The oxoruthenium(IV) complexes [RuIV(terpy)(6,6‘-Cl2-bpy)O](ClO4)2 (1a; terpy = 2,2‘:6‘,2‘ ‘-terpyridine; 6,6‘-Cl2-bpy = 6,6‘-dichloro-2,2‘-bipyridine), [RuIV(terpy)(tmeda)O](ClO4)2 (1b; tmeda =
N,N,N‘,N‘-tetramethylethylenediamine), [RuIV(Cn)(bpy)O](ClO4)2 (1c; Cn = 1,4,7-trimethyl-1,4,7-triazacyclononane), and [RuIV(PPz*)(bpy)O](ClO4)2 (1d; PPz* = 2,6-bis[(4S,7R)-7,8,8-trimethyl-4,5,6,7-tetrahydro-4,7-methanoindazol-2-yl]pyridine) are effective for the epoxidation of aromatic
alkenes in acetonitrile at ambient conditions. Their reactions with cis-alkenes such as cis-β-methylstyrene and cis-β-deuteriostyrene afford epoxides nonstereospecifically. The observation of
the inverse secondary kinetic isotope effect for the β-d
2-styrene oxidations [k
H/k
D = 0.87 (1b), 0.86
(1d)], but not for α-deuteriostyrene (k
H/k
D = 0.98 for 1b and 1d), indicates that C−O bond formation
is more advanced at the β-carbon atom than at the α carbon, i.e., a stepwise mechanism. The
second-order rate constants (k
2) for the styrene oxidations are weakly dependent on the E°(RuIV/III)
values of the oxoruthenium(IV) complexes, and both electron-withdrawing and -donating para
substituents mildly accelerate the oxidation reaction of styrene. These findings discount strongly
the intermediaries of an alkene-derived cation radical and a carbocation. A linear free-energy
relationship between the second-order rate constants for the para-substituted styrene oxidations
and the total substituent effect (TE) parameters has been established: ρTE
• = +0.43 (R = 0.99) for
1b, +0.50 (R = 0.98) for 1c, and +0.37 (R = 0.99) for 1d (Wu, Y.-D.; Wong, C.-L.; Chan, K. W.; Ji,
G.-Z.; Jiang, X.-K. J. Org. Chem.
1996, 61, 746). This suggests that the oxidation of aromatic alkenes
by oxoruthenium(IV) complexes should proceed via the rate-limiting formation of a benzylic radical
intermediate. Oxidation of styrene and cis- and trans-β-methylstyrenes by the chiral oxoruthenium(IV) complex 1d attains moderate enantioselectivities, in which the production of cis-epoxide is
more enantioselective than the trans counterpart. The ligand dissymmetry of PPz* together with
the bipyridine ligand create a “chiral pocket” around the RuIVO moiety, leading to enantiofacial
discrimination through nonbonding interaction. Because the acyclic benzylic radical intermediate
would undergo cis−trans isomerization before the second C−O bond formation, the overall product
enantioselectivity (% eeobs) cannot be determined exclusively by facial selectivity (eefacial) of the first
irreversible C−O bond formation step. The extent of the isomerization, measured by the cis−trans-epoxide selectivity or diastereoselectivity of epoxide ring closure, is an important element in
controlling the enantiomeric excess of the epoxides.