . . 83 complexes with metals and the normal diene Di-n-butyl ether 155 281 reactions. of a series of compounds having the conjugated
Di-72-butvl sulfide 155 258 Tri-n-butylamine“ 396 .. Tri-n-butylphosphine” 267 .. THEDow CHEMICAL COMPANY F. C. LEAVITT 310 .. Tri-ethylarsine’ EASTERNRESEARCHLABORATORY T. A . MANUEL a Solution in C2C14. * Positions of the free peaks: 3610 FRAMXNGHAM, MASSACHUSETTS F. JOHNSON cm.-’ for phenol; 3643 cm.-’ for methanol in both CCIJ and RECEIVED APRIL20, 1959 czc14. (2) H D Hartough, “Thiophene and I t s Derivatives,” Interscience Publishers, Kew York, N Y , 1952, pp. 473 f .
COMPARISON OF COVALENTLY BONDED ELECTRONEGATIVE ATOMS AS PROTON ACCEPTOR GROUPS IN HYDROGEN BONDING Sir :
Generalization,lv2widely held and widely quoted, concerning the requirements for hydrogen bonding, has emphasized the necessity of having as the (1) These references may serve as illustrations: M. Davies, Ann. R e p f s . o n Prog. Chem. (Chem. SOC.London), 43, 5 (1945); L. Hunter, rbtd., p. 141; L. Pauling, “The Nature of the Chemical Bond.’’ Corne11 University Press, Ithaca, N. y.,Second Edition, 1948, pp. 286 f . ; C. A. Coulson, “Valence,” Oxford University Press, London, 1952, pp. 298 f . ; T. Moeller, “Inorganic Chemistry,” John Wiley and Sons, New York, N. Y., 1952, pp. 188-189; G. W. Wheland, “Resonance in Orpanic Chemistry,” John Wiley and Sons, New York, N. Y . , 1955, pp. 49 f.; E. Cartmell and G. W. A. Fowles, “Valency and Molecular Structure,” Butterworths Scientific Publications, London, l O X , p. 138; G. M. Badger, Reus. Pure A p p l . Chcm. (Australia), 7 ,
55 (1957) ; J. A. A. Ketalaar, “Chemical Constitution,” Elsevier, Publ. Co., New York, K,Y . , Second Edition, 1958, p. 408. (2) For recent reviews on hydrogen bonding not referred to above: L. Kellner, Repts. Pvogr. in Phys., 1 5 , 1 (1952); L. Hunter in W.Klyne. Ed., “Progress in Stereochemistry,” Vol. I, Academic Press, Inc.. New York, N. Y . . 1954, pp. 223 f . ; C. A. Coulson, Research, 10, 119 (1957); C. G. Cannon, Speclrochimica A c t a , 10, 341 (1958). (3) 0. R. Wulf, U. Liddel and S . B. Hendricks, THIS JOURNAL, 68, 2287 (1936); 0. R . Wulf and E . J. Jones, J . C h c n . P h y s . , 8 , 745 (1940); G. Rossmy, W. Liittke and R . hlecke, ibid., 21, 1606 (1953); and especially A. W, Baker and A. T. Shulgin, THISJOURNAL, 80, 5358 (1958): A. W. Baker, ;bid.3 80, 3598 (1958). (4) R. M.Badger and S . H. Bauer, J . Chem. Phys., 6 , 839 (1937); R. M.Badger, ibid.,8, 288 (1940); R. C. Lord and R. E. Merrifield, ibid., 21, 160 (1953); K. Sakamoto, M. Margoshes and R. E. Rundle, THISJOURNAL, 7 7 , 6480 (1955); G. C. Pimentel and C. H. Sederholm, J . Che?n. P h y s . , 24, 639 (1956). ( 5 ) Perkin-Elmer Model 2 1 and Model 112 Spectrophotometers, LiF prisms, CClr solutions except where noted. Intramolecular concentrations were about 0.005 M . Intermolecular concentrations were about 1 M for the proton acceptor and 0.01 M for the donor.
June 20, 1959
COMMUNICATIONS TO THE EDITOR
3165
pletion of the reduction of azobenzene to hydrazobenzene takes thirty minutes a t room temperature. Similar to azobenzene itself, substituted derivatives such as 4-fluoroazobenzene, 4,4’-difluoroazobenzene, 4,4‘-dichloroazobenzene, 3-trifluoromethylazobenzene and 4-fluoro-3’-trifluoromethylazobenzene were reduced by the catalytic LAH reduction to the corresponding hydrazobenzenes with 90% yields or better. It is possible to use instead of the metal halides other metal salts (such as nitrates, carbonates, sulfates, acetates) or metal oxides. However, metal oxides as Moos, V206, WOa, TiO2, Cr03, CuO and NiO are considerably less active catalysts than the corresponding metal halides and the completion of the reduction of 0.1 FRICK LABORATORY mole of azobenzene in ethereal solution a t room PRIXCETON UNIVERSITY PAULv o s R. SCHLEYERtemperature takes from 2-6 hr. PRIXCETON, XEW JERSEY DEPARTMEST OF CHEMISTRY Nitrobenzenes are reduced by the catalytic OF WISCOXSIX ROBERT WEST UNIVERSITY LAH reduction in ether solution directly and with MADISOS,WISCONSIS 90% yield or better to hydrazobenzenes. AS the RECEIVED MAY1, 1969 reduction of nitrobenzenes by LAH to the corresponding azobenzenes in itself is quite vigorous, CATALYTIC EFFECT OF METAL HALIDES AND it is advantageous to add the ethereal solution of OXIDES ON T H E LITHIUM ALUMINUM HYDRIDE the azobenzene first to the ethereal LAH solution REDUCTION O F AZOBENZENES AND NITROBENand then the catalyst. ZENES T O HYDRAZOBENZENES On addition of an ethereal solution or suspension Sir : of the catalytic halide to the ethereal LAH s o h The interaction of metal halides and oxides tion generally a dark colour and some precipitate with lithium aluminum hydride (LAH) is known.’ is formed. The color of the solution is due to the A reduction method based on the use of mixed formed intermediate, unstable but highly active hydrides prepared from equimolar quantities of mixed metal hydride. The precipitate is finely LAH and aluminum halides also was developed.2 dispersed metal from the decomposition of the metal Different metal halides were used as additives in hydride. Similarly using metal oxides as catalyst ether cleavage reactions with LAH.a a marked change of color is observed. The presLAH alone can reduce azobenzenes to hydrazo- ence of metal-hydrogen bonds in the intermediate benzenes only with considerable d i f f i c ~ l t y . ~Large metal compounds has been proved. excess of the hydride, higher temperatures and proIt is interesting to note that aluminum halides longed reaction times (some days) are needed. are quite inactive as catalysts in the above menNitrobenzenes were known to be reduced by LAH tioned reduction. only to the corresponding azobenzenes, a reaction CONTRIBUTIOX KO.18 which also is used for the qualitative analytical EXPLORATORY RESEARCH LABORATORY G. A. OLAH Dovv CHEMICAL OF CANADA, LIMITED determination of nitro group^.^ It has now been observed that the addition of SARNIA,ONTARIO,CANADA RECEIVEDAPRIL3, 1959 catalytic amounts of certain metal halides such as MoC15, TiC14, TiC13, TiC12, vc13, WC16, CrCI3, FeCl3, Cu2C12, CoBr2, SnC14, CdI2, CbC15, BiC13, FLUORINATION REACTIONS O F SbC16 PbC12 has a very marked effect on the SULFUR TETRAFLUORIDE reduction of azobenzenes with LAH to hydrazo- Sir: benzenes. In ethereal solution and a t room temIn an investigation of the little-known chemistry perature 0.1 mole of azobenzene was reduced of sulfur tetrafluoride (SF4),1 we have found this quantitatively in the presence of less than 0.001 compound to be an unusual fluorinating agent, mole of the above mentioned metal halide catalysts especially for the unique replacement of carbonyl to hydrazobenzene by LAH. The reductions are oxygen with fluorine, The reaction is effective very fast and generally are completed in less than ten /C=O \ SFI +>CF2 SOFz minutes. Some less active metal halide catalysts are ZrC4, TaC15 and PtC12. With them the com-
prevent optimum interaction with the adjacent hydroxyl group. The same objections hold for the results with p-haloethanols (Table I). Completely unambiguous data confirming the same order, I > Br > C1 > F, is provided by intermolecular studies (Table 11). Data for analogous oxygen and nitrogen compounds have been included for comparison. Clearly, electronegativity does not play a dominant role. I n addition, results for similar derivatives of the second and third row elements, sulfur, phosphorus and arsenic, have been added to the tables where available. It is particularly noteworthy that, as proton acceptors, they rival their first row congeners despite their larger covalent radii.
+
(1) N. G. Gaylord, “Reduction with Complex Metal Hydrides,” Interscience Publishers, Inc., New York, N. Y.,1956, pp. 41-60. 77,2644 (1955); R. F.Nystrom ( 2 ) R. F.Nystrom, THISJOURNAL, and C. R. Berger, ibid., 80,2898 (1958); R. F.Nystrom, ibid., 81, 610
(1959). (3) P. Karrer and 0. Rutter, Helv. Chim. Acta, 38, 812 (1950); V. L. Tweede and M. Cuscurida, THISJOURNAL, 79, 5463 (1957). (4) R. F. Nystrom and W .G. Brown. ibid., 70,3738 (1948); F. Bohlmann, Chem. Ber., 85, 390 (1952); G.A. Olah, paper presented a t the Organic Chemistry Symposium of the Chemical Institute of Canada, December 8, 1958, Ottawa (Ontario). ( 5 ) H. Gilman and T. N. Goreau. THISJOURNAL, 73, 2939 (1951).
-C
