COMMUNICATIONS TO THE EDITOR
July 5, 1961
2967
t o us that such a high-energy system might be formed by irradiation of an acid salt of a 2-aminopyridine, since the formation of a cyclic amidine salt might give partial diene character to the remaining aromatic bonds. We wish to report in this Communication the successful realization of this prediction with the isolation and characterization of a valence-bond tautomer of 2-amino-5chloropyridine (I). Irradiation of I ( p K , 4.',7 in water: HC1 salt, m.p. 192-194', AEz 238, 315 m p ; E , 6780, 2770) in dilute hydrochloric acid with sunlight through Pyrex glass resulted in the gradual separation of colorless crystals (pK, 8.4 in water, m.p. 205.5207' with rapid heating, m.p. 202.5-204' with moderate heating, m.p. 192-194' with very slow heating, indicating complete reconversion to I.HCI before melting) which were characterized both as a monohydrate (Anal. Calcd. for C&N2CI.H20. HC1: C, 32.9; H, 4.4; N, 15.3; C1 (total), 38.8. Found: C, 32.9; H, 4.4; N, 15.8; C1 (total), 38.1) and as a hemihydra.te (Anal. Calcd. for C S H ~ N ~ C ~ - ' / ~ H ~ O .C, HC 34.5; I : H, 4.1; N, 16.1; C1 (ionic), 20.4. Found: C, 34.0; H, 4.5; N, 16.1; C1 (ionic), 15.7) depending upon drying conditions. An aqueous solution of this material was transparent to u1traviob.t light above 260 mp, with only end absorption below 260 mp ( E , 3230 a t 220 mp). The spectrum of I was gradually restored when this solution was allowed to stand a t room temperature. Rate constants for the reconversion (in spectroscopic concentrations) of the min.-l (27 + 2') photoisomer to I were 6 X in water and 140 X min.-' (27 2') in 0.1 N (7) Ref. 5 ; J. Inorr. Nucl. Chem., 10, 49 (1959); IS, 212 (1960). sodium carbonate. In non-spectroscopic concen(8) E. 0. Fischer and R. D. Fischer, Angew. Chem., 7 2 , 919 (1960). (9) H. J. Dauben and D. J. Bertelli, J . A m . Chem. SOC., 89, 497 trations, reconversion of the photoisomer to I (1961). upon neutralization was essentially instantaneous (10) G. Winkhaus and G. Wilkinson, Proc. Chem. SOC.,312 (1960). and quantitative. INSTITUT FUR ANORGANISCHE CHEMIE The n.m.r. spectrum (identical for the monoDER U N I V E R S I TMUNCHEN ~T G. N. SCHRAUZER hydrate and the hemihydrate) in D2O of the photoMUNICH,GERMANY isomer shows an equal intensity doublet in the RECEIVED MAY3, 1961 vinyl hydrogen region (7 == 3.45, 3.59, coupling constant, 8 cps.) .12 These values compare favorably A PHOTOCHEMICALLY-INDUCED, VALENCE-BOND with those recently reported by Dauben and coTAUTOMERISM IN A SIX-MEMBERED AROMATIC workers6,'' for the vinyl hydrogens in certain cycloSYSTEM butene derivatives. Furthermore, each comSir: ponent of the original doublet could be further rePhotochemically-induced valence-bond tautom- solved into a sharp doublet with a coupling conerisms, or "bond-switching'' reactions, have re- stant of approximately 2 CFS. The spectrum also ceived considerable recent attention, particularly in shows peaks consistent with the presence of bridgecyclic trienones, l 4 cyclic die none^'-^ and cyclic head hydrogens in a cyclob.utane ring adjacent to Analogous reactions in an electron-withdrawing group ( T = 5.9, 5.4, 5.2, dienesl0." and trienes." six-membered aromatic systems, which would lead with other peaks possibly obscured by a strong to products corresponding to the 'iDewar" structure water peak). These values are comparable to for benzene, have not been observed. It seemed those found for the cyclob-ityl hydrogens in the dimer of N-1nethyl-2-pyridone (7 = 5.6 and (1) F. Santavy, B i d . L i s l y , 91, 246 (1950). 5.8).13 (2) R. Grewe and W. Wolf, Ber., 84, 621 (1951). (3) E. J. Forbes, J. Chem. Soc., 3864 (1955). Catalytic reduction of the photoisomer with (4) 0. L. Chapman and D. J. Pasto, J. A m . Chem. S a . , 80, 6685 hydrogen and platinum oxide in aqueous solution (1958). yielded a product (m.p. 333-335' dec.; Anal. (5) 0. L. Chapman and D. J. Pasto, i b i d . , 82, 3642 (1960). Calcd. for C&N*.HCl: C, 45.3; H, 6.85; N, (6) W. G. Dauben, K. Koch, 0. L. Chapman and S. L. Smith. ibid., 89, 1768 (1961). 21.1; C1, 26.75. Found: C, 45.1; H, 6.7; N, (7) D. H. R. Barton, Hclu. Chim. A d a , 42, 2604 (1959). 20.7; C1, 26.3) which was transparent t o ultra(8) G. Buchi and E. M. Burgess, J . A m . Chem. SOC.,8 2 , 4333
(C0)s slowly and with decomposition dissolved in perchloric acid; attempts to isolate salts were unsuccessful. Structure I assigned to these salts is supported by the increased carbonyl frequencies in the spectra. Similar shifts generally are caused by a positive charge in the metal carbonyl moiety. It is most straightforward to assume that the seven olefinic carbon atoms are lying in one plane, whereas the CH2-group in the molecule is pointed upward or downward. The n.m.r. spectrum of the C8HgFe(C0)3+cation in sulfuric acid shows four absorptions, a t - lG8, - 17, lG2 and +225 C.P.S. (relative to water), with intensity ratios of 1:4:2:2. This result can be rationalized readily in terms of the proposed structure. In general, the unoccupied orbitals of all stable conjugated cyclic hydrocarbons are antibonding. Protonation will produce an unoccupied non-bonding orbital which when formed in a transition metal complex may be used for electron back-donation. I n accord with theoretical conclusions by Brown7 the presence of low lying orbitals will stabilize metal-ligand bonding, especially of the last members of the first transition metal series. Recently a number of similar salts of the cyclohexadieniumand cycloheptadienium- iron tricarbonyl cation have been d e s ~ r i b e d . ~The , ~ ~cyclooctatrienium~~ iron-tricarbonyl-cation is thus a higher member of a class of transition metal complexes with resonance stabilized carbonium ions. Acknowledgment.-1 wish to thank Miss U. Merkel for her skillful experimental assistance.
+
(1960). (9) J. J. Hurst and G. H. Whitham, Proc. Chcm. Soc., 116 (1961). (10) 0. L. Chapman and D. J. Pasto, Chem. and I n d . , 53 (1961). (11) W. G. Dauben and R. L. Cargill. Tefrahedron,la, 186 (1961).
*
(12) All n.m.r. spectra were determined on both 40 and 60 megacycle Varian instr$iments. (13) E. C . Taylor and W. U'. Paudler, Tetrahedmn Letters, 2 5 , - 1 (1960), and unpublished results.
2968
COhlhIUNICATIONS TO THE
EDITOR
Vol.
s3
violet light above 240 mp, with only end absorption might be achieved in suitable systems. The below 240 mp ( E , 3600 a t 220 mp). The n.m.r. first of these is the base catalyzed exchange bespectrum of this reduction product shows a broad tween DPand water,2which is pictured most easily hydrogen peak (7 = 7.89) and a pair of doublets in terms of the (admittedly very unfavorable) ( r = 5.79, 5.66 and 5.29, 5.15), typical of an AB equilibrium pattern. Integration showed that there are twice H2 + OH- F? HH20 (1) as many hydrogen atoms under the broad peak as Conceivably, the small amount of hydride in the combined doublets. Although the reduction product (as the hydrochloride), in contrast ion present in strongly basic systems might acto the photoisomer, was stable upon prolonged complish the reduction of an organic molecule via standing in aqueous solution, addition of 0.1 N a sequence such as 0 0sodium hydroxide a t room temperature resulted in I I/ the separation within two minutes and in essentially H- + RCR .-t R-CH-R (2) quantitative yield of a compound which, on the 0OH basis of all available evidence, appears to be cis-2I aminocyclobutanecarboxamzde IV, (m.p. 233-255'; R-cH-R ROH +.RAHR + RO(3) Anal. Calcd. for C5H10N20: C, 52.6; H , 8.8; Second, a considerable number of reactions are N, 24.5: mol. wt., 114. Found: C, 52.95; H , known in which the treatment of organic molecules 9.0; N, 24.9; mol wt. (in glacial HOXc), 99). with strong bases a t elevated temperatures leads On the basis of the above considerations, we to oxidation with evolution of molecular hydrogen. feel that the photoisomer of I can only be repre- Perhaps the oldest is the Varrentrapp reaction in sented by structure 11, and that the reduction which caustic fusion of oleic acid produces hydroproduct must be represented by structure 111. gen, palmitic acid and acetic acid.3 Since a t least The conversion of I t o I1 represents the first one step in these processes involves dehydrogenaexample of a valence-bond tautomerism in a six- tion, they may be considered as simply the back membered aromatic system. The reaction appears reactions of a based catalyzed hydrogenations. to be general, for similar photoisomers were obtained from 2-aminopyridine and %amino -5-bromoTABLE I pyridine ; catalytic reduction in each case yielded BASE CATALYZED HYDROGENATIONS IN teYt-BUTYL ALCOHOL compound 111. Furthermore, the conversion of I1 T H2 Time, Yield,b Substrate ("(2.) (Ib./in.*) hr. KOCaHgQ 7a to IV in an over-all yield of greater than 93% 200 1300 70 0.5 50-60 promises to be representative of an extraordinarily Benzophenone 180 2000 16 0 simple route to cyclobutane derivatives. Exten- Benzophenone 195 1900 31 0.25 40-50 sions of this new photoisomerization reaction to Benzophenone 230 2560 15 0 other systems and the chemistry of the photo Benzophenone Benzophenone 230 2560 15 0.2 +d products are under active exploration.
+
+
,--"*
i--NH,
e1
1~
r n ' + c 1 -
" I1
+ a-
I11
Acknowledgment.-This work was supported by a research grant (CY-2551) to Princeton University from the National Cancer Institute, Pu'ational Institutes of Health, U. S. Public Health Service.
Benzophenone 170 2100 25 0.05 (-)' Sitrobenzene 160 1900 45 0.1 Sitrobenzene 160 2000 49 0 ( + ) indicates reduction, but a Moles/mole substrate. yield not determined; ( - ) indicates no reaction. Small pressure drop, but no reduction products definitely identified. d Teflon liner in vessel.
+
We now report such a base-catalyzed homogeneous hydrogenation, achieved by heating benzophenone in tert-butyl alcohol solution containing under 1300FRICKCHEMICAL LABORATORY EDWARD C TAYLORpotassium tert-butoxide a t 200' PRINCETON UNIVERSITY LYILLIAM U' PAVDLER2000 1b.,h2 hydrogen pressure. Hydrogen is IRWI?; KUNTZ, J a slowly absorbed and benzhydrol obtained in PRIUCETOU, N. J. RECEIVED MAY12, 1961 40-60% yjeld, determined by gas chromatography, and identified by retention time, infrared spectra, and mixed melting point. No further reduction BASE CATALYZED HOMOGENEOUS to diphenylmethane is detected. HYDROGENATION Experiments were carried out in a small conSir. ventional stirred autoclave with nickel liner, Although the reaction of molecular hydrogen and the effects of variables are shown in Table with unsaturated organic molecules a t moderate I . Hydrogen, tert-butoxide and an elevated temperatures is generally energetically feasible, temperature are all required for the reaction. a potential barrier to the process exists which is The possibility of heterogeneous catalysis by the only overcome by the use of heterogeneous cata- nickel linear is eliminated by the successful exlysts or certain transition-metal complexes in periment using a Teflon liner. homogeneous systems. (2) K . Wirtz and K. F. Bonhoeffer, 2 . p h y s i k Chenz., 1778, 1 Two groups of observations of long standing in (193G); 'A', K. Wilmarth, J. C. Dayton and J . 11,Flournoy, J . A m . the literature suggested to us that simple basic Chem. Sos., 7 6 , 4549 (1953); S. L. Rliller and D Rittenberg, ibid , catalysis of homogeneous hydrogenation also 80, 61 (1958). (1) For a review and discussion,
u, aoz
(1957).
if
J , Halpern, A d v a n c e s i n C n t o l g r r s ,
(3) F. Varrentrapp, Liebiss Ann., 36, 196 (1840). For more recent work, cf. R.G Ackman, P. Linstead, I3 J . Wakefield and B. C . L,LinstePd, Tdviihedron, E, 291 (l'J(i0).
