COMMUNICATIONS TO THE EDITOR
Aug. 5, 1961 CH3
CH3
& c
o'8
t
1
0.6 W
While there are a number of aromatic hydrocarbon charge-transfer complexes, the tropylium ion-aromatic complexes appear to be the first directly observed in which the acceptor is a carbonium ion. On the basis of the observed K value for mesitylene-tropylium ion complex formation, tropylium ion as an acceptor compares favorably with other n-acids. The present charge-transfer complexes of the stable tropylium ion are instructive models for possible reaction intermediates containing x , n-interacting aromatic hydrocarbon and carbonium ion species." In principle, olefins should conceivably be able to replace aromatic hydrocarbons as donors toward tropylium ion. While such cornplexing of tropylium ion with olefins has not yet been examined thoroughly, visible colors are produced on addition of tropylium salt to a solution of tmns-stilbene, isoprene or tropilidene. Regarding charge-transf er complexes of tropyliurn ion with simpler donors such as halide ion, Doering' has suggested that the sequence of colors of the crystalline tropyliurn halides indicates a has ascharge-transfer process. Also, cribed a band for tropylium bromide in methylene chloride a t 401 mp to a charge-transfer transition. However, with halide ion donors charge-transfer complex formation tends to be followed or accompanied by more deep-seated reactions. For example, Dauben2 has reported that tropylium iodide decomposes on standing to tropylium triiodide and other products. Allso, we have observed that addition of tetrabutylammonium bromide to tropylium salt in ethylene dichloride gives rise to rapid formation of tribromide ion. We have similar indications that thiocyanate ion is oxidized by tropylium ion.
s
em z
LLI 0 _I
0.4
a
9
k0 0.2
01
I
50 0
I
600
700
WAVELENGTH ( m l ) .
Fig. 1.-Absorption spectra of solutioris 0.05 m M in heme a t pH 6.8. The H b + spectrum has maxima a t 631 and 500 r n M . The other spectra are those of the compounds of oxidized phenylhydrazine with H b (broken line) and Hb+ (solid fine).
oxidation of phenylhydrazine to benzene and nitrogen.' We observed, however, that addition of excess phenylhydrazine to a solution of Hb+ resulted in a brownish solution, the absorption spectrum of which differed from those of Hb, Hb+, or of a mixture of only these compounds. This observation led to experiments reported below, which indicate that an oxidation product of phenylhydrazine coordinates with both H b and Hbf to form compounds having characteristic absorption spectra. Bovine hemoglobin in phosphate buffer of pH 6.8, I'/2 0.1, was used in these experiments. Solutions of Hb, Hb+, and HbCO were prepared as previously described. Absorption spectra of samples 0.05 n i X in heme were recorded with the Cary Nodel 14M spectrophotometer. Oxidized i l l ) Of related interest is the demonstration by .I. K . Colter and S. S. Wang t h a t acetolysis of 2.4,7-trinitro-9-fluorenyl p-tduenesulfonphenylhydrazine was prepared by mixing solutions a t e is accelerated b y small concentrations of aromatic hydrocarbons of phenylhydrazine and KaFe(CMe a t concen[page 2 2 - 0 of Abstracts, 139th Meeting of t h e American Chemical trations up to 10 m M and 20 mM. respectively. Society. St. Louis, Mo.,March 21-30, 19611. At higher concentrations the mixed solutions be(12) Immediate colos formation from tropylium ion and 7-methyltropilidene already has been noted by Kenneth Conrow during tropylcame turbid. I t was found that two moles of ium ion-catalyzed isomerization of 7-methyltropilideoe IJ. Ana. C h u m ferricyanide are reduced per mole of phenylhydraChem. Soc., 83, 2343 (1961)l. zine oxidized. In order to prevent intercon(13) K. M. Harmon and A. B. Harmon, i b i d . . 83, 86.5 il9til). version between the ferrous and ferric compounds (14) X.S.F. Predoctoral Fellow, 1958-1962. LIEPARTMEST OF CHEMISTRY MARTIXFELDMAS'!of hemoglobin, excess of phenylhydrazine and ~ X I V E R S I T YOF CALIFORNIA S. ~ ' I X S T E I S ferricyanide were added to solutions for experiments Los ANGELES24, CALIFORSIA with Hb and H b t , respectively. RECEIVED J U N E 26, 1961 Addition of oxidized phenylhydrazine to H b + in presence of oxygen resulted in a compound with an absorption maximum a t 541 mp. This comEVIDENCE FOR COORDINATION OF MONOPHENYL pound and H b + have isosbestic points a t 609 and DIIMIDE WITH HEME PROTEINS 516 mp. The same compound was the major Sir:
I t has been reported that the reaction of ferrihemoglobin (Hb+) and phenylhydrazine results in reduction of HbT t o ferrohetnoglobin (Hb) and
(1) H. H . Rostorffer a n d J. R . Totter, .T. B i d . Chcm., 221, 1047 (1966). ( 2 ) H. .4. I t a n o and E. Robiiimn. Hiochitrz P I Hiophyr. Acta, 29, 545 (1958).
3340
COMMUNICATIONS TO THE EDITOR
product when untreated phenylhydrazine was added to H b + in the presence of oxygen. H b and phenylhydrazine did not react in the absence of oxygen or ferricyanide. However, addition of oxidized phenylhydrazine to H b resulted in a conipound different from the compound of H b + and oxidized phenylhydrazine. The compound of H b and oxidized phenylhydrazirie has absorption niaxima a t 637 and 539 mp. This compound and H b have isosbestic points a t 506 and 537 nip. Failure of cyanide ion to alter its spectrum ruled out the presence of Hb+. Absorption spectra of the two new compounds are compared with that of H b + in Fig. 1. The spectrum obtained by the reaction of untreated phenylhydrazine with H b + under nitrogen appeared to be a mixture of H b and the compounds of oxidized phenylhydrazine with Hb and Hhf. Addition of oxidized phenylhydrazine to HbCO under carbon monoxide resulted in a mixture of HbCO and the H b compound. According to Rekasheva and Lliklukhin; oxidation of phenylhydrazine by ferricyanide to benzene and nitrogen involves intermediate formation of the unstable coriipourid, nionophenyl diimide (C6HbN=NH). Beaven :ind White4 suggested the same compound as a possible intermediate procluct in the oxidation of phenylhydrazine in the presence of HbOa and noted the analogy between phenylhydrazine and phenylhydroxylamine. Phenylhydroxylamine is oxidized by oxygen or H b + to nitrosobenzene, which coordinates with I-Ib.5,6 Pauling7 has postulated that coordination with H b is restricted to “molecules with such electronic structure thdt they are able to conibine with the electrically neutral (iron) atoni of the ferroheme group without. changing its electronic charge.” Nitrosobenzene arid nionophenyl diimide have structures that meet this restriction,
Vol. 83
Preliminary experiments have indicated that oxidized phenylhydrazine coordinates with ferriheme, ferrimyoglobin and ferricytochrome c, and that other derivatives of hydrazine such as naphthylhydrazine, methylhydrazine, and dimethylhydrazine coordinate with H b + in the presence of ferricyanide. SATIONAL INSTITUTE OF ARTHRITIS
METABOLIC DISEASES HARVEY A. ITANO BETHESDA14,MARYLAXD ELIZABETH A. ROBINSON RECEIVED JUNE 15, 1961 AND
CORRELATION OF INDICATOR RATIOS OF AZULENES WITH RATES OF ACID CATALYZED HYDROGEN EXCHANGE’
Sir:
In a recent paper, Kresge and Chiang2 reported that in aqueous perchloric acid the indicator ratio, I = C B ~ of 1,3,5-trimethoxybenzene correlates with the I ~ acidity R function or equivalently that a plot of log I versus -110 is linear with a slope 2.0. Since the rate of the acid catalyzed detritiation of tritiated trimethoxybenzene follows Ho with a slope close to ~ n i t y ,Kresge ~ , ~ and Chiang drew the conclusion that the transition state for exchange was only part way along toward the conjugate acid. There are two areas of uncertainty in these indicator ratio studies. One is that protonation may be on oxygen rather than on carbon; a second is that diprotonation may be occurring. We wish to report kinetic and equilibrium studies with a system for which these points can be given specific consideration, the hydrocarbon azulene4 and some of its substitution products. The most basic site of azulene is the 1 (or 3 ) carbon; the spectrum of the conjugate acid and also its proton exchange properties are consistent .. + ..with a conjugate acid which involves a tetrahedrally and the structures Fe=-O--n’--C~H6 and Fe= + .. bonded carbon a t this site,j ;.e., protonation on NH--N-C~H6, respectively, can be written with carbon. Recent studies of the n.m.r. spectrum .. use of two unpaired electrons of the iron atoni. support this and offer evidence that only monoLye have confirmed this last H b + forms bonds with molecules in which the elec- protonation point by making conductivity studies of azulene tron pair used in the bond is contributed by the attached molecule.7 Ir~.accorda:ice with this prop- in ariliyclrous sulfuric acid. At two concentrations erty the structure Fe--N=N---C6Ha can be written the cmductivity (which is due almost entirely to for the compound of H b + and oxidized phenyl- the bisulfate ions that are formed) is very close to hydrazine. Dissociation of a hydrogen ion from that of solutes which monoprotonate, e.g., benzoic and p-nitroaniline, and is only about half that rnonophenyl diimide results in a neutral com- acid of a solute, p-phenylenediamine, which diprotopound. The proposed structures are consistent nates. with the electrophoretic behavior of the conlpoullds Indicator ratios for aqueous solutions are shown of oxidized Phenylhydrazine with FIb and Hb. +8 in Fig. 1. These were measured a t 350 arid 274 Both have lower cationic mobilities than I-Ib+.’ nip for azulene, a t 279 uip for 1-inethylazulene and Neither benzene nor nitrogen, the final 1)roducts a t 370 nip fnr 1-nitroazulene. For all three d(10g of oxidation of phenylhydrazine by ferricyanide, I ) / c i C a +is 0.8 0.1. However because of their coordinates with H b or Hb+. Lye therefore pro- tliflerent base strengths and because of the charpose that the absorption spectra of Fig. 1 result xteristics of‘ the $10 scale, the values of -dfrom the coordination of monophenyl diimide with (1) V;osk supported by a grant from t h e Atomic Energy COmmission. H b and Hb+.
*
(3) A. F. Rekasheva a n d G. P. Miklukhin, J. Gen. Chew. U.S.S.K., 24, 105 (1954) (English translation). (4) G. H. Braven and J , C . White, N a t u r e , 113, 389 (1954). ( 5 ) F. Jung, Biochem. J.. 305, 248 (1940). (6) L). Keilin and E. P. liartree, S n i u r e , 161, 340 (1943). (7) L. Pauling, Stanford M e d . BidL, 6, 215 (1948). (8) E.A. Robinson and H . A . Itano, unpiihlished stiiJies.
(2) A. J. Kresge and Y . Chiang, PYOC. Chcm. Soc., 81 (1961). (3) A. J. Kresge a n d Y .Chiang, J . A m . Chem. Sac., 81, 5509 (1959). (4) J . Colapietro a n d F. A. Long, Chem. and Ind., 1056 (1960). ( 5 ) For details and references see E. Heilbronner, “Azulenes, in NonBenzenoid Aromatic Compounds.” Interscience Publishers, Inc.. New York, N. Y . , 1959, Chap. V. (6) S. S. Danyluk and W. G . Schneider, J . A m . Chcm. Soc., 82, 997 (1980).
