A mechanistic investigation of alkene epoxidation by sterically encumbered trans-dioxoruthenium(VI) porphyrins

Article abstract of DOI:10.1021/jo990517x

The highly substituted dioxoruthenium(VI) porphyrins [Ru(VI)(DPP)O₂] (1a; H₂DPP = 2,3,5,7,8,10,12,13,15,17,18,20-dodecaphenylporphyrin), [Ru(VI)(TDCPP)O₂] (1b; H₂TDCPP = meso-tetrakis(2,6- dichlorophenyl)porphyrin), and [Ru(VI)(TMOPP)O₂] (1c; H₂TMOPP = meso- tetrakis(2,4,6-trimethoxyphenyl)porphyrin) are competent oxidants for alkene epoxidation. The oxidations were carried out in a CH₂Cl₂/Hpz solution, and a paramagnetic bis(pyrazolato)ruthenium(IV) porphyrin, [Ru(IV)(Por)(pz)₂] (2; H₂Por = H₂DPP, H₂TDCPP, H₂TMOPP), was isolated and characterized. For the oxidation of cis-alkenes, stereoselectivity is dependent upon both the alkenes and the ruthenium oxidants, and it decreases in the order: cis- stilbene > cis-β-methylstyrene > cis-β-deuteriostyrene. The observation of inverse secondary KIE for the oxidation of β-d₂-styrene [k(H)/k(D) = 0.87 (1a); 0.86 (1b)] but not for the α-deuteriostyrene oxidations suggests that the C-O bond formation is more advanced at the C(β) atom than at the C(α) atom of styrene, consistent with a nonconcerted mechanism. By consideration of spin delocalization and polar effects, the second-order rate constants for the oxidation of para-substituted styrenes by complexes 1a-c can linearly correlate with the carboradical substituent constants σ(mb) and σ(JJ)· (Jiang, X.-K. Acc. Chem. Res. 1997, 30, 283). This implies that the styrene oxidation by the dioxoruthenium(VI) porphyrins should involve rate-limiting generation of a benzylic radical intermediate, and the magnitude of |ρ·(JJ)/ρ(mb)| > 1 suggests that the spin delocalization effect is more important than the polar effect in the epoxidation reactions. The spontaneous epoxidation of trans-β-methylstyrene by the sterically encumbered [Ru(VI)(TDCPP)O₂] and [Ru(VI)(TMOPP)O₂] complexes and the comparable ΔS((+)) values for their reactions with trans-β-methylstyrene and styrene are incompatible with the 'side-on approach' model; a 'head-on approach' model is implicated.

Full text of DOI:10.1021/jo990517x

J . Org. Chem. 1999, 64, 7365-7374
7365
A Mech a n istic In vestiga tion of Alk en e Ep oxid a tion by Ster ica lly
En cu m ber ed tr a n s-Dioxor u th en iu m (VI) P or p h yr in s
Chun-J in Liu, Wing-Yiu Yu,* Chi-Ming Che,* and Chi-Hung Yeung
Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong
Received March 23, 1999
The highly substituted dioxoruthenium(VI) porphyrins [RuVI(DPP)O2] (1a ; H2DPP ) 2,3,5,7,8,10,-
12,13,15,17,18,20-dodecaphenylporphyrin), [RuVI(TDCPP)O2] (1b; H2TDCPP ) meso-tetrakis(2,6-
dichlorophenyl)porphyrin), and [RuVI(TMOPP)O2] (1c; H2TMOPP ) meso-tetrakis(2,4,6-trimethoxy-
phenyl)porphyrin) are competent oxidants for alkene epoxidation. The oxidations were carried out
in a CH2Cl2/Hpz solution, and a paramagnetic bis(pyrazolato)ruthenium(IV) porphyrin, [RuIV(Por)-
(pz)2] (2; H2Por ) H2DPP, H2TDCPP, H2TMOPP), was isolated and characterized. For the oxidation
of cis-alkenes, stereoselectivity is dependent upon both the alkenes and the ruthenium oxidants,
and it decreases in the order: cis-stilbene > cis-â-methylstyrene > cis-â-deuteriostyrene. The
observation of inverse secondary KIE for the oxidation of â-d2-styrene [kH/kD ) 0.87 (1a ); 0.86 (1b)]
but not for the R-deuteriostyrene oxidations suggests that the C-O bond formation is more advanced
at the C(â) atom than at the C(R) atom of styrene, consistent with a nonconcerted mechanism. By
consideration of spin delocalization and polar effects, the second-order rate constants for the
oxidation of para-substituted styrenes by complexes 1a -c can linearly correlate with the
•
carboradical substituent constants σmb and σJ J (J iang, X.-K. Acc. Chem. Res. 1997, 30, 283). This
implies that the styrene oxidation by the dioxoruthenium(VI) porphyrins should involve rate-limiting
generation of a benzylic radical intermediate, and the magnitude of |F•J J /Fmb| > 1 suggests that the
spin delocalization effect is more important than the polar effect in the epoxidation reactions. The
spontaneous epoxidation of trans-â-methylstyrene by the sterically encumbered [RuVI(TDCPP)O2]
and [RuVI(TMOPP)O2] complexes and the comparable ∆S‡ values for their reactions with trans-â-
methylstyrene and styrene are incompatible with the “side-on approach” model; a “head-on
approach” model is implicated.
In tr od u ction
loporphyrin complexes, however, remains less under-
stood. Toward this end, our approach is to scrutinize
closely the stoichiometric reactions of organic substrates
with well-characterized and highly reactive oxo-metal
complexes.7-12
Metalloporphyrin-catalyzed alkene epoxidations and
alkane hydroxylations have been widely invoked as the
biomimetic reactions of cytochrome P450.1 These reac-
tions are often characterized by remarkable regio- and
stereoselectivities, and systematic structural variation of
the porphyrin ligands has proved to be a useful strategy
to achieve electronic and steric tuning of the catalysts.2
Highly enantioselective epoxidation of unfunctionalized
alkenes relying solely on nonbonding interactions has
also been developed by using chiral metalloporphyrin
catalysts.3,4 Reactive metal-oxo (MdO) complexes are
often invoked as reactive intermediates in the metal-
loporphyrin-catalyzed organic oxidations,2,5 and isolation
and/or characterization of some oxometalloporphyrin
complexes of Cr, Mn, and Fe have been reported.6 The
structure-reactivity relationship of reactive oxometal-
(5) (a) Chang, C. K.; Kuo, M.-S. J . Am. Chem. Soc. 1979, 101, 3413.
(b) Groves, J . T.; Haushalter, R. C.; Nakamura, M.; Nemo, T. E.; Evans,
B. J . J . Am. Chem. Soc. 1981, 103, 2884. (c) Yamaguchi, K.; Watanabe,
Y.; Morishima, I. J . Chem. Soc., Chem. Commun. 1992, 1721. (d)
Ostovic, D.; He, G.-X.; Bruice, T. C. In Metalloporphyrins in Catalytic
Oxidations; Sheldon, R. A., Ed.; Marcel Dekker: New York, 1994;
Chapter 2, p 29. (e) Groves, J . T.; Lee, J .; Marla, S. S. J . Am. Chem.
Soc. 1997, 119, 6269.
(6) For [Fe(IV)dO]: (a) Chin, D.-H.; Balch, A. L.; La Mar, G. N. J .
Am. Chem. Soc. 1980, 102, 1446, 5945. (b) Balch, A. L.; Chan, Y.-W.;
Cheng, R.-J .; La Mar, G. N.; Latos-Grazynski, L.; Renner, M. W. J .
Am. Chem. Soc. 1984, 106, 7779. For [Cr(IV)dO]: (c) Groves, J . T.;
Kruper, W. J .; Haushalter, R. C.; Butler, W. M. Inorg. Chem. 1982,
21, 1363. For [Mn(IV)dO]: (d) Groves, J . T.; Stern, M. K. J . Am. Chem.
Soc. 1988, 110, 8628.
(7) (a) Ho, C.; Lau, T.-C.; Che, C.-M. J . Chem. Soc., Dalton Trans.
1991, 1901. (b) Ho, C.; Leung, W.-H.; Che, C.-M. J . Chem. Soc., Dalton
Trans. 1991, 2933. (c) Che, C.-M.; Li, C.-K.; Tang, W.-T.; Yu, W.-Y. J .
Chem. Soc., Dalton Trans. 1992, 3153. (d) Cheng, W.-C.; Yu, W.-Y.;
Li, C.-K.; Che, C.-M. J . Org. Chem. 1995, 60, 6840. (e) Fung, W.-H.;
Yu, W.-Y.; Che, C.-M. J . Org. Chem. 1998, 63, 7715.
(1) Cytochrome P-450: Structure, Mechanism and Biochemistry, 2nd
ed.; Oritz de Montellano, P. R., Ed.; Plenum Press: New York, 1995.
(2) (a) Meunier, B. Chem. Rev. 1992, 92, 1411. (b) Dolphin, D.;
Traylor, T. G.; Xie, L. Y. Acc. Chem. Res. 1997, 30, 251.
(3) (a) Groves, J . T.; Myers, R. S. J . Am. Chem. Soc. 1983, 105, 5791.
(b) Collman, J . P.; Zhang, X.; Lee, V. J .; Uffelman, E. S. Brauman, J .
I. Science 1993, 261, 1404. (c) Naruta, Y. In Metalloporphyrins In
Catalytic Oxidations; Sheldon, R. A., Ed.; Marcel Dekker: New York,
1994; Chapter 8, p 241. (d) Campbell, L. A.; Kodadek, T. J . J . Mol.
Catal. 1996, 113, 293. (e) Gross, Z.; Ini, S. J . Org. Chem. 1997, 62,
5514. (f) Collman, J . P.; Wang, Z.; Straumanis, A.; Quelquejeu, M. J .
Am. Chem. Soc. 1999, 121, 460.
(4) (a) Lai, T.-S.; Zhang, R.; Cheung, K.-K.; Kwong, H.-L.; Che, C.-
M. Chem. Commun. 1998, 1583. (b) Lai, T.-S.; Kwong, H.-L.; Zhang,
R.; Che, C.-M. J . Chem. Soc., Dalton Trans. 1998, 3559. (c) Zhang, R.;
Yu, W.-Y.; Lai, T.-S.; Che, C.-M. Chem. Commun. 1999, 409.
(8) (a) Stultz, L. K.; Binstead, R. A.; Reynolds, M. S.; Meyer, T. J .
J . Am. Chem. Soc. 1995, 117, 2520. (b) Dobson, J . C.; Seok, W. K.;
Meyer, T. J . Inorg. Chem. 1986, 25, 1325.
(9) Samsel, E. G.; Srinivasan, K.; Kochi, J . K. J . Am. Chem. Soc.
1985, 107, 7606.
(10) Mayer, J . M. Acc. Chem. Res. 1998, 31, 441.
(11) (a) Bailey, C. L.; Drago, R. S. J . Chem. Soc., Chem. Commun.
1987, 179. (b) Goldstein, A. S.; Drago, R. S. J . Chem. Soc., Chem.
Commun. 1991, 21. (c) Goldstein, A. S.; Beer, R. H.; Drago, R. S. J .
Am. Chem. Soc. 1994, 116, 2424.
(12) Marmion, M. E.; Leising, R. A.; Takeuchi, K. J . J . Coord. Chem.
1988, 19, 1.
10.1021/jo990517x CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/11/1999

7366 J . Org. Chem., Vol. 64, No. 20, 1999
Liu et al.
F igu r e 1.
Recently, ruthenium porphyrins were shown to exhibit
excellent activities and selectivities for catalytic hydro-
carbon oxidations in homogeneous13-16 and heteroge-
neous17 conditions using mild and inexpensive oxidants
such as N2O,15 2,6-dichloropyridine N-oxide,15 and O2.4a,16
More importantly, high-valent dioxoruthenium(VI) por-
phyrin complexes can be isolated and have been found
to be competent oxidants for hydrocarbon oxidations.18
We previously prepared two dioxoruthenium(VI) com-
plexes of tetraphenylporphyrin (H2TPP) and 2,3,7,8,12,-
13,17,18-octaethylporphyrin (H2OEP),18b and they can
bring about selective alkene epoxidations and saturated
C-H bond oxidations.7b Kinetic studies revealed that the
oxidation reactions involve the rate-limiting association
of oxoruthenium and substrates. This suggests that it is
possible to exercise regio- and stereocontrol of the reac-
tions by introducing bulky substituents near the OdRu
moiety. As anticipated, a well-characterized [RuVI(D4-
Por*)O2] complex [D4-H2Por* ) 5,10,15,20-tetrakis-
(1S,4R,5R,8S-1,2,3,4,5,6,7,8-octahydro-1,4:5,8-dimethano-
anthracen-9-yl)porphyrin] was found to effect enantio-
selective epoxidation of styrene and derivatives in up to
70% ee.4a,b To further investigate the role of sterically
encumbered porphyrins in the oxidation chemistry of
oxometalloporphyrins, we have prepared and examined
the reactivities of a series of dioxoruthenium(VI) com-
plexes supported by highly substituted porphyrins (Fig-
ure 1), [RuVI(DPP)O2] (1a ; H2DPP ) 2,3,5,7,8,10,12,13,-
15,17,18,20-dodecaphenylporphyrin), [RuVI(TDCPP)O2]
(1b; H2TDCPP ) meso-tetrakis(2,6-dichlorophenyl)por-
phyrin), and [RuVI(TMOPP)O2] (1c; H2TMOPP ) meso-
tetrakis(2,4,6-trimethoxyphenyl)porphyrin). The porphy-
rin macrocycle of [RuVI(DPP)O2] is saddle-shaped distorted
as a result of steric interaction among the phenyl
substituents; preliminary studies on its preparation and
reactivities have been reported.18d The facile oxidation
of trans-â-methylstyrene by the bulky [RuVI(TDCPP)O2]
and [RuVI(TMOPP)O2] complexes are incompatible with
the commonly accepted “side-on approach” model for the
metalloporphyrin-catalyzed epoxidations, and a “head-
on approach” model is implicated.
Resu lts a n d Discu ssion
P r ep a r a tion a n d Ch a r a cter iza tion of [Ru VI(P or )-
O2] (H2P or ) H2DP P , H2TDCP P , a n d H2TMOP P ).
The dioxoruthenium(VI) porphyrins were prepared by the
oxidation of the ruthenium(II) carbonyl porphyrins in
methylene chloride using m-chloroperoxybenzoic acid.
After chromatographic purification on an alumina column
using CH2Cl2 as the eluant, the complexes were isolated
1
in ca. 75% yield. The H NMR spectra of the [RuVI(Por)-
O2] products are typical of a D4-symmetric diamagnetic
complex, and the diamagnetism is in agreement with a
singlet (dxy)2 electronic ground state (considering OdRud
O axis as the z-direction). The eight pyrrolic protons of
[RuVI(TMOPP)O2] (1c) are located at δH ) 8.97 ppm.
Infrared analyses of the complexes revealed intense
absorptions at 818 (1a ), 825 (1b), and 820 (1c) cm-1
assignable as the asymmetric OdRudO stretches, and
the oxidation state marker bands are located at 1012 (1a ),
1019 (1b), and 1018 (1c) cm-1 in accord with a Ru(VI)
formulation. The UV-vis spectra of [RuVI(Por)O2] are
characterized mainly by the Soret band (B band) and a
less intense Q-band typical of a normal porphyrin.18 As
noted earlier, both the Soret and Q-bands of [RuVI(DPP)-
O2] were found to be red-shifted compared to those of the
planar [RuVI(TPP)O2] analogue;18d the steric-induced
macrocyclic distortion has led to a larger destablization
of the ligand’s HOMOs than of its LUMOs, thereby
narrowing the HOMO-LUMO gap.19,20
(13) (a) Mlodnicka, T.; J ames, B. R. Metalloporphyrins Catalyzed
Oxidations; Montanari, F., Casella, L., Eds.; Kluwer: Dordrecht, The
Netherlands, 1994; p 121. (b) Gross, Z.; Ini, S.; Kapon, M.; Cohen, S.
Tetrahedron Lett. 1996, 37, 7325.
(14) Groves, J . T.; Roman, J . S. J . Am. Chem. Soc. 1995, 117, 5594.
(15) (a) Higuchi, T.; Ohtake, H.; Hirobe, M. Tetrahedron Lett. 1989,
30, 6545. (b) Higuchi, T.; Ohtake, H.; Hirobe, M. Tetrahedron Lett.
1991, 32, 7435. (c) Ohtake, H.; Higuchi, T.; Hirobe, M. Tetrahedron
Lett. 1992, 33, 2521. (d) Groves, J . T.; Bonchio, M.; Carofiglio, T.;
Shalyaev, K. J . Am. Chem. Soc. 1996, 118, 8961.
(16) (a) Groves, J . T.; Quinn, R. J . Am. Chem. Soc. 1985, 107, 5790.
(b) Marchon, J .-C.; Ramasseul, R. Synthesis 1989, 389.
(17) (a) Liu, C.-J .; Li, S.-G.; Peng, W.-Q.; Che, C.-M. Chem. Commun.
1997, 65. (b) Liu, C.-J .; Yu, W.-Y.; Li, S.-G.; Che, C.-M. J . Org. Chem.
1998, 63, 7364.
(18) (a) Groves, J . T.; Quinn, R. Inorg. Chem. 1986, 25, 3845. (b)
Leung, W.-H.; Che, C.-M. J . Am. Chem. Soc. 1989, 111, 8812. (c) Maux,
P. L.; Bahri, H.; Simonneaux, G. J . Chem. Soc., Chem. Commun. 1994,
1287. (d) Liu, C.-J .; Yu, W.-Y.; Peng, S.-M.; Mak, T. C.-W.; Che, C.-M.
J . Chem. Soc., Dalton Trans. 1998, 1805.
(19) Gouterman, M. In The Porphyrins; Dolphin, D., Ed.; Academic
Press: New York, 1978; Vol. 3, Chapter 1.
(20) (a) Medforth, C. J .; Berber, M. D.; Smith, K. M.; Shelnutt, J .
A. Tetrahedron Lett. 1990, 31, 3719. (b) Takeuchi, T.; Gray, H. B.;
Goddard, W. A., III J . Am. Chem. Soc. 1994, 116, 9730.

Epoxidation by trans-Dioxoruthenium(VI) Porphyrins
J . Org. Chem., Vol. 64, No. 20, 1999 7367
Ta ble 1. Stoich iom etr ic Alk en e Oxid a tion s by Dioxor u th en iu m (VI) P or p h yr in s
Alk en e Oxid a tion s by tr a n s-Dioxor u th en iu m (VI)
P or p h yr in s a n d th e Isola tion of th e Bis(p yr a zola -
to)r u th en iu m (IV) P or p h yr in s. The [RuVI(Por)O2] com-
plexes 1a -c are competent oxidants for alkene oxidations
at ambient conditons. The results of the stoichiometric
oxidations are depicted in Table 1. In a typical reaction,
styrene (2 mmol) in a degassed CH2Cl2 solution [3 mL,
containing 2% w/w pyrazole (Hpz)] was treated with
[RuVI(Por)O2] (30 µmol) at room temperature, and the
mixture was stirred for 12 h. Styrene oxide was produced
in >90% yield for 1b and 1c. A trace amount of benzal-
dehyde (CdC bond cleavage) was also formed, and
phenylacetaldehyde (rearrangement) was not detected
(entry 1). At the end of the reactions, a bis(pyrazolato)-
ruthenium(IV) porphyrin complex, [RuIV(Por)(pz)2] (2),
was isolated and characterized. The molecular structure
of [RuIV(DPP)(pz)2] (2a ) had previously been established
by X-ray diffraction method.18d The [RuIV(Por)(pz)2] com-
plexes are paramagnetic, and the effective magnetic
moments (µeff) were measured to be 3.24 (2a ), 2.68 (2b)
and 2.88 µB (2c) (for spectral characterization data, see
the Experimental Section for details), consistent with the
Ru(IV) formulation.21
In a CH2Cl2/ethanol mixture, the [RuVI(Por)O2] com-
plexes reacted with norbornene or styrene to afford the
epoxides in comparable yields (see Table 1, entries 1 and
7) as in the cases when excess Hpz was used. However,
dark precipitates that are insoluble in common organic
solvents were formed at the end of the reactions. Infrared
spectral analyses of these solids did not reveal the
asymmetric OdRudO stretch, and a distinct absorption
around 1010 cm-1 was readily assignable as the oxidation
marker band of Ru(IV) porphyrins. Assuming a [Ru(Por)-
(OEt)2] formulation,21c their effective magnetic moments
were evaluated as 2.6-2.8 µB (solid sample).
In the absence of Hpz or ethanol, the reaction of [RuVI
-
(TDCPP)O2] (1b) with excess norbornene in CH2Cl2 was
monitored by infrared spectroscopy (1100-780 cm-1). The
intensities of the oxidation state marker band (1019 cm-1
)
and the asymmetric OdRudO stretch (825 cm-1) of the
starting Ru(VI) complex decreased isosbestically over 0.5
h with a concurrent development of an intense absorption
at 1010 cm-1, i.e., the formation of a Ru(IV) porphyrin
complex (Figure 2a). Upon prolonged standing of the
reaction mixture, this spectral feature was subsequently
replaced by a new absorption band at 1007 cm-1 assign-
able to the oxidation marker band of a Ru(II) porphyrin
complex (Figure 2b). The result is consistent with the
findings by Groves and co-workers that oxoruthenium-
(IV) porphyrin intermediate is unstable toward dispro-
portionation into Ru(VI) and Ru(II).22 Indeed, without
excess pyrazole or ethanol, the norbornene and cy-
(21) For examples of some structurally characterized paramagnetic
ruthenium(IV) porphyrins, see: (a) Huang, J .-S.; Che, C.-M.; Li, Z.-
Y.; Poon, C.-K. Inorg. Chem. 1992, 31, 1315. (b) Au, S.-M.; Fung, W.-
H.; Cheng, M.-C.; Peng, S.-M.; Che, C.-M. J . Chem. Soc., Chem.
Commun. 1997, 1655. (c) Cheng, S. Y. S.; Rajapakse, N.; Rettig, S. J .;
J ames, B. R. J . Chem. Soc., Chem. Commun. 1994, 2669.

7368 J . Org. Chem., Vol. 64, No. 20, 1999
Liu et al.
F igu r e 2. Infrared spectral traces for norbornene oxidation by [RuVI(TDCPP)O2] (1100-780 cm-1): (a) first 30 min of the reaction
and (b) over 14 h of prolonged standing of the reaction mixture. The absorption bands due to norbornene and CH2Cl2 have been
subtracted.
clooctene oxidations by 1 equiv of [RuVI(TDCPP)O2] in
degassed benzene gave 2 equiv of exo-2,3-epoxynorbor-
nane and 1.4 equiv of cyclooctene oxide, respectively.
Under such conditions, a carbonylruthenium(II) porphy-
rin (νCO ) 1950 cm-1) and some unidentified ruthenium
complexes were isolated at the end of the reactions.
P r od u ct An a lysis a n d Kin etic Mea su r em en ts.
Styrene, norbornene, and cyclooctene were converted to
the corresponding epoxides by 1a and 1b in 71-99%
yields, whereas lower epoxide yields (70-50%) were
obtained when 1c was used as the oxidant (Table 1,
entries 1, 7, and 8, respectively). Allylic C-H bond
oxidation prevailed for the cyclohexene oxidations by
[RuVI(Por)O2], and the combined yields of 2-cyclohexe-1-
ol and 2-cyclohexen-1-one were 83% (1a ), 44% (1b), and
42% (1c) (see entry 10).
All of the dioxoruthenium(VI) porphyrins studied here
are unreactive to trans-stilbene oxidation; however, they
readily oxidized trans-â-methylstyrene to afford trans-
epoxide selectively (entry 6). The cis-stilbene oxidations
by 1a -c were found to be highly stereoretentive (entry
3). In contrast, the stereoselectivity of the cis-â-methyl-
styrene oxidation varies with the nature of the ruthenium
oxidants. For example, when [RuVI(TDCPP)O2] (1b) and
[RuVI(TMOPP)O2] (1c) reacted with cis-â-methylstyrene,
only cis-â-methylstyrene oxide was produced in 84% and
58% yields, respectively. On the other hand, a significant
amount of the trans-â-methylstyrene oxide (14%) along
with the cis-epoxide (85%) were found when [RuVI(DPP)-
O2] (1a ) was the oxidant (entry 4).
erentially acts on the trisubstituted CdC bond. It is
known that the (+)-limonene oxidation by m-chloroper-
oxybenzoic acid would furnish a mixture of 1,2- and 8,9-
epoxides (ratio ) 8.9:1),23 and the reaction would result
in an equimolar amount of cis- and trans-1,2-epoxides
nonstereoselectively.23 In this work, we found that the
stoichiometric (+)-limonene oxidations by the dioxoru-
thenium(VI) porphyrins afforded cis-1,2-epoxide as the
major product; cis-: trans-epoxide ratios ) 55:27 (1a ), 55:
28 (1b), and 40:19 (1c) (see entry 9). The observed
stereoselectivity is indeed not inconceivable because the
facial approach leading to trans-1,2-epoxide is more
hindered as a result of the steric interaction between the
vinyl groups of the alkene and the substituents on the
porphyrin ring of the ruthenium complexes.
Figure 3 depicts the UV-vis spectral change for the
styrene oxidation by [RuVI(TDCPP)O2] (1b) in CH2Cl2
with 2% w/w pyrazole showing the isosbestic transfor-
mation from Ru(VI) to a Ru(IV) porphyrin (2b). Similar
spectral features were also observed for the reactions
involving the other dioxoruthenium(VI) porphyrin com-
plexes 1a and 1c. Isosbestic points were located at 431
and 468 nm for 1a (411 and 435 nm for 1b and 415 and
442 nm for 1c), suggesting that Hpz can spontaneously
react with the putative Ru(IV)-oxo intermediates to form
[RuVI(Por)(pz)2] without accumulation of intermediates.
For the styrene oxidation by [RuVI(DPP)O2] and [RuVI
-
(TDCPP)O2], ethanol was found to be equally effective
as Hpz for obtaining clean isosbestic spectral changes.
However, without pyrazole, the reaction involving the
less reactive [RuVI(TMOPP)O2] displayed complicated
kinetics in a CH2Cl2/EtOH mixture. It should be noted
that the reactions of all of the olefinic substrates utilized
When (+)-limonene containing an internal trisubsti-
tuted and an isolated terminal CdC bond reacted with
the dioxoruthenium(VI) complexes, 1,2-epoxides were
selectively produced and only a trace amount of 8,9-
epoxide was detected (entry 9), i.e., [RuVI(Por)O2] pref-
(23) (a) Groves, J . T.; Nemo, T. E. J . Am. Chem. Soc. 1983, 105,
5786. (b) Battioni, P.; Renaud, J . P.; Bartoli, J . F.; Reina-Artiles, M.;
Fort, M.; Mansuy, D. J . Am. Chem. Soc. 1988, 110, 8462.
(22) Groves, J . T.; Quinn, R. J . Am. Chem. Soc. 1985, 107, 5790.

Epoxidation by trans-Dioxoruthenium(VI) Porphyrins
J . Org. Chem., Vol. 64, No. 20, 1999 7369
Ta ble 2. Secon d -Or d er Ra te Con sta n ts (k2) for th e Oxid a tion of Alk en es by Dioxor u th en iu m (VI) P or p h yr in s
k2 × 103 (dm3 mol-1 s-1
)
entry
1
alkenes
1a
1b
1c
styrene
4.78 ( 0.11
3.90 ( 0.14
(3.63 ( 0.12)a
18.8 ( 1.1
0.61 ( 0.03
(0.50 ( 0.01)a
5.77 ( 0.14
1.70 ( 0.06
1.01 ( 0.06
1.26 ( 0.06
1.34 ( 0.08
0.57 ( 0.03
0.46 ( 0.02
0.39 ( 0.02
0.87 ( 0.04l
(0.53 ( 0.03)a
0.19 ( 0.01
0.29 ( 0.02
(0.22 ( 0.01)a
2
3
4
5
6
7
8
9
10
4-methoxystyrene
4-methylstyrene
4-fluorostyrene
4-chlorostyrene
3-nitrostyrene
cis-â-methylstyrene
trans-â-methylstyrene
R-methylstyrene
norbornene
25.2 ( 1.2
10.0 ( 0.4
6.01 ( 0.14
6.78 ( 0.18
5.55 ( 0.17
3.28 ( 0.18
2.59 ( 0.12
2.07 ( 0.10
3.76 ( 0.14
8.6 ( 0.2
5.06 ( 0.18
5.72 ( 0.15
5.18 ( 0.16
3.40 ( 0.14
1.69 ( 0.08
1.44 ( 0.07
12.7 ( 0.6
(15.0 ( 0.05)a
1.77 ( 0.09
1.21 ( 0.06
(1.04 ( 0.04)a
4.53 ( 0.12
3.86 ( 0.13
11
12
cis-stilbene
cyclooctene
1.73 ( 0.06
1.89 ( 0.06
13
14
â-d2-styrene
R-deuteriostyrene
5.49 ( 0.13
4.74 ( 0.11
a
The values in parentheses were obtained when the kinetic experiments were conducted in a CH2Cl2/EtOH mixture at 298 K.
Ta ble 3. Activa tion P a r a m eter s for th e Oxid a tion of Styr en e, cis-â-Meth ylstyr en e, tr a n s-â-Meth ylstyr en e, a n d
Nor bor n en e
1a
∆Hq (kcal mol-1
1b
1c
alkenes
)
∆Sq (eu)
∆Hq (kcal mol-1
)
∆Sq (eu)
∆Hq (kcal mol-1
)
∆Sq (eu)
styrene
14.6 ( 1.0
-(20.1 ( 2.0)
14.3 ( 1.0
-(21.9 ( 2.3)
-(25.6 ( 2.6)a
-(27.6 ( 2.1)
-(26.1 ( 1.9)a
-(32.4 ( 2.8)
-(23.7 ( 2.5)
14.0 ( 1.4
13.4 ( 1.2a
12.2 ( 0.9
-(27.3 ( 2.8)
-(29.0 ( 2.7)a
-(31.6 ( 2.4)
a
13.2 ( 1.0
norbornene
13.3 ( 1.0
-(25.0 ( 2.2)
11.8 ( 1.0
a
12.2 ( 1.0
cis-â-methylstyrene
trans-â-methylstyrene
11.2 ( 0.9
14.3 ( 1.2
13.8 ( 1.2
14.5 ( 0.9
-(27.3 ( 2.0)
-(23.7 ( 2.8)
a
The kinetic experiments were conducted in a CH2Cl2/EtOH solvent.
here, para-substituted styrenes, R-methylstyrene, cis-
and trans-â-methylstyrenes, and norbornene, exhibited
isosbestic spectral changes and followed clean first-order
kinetics under pseudo-first-order conditions, i.e., [alkene]
. [Ru], in a CH2Cl2 / 2% w/w Hpz solution.
The pseudo-first-order rate constants, kobs, were deter-
mined by monitoring the disappearance of the Soret band
of the dioxoruthenium(VI) complexes in CH2Cl2 with 2%
w/w Hpz; λmax ) 448 nm for 1a , 420 nm for 1b, and 424
nm for 1c. On the basis of the proposed reaction in
Scheme 1, the second-order rate constants, k2 (see Table
2), were evaluated from the linear plots of kobs versus
[alkene]. Rate saturation was not observed over the
alkene concentrations employed in this work. It is
noteworthy that all of the k2 values were unaffected by
changing the amount of Hpz from 1% to 10% w/w in CH2-
Cl2. Indeed, comparable k2 values were obtained for the
styrene, norbornene, and cyclooctene oxidations by [RuVI-
(TDCPP)O2] (1b) and [RuVI(TMOPP)O2] (1c) at 298 K,
regardless of employment of either the Hpz/CH2Cl2 or
EtOH/CH2Cl2 conditions (see Table 2). These verify that
Hpz or EtOH does not interfere with the rate-limiting
step of the epoxidation reactions.
F igu r e 3. UV-vis spectral trace (300-600 nm) for the
reaction of styrene (1.0 M) with [RuVI(TDCPP)O2] (1b) (10 µM)
in CH2Cl2 (with 2% w/w Hpz) at room temperature. Scan
interval: 2 min.
stoichiometric alkene epoxidations are consistent with
association of the reactants in the transition states.
Cis-Tr a n s Alk en e Selectivity of Ster ica lly En -
cu m ber ed Dioxor u th en iu m (VI) P or p h yr in ss“Sid e-
On ” ver su s “Hea d -On ” Ap p r oa ch . Stereoselective cis-
alkene oxidation is well documented for the metallo-
porphyrin-catalyzed epoxidations,2a and a side-on ap-
proach of alkene to a putative oxo-metal complex was
first proposed by Groves and co-workers.3a,23a This olefinic
approach is believed to be favored by a more effective
interaction between the filled π-orbital of alkene and the
dπ-pπ M-O antibonding orbitals. Later, Groves had
The temperature effect on the k2 values for the oxida-
tion of some representative alkenes has been studied. The
Eyring plots are linear over a temperature range of 19-
42 °C. The activation enthalpies ∆H‡ and entropies ∆S‡
for the styrene, trans-â-methylstyrene, cis-â-methylsty-
rene, and norbornene oxidations by 1a -c are shown in
Table 3. For the oxidation of styrene and norbornene,
small variations of the ∆H‡ and ∆S‡ values were noted
upon changing the solvent system from CH2Cl2/Hpz to
CH2Cl2/EtOH. The large and negative ∆S‡ values for the

7370 J . Org. Chem., Vol. 64, No. 20, 1999
Liu et al.
Sch em e 2. “Hea d -On Ap p r oa ch ” Mod el for th e
Oxid a tion of tr a n s-â-Meth ylstyr en e by Ster ica lly
En cu m ber ed Dioxor u th en iu m (VI) P or p h yr in s
Sch em e 1. Oxid a tion of Alk en es by
Dioxor u th en iu m (VI) P or p h yr in s
employed the molecular structure of [RuII(TDCPP)(CO)-
(styrene oxide)] complex, wherein the bound epoxide was
inclined at a 49° angle with respect to the mean porphy-
rinato plane, to model the transition state stereochem-
istry of the epoxidation.24 From the X-ray structure, a
close nonbonding interaction was noted between the
epoxide protons and the porphyrin ring, as well as
between the phenyl carbons of the styrene oxide and one
of the ortho-chloro substituents. It was, therefore, pur-
ported that the porphyrin cavity encircled by the chlorine
atoms would not have enough room to accommodate a
trans-1,2-disubstituted epoxide. This deliberation was
supported by the failure of trans-â-methylstyrene oxide
to coordinate effectively to bulky ruthenium(II) porphy-
rins.24 Indeed, the side-on approach model has gained a
widespread acceptance in the oxidation chemistry of
metalloporphyrins and has been successfully applied to
explain the enantioselectivity of the chiral Mn(salen)-
catalyzed asymmetric epoxidation of unfunctionalized
alkenes.25
To test the generality of the side-on approach as a
transition state model for the oxometalloporphyrin-
mediated epoxidations, we examined the reactions of two
sterically congested dioxoruthenium(VI) porphyrins,
[RuVI(TDCPP)O2] (1b) and [RuVI(TMOPP)O2] (1c), with
cis- and trans-â-methylstyrenes. The bulky ortho sub-
stituents would prevent a side-on approach of the trans-
alkene to the OdRu moiety. Therefore, if the side-on
approach is an obligatory route for epoxidation, the
encumbered dioxoruthenium(VI) porphyrins 1b and 1c
would be inactive toward trans-â-methylstyrene oxida-
tion. And yet, in this work, we found that the cis- and
trans-alkenes were readily epoxidized by both ruthenium
oxidants to give the corresponding epoxides stereoselec-
tively (see Table 1, entry 6). Furthermore, the ∆S‡ values
for their reactions with styrene (-21.9 eu for 1b and
-27.3 eu for 1c) and trans-â-methylstyrene (-23.7 eu for
1b and 1c) are comparable (see Table 3), indicative of a
similar degree of oxidant-alkene association in the
transition states for the styrene and trans-â-methylsty-
rene oxidations. Indeed, the facile oxidation of the trans-
alkene by the bulky dioxoruthenium(VI) porphyrins
cannot be rationalized by the side-on approach model.
Apart from electronic effect, steric hindrance by the
bulky substituents had been shown to play an important
role in determining the optimal angles of approach of the
CdC bond to the CrVdO moiety on the basis of the
detailed docking studies on the approach of several cis-
and trans-alkenes to [(Br8TPP)CrV(O)(X)] (H2(Br8TPP) )
meso-tetrakis(2,6-dibromophenyl)porphyrin) by Bruice
and co-workers.26 For the trans-â-methystyrene oxidation
by the oxochromium(V) porphyrin, the CdC bond would
be directed preferably from the top of the CrVdO group.
A similar head-on approach model can be implicated to
rationalize the facile trans-â-methylstyrene oxidations by
the sterically encumbered [RuVI(TDCPP)O2] and [RuVI
-
4c
(TMOPP)O2] complexes (Scheme 2).
Con cer ted ver su s Non con cer ted Mech a n ism . cis-
Alkenes have been widely employed as mechanistic
probes for the concertedness of oxo-metal-mediated
epoxidation reactions.27 If the epoxidation of cis-alkene
involves breakage of the CdC π bond resulting in the
formation of an acyclic intermediate, i.e., a nonconcerted
pathway, then isomerization via the unhindered C-C
bond rotation to form the trans-epoxide product would
occur. In the case of cis-stilbene oxidation, the supposed
acyclic intermediate is particularly prone to cis-trans
isomerization because of severe steric interaction between
the two phenyl rings. In this work, the reactions of 1a -c
with cis-stilbene were found to be highly stereoretentive
with no detectable formation of trans-epoxide. cis-Stil-
bene oxide was formed in 99% and 77% yield for 1a and
1b as the oxidants, respectively, whereas a moderate
epoxide yield (46%) was obtained for 1c.
As mentioned earlier, stereospecificities of the cis-
alkene oxidations were found to vary with both the
alkenes and the ruthenium porphyrins. Unlike the cis-
stilbene oxidations by the sterically congested dioxoru-
thenium(VI) porphyrins, the analogous reaction with
[RuVI(OEP)O2] (H2OEP ) 2,3,7,8,12,13,17,18-octaeth-
ylporphyrin) afforded trans-stilbene oxide as the major
product with a trans-:cis-oxide ratio of 44:16.18b Although
a complete stereoretention was attained for the cis-â-
methylstyrene oxidations by 1b and 1c, a substantial
amount of trans-â-methylstyrene oxide (14%) was formed
when [RuVI(DPP)O2] (1a ) was the oxidant.
In this work, the stereoselectivity of the epoxidation
reaction was further examined by employing cis-â-
deuteriostyrene as a mechanistic probe.7e,28 When cis-â-
deuteriostyrene was subjected to the standard reaction
conditions, cis-alkene (1 mmol), [RuVI(Por)O2] (30 µmol)
in 2% w/w Hpz/CH2Cl2 under an argon atmosphere, the
epoxidation proceeded nonstereospecifically, and the cis-:
trans-epoxide ratios were found to be 61:39 (1a ), 87:13
(1b), and 62:14 (1c) (see Table 1, entry 5). The oxidations
had been repeated three times for each ruthenium
oxidant, and the cis-trans selectivities were obtained in
a reproducible manner and reported as average values.
(24) Groves, J . T.; Han, Y.; Van Engen, D. J . Chem. Soc., Chem.
Commun. 1990, 436.
(25) J acobsen, E. N. In Comprehensive Organometallic Chemistry
II; Wilkinson, G. W., Stone, F. G. A., Abel, E. W., Hegedus, L. S., Eds.;
Pergamon: New York, 1995; Vol. 12, Chapter 11.1.
(26) He, G.-X.; Mei, H.-Y.; Bruice, T. C. J . Am. Chem. Soc. 1991,
113, 5644.
(27) (a) Castellino, A. J .; Bruice, T. C. J . Am. Chem. Soc. 1988, 110,
158. (b) Groves, J . T.; Stern, M. K. J . Am. Chem. Soc. 1988, 110, 8628.
(28) Palucki, M.; Pospisil, P. J .; Zhang, W.; J acobsen, E. N. J . Am.
Chem. Soc. 1994, 116, 9333.

Epoxidation by trans-Dioxoruthenium(VI) Porphyrins
J . Org. Chem., Vol. 64, No. 20, 1999 7371
Sch em e 3
Sch em e 4. In ver se Secon d a r y KIE for th e
Oxid a tion of Styr en e
range reported in a related study on asymmetric chiral
Mn(salen)-catalyzed alkene epoxidation.32 The secondary
KIE results favor a nonsymmetrical transition state
wherein the C-O bond formation is more advanced at
the â-carbon atom than at the R-carbon atom with
concomitant rehybridization of the C(â) atom from sp2
to sp3, whereas the latter remains more or less sp2
hybridized. It is more consistent with the formation of
an acyclic intermediate at the rate-limiting step (Scheme
4). By the same token, the KIE results should also
discount a concerted oxene insertion mechanism.
Na tu r e of Tr a n sition Sta te a n d Ha m m ett Cor -
r ela tion Stu d ies. The effect of para-substituents on the
rate of styrene oxidations by 1a -c has been investigated
(Table 2, entries 1-6). It is noteworthy that both electron-
releasing and -withdrawing substituents can moderately
accelerate the reactions (∼9 to 5-fold for 4-methoxysty-
rene and 2 to 1.1-fold for 3-nitrostyrene). The correlation
of log krel vs σ+ [krel ) k2(substituted styrene)/(k2(styrene)]
gave rise to concave Hammett curves resembling the
cases involving [RuVI(TPP)O2]7b and some cationic oxo-
ruthenium(IV)7a,e complexes as oxidants. This is contrary
The purity of cis-â-deuteriostyrene (>99%) was checked
1
by H NMR spectroscopy, and the trans-alkene was not
to the linear Hammett log krel vs σ+ correlations (F+
-1.9 to -2.1) for the styrene oxidations involving the
[FeIV(TMP•+)O] (H2TMP tetramesitylporphyrin),33
)
detected at the end of the reactions. A control experiment
was performed by stirring a mixture of pure cis-â-
deuteriostyrene oxide and [RuIV(Por)(pz)2] in methylene
chloride at room temperature for 12 h, and the starting
cis-epoxide was fully recovered without trans-epoxide
being detected. This confirms that the formation of the
cis- and trans-epoxides associated with the cis-â-deute-
riostyrene oxidation did not arise from the Ru-catalyzed
epoxide isomerization.29 Evidently, the loss of stereospec-
ificity is inconsistent with a concerted reaction pathway
such as oxene insertion (Scheme 3, pathway a), and a
nonconcerted mechanism is more plausible.
)
[(Br8TPP)CrV(O)(X)],34 and [RuVI(N4)O2]2+ (N4 ) macro-
cyclic tertiary amines)7c,d complexes. Because the rate-
limiting carbocation formation for the electrophilic ad-
dition to the CdC bond would have F+ values as large as
-3.5 (hydration)35 and -4.1 (bromination),36 the minor
influence of the electronic substituent effect on the k2
values discounts the participation of an alkene-derived
cation radical (Scheme 3, pathway c), as well as a
carbocation intermediate (Scheme 3, pathway d). The
insensitivity of the k2 values to the para-substituent effect
would be more compatible with the rate-limiting forma-
tion of a carboradical intermediate (Scheme 3, pathway
e).
The rate-limiting formation of a metallaoxetane inter-
mediate (Scheme 3, pathway b) by (2 + 2) cycloaddition
between the RudO and CdC bonds is unlikely.30 This
pathway requires the simultaneous rehybridization of
both the R- and â-olefinic carbon atoms from sp2 to sp3
at the rate-determining step, which is incompatible with
the fact that inverse secondary kinetic isotope effect
(KIE)31 has only been observed for the â-d2-styrene
oxidations (kH/kD ) 0.87 for 1a and 0.86 for 1b) and not
for R-deuteriostyrene (kH/kD ) 1.01 for 1a and 1b); see
Table 2 entries 13 and 14. The magnitude of the inverse
secondary KIE observed here falls within the 0.82-0.93
The alkene-derived cation radical intermediate is
expected to be susceptible to skeletal rearrangement. For
example, the oxidation of norbornene by some highly
oxidizing oxometalloporphyrins produced exo- and endo-
epoxynorbornane, norcamphor, and cyclohexen-4-car-
boxyaldehyde. A cation radical intermediate was invoked
by Traylor and co-workers37 to explain this finding. In
(32) Paluki, M.; Finney, N. S.; Pospisil, P. J .; Gu¨ler, M. L.; Ishida,
T.; J acobsen, E. N. J . Am. Chem. Soc. 1998, 120, 948.
(33) Groves, J . T.; Watanabe, Y. J . Am. Chem. Soc. 1986, 108, 507.
(34) Garrison, J . M.; Ostovic, D.; Bruice, T. C. J . Am. Chem. Soc.
1989, 111, 4960.
(35) Schubert, W. M.; Keeffe, J . R. J . Am. Chem. Soc. 1972, 94, 559.
(36) Yates, K.; McDonald, R. S.; Shapiro, S. A. J . Org. Chem. 1973,
38, 2460.
(29) Groves, J . T.; Ahn, K.-H.; Quinn, R. J . Am. Chem. Soc. 1988,
110, 4217.
(30) For a comprehensive review of metallaoxetane, see: J ørgensen,
K. A. Chem. Rev. 1983, 83, 431.
(31) Issacs, N. Physical Organic Chemistry, 2nd ed.; J ohn Wiley &
Sons: New York, 1995; p 296-302.

7372 J . Org. Chem., Vol. 64, No. 20, 1999
Liu et al.
Sch em e 5. P r op osed Mech a n ism for th e
Oxid a tion of Alk en es by Dioxor u th en iu m (VI)
P or p h yr in s
to establish a pure carboradical (σ•) substituent constant
for spin delocalization effect using spectroscopic and
kinetic methods. Notable examples are the works by
•
Arnold (σR , ESR studies of para-substituted benzylic
radicals),38 Creary (σc•, kinetic studies of the rearrange-
ment of methylenecyclopropanes),39 and J ackson (σJ ,
•
kinetic studies of the thermal decomposition of substi-
tuted dibenzyl mercurials).40 However, a complete sepa-
ration of spin delocalization effect from the residual polar
effect was achieved with limited success.
Recently, J iang and J i proposed a new carboradical
scale, σJJ•, based on the kinetic studies of para-substituent
effect on thermal cycloaddition reaction of R,â,â-trifluo-
rostyrenes.41 The polar and spin delocalization effects
were separated. A substituent constant, σmb, for the polar
effect based on the degree of ground state π-bond
polarization induced by the para-substituents was also
established. Taking into account the spin delocalization
•
and polar effects, we then applied the σJ J and σmb
substituent constants to our Hammett correlation analy-
•
sis (log krel ) Fmbσmb + FJ J σJ J •) for the styrene oxidations,
and three straight lines resulted (slope ) 1.00; R ) 0.98-
•
0.99) (Figure 4): log krel ) -0.62σmb + 1.01σJ J (1a ), log
krel ) -0.58σmb + 1.06σJ J • (1b), and log krel ) -0.77σmb
+
•
F igu r e 4. Linear dual-parameter Hammett correlations using
1.67σJ J (1c). The linear Hammett correlations coincide
with the rate-determining formation of a benzylic radical
intermediate (Scheme 5) for the styrene epoxidations by
dioxoruthenium(VI) porphyrins. The positive FJ J • values
suggest that the epoxidation reaction is promoted by
delocalizing spin density at the radical center. The
negative Fmb values are consistent with the electrophilic
nature of the oxoruthenium complexes.
•
σmb and σJ J for the ruthenium oxidation of para-substituted
styrenes: (a) [RuVI(DPP)O2], (b) [RuVI(TDCPP)O2], and (c)
[RuVI(TMOPP)O2].
this work, exo-epoxynorbornane was found to be the
exclusive product of the norbornene oxidations (Table 1,
entry 7), suggesting that a cation radical species is not
necessarily the intermediate of the reaction.
Apart from polar substituent effect, radical reactivity
is also influenced by the substituent’s ability to delocalize
the spin density on the carbon radical center, namely spin
delocalization effect. There have been several attempts
(38) (a) Dust, J . M.; Arnold, D. R. J . Am. Chem. Soc. 1983, 105, 1221.
(b) Wayner, D. D. M.; Arnold, D. R. Can. J . Chem. 1985, 63, 2378.
(39) (a) Creary, X.; Mehrsheikh-Mohammadi, M. E. J . Org. Chem.
1986, 51, 1110. (b) Creary, X.; Mehrsheikh-Mohammadi, M. E. J . Org.
Chem. 1987, 52, 3254.
(40) Dinctu¨rk, S.; J ackson, R. A.; Townson, M.; Agirbas, H.; Norman,
C. J . Chem. Soc., Perkin Trans. II 1981, 1127.
(41) J iang, X.-K. Acc. Chem. Res. 1997, 30, 283.
(37) (a) Traylor, T. G.; Miksztal, A. R. J . Am. Chem. Soc. 1987, 109,
2770. (b) Traylor, T. G.; Xu, F. J . Am. Chem. Soc. 1988, 110, 1953.

Epoxidation by trans-Dioxoruthenium(VI) Porphyrins
J . Org. Chem., Vol. 64, No. 20, 1999 7373
In all three cases, the magnitude of |FJ J •/Fmb| > 1
suggests that the spin delocalization effect is more
important than the polar substituent effect in the dioxo-
ruthenuim-mediated alkene epoxidation.41
dehyde, the CdC bond cleavage product, should result
from the reaction of the carboradical species with dis-
solved dioxygen.44
For comparison, we have also applied to our correlation
studies another set of carboradical parameters, namely,
the total substituent effect (TE) proposed by Wu and co-
workers.42 The parameters were established by the
density functional calculation (BLYP/6-31*) based on the
influence of the para-substituents on the benzylic C-H
bond dissociation energies of toluenes. It should be noted
that the TE parameters reflect the combined substituent
effect of the change in spin (∆s) and charge (∆c) densities
at the radical center: TE ) -29.9∆s - 11.7∆c, where
the spin delocalization effect is dominant. As expected,
an excellent linear correlation (R ) 0.99) between log krel
for the substituted styrene oxidations by [RuVI(DPP)O2]
(1a ), [RuVI(TDCPP)O2] (1b), and [RuVI(TMOPP)O2] (1c)
with TE was attained with slopes of FTE• ) +0.67, +0.63,
and +0.86, respectively.43 Analogous to the FJ J • value, the
positive FTE• values also indicate that the styrene oxida-
tion is promoted by delocalizing spin density on the
radical center. Indeed, the results of the correlation
analyses based on two independent carboradical param-
eters consistently support the rate-limiting benzylic
radical formation for the dioxoruthenium(VI)-mediated
styrene epoxidation.
Con clu sion
With regard to the mechanism of alkene epoxidation
by bulky dioxoruthenium(VI) porphyrins, the following
findings are summarized:
(1) The stereoselectivity for the epoxide formation
varies with alkenes in the order of cis-stilbene > cis-â-
methylstyrene > cis-â-deuteriostyrene, whereas the ster-
ically bulky dioxoruthenium (VI) porphyrins such as
[RuVI(TDCPP)O2] and [RuVI(TMOPP)O2] can afford a
better stereoretention than [RuVI(DPP)O2].
(2) The observation of inverse secondary KIE for the
oxidation of â-d2-styrene [kH/kD ) 0.87 for 1a , 0.86 for
1b) and its absence in the R-deuteriostyrene oxidations
discounts strongly a rate-limiting formation of a metal-
laoxetane and a concerted oxene insertion mechanism.
(3) For the oxidation of para-substituted styrenes, a
linear dual-parameter Hammett correlation was estab-
lished using σmb and σJ J • parameters, which is consistent
with the rate-limiting formation of a benzylic radical
intermediate. The |FJ J •/Fmb| > 1 suggests that the genera-
tion of the carboradical intermediate is controlled prin-
cipally by spin delocalization effect.
The sterically encumbered [RuVI(TDCPP)O2] and [RuVI-
(TMOPP)O2] readily oxidize trans-â-methylstyrene in
good yields. The comparable ∆S‡ values for the trans-â-
methylstyrene and styrene oxidations by [RuVI(TDCPP)-
O2] and [RuVI(TMOPP)O2] suggest a similar degree of
oxidant-alkene association in the transition state. Be-
cause a side-on approach of trans-alkene is not favored
by the bulky substituents, a head-on approach model is
postulated.
In accord with the Hammond postulate, the alkene
epoxidation by the milder oxidant [RuVI(TMOPP)O2] (1c)
proceeds via a more product-like transition state. In
contrast, alkene oxidation by the more reactive oxidants
1a and 1b has led to a more reactant-like transition state.
For a more product-like transition state, i.e., more
carboradical-like, the rate of the alkene epoxidation by
[RuVI(TMOPP)O2] was found to be more sensitive to the
para-substituent effect manifested by a larger magnitude
•
of its FJ J •, Fmb (J iang and J i) and FTE (Wu) values.
P r op osed Mech a n ism . We can conclude that the
oxidation of aromatic alkenes by dioxoruthenium(VI)
porphyrins involves the rate-limiting formation of a
benzylic radical intermediate. The radical intermediate
would then partition between ring closure and a C-C
rotation/cyclization thereby producing cis- and trans-
epoxides, respectively. The participation of a carboradical
intermediate in the oxometal-mediated alkene epoxida-
tions is not without precedent; oxidation systems involv-
ing the [MnIV(O)(TMP)],44a [FeIV(O)(TMP)] (H2TMP )
tetramesitylporphyrin),44b [MnV(O)(salen)(X)],32 and [RuIV-
(O)(L1)(L2)]2+ (L1 and L2 ) polypyridine ligands)7e com-
plexes have already been proposed to proceed via a
radical intermediate.
Exp er im en ta l Section
All solvents were purified by the standard procedures before
use. Benzaldehyde, 2,6-dichlorobenzaldehyde, 2,4,6-trimethoxy-
benzaldehyde, and pyrrole were either distilled or purified by
the standard methods prior to use. Bromine, m-chloroperoxy-
benzoic acid (Merck), dodecacarbonyltriruthenium(0), and
pyrazole (Aldrich) were used as received. 2,3,5,7,10,12,13,15,-
18,20-Dodecaphenylporphyrin (H2DPP) was prepared using
the previously reported procedure.18d meso-Tetrakis(2,6-dichlo-
rophenyl)porphyrin (H2TDCPP) and meso-tetrakis(2,4,6-tri-
methoxyphenyl)porphyrin (H2TMOPP) were synthesized ac-
cording to Lindsey’s method.45 The alkene substrates were
purified by either vacuum distillation or recrystallization, and
their purities were checked by GC or 1H NMR analysis. cis-
â-Methylstyrene,46 cis-â-deuteriostyrene,46 â-d2-styrene,47a,b
and R-deuteriostyrene47a,c were prepared according to the
literature procedures. The preparation and characterization
of [RuVI(DPP)O2] and [RuIV(DPP)(pz)2] have been reported
elsewhere.18d
Apparently, the stereoretention of the cis-alkene oxida-
tions can be influenced by nearby bulky substituents. For
example, the cis-â-methylstyrene oxidation by [RuVI
-
(TDCPP)O2] and [RuVI(TMOPP)O2] are highly stereose-
lective, whereas the reaction with [RuVI(DPP)O2] results
a mixture of cis- and trans-epoxides. Presumably, the
bulky groups tends to hinder the C-C rotation of the
benzylic radical intermediate, resulting in a high degree
of stereoretention for the cis-alkene oxidations. Benzal-
[Ru II(P or )(CO)(py)], H2P or ) H2TDCP P an d H2TMOP P .
A mixture of Ru3(CO)12 (200 mg, 0.313 mmol) and free base
porphyrin (H2Por) was heated in molten naphthalene (220 °C)
(45) Lindsey, J . S.; Wagner, R. W. J . Org. Chem. 1989, 54, 828.
(46) Lindler, H.; Dubuis, R. Organic Syntheses; Wiley & Sons: New
York, 1973; Collect. Vol. V, p 880.
(47) (a) Overberger, G. H.; Saunders, J . H. Organic Syntheses; Wiley
& Sons: New York, 1955; Collect. Vol. III, p 204. (b) Hosokawa, T.;
Ohta, T.; Kanagama, S.; Murahashi, S.-I. J . Org. Chem. 1987, 52, 1758.
(c) Wesener, J . R.; Moskau, D.; Gu¨nther, H. J . Am. Chem. Soc. 1985,
107, 7307.
(42) Wu, Y.-D.; Wong, C.-L.; Chan, K. W. K.; J i, G.-Z.; J iang, X.-K.
J . Org. Chem. 1996, 61, 746.
(43) Liu, C.-J . Ph.D. Thesis, The University of Hong Kong, 1998.
(44) (a) Arasasingham, R. D.; He, G.-X.; Bruice, T. C. J . Am. Chem.
Soc. 1993, 115, 7985. (b) Groves, J . T.; Gross, Z.; Stern, M. K. Inorg.
Chem. 1994, 33, 5065.

7374 J . Org. Chem., Vol. 64, No. 20, 1999
Liu et al.
under an argon atmosphere for 24 h. After cooling to room
temperature, the mixture was dissolved in toluene, and the
solution was loaded directly onto an alumina column. Naph-
thalene, unreacted Ru3(CO)12, and H2Por were eluted by
toluene. The desired ruthenium porphyrin complex was col-
lected using a chloroform/ethyl acetate mixture (95:5 v/v for
H2Por ) H2TDCPP) or chloroform/methylene chloride mixture
(1:1 v/v for H2Por ) H2TMOPP) as the eluant. Upon addition
of methanol (ca. 20 mL), the solution was concentrated until
the [RuII(Por)(CO)(MeOH)] complex started to precipitate as
a red solid. Recrystallization of the [RuII(Por)(CO)(MeOH)]
complex in a pyridine/dichloromethane/benzene solution [6:3:
1 (v/v), 10 mL] afforded the titled complex as a dark purple
crystal.
The organic products were then analyzed and quantified, after
addition of an internal standard, by either gas chromatography
or 1H NMR spectroscopy.
The product ruthenium(IV) porphyrin was eluted by meth-
ylene chloride, and addition of acetonitrile led to the isolation
of [RuIV(Por)(pz)2] (H2Por ) H2DPP (2a ), H2TDCPP (2b), and
TMOPP (2c)) as a dark purple solid. Yield: 70-80%. For the
characterization data of [RuIV(DPP)(pz)2], see ref 6k.
[Ru IV(TDCP P )(p z)2] (2b). Yield: 75%. UV-vis (CH2Cl2)
λ
max/nm (log ꢀ): 401 (5.19), 520 (4.01). IR (KBr): 1006 cm-1
(oxidation state marker band). FAB-MS: m/z 1123 (M+), 989
(M+ - 2pz). Anal. Calcd for C50H32Cl8N8Ru: C, 53.42; H, 2.32;
N, 9.97. Found: C, 53.20; H, 2.62; N, 9.68. µeff (Evan’s method)
) 2.68 µB (CHCl3, room temperature).
Yield for [RuII(TDCPP)(CO)(py)]: 90% based on H2TDCPP.
IR(KBr): 1947 cm-1 (νCdO). UV-vis (CH2Cl2) λmax/nm (log ꢀ):
410 (5.14), 530 (4.23), 558 (sh) (3.59). 1H NMR (300 MHz,
CDCl3, TMS): δ 7.73 (m, 12H), 8.44 (s, 8H). FAB-MS: m/z
1023 (M+). Anal. Calcd for C46H24Cl8N4O2Ru: C, 53.06; H, 2.46;
N, 5.27. Found: C, 53.27; H, 2.60; N, 5.10.
[Ru IV(TMOP P )(p z)2] (2c). Yield: 65%. UV-vis (CH2Cl2)
λ
max/nm (log ꢀ): 406 (5.06), 521 (3.84). IR (KBr): 1004 cm-1
(oxidation state marker band). FAB-MS: m/z 1207 (M+), 1140
(M+ - pz), 1074 (M+ + 1 - 2pz). Anal. Calcd for C62H58N8O12
-
Ru: C, 61.64; H, 4.81; N, 9.28. Found: C, 61.41; H, 4.96; N,
9.09. µeff (Evan’s method) ) 2.88 µB (CHCl3, room temperature).
Kin etic Mea su r em en ts for Alk en e Ep oxid a tion s by
Dioxor u th en iu m (VI) P or ph yr in s. Kinetic experiments were
performed on a Hewlett-Packard HP8452A UV-vis diode
array spectrophotometer interfaced with an IBM-compatible
PC. The measurements were made in a standard 1.0 cm quartz
cuvettes. The temperature of the solution was maintained to
within (0.2 °C during the kinetic studies with the use of a
Lauda RM6 circulating water bath. The oxidation of alkenes
by [RuVI(Por)O2] was followed by monitoring the decrease of
the Soret absorption band under the condition that the alkene
concentration was at least 100-fold in excess of [RuVI(Por)O2].
The pseudo-first-order rate constants, kobs, were obtained by
nonlinear least-squares fits of (Af - At) to time t according to
the equation
Yield for [RuII(TMOPP)(CO)(py)]: 80% based on H2TMOPP.
IR(KBr): 1938 cm-1 (νCdO). UV-vis (CH2Cl2) λmax/nm (log ꢀ):
412 (5.46), 528 (4.40), 560 (sh) (3.40). 1H NMR (300 MHz,
CDCl3, TMS): δ 3.55 (s, 12H), 3.67 (s, 12H), 4.10 (s, 12H), 6.61
(m, 8H), 8.57 (s, 8H). FAB-MS: m/z 1131 (M+ + 1). Anal. Calcd
for C58H56N4O14Ru: C, 61.42; H, 4.94; N, 4.94. Found: C, 60.21;
H, 50.03; N, 4.87.
[Ru VI(P or )O2], H2P or ) H2TDCP P (1b) a n d H2TMOP P
(1c). A methylene chloride solution (20 mL) of [RuII(Por)(CO)-
(MeOH)] (40 µmol) was treated with m-chloroperoxybenzoic
acid (100 mg, 58 µmol). The mixture was stirred for 5 min,
and the resultant brown solution was loaded onto an alumina
column. The product ruthenium complex was then eluted with
methylene chloride (ca. 20 mL). Addition of purified n-hexanes
induced precipitation of the complex as a purple microcrys-
talline solid. The solid was then collected on a frit, washed
with anhydrous methanol, and then dried in vacuo.
(Af - At) ) (Af - Ai) exp(-kobst)
Yield for 1b: 80%. IR (KBr): 825 cm-1 (νRuO2), 1019 cm-1
(oxidation state marker band). UV-vis (CH2Cl2) λmax/nm (log
ꢀ): 420 (5.42), 521 (4.36), 572 (3.86). 1H NMR (300 MHz,
CDCl3, TMS): δ 7.81 (m, 12H), 8.89 (s, 8H). FAB-MS: m/z
1012 (M+), 989 (M+ - 2O). Anal. Calcd for C44H20Cl2N4O2Ru:
C, 51.71; H, 1.97; N, 5.49. Found: C, 51.80; H, 2.20; N, 5.17.
Yield for 1c: 75%. IR (KBr): 820 cm-1 (νRuO2), 1018 cm-1
(oxidation state marker band). UV-vis (CH2Cl2) λmax/nm (log
ꢀ): 424 (5.42), 521 (4.36), 563 (3.86). 1H NMR (300 MHz,
CDCl3, TMS): δ 3.61 (s, 24H), 4.12 (s, 12H), 6.66 (s, 8H), 8.97
(s, 8H). FAB-MS: m/z 1132 (M+ + 1), 1101 (M+ - 2O). Anal.
Calcd for C56H52N4O14Ru: C, 60.80; H, 4.74; N, 5.07. Found:
C, 61.11; H, 4.36; N, 4.89.
where Ai and Af are the initial and final absorbances, respec-
tively, and At is the absorbance at time t. Kinetic data over
four half-lives (t1/2) were used for the fitting. The second-order
rate constants, k2, were evaluated from the linear fit of the
kobs values to the alkene concentrations.
The activation enthalpies (∆H‡) and entropies (∆S‡) were
calculated from the linear plot of ln(k2/T) versus (1/T) according
to the Eyring equation:
ln(k2/T) ) ln(R/NAp) + ∆S‡/R - ∆H‡/RT
where T is the temperature (K), NA is Avogadro’s number, R
is the universal gas constant, and p is Planck’s constant.
Stoich iom etr ic Oxid a tion of Alk en es by Dioxor u th e-
n iu m (VI) P or p h yr in s. To a degassed methylene chloride
solution (5 mL) containing pyrazole (2% w/w) and alkenes (2
mmol) was added [RuVI(Por)O2] (30 µmol) under an argon
atmosphere. After stirring for 12 h, the reaction mixture was
filtered through a short alumina column with hexanes/ethyl
acetate (9:1) as the eluant to remove the ruthenium complex.
Ack n ow led gm en t. We acknowledge support by The
University of Hong Kong and the Hong Kong Research
Grants Council. We are also grateful to Dr. Wai-Hong
Fung for assistance in kinetic measurements.
J O990517X