Comments on the Proposed Dibenzocyclobutadiene Dianion - Journal

Comments on the Proposed Dibenzocyclobutadiene Dianion. Richard Waack, Mary A. Doran, and Peter West. J. Am. Chem. Soc. , 1965, 87 (23), pp 5508–551...
0 downloads 0 Views 1MB Size
Mechanisms in which the transition state contains the hydroxide ion and the elements of SH+ (such as in an S N ~attack of OH- on the sulfur-conjugate acid) are not ruled out by the acidity dependence of kobsd. However, they are inconsistent with the other results obtained. A literature search failed to reveal any unambiguous example of the S N ~solvolysis of a neutral sulfide. Carbon-sulfur bond cleavage in general is much more difficult to achieve than carbon-oxygen bond ~ l e a v a g e . l ~ - ~ ~ However, in this instance the “carbonium” ion formed, I+, has a relatively high stability. The iminium ion character of the transition state no doubt is of great assistance in promoting the SN1 heterolysis of the carbon-sulfur bond.

-2.0

+ -3.0 y” cn

-

-4.0

-1.0

Figure 1. Plot of log

-0.5 0.0 log [H30+]. kobsd

0.5

against log [HaO+] for the reaction f

PhCH(&HMez)SEt

-+

PhCH-NMe,

+ EtSH

(12) D. S. Tarbell and D. P. Harnish, Chem. Rev., 49, 1 (1951). (13) N. Kharasch, Ed., “Organic Sulfur Compounds,” Vol. I, Pergamon Press, New York, N. Y., 1961. (14) W. A. Pryor, “Mechanisms of Sulfur Reactions,” McGraw-Hill Book Co., Inc., New York, N. Y., 1962.

W.M. Schubert, Yoshiaki Motoyama Department of Chemistry, University 01 Washington Seattle, Washington Received September 16, 1965

by eq. 3, where KI = a S a H 8 O f / a S H + a H z O . The term aI120fSH’/fHsO-ft, should change little in the predominantly aqueous solutions used, and hence kobsd should be inversely proportional to [H30+], as observed. Comments on the Proposed The assigned mechanism also accounts for the lack of Dibenzocyclobutadiene Dianion catalysis by formate ion or molecular formic acid. Sir : In complete agreement with the mechanism is the of finding of a solvent isotope effect We wish to report results of our spectroscopic meas5.9 ~-t 0.1 in several acidities for the ethyl sulfide urements of equilibrium constants for disproportiona(Figure 1). Since equilibrium 1 lies far to the left, tion of anthracene (ao) and biphenylene (bo) radical the concentration of S should be significantly greater in anions in tetrahydrofuran (THF) solution (sodium protio than in deuterio acid of the same m ~ l a r i t y . ~ gegenion) It can reasonably be assumed that the solvent isotope K 2H- + -+ HO + H 2effect on [SI is comparable to that on the acid dis(1) sociation constant of trimethylammonium ion. This is Equilibrium data have been reported for disproporreported to be 4.0 but is more likely 5.0.6 The second tionation of the radical anions of stilbene, tetraphenstep probably has a small positive isotope e f f e ~ t . ~ t l ~ 1,2,3,4-tetraphenylbutadiene, and cycloylethylene,2 The closest model is the reaction octatetraene, but no direct measurements of equilibrium constants for disproportionation. of aromatic PhCH(SEt)fjMe8+PhCH==&Et Me3N radical anions have been reported. P o l a r o g r a p h i ~ ~ ~ ~ for which kHzO/kD20is 1.2.” Thus the estimated total and potentiometric titration^^,^ indicate disproporsolvent isotope effect agrees well with that observed. tionation of aromatic radical anions is small. The transition state of the rate-controlling step of the It was recently r e p ~ r t e d ,based ~ on visible spectral assigned mechanism has sulfide ion character. In measurements, that in THF solutions biphenylene agreement with this, the relative rates for various radical anion (b’) (sodium gegenion) undergoes exa-dimethylaminobenzyl thio ethers are: Ph (398), tensive disproportionation, but the equilibrium conPhCI-I, (4.6), Et (l,O), t-Bu (0.8). A plot of log k stant was not measured. The authors interpreted their YS. log K R S H , where K R S H is the acidity constant of the observed disproportionation of b’ as evidence that thiol in water, is linear with slope -0.57. biphenylene dianion (b ,--) derives a special stability 2 n-electron molecule from being the closed shell 4n (4) I(. B. Wiberg, Chem. Rev., 55, 713 (1955). (5) In 1936, Schwarzenbach and Epprecht reported the following dibenzocyclobutadiene dianion.

+

+

KHA/KD& values: HzO (5.4), NH4’ (3.1), (CHa)aNH+ (4.0), CHsCOzH (2.9), and H2POd- (2.9).6 No later values for (CH3)3NH+ were found, but the generally accepted values for all the other acids are consistently larger than the earlier values, by a factor of 1.25 i 0.05: HzO (6.5),’-* NHa+ (4.06),Q CHsCOzH (3.33),’39 and HzP04- (3.62).8 Applying this factor to the early value for (CH&NH+ gives KHPJKHA= 5.0. (6) G. Schwarzenbach, A. Epprecht, and H. Erlenmeyer, Helv. Chim. Acm, 19, 1292 (1936). (7) C. A. Bunton and V. J. Shiner, Jr., J . Am. Chem. SOC.,83, 42 (1961). (8) R. P. Bell, “The Proton 1 ~ .Chemistry,” Cornell University Press, Ithaca, N. Y., 1959, p. 188. (9) F. A. Long, P. Salomaa, and L. L. Schlaeger, J . Phys. Chem., 68, 410 (1964). (10) C. A. Bunton and V. J. Shiner, J. Am. Chem. SOC.,83, 3207, 3214 (1961). (11) Y . Motoyama, unpublished results.

5508

Journal o j t h e American Chemical Society

(1) E. R. Zabolotny and J. F. Garst, J. Am. Chem. SOC.,86, 1645 (1964). (2) J. F. Garst, E. R. Zabolotny, and R. S. Cole, ibid., 86, 2257 (1964). (3) M. A. Doran and R. Waack, J . OrganomernL Chem., 3,94 (1965). (4) H. L. Straws, T. J. Katz, and G. K. Fraenkel, J. Am. Chem. SOC., 85, 2360 (1963). (5) G. J. Hoijtink, J. van Schooten, E. deBoer, and W. Y . Aalbersberg, Rec. Truv. Chim., 73,355 (1954). (6) S. Wanzonek, E. W. Blaha, R. Berkey, and M. E. Runney, J . Electrochem. SOC.,102, 234 (1955). (7) G. J. Hoijtink, E. deBoer, P. H. van der Meij, and W. P. Weilland, Rec. Trav. Chim., 75, 487 (1956). ( 8 ) J. Jaguar-Grodzinski, M. Feld, S. L. Yang, and hf. Szwarc, J . Phys. Chem., 69, 628 (1965). (9) N. L. Bauld and D. Banks, J. Am. Chem. SOC.,87, 128 (1965).

/ 87:23 / December

5, 1965

The equilibrium constant measurements given here are based on electronic absorption spectra in the ultraviolet and visible using methods described previouslv.'o THF solutions of bo and ao were exDosed to a Na mirror, then poured off the metal &face prior to making spectral readings. Biphenylene radical anion has strong absorption maxima at 382 floe e 4.72) and 270 mu floe e 4.69)." BiDhenvlene dianion12 has an absorption maximum o i 3 4 i mp (log e 4.64). The strong absorptions of biphenylene at 249 (log e 4.99) and 241 mp (log e 4.74) were used to obtain the concentration of bo in equilibrium with band bZ-. Similarly the absorptions of anthracene radical anion (a-) at 258 (log e 4.90), 325 (log e 4.54), and 364 mp (log e 4.38), of anthracene dianion (a") at 325 (log e 4.77) and 605 mp (log e 4.28), and of the absorption maxima of anthracene at 245 (log e 4.97) and 253 mp (log e 5.29) were used to evaluate equilibrium data for the anthracene system.'* Because we find that the two low extinction absorption maxima in the visible (Le., 595 and 565 mp) are characteristic only of b', which is contrary to the recent assignments that the shorter wave length 565 mp absorption arises from b*, we have illustrated our spectra forb', b*, and bo in Figure 1. Successive spectral measurements following the progress of these reactions have established that in experiments having initial bo or ao concentrations of 4 X le3 to 2 X 1W2 M disproportionation of both b and a- is so small it is not possible to simultaneously measure spectroscopically all three species i.e., Ha, H I , and H". These experimentsestablishtheeffective K (disproportionation) of b- must be less than 1.5 X 1W2,which is in agreement with previous conclusions." A similar small value of K (Le.,