Aug. 20, 1962
COMNUNICATIONS TO
higher than any found for phosphorescent hydroIn thc nitrenes there are two unpaired electrons a t least partially on the same atom. Such localization causes considerable interaction. This arrangement is forbidden in the excited states of aromatic hydrocarbons where there is only one available atomic orbital in each atom. In addition, these values are higher than any likely to be required for diphenylmethylene.Y In the latter the presence of the second phenyl group aids delocalization, and the smaller nuclear charge of the carbon atom permits a larger interelectronic separation. Both effects should reduce the spin-spin dipole energv and lead to smaller values of D.
THE
EDITOR
3221
proton magnetic resonance spectrum of I in CSS solution consists of but one signal of the phenyl protons. From these observations a structure is deduced in which a nickel atom is surrounded by two molecules of dithiobenzil in a square planar arrangement.
Magnetic measurements3 indicate a weak paramagnetism of I a t room temperature (Pfound = (16) Computations by J. Higuchi have given values of D of 1.0 -0.8 B.hl ) . The bis-piperidine adduct or the a n d 1.8 cm.-1 for CHr a n d N H , respectively (J. Hipuchi, private solutions of I in pyridine are more strongly paracommunication), 0.1 and 2.5 f 0.2 BELLTELEPHONE G. SMOLINSKY magnetic with moments of 1.9 LABORATORIES, INCORPORATED E. WASSERMANB . K , respectively. The magnetic properties of I W. rl. YAGER thus are consistent with those frequently observed MURRAY HILL,NEWJERSEY RECEIVED JULY 7 , 1962 for square planar Ni (11) complexes. However, the particular nature of the ligands in I forces us REACTION OF DIPHENYLACETYLENE WITH NICKEL to assume that two electrons must be placed into SULFIDES a low-lying unoccupied molecular orbital of the complex as I would otherwise hare to be formuSir: lated as a Si(0) complex. I is also isolated in The hypothesis that IT.Steinkopf’s well-known synthesis of thiophene from acetylene and pyrite1 yields up to l4Yc per run if Ni(C0)4 or finely dimight yield organometallic complexes if conducted vided metallic nickel is refluxed with a solution of under milder conditions has led us to investigate diphenylacetylene and sulfur in toluene. This mode the reaction of metallic sulfides with acetylenes. of preparation of I implies that the intermediate Technical-grade nickel sulfide, prepared by the formation of a highly reactive, possibly free-radicaladdition of ammonium sulfide to neutral or slightly type, nickel sulfide (NiS4 ’) may be responsible for acidic solutions of a nickel salt, usually contains this unusual reaction. A similarly reactive nickel a non-stoichiometric excess of sulfur. This nickel sulfide may also be present as an “impurity” a t sulfide was found to react with diphenylacetylene the surface of the sulfur-rich nickelous sulfide in toluene in a closed tube a t 160’ to produce large used in the initial experiments. We are currently amounts of tetraphenylthiophene. If, however, investigating the full scope of this reaction. (3) T h e magnetic measurements a e r e performed b y H. hlbdl, the reaction was conducted a t 120’, the toluene solution became dark green within 24 hours and Technische Hochschule a t 3lunich FUR ANORGATISCHE CHEMIE G X. SCHRAUZER upon cooling deposited a nearly black, crystalline INSTITUT DER UNIVERSITAT MUNCHEN, MEISERSTRASSE 1 complex (I) of composition NiCZ8Hz0S4(Calcd. ,MUSICH2 , GERMANY V. MAYWEG C, 61.89; H, 3.70; S, 23.61; Ni, 10.80; mol. RECEIVED JUNE2, 1962 wt. 543. Found C, 61.9; H, 3.7; S, 23.5; Ni, 10.5; mol. wt. 378). Under these conditions only RELATIVE STRENGTHS OF ALKYL HALIDES small amounts of tetraphenylthiophene are formed THE AS PROTON ACCEPTOR GROUPS IN HYDROGEN suggesting that I is an intermediate in the tetraBONDING’ phenylthiophene formation. I is completely stable Sir : to air and decomposes a t 290’ into nickelous sulfide, I n 1959 data were published on the shifts in fresulfur and (exclusively) 2-phenylthianaphthene. quency of the 0-H stretching absorption bands of Reduction with HI-phosphorus in acetic acid a t phenol and methanol upon hydrogen bond forma140’ yields mainly desoxybenzoin and small tion to alkyl halides.2 The frequency shifts, hv, amounts of bibenzyl. Reaction with diphenyl- increased as the halogen atom was changed in the acetylene a t 140’ in toluene afforded up to 76% order F < C1 < Br < I (Table I, column 2).5 By of the theoretical amount of tetraphenylthiophene. application of the Badger-Rauer rule5 relating Diethylacetylenecarboxylate reacted even a t 80 ’ Acknowledgment of partial support of this research is made t o producing diethyl 4,5- diphenylthiophene- 2,3-di- t h e(1)donors of t h e Petroleum Research Fund, administered by tho carboxylate (m.p. 98’). I is moderately soluble in American Chemical Society (R.W.) and t o t h e National Science most nonpolar solvents, forming deeply green solu- Foundation (P.S.). (2) P. von R . Schleyer and R. West, J . A m . Chem. Soc., 81, 3164 tions. Enhanced solubility in bases such as pyriAlso see 31.-L. Josien, et ai., Bull. m c . chim. France, 423 dine and piperidine and a color change to brown- (1959). (1957); 188 (1958): G. J. Korinek and W. G. Schneider, Con. J . red suggest the formation of complexes of I with Chem., 36, 1157 (1957). (3) T h e same order of basicity holds for charge-transfer interactions solvent molecules. Indeed, labile bis-adducts of alkyl halides with halogens.4 both bases were isolated in crystalline form. The of (4) R. West, D. L. Powell, L. S. XVhatley and N.K. T. I,ee, un-
*
(1) W. Steinkopf and G. Kirchhoff, A n n . , 403, 1 (1914); W. Steiokopf, i b i d . , 403, 11 (1814). (2) E. Donges, Z. il’ufurforschunp. 1, 221 (1946).
pi1 bli shed
data. ( 5 ) R. M. Badger and S. H . Bauer, J . Chem. P k y s . , 6 , 839 (1837); R. >I. Badger, i b i d . . 8 , 288 (1940); ref. 7 , pp. 82-84.
3222
COMMUNICATIONS TO THE EDITOR
Vol. 84
spectral shifts and enthalpies of hydrogen bonds, or nitrogen bases (ca. 5-8 kc;tl/rn~le).~.~J Table I our evidence was taken to indicate that proton ac- includes data for the enthalpy of hydrogen bonding ceptor ability of covalently-bound halogen atoms of phenol to the di-n-butyl derivatives of oxygen, increased in the same sequence. This order has sulfur and selenium; a decrease in - A H o in the been used by others in discussing the relative order given was found. The trend thus parallels basicity of alkyl halides toward protons6 and it that found for the alkyl halides, but i t is to be noted therefore seems iniportant to set forth our present that both sulfur and selenium are capable of formmodified views. ing hydrogen bonds of appreciable strength, a fact The recent accumulation of reliable thermo- not fully recognized heretofore. chemical data on the strength of hydrogen bonds The phenomenon of hydrogen bonding is not has made apparent the invalidity of the Badger- sufficiently understood a t present to explain the Bauer rule when applied to more than a limited data of Table I. However, i t is not hard to underseries of closely related compo~nds.~~'.8 This evi- stand in qualitative terms why Av and -AHo dence has led us t o determine the thermodynamic should not correlate.l1 - AHQmeasures the total properties of the interaction of phenol with alkyl energy of the interaction A-H.. .B-Y, i.e., the halides; the results (Table I) quite contradict the strength of the H. . .B bond partially compensated earlier conclusion concerning the order of hydrogen by the weakening of the A-H and the B-Y bonds; bond forming ability of covalently bound halogens. Av measures the weakening of the A-H bond only. Both the free energy and enthalpy of interaction If the proton donor, A-H, is kept the same, A v and tend to decrease in the order F > C1 > Br > I, the - A H o may correlate for minor structural changes in reverse of the spectral shift order. Baker and the proton acceptor, e.g., in Y , but not necessarily KaedinglO have shown that the free energies of in- for major changes, e.g., in the acceptor atom, B. tramolecular hydrogen bonding between hydroxyl (11) We ore indebted to Dr. H. J. Bernstein, National Research groups and halogens in 2,6-dihalophenols increase Council, Ottawa, for this observation. Dr. Bernstein (private comhas demonstrated a simple relationship between A P / A H ~ munication) in the order I < F < Br < C1; however, steric a given acceptor and the ionization potential for acceptor atoms factors are expected to influence the free energy of for in related molecules. interaction in these compounds. Q
TABLE I THERMODYNAMIC PROPERTIES AND SPECTRAL SHIFTSOB HYDROGEN BONDSOF PHENOLTO ALKYL HALIDESAND ALKYLCHALCOGENIDES IN CClr SOLUTION'
- AFQ.
Av,
Compound
cm.-I*
-AHo, kcal./ mole
- ASQ,
25'.
25',
kcal./ mole
cal./deg. mole
Cyclohexyl fluoride 53 6.1 3.13 1.31 Cyclohexyl chloride 66 2.21 0.87 4.5 Cyclohexyl bromide 82 4.0 2.05 0.85 Cyclohexyl iodide 0.82 3.0 86 1.72 %Butyl ether 5.98 2.45 11.8 278" n-Butyl sulfide 254* 4.26 1.59 9.0 n-Butyl selenide 240' 3.72 1.46 7.6 a Thermodynamic properties determined in the near nfrared. See D. L. Powell, Ph.D. Thesis, University of Wisconsin, 1962, for the method of calculation. * Taken from ref. 2, or the present work, unless noted. O P . von R. Schleyer and L. Robinson, Abstracts, Fourth Delaware Valley Regional Meeting, Am. Chem. SOC., Philadelphia, Pa., Jan., 1962, p. 76. The corresponding spectral shift for phenol-n-butyl telluride is 220 cm.-1.
In the previous communication,a it was also concluded that alkyl halides were relatively weak proton acceptors in hydrogen bonding. This conclusion is unchanged by the thermodynamic data, which show that such hydrogen bonds are quite weak compared with those from phenol to oxygen (6) S. Andreades and D. C. England, J . A m . Chcm. SOL, 88, 4670 (1961): J. R. Gerslen, J . Org. Chcm., 18,758 (1901). Cf.J. E.Gordon, J Ore. Chcm., 38, 738 (1981). (7) G. C. Pimentel and A. L. McClellan, "The Hydrogen Bond," W. H. Freeman and Co., Sam Francisco, 1960, pp. 82-85 and 348 ff. (8) E. D. Becker, Sgect. Acto, 17, 436 (1961); H. Dunken and H. Fritsche, Zeit. Chcm., 1, 127, 249 (1961). (9) Added in proof: some measurements dmllsr to those reported here have recently been presented in preliminary form, M.-L. Josies,
Purr Appl. Chem., 4, 33 (1962). (10) A. W. Baker and W. W. Kaeding, J . A m . Chem. SOC.,81,5904 (1959); dso see J. H. RIchards and S. Walker, Trans. Faraday Sac,, 67, 412 (1961).
DEPARTMENT OF CHEMISTRY UNIVERSITY OF WISCONSIN MADISON,WISCONSKN
ROBERTWEST DAVIDL. POWELL LINDAS. WHATLFJY MARGARET K. T. LEE PAULVON R. SCHLEYER
FRICK LABORATORY PRINCETON UNIVERSITY PRINCETON, NEW JERSEY RECEIVED MAY31, 1962
THE TOTAL SYNTHESIS OF 6-DEMETHYL-6-DEOXYTETRACYCLINE
Sir: The molecular structures of oxytetracycline (Ia) and chlorotetracycline (Ib) were elucidated in our laboratories a decade ago.' Since that
CONH;! HO
0
0
Ia
Ib
IC
H OH Me OH
CI (7) OH Me H
OH
I
R, R2 R3 Rq
H H
H
H
time, the tetracycline antibiotics have emerged as a unique class, whose characteristic chemotherapeutic activity is strictly dependent upon the main(1) (a) F. A. Hochstein, C. R. Stephens, L. H. Conover, P. P. Regna, R. Pasternack, K. J. Brunings and R. B. Woodward. J . Am. Chcm. SOC.,74, 3708 (1952); F. A. Hochstein, C. R. Stephens, L. H. Conover, P. P. Regna, R. Pasternack, P. N. Gordon, F. J. Pilgrim, K. J. Brunings and R. B. Woodward, ibid., 16, 6455 (1953). Cf. also S. Hirokawa, Y.Okaya, F. M. Love11 and R. Pepinsky, 2. Krht., 114,439 (1959). (b) C. R. Stephens, L. H. Conover, F. A. Hochstein, P. P. Regna, F. J. Pilgrim, K . J. Bruningr and R. B. Woodword, J . A m . Chem. SOC.,74, 4970 (1952); C. R. Stephens, L. E. Conover, R. Pasternack, F. A. Hochstein, W. T. Mordand, P. P. R e p , F. J. Pilgrim K. J. Brunings and R. B. Woodward, ibid., 76,3568 (1854).
