Saddle-shaped dioxo-ruthenium(VI) and -osmium(VI) 2,3,5,7,8,10,12,13,15,17,18,20-dodecaphenylporphyrin (H2dpp) complexes. Synthesis, spectral characterisation and alkene oxidation by [RuVI(dpp)O2]

Article abstract of DOI:10.1039/a800535d

An improved procedure for the preparation of the saddle-distorted porphyrin 2,3,5,7,8,10,12,13,15,17,18,20-dodecaphenylporphyrin (H₂dpp) (yield = 75%) based on the Suzuki cross-coupling reaction between phenylboronic acid PhB(OH)₂ an

Full text of DOI:10.1039/a800535d

View Article Online / Journal Homepage / Table of Contents for this issue
Saddle-shaped dioxo-ruthenium(VI) and -osmium(VI)
2,3,5,7,8,10,12,13,15,17,18,20-dodecaphenylporphyrin (H2dpp)
complexes. Synthesis, spectral characterisation and alkene oxidation
by [RuVI(dpp)O2]†
Chun-Jing Liu,a Wing-Yiu Yu,a Shie-Ming Peng,b Thomas C. W. Makc and Chi-Ming Che*,a
a Department of Chemistry, The University of Hong Kong, Pokfulam Road, Hong Kong
b Department of Chemistry, National Taiwan University, Taipei, Taiwan
c Department of Chemistry, The Chinese University of Hong Kong, Shatin, Hong Kong
An improved procedure for the preparation of the saddle-distorted porphyrin 2,3,5,7,8,10,12,13,15,17,18,20-
dodecaphenylporphyrin (H2dpp) (yield = 75%) based on the Suzuki cross-coupling reaction between phenyl-
boronic acid PhB(OH)2 and [2,3,7,8,12,13,17,18-octabromo-5,10,15,20-tetraphenylporphyrin] has been developed.
X-Ray diffraction studies of [MII(dpp)(CO)(py)] (M = Ru 1 or Os 3) showed that 1 and 3 are isostructural, and
the porphyrin macrocycles exhibit severe out-of-plane saddle and ruffle distortions. In both 1 and 3 the pyrrole
rings are alternately tilted up and down with respect to the least-squares plane of the 25-atom porphyrin core,
and the pyrrole carbons experience an average displacement of 0.769 Å from the least-squares plane compared
to 0.78 Å for free H2dpp, whereas the Ru and Os atoms are displaced by 0.1006 and 0.0792 Å from the 25-atom
porphyrin core respectively. The complex [RuVI(dpp)O2] 2, prepared by m-chloroperoxybenzoic acid oxidation,
is an active oxidant for alkene epoxidations. In CH2Cl2 [containing 2%(w/w) pyrazole], styrene, norbornene and
cis-stilbene were oxidised selectively to their respective epoxides in excellent yield. Complete stereoretention was
observed for the oxidation of cis-stilbene, however oxidation of cis-β-methylstyrene afforded significant amounts
of trans-epoxide suggesting that a carboradical mechanism is operative. The crystal structure of the complex
[RuIV(dpp)(pz)2] (5), the product of the stoichiometric alkene oxidations, was determined. Magnetic susceptibility
measurement (µeff = 3.24 µB) suggests the formulation of RuIV with two unpaired electrons in its electronic ground
state. The Ru᎐N (pz) bond distances are 2.022(13) and 2.083(12) Å. The reactions of 2 with alkenes in CH2Cl2
(with 2% Hpz) follow second-order kinetics: rate = k1[2][alkene]. For norbornene and styrene, the second-order
rate constants, k1, in CH2Cl2 at 25.9 ЊC are (3.79 0.04) × 10Ϫ3 and (4.78 0.09) × 10Ϫ3 dm3 molϪ1 sϪ1
respectively.
The role of non-planar conformation in modifying the bio-
logical properties and functions of some metalloporphyrins
has recently come under rigorous investigations.1 Out-of-
plane distortion is observed in the crystal structures of the
bacteriochlorophylls in the photosynthetic reaction centres of
Rhodopseudomonas viridis2 and the light-harvesting antenna
bacteriochlorophyll a of Prosthecochloris aestuarii.3 Variations
in the extent of the distortion have been proposed as a possible
reason for the unidirectionality of electron transfer.4 On the
other hand, the steric congestion among the peripheral sub-
stituents in some synthetic polyhalogenated porphyrins such as
octa-β-halogenotetrakis(pentafluorophenyl)porphyrins (X = Cl
or Br) has created an out-of-plane distortion of the porphyrin
ligands, which is proposed to play a crucial role in enhancing
the catalytic activities of the iron() complexes by lowering
their oxidation potentials.5 In this respect, we are interested to
examine how the reactivities/catalytic activities of ruthenium/
osmium complexes could be influenced by the conformational
distortion of the porphyrin macrocycles.6 Here we report the
first synthesis and characterisation of the saddle-distorted
dioxo-ruthenium() and -osmium() complexes of 2,3,5,7,8,
10,12,13,15,17,18,20-dodecaphenylporphyrin (H2dpp), and the
reactivities, as well as the catalytic activities, of the dioxo-
ruthenium() complex towards alkene oxidations.
Experimental
Materials
All solvents were purified by the standard procedures prior
to use. Benzaldehyde, tert-butylamine, propionic acid and
pyrrole were either distilled or purified by standard methods
before use. Bromine and m-chloroperoxybenzoic acid
(Merck), palladium() chloride, dodecacarbonyltriruthenium
and dodecacarbonyltriosmium (Aldrich) were used as received.
Tetrakis(triphenylphosphine)palladium(0)
was
prepared
according to the reported procedure.7 2,3,7,8,12,13,17,18-
Octabromo-5,10,15,20-tetraphenylporphyrin, [H2obtpp], was
obtained by bromination of [CuII(tpp)]8 following the pro-
cedures described by Bhyrappa and Krishnan.9
Physical measurements
The UV/VIS spectra were acquired on a Perkin-Elmer Lambda
19 spectrophotometer and infrared spectra on a Shimadzu IR-
470 spectrophotometer. Cyclic voltammograms were recorded
on a Princeton Applied Research model 273 potentiostat using
a glassy carbon electrode and Ag–AgNO3 (0.1 mol dmϪ3 in
MeCN) as the reference electrode. Gas chromatography was
performed on a Hewlett-Packard 5890 Series II gas chromato-
graph equipped with a SPB-5 capillary column (30 m) using
nitrogen as the carrier gas, a flame ionisation detector and a
† Supplementary data available: rate constant concentration dependen-
cies, cyclic voltammograms. For direct electronic access see http://
www.rsc.org/suppdata/dt/1998/1805/, otherwise available from BLDSC
(No. SUP 57369, 7 pp.) or the RSC Library. See Instructions for
Authors, 1998, Issue 1 (http://www.rsc.org/dalton).
Non-SI units employed: µB ≈ 9.27 × 10Ϫ24 J TϪ1, atm = 101 325 Pa, cal =
4.184 J.
1
3396 Series II integrator. The H and 13C NMR spectra were
J. Chem. Soc., Dalton Trans., 1998, Pages 1805–1812
1805

View Article Online
recorded on Bruker 300 and 500 NMR spectrometers. All the
chemical shifts are given in ppm vs. SiMe4. Elemental analyses
were performed by Butterworth Co. Ltd., UK.
(50 cm3) after cooling to room temperature, and then poured
into brine (200 cm3) and extracted with benzene (3 × 100 cm3).
The combined organic extracts were washed with water (3 × 50
cm3) and dried over Na2SO4, and then evaporated to dryness.
The residue was redissolved in dichloromethane (15 cm3), and
chromatographed on a basic alumina column using dichlo-
romethane as the eluent. The leading purple band was collected
and evaporated to dryness to afford dark red [OsII(dpp)(CO)-
(MeOH)]. The solid was redissolved in pyridine–dichloro-
methane–ethanol (6:3:1, 10 cm3) to yield complex 3 as dark
purple crystals. Yield: 45% based on H2dpp. IR(KBr) νC᎐O 1923,
1008 cmϪ1 (oxidation state marker band). 1H NMR (500 MHz,
CD2Cl2): δ 3.08 (d, J = 6.8, 2 H), 5.87 (t, J = 7.3, 3 H), 6.42
(d, J = 7.50, 8 H), 6.53 (t, J = 7.07, 4 H), 6.59 (t, J = 1.85, 8 H),
6.69 (m, 32 H), 7.18 (d, J = 7.11, 4 H) and 7.42 (d, J = 6.94 Hz, 4
H). 13C NMR (75.6 MHz, CD2Cl2): δ 122.3, 124.9, 125.5, 125.8,
126.0, 126.2, 126.5, 126.9, 131.4, 132.1, 136.2, 138.4, 139.0,
144.7, 144.9 and 146.8. FAB mass spectrum: 1520 (Mϩ), 1441
(Mϩ Ϫ py) and 1413 (Mϩ Ϫ py Ϫ CO) (Found: C, 75.4; H,
4.29; N, 3.70. Calc. for C98H65N5OOsؒ2C5H5N: C, 76.7; H, 4.35;
N, 3.80%).
Synthesis of 2,3,5,7,8,10,12,13,15,18,20-dodecaphenylporphyrin
(H2dpp) by Suzuki cross-coupling reaction
A Teflon-stoppered flask (2 l) was charged with H2obtpp (5.7 g,
4.58 mmol), [Pd(PPh3)4] (800 mg, 0.69 mmol), toluene (1.3 l),
dry K2CO3 (25.3 g, 183.2 mmol) and phenylboronic acid
PhB(OH)2 (11.17 g, 91.76 mmol). The green suspension was
heated at 98 ЊC under nitrogen for 7 d. After cooling to room
temperature the mixture was diluted with dichloromethane (1.3
l) and washed with 4% sodium hydrogencarbonate solution
(3 × 600 cm3). The organic layer was then dried (Na2SO4) and
evaporated to dryness by rotary evaporation to afford the crude
product, which was purified by chromatography with a silica gel
column using chloroform as the eluent. The slowest-moving
green band was collected; after solvent evaporation H2dpp was
isolated as a green solid. The compound was further recrystal-
lised from CH2Cl2–EtOH (1:2) to afford a green crystalline
solid. Yield: 75%. 1H NMR (300 MHz, CDCl3): δ 7.58 (d,
J = 6.76 Hz, 8 H) and 6.72 (m, 52 H). 13C NMR (75.6 MHz,
CD2Cl2): δ 125.1, 125.7, 126.1, 126.8, 131.4, 136.6 and 138.4.
FAB mass spectrum: m/z 1224 (Mϩ ϩ 2) (Found: C, 89.46; H,
4.84; N, 4.61. Calc. for C92H62N4: C, 90.34; H, 5.07; N, 4.58%).
[OsVI(dpp)O2] 4. To a dichloromethane solution (25 cm3) of
[OsII(dpp)(CO)(MeOH)] (50 mg, 0.037 mmol) was added m-
chloroperoxybenzoic acid (100 mg, 0.058 mmol). A change from
red to red-green occurred instantly, and the solution was stirred
for 3 min. It was then concentrated by rotary evaporation to
ca. 5 cm3, and chromatographed on a basic alumina column
using dichloromethane as the eluent. The green band was col-
lected; after solvent removal and subsequently recrystallization
(CH2Cl2–MeCN) analytically pure complex 4 was obtained as
Syntheses of ruthenium complexes
[RuII(dpp)(CO)(py)] 1. The compound H2dpp (200 mg, 0.164
mmol) was refluxed with [Ru3(CO)12] (200 mg, 0.313 mmol) in
toluene (120 cm3) under nitrogen for 4 h. The resulting red
solution was evaporated at reduced pressure to dryness. The
residue was then chromatographed on a basic alumina column
using dichloromethane as the eluent. The second brick red band
was collected. Upon addition of methanol (20 cm3), the solu-
tion was concentrated until the [RuII(dpp)(CO)(MeOH)] com-
plex started to precipitate as a red solid. The compound was
recrystallised in pyridine–dichloromethane–benzene (6:3:1, 10
cm3), and 1 was obtained as dark purple crystals. Yield ≈50%
dark green crystalline solids. Yield: 80%. IR(KBr): νOs᎐O 838,
᎐
1011 cmϪ1 (oxidation state marker band). 1H NMR (500 MHz,
CDCl3): δ 6.67 (m, 52 H) and 7.44 (d, J = 7.0 Hz, 8 H). FAB
mass spectrum: 1446 (Mϩ) (Found: C, 75.8; H, 4.31; N, 3.72.
Calc. for C92H60N4O2Os: C, 76.2; H, 4.33; N, 3.76%).
X-Ray crystallography
X-Ray diffraction data were either collected on an Enraf-
Nonius CAD-4 (for complexes 1 and 5) or a Rigaku AFC7R
(for 3) four-circle diffractometer (graphite-monochromatised
Mo-Kα radiation, λ = 0.7107 Å) using the θ–2θ scan mode at
298 K. The cell dimensions were obtained from a least-squares
fit of 25 reflections in the range 11 < 2θ < 19.5Њ for 1 and
15.32 < 2θ < 30.12Њ for 5. The data were corrected for Ψ-scan
absorption. All the data reduction and structure refinement
were performed using the NRCC-SDP-VAX packages or
TEXSAN, SHELXL 93 programs.10 The structures were solved
by the Patterson method and refined by least-squares cycles. All
non-hydrogen atoms were refined with anisotropic thermal
parameters, hydrogen atoms were included at idealised posi-
tions with a fixed isotropic thermal parameter UH = UC ϩ 0.01
Å2. Crystallographic data and experimental details for com-
plexes 1, 3 and 5 are summarised in Table 1.
based on H2dpp. IR(KBr): νC᎐O 1934 cmϪ1. 1H NMR (500 MHz,
᎐
CD2Cl2): δ 3.08 (d, J = 6.77, 2 H), 5.77 (t, J = 7.02, 3 H), 6.36 (d,
J = 7.44, 8 H), 6.50 (t, J = 7.2, 4 H), 6.53 (t, J = 1.82, 8 H), 6.59
(m, 32 H), 7.18 (d, J = 7.15, 4 H) and 7.38 (d, J = 6.71 Hz, 4 H).
13C NMR (75.6 MHz, CD2Cl2): δ 120.6, 122.8, 123.8, 124.7,
124.9, 125.0, 125.2, 125.7, 130.2, 130.7, 135.4, 135.5, 137.1,
138.4, 143.9, 144.3 and 144.8. FAB mass spectrum: 1428 (Mϩ),
1351 (Mϩ Ϫ py), 1323 and 1321 (Mϩ Ϫ py Ϫ CO) (Found:
C, 82.97; H, 4.49; N, 4.48. Calc. for C98H65N5ORuؒ2C6H6: C,
83.56; H, 4.5; N, 4.29%).
[RuVI(dpp)O2] 2. The complex [RuII(dpp)(CO)(MeOH)] (50
mg, 0.038 mmol) was completely dissolved in dichloromethane–
chloroform (1:1, 15 cm3). m-Chloroperoxybenzoic acid (200
mg, 0.116 mmol) and absolute ethanol (5 cm3) were added
sequentially, and stirring was continued for 5 min. The reaction
mixture was then poured into methanol (50 cm3) with stirring,
and the yellowish brown complex 2 gradually precipitated. The
solid was then collected on a frit, washed with dry methanol
CCDC reference number 186/952.
Stoichiometric oxidations of alkenes by [RuVI(dpp)O2] and the
isolation of [RuIV(dpp)(pz)2]
and dried in vacuo. Yield: 75%. IR(KBr): νRu᎐O 818, 1012 cmϪ1
᎐
To a degassed dichloromethane solution containing pyrazole
(2% w/w) and alkenes (2 mmol) was added complex 2 (30 mg,
22 µmol) under argon. After stirring for 12 h the reaction
mixture was filtered through a short column of neutral alumina
with hexanes–ethyl acetate (9:1) as the eluent to remove the
ruthenium complex. The organic products were then analysed
and quantified, after addition of internal standards, by either
GLC or 1H NMR spectroscopy.
1
(oxidation state marker band). H NMR (300 MHz, CDCl3,
SiMe4): δ 6.68 (m, 52 H) and 7.45 (d, J = 7.0 Hz, 8 H). FAB
mass spectrum: 1354 (Mϩ) (Found: C, 80.9; H, 4.19; N, 4.04.
Calc. for C92H60N4O2Ru: C, 81.5; H, 4.43; N, 4.13%).
Syntheses of osmium complexes
[OsII(dpp)(CO)(py)] 3. A mixture of [Os3(CO)12] (200 mg,
0.221 mmol) and H2dpp (200 mg, 0.164 mmol) in degassed
diethylene glycol monoethyl ether (60 cm3) was heated at reflux
for 1.5 h. The resulting red solution was diluted with benzene
The ruthenium complex was then eluted by dichloromethane,
after addition of acetonitrile led to the isolation of [RuIV-
(dpp)(pz)2] 5 as a dark purple crystalline solid. Yield: 80%.
1806
J. Chem. Soc., Dalton Trans., 1998, Pages 1805–1812

View Article Online
IR(KBr): 1006 cmϪ1 (oxidation state marker band). FAB mass
spectrum: 1458 (Mϩ), 1390 (Mϩ Ϫ pz) and 1322 (Mϩ Ϫ 2pz)
(Found: C, 79.90; H, 4.23; N, 7.57. Calc. for C98H66N8Ru: C,
80.82; H, 4.45; N, 7.70%). µeff (Evans’ method) = 3.24 µB.
Br
Br
Br
Br
Br
Br
Aerobic alkene oxidations catalysed by [RuVI(dpp)O2]
NH
N
In a typical experiment complex 2 (15 mg, 11 µmol) was stirred
with an excess of alkenes (2 mmol) in benzene–acetonitrile
(9:1), 2 cm3 under 1 atm oxygen for 4 h at 25 ЊC. The crude
reaction mixture was then passed through a short alumina col-
umn using hexanes–ethyl acetate (9:1) as the eluent to remove
the ruthenium complex. The organic products were character-
HN
N
Br
Br
1
ised and quantified by either GLC or H NMR spectroscopy
+
using internal standard methods.
Ph B(OH)2
Kinetic measurements
toluene, 98 °C, 7d
-KBr, -K3BO3, -CO2
[Pd(PPh3)4], K2CO3
Dichloromethane was distilled over CaH2 before use. Pyrazole
(99%, Aldrich) was used as received. Styrene and norbornene
(bicyclo[2.2.1]hept-2-ene) were purified using standard pro-
cedures. The rates of reduction of [RuIV(dpp)O2] by alkenes
were measured by monitoring the decrease of absorbance of the
ruthenium complex at 450 nm. The reactions were carried out
with [alkenes] ӷ [RuVI]. Plots of ln|A∞ Ϫ |At| vs. time are linear
Ph
Ph
Ph
Ph
over four half-lives. The pseudo-first-order rate constants (kobs
were determined on the basis of a least-squares fit by relation
(1) where A∞ and At are the absorbance at the completion of the
)
NH
N
HN
N
Ph
ln|A∞ Ϫ At| = Ϫkobst Ϫ ln|A∞ Ϫ A0|
(1)
Ph
Ph
Ph
reaction and at time t respectively; A∞ readings were obtained
for at least four half-lives. Second-order rate constants (k1) were
determined from the slopes of plots of kobs vs. alkene
concentration.
Yield: 75%
Scheme 1 Suzuki cross-coupling reaction
Results and Discussion
Synthesis and characterisation of [MVI(dpp)O2] (M ؍ Ru or
Os)
2,3,5,7,8,10,12,13,15,17,18,20-Dodecaphenylporphyrin
(H2dpp) was prepared in good yield (75%) using the Suzuki
cross-coupling reaction between phenylboronic acid and
2,3,7,8,12,13,17,18-octabromo-5,10,15,20-tetraphenylporph-
yrin, [H2obtpp], as the principal step (Scheme 1).11 The com-
pound H2obtpp was derived from bromination of [CuII(tpp)]
(yield = 76%), H2tpp = 5,10,15,20-tetraphenylporphyrin, fol-
lowed by demetallation using perchloric acid (yield = 91%).9
Compared with the procedure based on the condensation of
benzaldehyde and 3,4-diphenylpyrrole (yield = 5.7%),12 the
strategy employing the Suzuki cross-coupling reaction is
apparently more efficient and practical.
Insertion of Ru and Os into H2dpp occurs readily upon treat-
ing [M3(CO)12] (M = Ru or Os) with the free-base porphyrin
in toluene (Ru) and diethylene glycol monomethyl ether (Os).
The metal-carbonyl complexes [MII(dpp)(CO)(MeOH)] were
obtained after purification by column chromatography using
alumina. Recrystallisation of the metal carbonyl complexes in a
mixture of pyridine–dichloromethane–benzene led to quality
crystals of [MII(dpp)(CO)(py)] suitable for X-ray diffraction
studies.
Complexes 1 (Fig. 1) and 3 are isostructural, and their por-
phyrin macrocycles exhibit both saddle and ruffle distortions.
The edge-on-view of 1 (Fig. 2) shows that the pyrrole rings are
alternately tilted up and down with respect to the least-squares
plane of the 25-atom porphyrin core, and the phenyl rings are
rotated into the macrocycle plane to minimise contact between
the substituents. The saddle deformations in 1 and 3 are equally
striking in that the pyrrole carbons experience an average dis-
placement of 0.769 Å from the least-squares plane compared to
0.78 Å for free H2dpp13 and 0.99 for [NiII(dpp)].14 The angle
Fig. 1 Molecular structure of [RuII(dpp)(CO)(py)] with atom labelling
scheme
between the least-squares planes of the pyrrole rings and that
of the meso carbon atoms is 21Њ for both complexes, and the Ru
and Os atoms are displaced by 0.1006 and 0.0792 Å (Table 2)
from the 25-atom porphyrin core respectively. The Ru᎐N (py)
(2.240 Å) (Table 3) and the Os᎐N (py) (2.151 Å) bond distances
fall within the range found in other pyridine-ligated metallo-
porphyrins.
Treatment of [M(dpp)(CO)(MeOH)] (M = Ru or Os) com-
plexes with m-chloroperoxybenzoic acid in CH2Cl2–EtOH at
room temperature afforded the saddle-distorted oxometallo-
porphyrins 2 and 4 in good yield. The [MVI(dpp)O2] complexes
J. Chem. Soc., Dalton Trans., 1998, Pages 1805–1812
1807

View Article Online
Table 1 Crystallographic data and structure refinement for complexes 1, 3 and 5
1
3
5
Empirical formula
Crystal size/mm
M
C98H65N5ORuؒ2C6H6
0.10 × 0.10 × 0.40
1585.92
C98H65N5OOsؒ2C5H5N
0.06 × 0.20 × 0.24
1677.02
C98H66N8Ruؒ3CH3CNؒCH2Cl2ؒ7H2O
0.90 × 1.20 × 1.20
1791.0
Crystal symmetry
Space group
a/Å
b/Å
c/Å
α/Њ
β/Њ
Triclinic
P1 (no. 2)
Triclinic
P1 (no. 2)
Orthorhombic
Cmc21 (no. 36)
24.499(5)
16.391(3)
24.527(8)
¯
¯
12.662(3)
17.600(7)
18.904(5)
83.74(3)
83.34(2)
85.94(3)
4152.1(23)
2
1.269
2.372
10 816
5003
10 797
0.069
12.673(3)
17.649(4)
18.937(4)
84.08(3)
83.52(3)
85.95(3)
4179(20)
2
1.333
15.78
15 231
9651 [I > 6.0σ(I)]
14 113
γ/Њ
U/Å3
9849(4)
4
Z
Dc/g cmϪ3
1.208
3.911
4553
2975
4553
0.063
0.074
1.99
µ/cmϪ1
No. measured reflections
No. observed reflections [I > 2.0σ(I)]
No. unique reflections
RF
0.078
0.115
1.48
RЈ
0.065
1.70
Goodness of fit
Table 2 Average out-of-plane displacements and dihedral angles
between least-squares planes
Cα′
Cα
Cβ′
Cβ
Ph
Ph
Ph
Ph
Ph
Ph
N2
N1
N4
M
N3
Ph
Ph
M = Ru, Os
[RuII(dpp)(CO)(py)]
[RuIV(dpp)(pz)2]
Displacement (Å)a
Ru
N1,3
N2,4
Cα
CαЈ
Cβ
0.101
ϩ0.077
Ϫ0.0926
0.4561
0.132
0.890
0.649
0.039
ϩ0.066
Ϫ0.068
0.379
Fig. 2 Edge-on-view of [RuII(dpp)(CO)(py)] (all phenyl groups are
removed for clarity)
1.06
CβЈ
Dihedral angle/Њ
were isolated as air-stable solids. The presence of the weakly co-
ordinating alcohol has proved crucial for the success of the
synthesis; no complications such as dimerisation were
observed. The absence of the CO stretch indicates the starting
carbonyl complexes have been completely consumed. The
dioxo-ruthenium() and -osmium() complexes show intense
IR absorptions at 818 and 838 cmϪ1 assigned to asymmetric
Pyrrole tiltingb
20.93
75.50
66.48
<0.05
27.78
66.95
47.21
<0.05
Pyrrole phenyl twistc
meso-Phenyl twistd
Pyrrole planaritye
a From the least-squares plane of the 25-atom porphyrin core.
b Dihedral angle between the least-squares plane of a pyrrole ring and
that of the four nitrogen atoms. c Dihedral angle between the least-
squares planes of the pyrrole and phenyl rings. d Dihedral angle
between the least-squares plane of a meso-phenyl ring and that of the
meso-carbon atoms. e Largest deviation of any atoms in a pyrrole ring
from the least-squares plane of the pyrrole ring.
O᎐Ru᎐O and O᎐Os᎐O stretches respectively, and the oxidation
᎐
᎐
᎐
᎐
state marker bands at 1020 (Ru) and 1011 cmϪ1 (Os) suggested
that the ruthenium (2) and the osmium (4) centres are at ϩ6
oxidation state.15
Magnetic susceptibility measurement established that com-
plexes 2 and 4 are diamagnetic in accordance with a (dxy)2 elec-
tronic ground state (here the O᎐M᎐O axis is taken as the z axis).
᎐
᎐
The 1H NMR spectra (298 K) of 2 and 4 show a broad doublet
at δ 7.44 (J = 7.0 Hz, 8 H) assignable to the ortho protons at the
meso-substituted phenyl rings and two sets of multiplets corre-
sponding to the remaining 52 protons. The free H2dpp is known
to undergo two dynamic processes: NH tautomerism and
macrocyclic inversion.12,13 Using variable-temperature NMR
the activation free energy (∆G‡) for the ring inversion process
was estimated to be 10.9 kcal molϪ1. No dynamic processes
were observed for 1–4 (∆G‡ < 7.0 kcal molϪ1), although the
porphyrin ligand must be non-planar. Presumably, the macro-
1808
J. Chem. Soc., Dalton Trans., 1998, Pages 1805–1812

View Article Online
cyclic inversion barriers for the ruthenium and osmium com-
plexes are too low to be detected. The small inversion barrier is
not inconceivable since the macrocyclic inversion barrier
observed in the free H2dpp porphyrin should arise from the
repulsion between the NH protons.16
As H2dpp adopts a non-planar conformation in solution, the
macrocyclic distortion should have an effect on the UV/VIS
absorption spectra of the ruthenium and osmium complexes.
The absorption maxima for 1–4 and the analogous complexes
with the essentially ‘planar’ tpp ligand are summarised in Table
4. The optical data show that the absorption maxima of the
Soret bands and the Q bands for 1–4 are all red shifted com-
pared with those of the complexes containing tpp. Theoretical
calculations1e,18 performed by other workers have indeed sug-
gested that the distortion of the macrocycle would lead to a
large destabilisation of the ligand’s HOMOs together with a
small destabilisation of the LUMOs, hence the smaller
HOMO Ϫ LUMO gap results in the red shift of the optical
spectra.
The electrochemical potentials of complexes 1–4 (0.1 
NBu4PF6–CH2Cl2) listed in Table 5 also reveal that the steric-
ally induced macrocyclic distortion promotes oxidation by rais-
ing the HOMO energies. For instance, the cyclic voltammogram
of [RuVI(dpp)O2] in dichloromethane shows one quasi-
reversible and one irreversible oxidation couple together with
one irreversible reduction wave. With reference to previous
studies,19 the quasi-reversible oxidation couple is tentatively
assigned as the ligand–centred oxidation [RuVI(dpp)O2] Ϫ
ϩ
eϪ → [RuVI(dpp )O2], E = 0.55 V vs. ferrocenium–ferrocene.
₁
᝽
₂
A noticeable drop in the magnitude of the oxidation potential of
240 mV due to the conformational distortion of the porphyrin
ring is observed when compared with that of [RuVI(tpp)O2]
Table 3 Selected bond distances (Å) and angles (Њ) for [RuII(dpp)-
(CO)(py)] and [RuIV(dpp)(pz)2]
₁
(E = 0.79 V) (entry 3 vs. 4). A comparable drop in the oxidation
₂
potential (180 mV) is also noticed for the oxidation of
[OsVI(dpp)O2] in dichloromethane when compared with the
analogous tpp complex (entry 7 vs. 8). Since 1 and 3 are iso-
structural, we would anticipate that the dodecaphenylporphyrin
ligand in 2 and 4 should exhibit similar degrees of conform-
ational distortion; the HOMOs have experienced similar
degrees of destabilisation vs. the analogous tpp complexes of
Ru and Os.
[RuII(dpp)(CO)(py)]
[RuIV(dpp)(pz)2]
Ru᎐N(1)
Ru᎐N(5)
Ru᎐C
N(1)᎐C(1)
C(1)᎐C(2)
C(2)᎐C(3)
C(1)᎐C(20)
C(2)᎐C(21)
C(21)᎐C(22)
C(20)᎐C(87)
2.048(7)
2.240(8)
1.933(12) Ru᎐N(5)
1.363(10) N(1)᎐C(2)
1.464(12) N(4)᎐N/C(47)
1.338(13) C(1)᎐C(1a)
1.424(13) C(1)᎐C(2)
1.469(12) C(1)᎐C(11)
1.388(15) C(11)᎐C(12)
1.468(11) C(2)᎐C(3)
Ru᎐N(1)
Ru᎐N(4)
2.042(11)
2.022(13)
2.083(12)
1.385(13)
1.288(17)
1.386(23)
1.422(16)
1.565(15)
1.501(18)
1.342(15)
Stoichiometric alkene oxidations by [RuVI(dpp)O2]
The complex [RuVI(dpp)O2] 2 is a competant oxidant of alkenes
(Table 6), and as expected the analogous dioxoosmium()
complex 4 is unreactive towards organic oxidations. Stoichio-
metric oxidation of some representative alkenes such as nor-
N(1)᎐Ru᎐N(2)
N(1)᎐Ru᎐N(3)
N(1)᎐Ru᎐N(5)
N(1)᎐Ru᎐C
N(2)᎐Ru᎐N(3)
N(2)᎐Ru᎐N(4)
N(2)᎐Ru᎐N(5)
N(2)᎐Ru᎐C
N(3)᎐Ru᎐N(4)
N(3)᎐Ru᎐N(5)
N(3)᎐Ru᎐C
N(4)᎐Ru᎐N(5)
N(4)᎐Ru᎐C
89.7(3)
170.0(3)
83.2(3)
90.9(3)
89.7(3)
179.6(3)
90.8(3)
87.5(4)
90.4(3)
86.9(3)
99.1(3)
88.8(3)
92.9(4)
173.8(3)
N(1)᎐Ru᎐N(2)
N(1)᎐Ru᎐N(3)
N(1)᎐Ru᎐N(4)
N(2)᎐Ru᎐N(4)
N(2)᎐Ru᎐N(2a)
N(2)᎐Ru᎐N(3)
N(2)᎐Ru᎐N(4)
N(2)᎐Ru᎐N(5)
N(2a)᎐Ru᎐N(3)
N(2a)᎐Ru᎐N(4)
N(2a)᎐Ru᎐N(5)
N(3)᎐Ru᎐N(4)
N(3)᎐Ru᎐N(5)
N(4)᎐Ru᎐N(5)
91.6(6)
173.9(6)
86.6(9)
89.4(3)
176.5(7)
88.5(6)
89.4(3)
90.8(3)
88.5(6)
89.4(3)
Table 5 Electrochemical data (in V vs. ferrocenium–ferrocene) for the
non-planar ruthenium and osmium dpp complexes
ox
₁
E
₂
Ered a
III
90.8(3)
99.5(9)
Entry
Compound
I
II
1
2
3
4
5
6
7
8
9
[RuII(dpp)(CO)(py)]
[RuII(tpp)(CO)(py)]
[RuVI(dpp)O2]d
0.51b
0.89c
0.90a
0.79c
0.74b
1.01b
0.97c
0.80c
0.78c
0.13b
0.37b
0.55c
86.1(11)
174.4(13)
120.6(12)
107.2(8)
126.4(10)
98.5(16)
109.0(10)
126.3(10)
131.4(10)
127.6(10)
129.6(13)
115.0(12)
117.4(9)
N(5)᎐Ru᎐C
Ru᎐C᎐O
Ϫ1.00
Ϫ0.83
173.8(10) Ru᎐N(2)᎐C(4)
[RuVI(tpp)O2]d
C(1)᎐N(1)᎐C(4)
N(1)᎐C(1)᎐C(2)
C(1)᎐C(2)᎐C(3)
C(1)᎐C(2)᎐C(21)
C(3)᎐C(2)᎐C(21)
N(1)᎐C(4)᎐C(5)
C(4)᎐C(5)᎐C(6)
C(4)᎐C(5)᎐C(33)
109.4(7)
107.9(7)
106.0(7)
126.5(8)
126.2(9)
123.7(8)
126.6(8)
114.2(7)
C(4)᎐N(2)᎐C(7)
[OsII(dpp)(CO)(py)]
[OsII(tpp)(CO)(py)]
[OsVI(dpp)O2]d
0.11b
0.26b
0.62c
Ru᎐N(4)᎐N/C(47)
N/C(47)᎐N(4)᎐N/C(47a)
N(2)᎐C(4)᎐C(5)
C(4)᎐C(5)᎐C(23)
C(6)᎐C(5)᎐C(23)
C(5)᎐C(23)᎐C(3)
C(5)᎐C(23)᎐C(24)
Ϫ1.20
Ϫ1.34
Ϫ0.41
[OsVI(tpp)O2]
[RuIV(dpp)(pz)2]d
0.45b
Conditions: in 0.1 mol dmϪ3 NBu4PF6 in CH2Cl2 using Ag–AgNO3
in MeCN as the reference electrode and a glassy carbon working
C(27)᎐C(28)᎐C(29) 119.0(13) C(24)᎐C(23)᎐C(28)
C(4)᎐C(3)᎐C(17)
electrode; scan rate 100 mV sϪ1 a Irreversible. b Reversible couple.
.
c Quasi-reversible couple. d Solvent: CH2Cl2–MeCN (10:1).
Table 4 The UV/VIS spectra data of non-planar ruthenium and osmium dpp complexes in CH2Cl2 at room temperature
λ/nm (log ε)
Compound
B bands
Q bands
H2dpp
H2tpp
468 (5.38)
416 (5.40)
564 (4.13), 618 (4.15), 718 (4.01)
514 (4.19), 548 (4.01), 589 (4.08), 645 (4.01)
505 (sh) (4.56), 578 (4.38)
5.32 (4.25), 568 (3.57), 495 (sh) (3.71)
550 (3.85), 586 (3.59)
518 (4.24), 545 (sh) (3.89)
474 (sh) (4.44), 556 (4.10)
518 (4.23)
[RuII(dpp)(CO)(py)]
[RuII(tpp)(CO)(py)]*
[RuVI(dpp)O2]
[RuVI(tpp)O2]
430 (5.26), 271 (4.82)
413 (5.45)
448 (4.78), 358 (sh) (3.98)
418 (5.29), 340 (sh) (4.19)
428 (5.03), 279 (4.68)
406 (4.45)
419 (4.68), 356 (4.27)
395 (4.82), 339 (4.15)
436 (5.24), 332 (4.71)
[OsII(dpp)(CO)(py)]
[OsII(tpp)(CO)(py)]
[OsVI(dpp)O2]
[OsVI(tpp)O2]
498 (sh) (4.23), 608 (3.89), 669 (sh) (3.67)
479 (sh) (4.08), 586 (3.71)
521 (4.66), 551 (4.21)
[RuIV(dpp)(pz)2]
* Data taken from ref. 17.
J. Chem. Soc., Dalton Trans., 1998, Pages 1805–1812
1809

View Article Online
Table 6 Stoichiometric oxidation of alkenes by [RuVI(dpp)O2]
Yield
(%)a
Entry
Alkene
Products
1
2
Norbornene
Styrene
2,3-exo-Epoxynorbornane
Styrene oxide
Benzaldehyde
>99b
>99b
n.d.c
Phenylacetylaldehyde
Cyclohexene oxide
Cyclohex-2-en-1-ol
Cyclohex-2-en-1-one
cis-Stilbene oxide
trans-Stilbene oxide
Benzaldehyde
n.d.
3
4
5
Cyclohexene
14b
80b
3b
cis-Stilbene
>99d
n.d.
n.d.
85d
cis-β-Methylstyrene cis-β-Methylstyrene oxide
trans-β-Methylstyrene oxide
14d
a Yield based on [RuVI(dpp)O2]. b Determined by gas chromatography.
c n.d. = Not detected. d Determined by 1H NMR spectroscopy using
1,1-diphenylethylene as the internal standard.
R1
R1
•
R2
R2
C
Fig. 3 Molecular structure of [RuIV(dpp)(pz)2] with atom labelling
scheme
O
O
O
C rotation
•
R1
R2
+
Ru(L)
O
Ru(L)
O
Ru(L)
O
L = porphyrin
R1
R2
R1
R2
O
O
cis-epoxide
trans-epoxide
Scheme 2
bornene (entry 1) and styrene (entry 2) in CH2Cl2 containing
2% (w/w) of pyrazole afforded almost quantitatively (>99%)
exo-epoxynorbornane and styrene oxide respectively without
C᎐C bond cleavages and rearrangement products. Allylic C᎐H
᎐
oxidation prevailed when cyclohexane (entry 3) was used as the
substrate; 80% of cyclohex-2-en-1-ol together with only 18% of
cyclohexene oxide were formed. Oxidation of cis-stilbene was
accompanied by full stereoretention giving rise to cis-stilbene
oxide exclusively. However [RuIV(dpp)O2] failed to react with
trans-stilbene, i.e. no trans-stilbene oxide was detected. The
preference of the dioxoruthenium() complex for cis-alkene
results from the unfavourable interaction of the phenyl rings of
trans-stilbene with the porphyrin ligand, based on the ‘side-on
approach’ model proposed first by Groves and co-workers.20
The product stereoselectivity for the oxidation of cis-alkenes
could be influenced by the steric bulkiness of the peripheral
substituents of the porphyrin ring. For instance, the sterically
less encumbered [RuVI(oep)O2]19 (H2oep = 2,3,7,8,12,13,17,18-
octaethylporphyrin) reacts with cis-stilbene to afford a mixture
of cis-(16%) and trans-stilbene oxide (44%). The predominant
formation of trans-epoxide was explained by the formation of
a carboradical intermediate, and the unhindered C᎐C bond
rotation causes cis to trans isomerisation (Scheme 2).21 Indeed,
reaction of 2 with cis-alkenes, which have smaller substituents
such as cis-β-methylstyrene, produces a significant amount of
trans-epoxide (14%) (entry 5).
Fig. 4 The UV/VIS spectral trace for the reaction of [RuVI(dpp)O2]
(1 × 10Ϫ5 mol dmϪ3) with styrene (5 × 10Ϫ2 mol dmϪ3) in CH2Cl2 (2%
pyrazole) in the range 300–700 nm
k1
[RuVI(dpp)O2]
+
alkene
[RuIV(dpp)O]
+ epoxide
2
2 x Hpz
–H2O
k2
[RuIV(dpp)(pz)2]
5
Scheme 3 k2 >> k1
paramagnetic species, the µeff of 3.24 µB (by Evans’ method in
CHCl3–MeOH at room temperature) is slightly greater than the
spin-only value (2.83 µB) required for two unpaired electrons.
The crystal data, structural distortion parameters, bond dis-
tances and angles for 5 are summarized in Tables 1–3. The
porphyrin macrocycle of 5 (Fig. 3) also exhibits out-of-plane
distortions creating a saddle-shape structure similar to that
found in 1 and 3. The average Ru᎐N (pz) bond distances are
found to be 2.022(13) and 2.083(12) Å, these values are com-
parable to 2.111(11) Å for Ru᎐N (pz) and 2.025(11) Å for Ru᎐N
(amide) in [RuIV(tpp)(pz)(NHSO2C6H4Me-p)],22b and the Ru᎐N
(amide) distances of 1.987–2.044(5) Å found in [RuIV(chbae)-
For all oxidation reactions, [RuIV(dpp)(pz)2]
5 (pz =
pyrazolate) was isolated as the final product and characterised
by crystallographic means. The monomeric complex 5 has
two axial pyrazolate ligands, and similar well characterised
ruthenium() porphyrin bis(amide) complexes have limited
precedents in the literature. The oxidation marker band
(IR spectroscopy) located at 1006 cmϪ1 is consistent with a
ruthenium() formulation.21 The complex is an air-stable and
(PPh3)(py)],23
[chbae = 1,2-bis(3,5-dichloro-2-hydroxybenz-
amido)ethane tetraanion]. The two pz ligands are trans to each
other with the N (pz)᎐Ru᎐N (pz) bond angle of 174.4(13)Њ.
1810
J. Chem. Soc., Dalton Trans., 1998, Pages 1805–1812

View Article Online
deactivation. Groves and Quinn19c showed that the sterically
encumbered [RuVI(tmp)O2] (H2tmp = 5,10,15,20-tetramesityl-
porphyrin) can avoid the µ-oxo dimer formation and give
higher catalyst turnover in a range of 22–46 for oxidation of
norbornene and cis-β-methylstyrene. In this work, the turnover
number and selectivity of the Ru᎐dpp system are comparable to
that of the Ru᎐tmp system. By examining the crystal structures
of [RuII(dpp)(CO)(py)] and [RuIV(dpp)(pz)2] complexes, the
dioxoruthenium() [RuVI(dpp)O2] and [RuVI(tpp)O2] com-
plexes are anticipated to have similar steric environments, and
hence their propensity to µ-oxo dimer formation under the
catalytic conditions would be alike. The apparent increase in
catalytic activity vs. [RuVI(tpp)O2] and [RuVI(oep)O2] could be
due to the lower oxidation potential of 2 induced by the non-
planar distortion of the porphyrin macrocycle.
Table 7 Aerobic oxidation of alkenes catalysed by [RuVI(dpp)O2]
Alkenes
Products
Turnovera
40b
32b
7b
Norbornene
Styrene
2,3-exo-Epoxynorbornane
Styrene oxide
Benzaldehyde
Phenylacetylaldehyde
Cyclohexene oxide
Cyclohex-2-en-1-ol
Cyclohex-2-en-1-one
Cyclooctene oxide
cis-Stilbene oxide
trans-Stilbene oxide
Benzaldehyde
Trace
Cyclohexene
8b
20b
11b
24b
10c
3c
Cyclooctene
cis-Stilbene
Trace
a Based on the amount of [RuVI(dpp)O2]. b Yield determined by gas
chromatography using 1,4-dichlorobenzene as the internal standard.
c Yield determined by 1H NMR spectroscopy using 1,1-diphenyl-
ethylene as the internal standard.
Conclusion
Two highly distorted oxometalloporphyrin (Ru and Os)
complexes of 2,3,5,7,8,10,12,13,15,17,18,20-dodecaphenylpor-
phyrin have been synthesized based on the procedure previously
developed for the preparation of [RuVI(por)O2] with suitable
modification. X-Ray analyses of [MII(dpp)(CO)(py)], M = Ru
or Os, and 5 revealed dpp exhibits both saddle and ruffle
distortions. Metalloporphyrins 2 and 4 are believed to adopt
similar conformational distortion based on the UV/VIS spec-
troscopic and electrochemical studies. Complex 2 is a com-
petent oxidant for alkene epoxidation and shows an enhanced
catalytic activity towards aerobic epoxidation when compared
with [RuVI(tpp)O2] and [RuVI(oep)O2].
Fig. 4 depicts the UV/VIS spectral change for the reaction of
complex 2 with styrene in dichloromethane containing 2%
pyrazole (Hpz). The observation of isosbestic points through-
out the reaction indicates that the putative oxoruthenium()
intermediate should be at very low concentration compared to 2
and 5. Scheme 3 is proposed for the oxidation reactions. Kinetic
experiments revealed that the reactions of 2 with alkenes in
dichloromethane (2% of Hpz) display clean first-order kinetics
and the observed rate constants, kobs, were determined by moni-
toring at 450 nm under pseudo-first-order conditions, i.e.
[alkene] ӷ [2]. A second-order rate law, rate = k1[2][alkene], is
established. With norbornene and styrene the second-order rate
constants (k1) in dichloromethane at 25.9 ЊC are (3.79 0.04) ×
10Ϫ3 and (4.78 0.09) × 10Ϫ3 dm3 molϪ1 sϪ1 respectively. These
values are close to the second-order rate constants for the oxi-
Acknowledgements
We acknowledge support from The University of Hong Kong
and Hong Kong Research Grant Council.
dation reactions of styrene [(4.30 0.3) × 10Ϫ3 dm3 molϪ1 sϪ1
]
and norbornene [(3.01 0.09) × 10Ϫ3 dm3 molϪ1 Ϫ1] with the
s
planar [RuVI(tpp)O2] complex.24 To account for the similar reac-
tivities of the planar and saddle-distorted dioxoruthenium()
porphyrin complexes, we propose that the macrocyclic distor-
tion should mainly destabilise the ligand’s HOMO energies;
References
1 (a) J. L. Hoard, in Porphyrins and Metalloporphyrins, ed. K. M.
Smith, Elsevier, Amsterdam, 1975, ch. 8; (b) W. R. Scheidt, in The
Porphyrins, ed. D. Dolphin, Academic Press, New York, 1979, vol. 3,
ch. 10; (c) W. R. Scheidt and Y. J. Lee, Struct. Bonding (Berlin), 1987,
64, 1; (d) K. M. Barkigia, L. Chantranupong, K. M. Smith and
J. Fajer, J. Am. Chem. Soc., 1988, 110, 7566; (e) M. Ravikanth and
T. K. Chandrashekar, Struct. Bonding (Berlin), 1995, 82, 105.
2 J. Deisenhofer and H. Michel, Science, 1989, 245, 1463.
3 D. E. Tronrud, M. F. Schmid and B. W. Matthews, J. Mol. Biol.,
1986, 188, 443.
however, the LUMOs related to the O᎐Ru᎐O moiety are
᎐
᎐
relatively less perturbed, and therefore their electrophilicities
would not differ significantly.25
Catalytic aerobic alkene oxidations by [RuVI(dpp)O2]
The complex [RuVI(dpp)O2] can mediate alkene epoxidation
with the use of molecular dioxygen (1 atm). In a typical experi-
ment, 2 (15 mg, 11 µmol) was stirred with 2 mmol of alkenes in
benzene–acetonitrile (9:1) at room temperature under dioxygen
(1 atm) for 4 h. The reaction products were quantitatively
analysed and identified by gas chromatography. A prolonged
reaction for 12 to 24 h did not afford higher turnover. After 5 h
of the reactions, traces of 2, [Ru(dpp)(CO)] and some unidenti-
fied ruthenium complex were isolated.
When norbornene and cyclooctene were used as the sub-
strates, the corresponding epoxides were the dominant prod-
ucts (Table 7). Allylic C᎐H oxidation is the major reaction
undertaken by cyclohexene leading to cyclohex-2-en-1-ol and
cyclohex-2-en-1-one, together with a minor production of
cyclohexene oxide. Oxidation of styrene afforded a mixture of
4 L. R. Furenlid, M. W. Renner and J. Fajer, J. Am. Chem. Soc., 1990,
112, 8987 and refs. therein.
5 M. W. Grinstaff, M. G. Hill, J. A. Labinger and H. B. Gray, Science,
1994, 264, 1311; J. A. Labinger, Catal. Lett., 1994, 26, 95.
6 T. Shingaki, K. Miura, T. Higuchi, M. Hirobe and T. Nagano,
Chem. Commun., 1997, 861; Z. Gross and S. Ini, J. Org. Chem., 1997,
62, 5514; E. Galardon, P. Le Maux and G. Simonneaux, Chem.
Commun., 1997, 927.
7 D. R. Coulson, Inorg. Synth., 1972, 13, 121.
8 A. D. Alder, J. Org. Chem., 1967, 32, 476; P. Rothemund and A. R.
Menotti, J. Am. Chem. Soc., 1948, 70, 1808.
9 P. Bhyrappa and V. Krishnan, Inorg. Chem., 1991, 30, 239.
10 TEXSAN-TEXRAY, Structure Analysis Package, Molecular
Structure Corporation, Houston, TX, 1985; SHELXL 93, G. M.
Sheldrick, University of Göttingen, 1993; NRCC-SDP-VAX, E. J.
Cabe, Y. Le Page, J. P. Charland, F. L. Lee and P. S. White, J. Appl.
Crystallogr., 1989, 22, 384.
styrene oxide and benzaldehyde (C᎐C cleavage product) in a
᎐
ratio of 4:1. The reaction of cis-stilbene under the oxidative
conditions remains largely stereoretentive albeit with a small
amount of trans-oxide formed.
When styrene and norbornene were used as the standard sub-
strates [RuVI(tpp)O2] and [RuVI(oep)O2] attained <5 turnovers
for the catalytic aerobic oxidation under identical reaction con-
ditions. The formation of the catalytically inactive µ-oxo
ruthenium dimers should be responsible for the catalyst
11 X. Zhou, Z. Y. Zhou, T. C. W. Mak and K. S. Chan, J. Chem. Soc.,
Perkin Trans. 1, 1994, 2519.
12 C. J. Medforth and K. M. Smith, Tetrahedron Lett., 1990, 31,
5583.
13 C. J. Medforth, M. O. Senge, K. M. Smith, L. D. Sparks and J. A.
Shelnutt, J. Am. Chem. Soc., 1992, 114, 9859.
14 K. M. Barkigia, M. W. Renner, L. R. Furenlid, C. J. Medforth,
K. M. Smith and J. Fajer, J. Am. Chem. Soc., 1993, 115, 3627.
15 J. T. Groves and K.-H. Ahn, Inorg. Chem., 1987, 26, 3831.
J. Chem. Soc., Dalton Trans., 1998, Pages 1805–1812
1811

View Article Online
16 C. J. Medforth, M. D. Berber, K. M. Smith and J. A. Shelnutt,
Tetrahedron Lett., 1990, 31, 3719.
17 J. J. Bonnet, S. S. Eaton, G. R. Eaton, R. H. Holm and J. A. Ibers,
J. Am. Chem. Soc., 1973, 95, 2141.
18 T. Takeuchi, H. B. Gray and W. A. Goddard, III. J. Am. Chem. Soc.,
1994, 116, 9730.
19 (a) W. H. Leung and C. M. Che, J. Am. Chem. Soc., 1989, 111, 8812;
(b) J. T. Groves and R. Quinn, Inorg. Chem., 1984, 23, 3844; (c) J. T.
Groves and R. Quinn, J. Am. Chem. Soc., 1985, 107, 5790.
20 J. T. Groves, J. Chem. Educ., 1985, 62, 928; J. T. Groves, Y. Hau and
D. V. van Engen, J. Chem. Soc., Chem. Commun., 1990, 436.
21 A. J. Castellino and T. C. Bruice, J. Am. Chem. Soc., 1988, 110, 158;
J. T. Groves, K. H. Ahn and R. Quinn, J. Am. Chem. Soc., 1988, 110,
4217.
22 J. S. Huang, C. M. Che, Z. Y. Li and C. K. Poon, Inorg. Chem., 1992,
31, 1315; S. M. Au, W. H. Fung, M. C. Cheng, S. M. Peng and C. M.
Che, Chem. Commun., 1997, 1655.
23 C. M. Che, W. K. Cheng, W. H. Leung and T. C. W. Mak, J. Chem.
Soc., Chem. Commun., 1987, 418.
24 C. Ho, W. H. Leung and C. M. Che, J. Chem. Soc., Dalton Trans.,
1991, 2933.
25 H. Fujii, J. Am. Chem. Soc., 1993, 115, 4641.
Received 20th January 1998; Paper 8/00535D
1812
J. Chem. Soc., Dalton Trans., 1998, Pages 1805–1812