Dec. 20, 1962
COMMUNICATIONS TO THE EDITOR
4985
rate of reaction 1 is very sensitive to C1-, and the sensitivity appears to be about the same as it is for the reaction of Cr++ with C O ( N H ~ ) ~ + + + . ~ Experimental. -R U ( N H ~ ) ~ Cwas & supplied by Johnson-Matthey & Co., London. Ru(NH~)~CI, was prepared by heating Ru(NH3)6Cls with concentrated hydrochloric acid.3 Ru(NH3)60H2(c104)3 was prepared from Ru(NH3)6C& by dissolving in NH3 aq. then acidifying with concenRU(NHs)6+++ f c r + + + RU(NHs)6++ Cr+++ (1) trated HC104. Ru was analyzed spectrophotoR U ( N H ~ ) ~ C I + +C r + + + metrically by the method of Woodhead and [ R U ( N H ~ ) ~ + + ]CrCl++ ~ (2) Fletcher4 and using the characteristic absorptions Ru(NH&0H2+++ Cr++ + of the various Ru(II1) ammine species.6 We Ru(NH&++ Cr+++ (3) have found E for Ru(NH3)60H2+++to be 757 M-l The 1:l stoichiometry has been established in cm.-I a t the absorption maximum, X = 268 mp. Acknowledgment. - This work was supported by reactions (1) and (2) with an accuracy of at least 10%. The Ru containing products of the reduc- the National Science Foundation under Grant tion of Ru(NH3)6Cl++ and Ru(NH8)60Hz+++ G20954. The spectrophotometer used was purby Cr++ appear to be identical. Both reactions chased under Grant G-22611 from the National produce a substance having e = 260 1M-l cm.-' Science Foundation. at 327.5 rnb (not a maximum for Ru(I1) but a (2) A. M. Zwickel and H. Taube, J . A m . Chem. Soc., 83, 793 (1961). (3) K.Gleu and K. Rehm, 2. anorg. Chem., 447, 237 (1936). maximum for R U ( N H ~ ) ~ C ~ +and + ) ,having a weak (4) J. L. Woodhead and J. M. Fletcher, J . Chem. Soc., 5039 (1961). lo2). The absorption maximum a t 420 mp ( e (5) Cr containing product of reaction 1 has been iden- (1957).H. Hartmann and C . Buschbeck, Z . physik. Chem., 11, 120 tified as Cr(&O)a+++, and CrCl++ has been shown DEPARTMENT OF CHEMISTRY to comprise at least 90% of the Cr product in reac- STANFORD UNIVERSITY JOHN F. ENDICOTT tion 2. STANFORD, CALIFORNIA HENRYTAUBE By reoxidizing the Ru(I1) species, it has been RECEIVED h?OVEblBER 15, 1962 shown that the Ru(II)-NH3 bonds remain intact in acidic solution (up to 0.2 M ) for at least one hour. SOME CARBENOID ADDITIONS A convenient oxidizing agent has proved to be STERIC COURSE OFTO OLEFINS' clod-. Rate studies were made of the reaction Sir: of Ru(NH3)6++ with clod- at 25" using a medium Additions of unsymmetrically substituted car(Na+, H+, C1-, clod-) at p = 0.14. The reaction is first order in Ru(NH3)6++ and ClO4- and zero benoid intermediates to olefins lacking a center of order in H + a t least in the range from to symmetry result in pairs of isomeric cyclopropanes. With the exception of carboethoxycarbene addiM , and the specific rate is 26 f 1 X M-I set.-'. When Ru(NH3)6++ is the reactant tions,2no proof of the configurations of the products has been given. For a number of cases, however, the reaction is slower by a factor of about 50. The half-life for the aquation of Ru(NH3)6Cl++ it has been assumed that steric hindrance in the is greater than IO5 sec., but when Ru(NH&++ transition state will be product controlling, and, is present, the reaction may be complete in a few being similar to that in the products, will lead to the minutes. Ru(NH3)6++is very efficient in bringing predominance of the isomer with the fewest nonbonded interactions. We wish to present into equilibrium a number of reactions of the type evidence that this assumption is not generally valid RU(NHI)SOHZ'++ X = Ru(NH~)L,"'X HnO and that the previously assumed configurations of Making use of this catalytic effect, the equilibrium some chlorocyclopropanes are in quotient in the reaction with X = C1- has been Treatment of benzal bromides 1-111 with alkyldetermined as 43 f 3 a t 25" and p = 0.1. The lithiums in the presence of olefins gave arylcycloreactions which bring about the equilibration are propanes in moderate yields. The same compounds also were obtained from photolysis of the Ru(NHJ)LOHP+++ Ru(NH~),++ C1-+ (4) corresponding aryldiazomethanes using olefins as Ru(NHa)bCl++ Ru(NHa)6+++ (5) solvents. With 1-butene (IV), cis-2-butene (V) and the specific rates have been determined ap- and 2-methyl-2-butehe (VI) as substrates the s e ~ . -and ~ 2 X lo2 expected isomers were formed in ratios as listed proximately as 4 X lo3 M-l =.-I. The ratio agrees within experimental in the table. error with the value which was measured for Assignment of configurations by n.m.r. and inthe equilibrium quotient. Substitution on Ru- dependent syntheses show that the predominantly (NHs)s++ apparently is not rate determining for the formed isomers have the configuration in which the catalysis under our conditions, and taking into larger number of alkyl groups and the aryl subaccount the concentration levels of the reagents, (1) This work was supported by a grant from The Petroleum Rewe conclude that for substitution on Ru(NH3)6++, search Fund, administered by The American Chemical Society. t l / 2 < 10 sec. (2) W. v. E. Doering and T. Mole, Telrohcdron, 10, 65 (19130); The s@c rates of reactions 1 and 2 are m. P.S.Skell and R. M. Etter, Proc. Chcm. Soc., 443 (1961). (3) (a) G. L. Closs and L. E. Closs, J . A m . Chcm. Soc., 84, 5723 1 X lo2 M-l sec.-I and 8 X lo8 M-l sec.-l. The
change and the comparison of the results with those
already obtained for the Co(II)-Co(III) ammines is of significance. Some of the observations made in the current study seem of particular interest and are reported herewith. In the reduction of Ru(NH&+++, Ru(NHs)hCl++ or R U ( N H ~ ) ~ O H ~ by + +Cr++, + we fmd that the net changes can be described by the equations
+ + +
+
+
-
+
+
+
+
+
(1) This species is almost certainly Ru(NHa)sOHr++, but experimental proof that it actually is heraco6rdinated is lacking.
(1960); (b) E.E. Schweizer and W. E. Parham, ibid., 84, 4085 (1960); (c) U. Sch6llkopf. and G. J. Lehmann,, Telrahedron L e f l e r s , 4, 1 G 5 (1962).
COMMUNICATIONS TO TH& EDITOR
4986
VOl. 84
stituent are located on the same side of the ring figurations to some chlorocyclopropanes, obtained (subsequently referred to as the cis configuration). from additions of chlorocarbene to olefins, are in Chemical shifts of the alkyl proton resonances of error.8P We have now synthesized the two adducts adducts VI1 (both methylene and methyl protons) of chlorocarbene to 1-butene from cis and truns and VI11 and of two methyl groups of adducts 1-chloro-1-butene v i a the Simmons-Smith reaction.' I X are 0.18 to 0.45 p.p.m. higher in the predominant Identity of the product obtained from truns-lisomers than the corresponding resonances of the chloro-1-butene with the minor adduct of the carminor adducts. Deriving geometrical parameters bene reaction and vice vena established that here from scale models, it can be shown that the cis again the major product of chlorocarbene addition oriented alkyl groups are located in the diamagnetic has the cis orientation and that the original assignshielding cone of the aryl ring' in or near those ments should be reversed. conformations of the aryl substituent which can be A possible explanation of the observed stereoexpected to be the most stable. In contrast the chemistry may be found in the assumption of a trans alkyl substituents find themselves in the transition state with considerable charge separaparamagnetic zone in any conformation. Sim- tion. Partial delocalization of opposite charges ilarly, the observed larger shielding of the benzylic over the alkyl groups by hyperconjugative effects and over the aryl and chlorine substituent, respecX-C6H,-CHBrZ tively, by inductive effects, will lead to smaller I, X = H, 11, X CHI, 111, X = C1 over-all charge separation in the cis transition state. Observed solvent dependence, leading to smaller R3 ratios in more polar solvents, such as ether, concur with this explanation. loThe striking similarity of the steric course of these carbene additions Ar R2 with the Diels-Alder reaction should be pointed out, and, perhaps, may be regarded to be more VIIa, R1 = CtHs, RZ = Rs = H VIIIa, RI = RJ = CH,, Rt = H than just a coincidence.l1
RJ
IXa, R1 = Rz = R3 = CH3 H VIIb, RI = GHb, RI = Rs VIIIb, R1 = Rs = CHa, RL 3 H IXb, R I = Rz = Ra = CH,
(9) This assumption is strongly supported by the established electrophilic nature of carbenoid intermediates: W. v. E. Doering and W. A. Henderson, 1.Am. Chcm. SOC.80, 5274 (1958); G. L. Closs and G. M. Schwartz, ibid.. 84, 5729 (1960). (10) Naturally, other factors, such as the degree of bond formation ISOMERRATIOSAS OBTAINEDFROM BENZALBROMIDES~in the transition state, determined in part by the ground state stability of A N D PHOTOLYSIS OF ARYLDIAZOMETHANES (IN PARENTHE- the carbenoid intermediate dil influence the magcitude of the isomer ratio as well. The small but Wperimentally significant preference for SES) WITH OLEFINS IV, AND VI* the formation of same isomer in the diazo compound photolysis indiAr = CsH8 .4r = P-CHa-CsH, Ar = fl-CI-CrHa cates t h a t the probable incorporation of a molecule of lithium halide i n the transition state of t h e =-elimination reaction is not the sole VIIa/VIIh 2.1 (1.0) 2.8(1.3) 2 . 1 (1.1) factor in determining the stereochemistry. VIIIa/VIIIb 2 . 4 (1.1) 4.5(1.4) 3.4 (1.2) (11) Satisfactory analyses have been obtained for all new comIXa/IXb 1.3 (1.1) 1.4 (1.3) 1.4(1.1) pounds. n-Pentane used as solvent. Reaction temperature in (12) A. P. Sloan Foundation Fellow, 1962-1964. both series, 10". (13) National Srience Foundation Predoctoral Cohperative Fellow, 1961-1963.
v
-
*
ring protons of the minor adducts are in line with expectations considering the diamagnetic effect of the cis oriented ring alkyl carbon carbon bonds6 Further support for this assignment is found in the larger spin-spin couplings of the vicinal ring protons with the benzylic protons in the predominant isomers. The stronger coupling has been demonstrated to be associated with cis orientation in a number of cyclopropanes with known configurat i o m 6 Finally, independent syntheses of the minor adducts to IV from trans-1-aryl-1-butenes v i a the stereospecific Simmons-Smith reaction' render the assignments of configurations unambiguous. The above mentioned correlations of relative magnitudes of vicinal proton spin spin coupling and of chemical shifts of ring protons with configurations8 on three membered rings made it very probable that the previous assignments of con(4) Determined according t o C. E. Johnson and F. A. Bovey, J . Chem. Phys., 29, 1012 (1958). (5) J. I. Musher, ibid., S6, 1159 (1961), and private communication. (6) J. D. Graham and M. T. Rogers, J. Am. Chcm. Soc., 84, 2249
(1962), and references cited therein. (7) H. E. Simmons and R . D. Smith, ibid., 81, 4256 (1959). (8) The predominantly formed adducts of chlorocarbene t o IV, V and VI show larger vicinal spin spin coupling and decreased shielding of the proton located on the carbon bearing the chlorine atom.&
DEPARTMENT OF CHEMISTRY G. L. C L O S S ~ ~ THEUNIVERSITY OF CHICAGO R. A. Moss1* CHICAGO 37, ILLINOIS J. J. COYLE RECEIVED SEPTEMBER 5, 1962
ON THE COPOLYMERIZATION OF BENZENE
Sir : Considerable interest was generated by claims that benzene copolymerizes with vinyl acetate,' methyl methacrylate,* and ~ t y r e n e . ~ More recently, however, the vinyl acetate work has been refuted4 and the styrene claim withdrawn.6 Both sets of workers4J found radioactive impurities in their benzene which led to erroneous results. Although the matter seems settled for these two monomers, the question of whether copolymeriza(1) W. H. Stockmayer and L. H . Peebles. J. A m . Chcm. Soc., 7 6 , 2278 (1953); L. H . Peebles. Jr., J. T. Clarke and W. H. Stockmayer, ibid., 82, 4780 (1960). (2) D. B. Anderson, G. M. Burnett and A. C. Gowan, IUPAC Symposium on Macromolecular Chemistry, Moscow, 1960, Sec. 11, p. 111. (3) G. Henrici-Olive and S. Olive, Makromol. Chem., 48, 237 (1Y61). (4) J. W. Breitenhach, G. Billek, G. Falthansl and E. Neber, Monafsh. Chem., 92, 1100 (1961). (5) G. Henrici-Olive and S. OlivC, Makromol. Chem., 61, 236 (1962).
