7542
and biphenyls can be reduced in the 0 to -3 V range, and physical properties the effects of the large nonbut trifluoroethoxy, phenoxy, o-dioxyphenyl, and bonded interactions between ortho substituents. Commethoxy units cannot. Benzene cannot be reduced in pound 1 (Figure l), mp 156-157”, was prepared from the 0 to -3 V range but, when a phenyl group is bonded dimesitylcarbinol and 1,3,5-trimethoxybenzene by treatdirectly to phosphorus in I or 11, the reduction potenment with sulfuric acid in acetic acid. It shows untial is lowered until it is comparable to that of free naphprecedentedly large interactiops of this sort (a) in nmr thalene. This result is consistent with observations refor the evidence for hindered rotation, (b) in JlaC-H ported by Santhanam and Bard for triphenylphosphine methane proton, and (c) in a structure determined by and triphenylphosphine oxide reductions. Thus, it X-ray crystallography. appears that the reducibility of the phenylcyclophosBelow -30” compound 1 shows three peaks (with phazenes I and I1 results from delocalization effects inareas 3:6:3, separated by about 12 Hz at 60 MHz) volving the phenyl groups and the skeleton. for four nonequivalent o-methyls and two peaks (of This interpretation is confirmed by the esr data. The area 3 and separated by 20 Hz) for two o-methoxyls. spectra derived from 111, IV, and V can be rationalized The methyl peaks coalesce to a single peak near -2O”, in terms of hyperfine splittings within each discrete side but the methoxyl peaks require a much higher temperagroup unit only. For example, the spectra are similar ture, near 145”. Line-shape analyses4 over the temto those of p-nitroanisole, or the appropriate disubperature range 118-176” for the nmr signals for the stituted naphthalenes. Presumably the oxygen atoms o-methoxyl groups show AH* = 17.7 f 0.4 kcal/mol effectively insulate the reduced organic aromatic comand AS* = -9.8 f 1.0 eu for the process interconponent from the phosphazene ring. This is further verting exo- and endo-methoxyl groups. These data confirmed by the nonreducibility of [NP(OC6H5)2]3and suggest the importance in solution of a distorted pro[NP(OC6H&], below - 3 V. peller conformation similar to that established by X-ray However, the esr singlet obtained from I and I1 analysis for crystalline 1. The higher energy barrier must indicate delocalization of the unpaired electron for the position exchange of methoxyl groups than for either within a C6H5-P-C6H5 unit or into the phosmethyl groups in 1 is in accord with the order of size phazene skeleton as a whole. Interaction within a (CH3 > OCH,) established from rate data for biphenyl CsHSP-C6H5 segment could give rise to as many as 150 racemization6 and from consideration of the relative lines (648 lines if the phenyl groups are nonequivalent), van der Waals radii of oxygen and methyl groups6 if and this is probably beyond the resolution limit for this the detailed mechanism for rotation about the C,-aryl system. Delocalization involving the whole s k e l e t ~ n ~ - ~bond is considered. Individual steps in this rotation is also possible, but the failure of the skeleton to reduce must involve a reversal of the pitch of the propeller, when aliphatic substituents are present, coupled with the with simultaneous 90” rotations of each ring about the almost identical results obtained from [NP(C6H&I3 and C,-aryl bond. [NP(C6H5)2]4, and the inability of long-chain species such The pictured process moves exo-methyl group ZI as [NP(OC6H5)z],and [NP(OCH2CF3)2],to stabilize an (in l a ) to an endo position (in lb) and moves Zz endo unpaired electron, force us to the view that extensive to exo. N o exo-endo interconversion of methoxyl skeletal delocalization does not occur under these congroups is effected by this process. It involves a “gearditions. It appears more likely that in phenylphosphameshing” correlation of the rotations of rings A and C zenes an unpaired electron can be delocalized within a and of rings B and C.’ The “gear-clashing” correlation C6H6-P-C6H5unit or, at the most, into a very short adof rotation of rings A and B which is necessary in this jacent skeletal segment. ls8 mechanism provides the largest steric interaction in the A more detailed discussion of these results and of transition state, that between endo substituents XZ additional work now in progress will be given in a sub(OCH,) and Y z(CH,). sequent publication. The related processes which would interconvert exoand endo-methoxyls in l a would involve rings B and C in Acknowledgment. We are indebted to Dr. M. D. a “gear-clashing” counterrotation with a resulting Morris for his suggestions concerning the polarographic large nonbonded transition-state interaction between technique. juxtaposed endo-methyls (e.g., Y 2and Z 2 ) in the transi(15) K. S. V. Santhanam and A. J. Bard, J . Amer. Chem. Soc., 90, tion state. This interaction would be expected to be 1118 (1968). less favorable than the methyl-methoxyl interaction H. R. Allcock, W. J. Birdsall in the pictured process ( l a -+ lb) which exchanges Department of Chemistry, The Pennsylvania State University methyls. University Park, Pennsylvania 16802 As expected, the rotation of the methoxylated rings Received September 3, 1969 in 2, which involves only the much smaller methoxylmethoxyl interactions, is much faster, AH* = 8 f 1 kcal/mol and AS* = - 10 f 5 eu (over the range -69 Steric Effects in ortho-Substituted Triarylmethanes to - 8 1O)* as established by the line-shape analysis of the Sir: Triarylmethane derivatives in which all ortho posi(4) H. S. Gutowsky and C. H. Holm, J . Chem Phys., 25, 1228 ( 1956). tions have bulky substituents reflect l-, in their chemical (1) J. C. Martin, J . Chem. Educ., 38, 286 (1961); J. C. Martin and R. G. Smith, J . Am. Chem. Soc., 86,2252 (1964); M. J. Sabacky, C . S. Johnson, Jr., R. G. Smith, H. S. Gutowsky, and J. C. Martin, ibid., 89, 2054 (1967). (2) F. A. Carey and H. S.Tremper, ibid., 90,2578 (1968). (3) N. Kessler, A. Moosmayer, and A. Rieker, Tetrahedron, 25, 287 (19 69).
Journal of the American Chemical Society
(5) R. Adams and H. C. Yuan, Chem. Rev., 12,261 (1933). (6) L. Pauling, “The Nature of the Chemical Bond,” Cornell University Press, 3rd ed, Ithaca, N. Y.,1960, pp 257-264. (7) For a discussion of such rotation in triphenylmethyl cations see A. K. Colter, I. I. Schuster, and R. J. Kurland, J . Am. Chem. Soc., 87, 2278 (1965); in diarylmethanes see H. Kwart and S. Alekman, ibid., 90, 4482 (1968). (8) In ref 3 the coalescence temperature for 2 was recently reported to
/ 91:26 / December 17, 1969
7543
x.
' a
x
