Fast catalytic hydroxylation of hydrocarbons with ruthenium porphyrins

Article abstract of DOI:10.1021/ic0520566

Ruthenium porphyrin complexes such as carbonylruthenium(II) tetrakispentafluorophenylporphyrin [Ru^II(TPFPP)-(CO)] were found to be efficient catalysts for the hydroxylation of alkanes in the presence of 2,6-dichloropyridine N-oxide as the oxidant under mild, nonacidic conditions. Up to 14 800 turnovers (TO) and rates of 800 TO/min were obtained for the hydroxylation of adamantane. The hydroxylation of cis-decalin afforded cis-9-decalol and cis-decalin-9,10-diol, exclusively, thus, excluding a long-lived radicals mechanism. The kinetics of product evolution in a typical catalytic oxygenation showed an initial induction period followed by a fast, apparently zero-order phase with maximum rates and high efficiencies. Deuterium isotope effects (K_H/k_D) in the range of 4.2-6.4 were found for the hydroxylation of alkanes. A Hammett treatment of the data for the oxidation of para-substituted toluene derivatives showed a linear correlation with a highly negative ρ⁺ value of -2.0. On the basis-of kinetic and spectroscopic evidence, Ru^VI(TPFPP)(O)₂, Ru ^II(TPFPP)(CO), and Ru^IV(TPFPP)Cl₂ observed during catalysis were ruled out as candidates for the active catalyst responsible for the high efficiencies and turnover rates in the oxidation reactions. The fastest rates of adamantane hydroxylation with 2,6-dichloropyridine N-oxide were achieved by the reductive activation of Ru^IV(TPFPP)Cl₂ with a zinc amalgam. This redox activation is consistent with the formation of an active Ru(III) intermediate in situ by a one-electron reduction of the Ru(IV) porphyrin. EPR spectra characteristic of Ru(III) have been observed upon the reduction of Ru^IV(TPFPP)Cl ₂ with a zinc amalgam. In the adamantane oxidation mediated with Ru^III(TPFPP)(OEt), it was found that, during this process, the Ru(III) porphyrin was gradually converted to a dioxoRu(VI) porphyrin. Concomitant with this conversion, the reaction rates decreased. Catalyst activation was also stimulated by autoxidation of the solvent CH ₂Cl₂. On the basis of these data, a mechanism is proposed that incorporates a fast cycle involving metastable Ru(III) and oxoRu(V) intermediates and a slow oxidation cycle, mediated by oxoRu(IV) and trans-dioxoRu(VI) porphyrin complexes.

Full text of DOI:10.1021/ic0520566

Inorg. Chem. 2006, 45, 4769−4782
Fast Catalytic Hydroxylation of Hydrocarbons with Ruthenium
Porphyrins
Chuanqing Wang, Kirill V. Shalyaev, Marcella Bonchio, Tommaso Carofiglio, and John T. Groves*
Department of Chemistry, Princeton UniVersity, Princeton, New Jersey 08544
Received November 30, 2005
Ruthenium porphyrin complexes such as carbonylruthenium(II) tetrakispentafluorophenylporphyrin [RuII(TPFPP)-
(CO)] were found to be efficient catalysts for the hydroxylation of alkanes in the presence of 2,6-dichloropyridine
N-oxide as the oxidant under mild, nonacidic conditions. Up to 14 800 turnovers (TO) and rates of 800 TO/min
were obtained for the hydroxylation of adamantane. The hydroxylation of cis-decalin afforded cis-9-decalol and
cis-decalin-9,10-diol, exclusively, thus, excluding a long-lived radicals mechanism. The kinetics of product evolution
in a typical catalytic oxygenation showed an initial induction period followed by a fast, apparently zero-order phase
with maximum rates and high efficiencies. Deuterium isotope effects (kH/kD) in the range of 4.2
the hydroxylation of alkanes. A Hammett treatment of the data for the oxidation of para-substituted toluene derivatives
showed a linear correlation with a highly negative F+ value of
2.0. On the basis of kinetic and spectroscopic
−6.4 were found for
−
evidence, RuVI(TPFPP)(O)2, RuII(TPFPP)(CO), and RuIV(TPFPP)Cl2 observed during catalysis were ruled out as
candidates for the active catalyst responsible for the high efficiencies and turnover rates in the oxidation reactions.
The fastest rates of adamantane hydroxylation with 2,6-dichloropyridine N-oxide were achieved by the reductive
activation of RuIV(TPFPP)Cl2 with a zinc amalgam. This redox activation is consistent with the formation of an
active Ru(III) intermediate in situ by a one-electron reduction of the Ru(IV) porphyrin. EPR spectra characteristic
of Ru(III) have been observed upon the reduction of RuIV(TPFPP)Cl2 with a zinc amalgam. In the adamantane
oxidation mediated with RuIII(TPFPP)(OEt), it was found that, during this process, the Ru(III) porphyrin was gradually
converted to a dioxoRu(VI) porphyrin. Concomitant with this conversion, the reaction rates decreased. Catalyst
activation was also stimulated by autoxidation of the solvent CH2Cl2. On the basis of these data, a mechanism is
proposed that incorporates a “fast” cycle involving metastable Ru(III) and oxoRu(V) intermediates and a “slow”
oxidation cycle, mediated by oxoRu(IV) and trans-dioxoRu(VI) porphyrin complexes.
Introduction
and related complexes have received considerable attention
because of the close periodic relationship to the biologically
significant iron and the rich coordination and redox chemistry
of ruthenium, which spans oxidation states from +2 to +7
in the porphyrin ligand environment.5-17
Nature selected iron porphyrins to mediate intrinsically
difficult oxidations with dioxygen at the heme center of
cytochrome P450 monooxygenases.1-4 An extensive variety
of cytochrome P450 models has been developed over the
past three decades and applied to hydrocarbon oxidations
with the aims of both developing useful catalysts and probing
the mechanisms of these processes. Ruthenium porphyrins
(5) Griffith, W. P. Chem. Soc. ReV. 1992, 21, 179-185.
(6) Mlodnicka, T.; James, B. R.; Montanari, F.; Casella, L. Metallopor-
phyrins Catalyzed Oxidations; Kluwer Academic Publishers: Neth-
erlands, 1994.
* Author to whom correspondence should be addressed. Phone: (609)
258-3593. Fax: (609) 258-0348. E-mail: jtgroves@princeton.edu.
(1) Groves, J. T. In Cytochrome P450: Structure, Mechanism, and
Biochemistry, 3rd ed.; Ortiz de Montellano, P. R., Ed.; Klewer
Academic/Plenum: New York, 2004; pp 1-44.
(7) Ito, R.; Umezawa, N.; Higuchi, T. J. Am. Chem. Soc. 2005, 127, 834-
835.
(8) Jitsukawa, K.; Oka, Y.; Yamaguchi, S.; Masuda, H. Inorg. Chem. 2004,
43, 8119-8129.
(9) Chen, J.; Che, C. M. Angew. Chem., Int. Ed. 2004, 43, 4950-4954.
(10) Chatterjee, D.; Mitra, A.; Shepherd, R. E. Inorg. Chim. Acta 2004,
357, 980-990.
(2) Groves, J. T. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 3569-3574.
(3) He, X.; Cryle, M. J.; DeVoss, J. J.; de Montellano, P. R. O. Drug
Metab. ReV. 2004, 36, 6.
(11) Le Maux, P.; Lukas, M.; Simonneaux, G. J. Mol. Catal. A 2003, 206,
95-103.
(4) Meunier, B.; de Visser, S. P.; Shaik, S. Chem. ReV. 2004, 104, 3947-
3980.
(12) Berkessel, A.; Kaiser, P.; Lex, J. Chem.sEur. J. 2003, 9, 4746-4756.
10.1021/ic0520566 CCC: $33.50
Published on Web 05/18/2006
© 2006 American Chemical Society
Inorganic Chemistry, Vol. 45, No. 12, 2006 4769

Wang et al.
Scheme 1. Hydroxylation of Hydrocarbons with 2,6-Dichloropyridine
N-Oxide Catalyzed by RuII(TPFPP)(CO)
Notably, ruthenium porphyrins have been shown to utilize
dioxygen in air for the clean epoxidation of olefins under
mild conditions,18 and they have also shown promise in the
activation of nitrous oxide.19-23 The mechanism of aerobic
epoxidation mediated by ruthenium porphyrins has been
examined, and a trans-dioxoruthenium(VI) complex was
found to be the active oxidant. Unfortunately, these reactions
afforded only ca. 50 turnovers (TO) and proceeded at
relatively slow rates of 1-2 turnovers per hour. Hirobe et
al. have reported the use of 2,6-disubstituted pyridine
N-oxides as oxygen donors with ruthenium porphyrins such
as RuII(TMP)(CO) and RuVI(TMP)(O)2. High turnover
numbers on the order of 10 000 were reported for this
system.24-26 We have reported on the high activity of
ruthenium pentafluorophenyl porphyrins in similar oxida-
tions.27,28 These studies have suggested that an active oxidant
other than the previously characterized trans-dioxorutheni-
um(VI) may be responsible for these highly efficient
oxygenations with 2,6-disubstituted pyridine N-oxides,29,30
but the mechanism and the nature of the active oxidant
species in this system has remained unclear.
Results and Discussion
Scope of Oxidative Catalysis by Ruthenium Porphy-
rins. Ruthenium porphyrin complexes such as carbonylru-
thenium(II) tetrakispentafluorophenylporphyrin [RuII(TPFPP)-
(CO)] efficiently catalyzed the hydroxylation of unactivated
alkanes, the cleavage of ethers, and the oxygenation of
benzene with 2,6-dichloropyridine N-oxide as the oxidant
under mild, nonacidic conditions (Scheme 1). The hydroxy-
lation of adamantane proceeded with turnover rates up to
800 TO/min at high N-oxide concentrations and gave 14 800
catalytic turnovers. Up to 3100 turnovers at the maximum
rate of 30 TO/min have been obtained in the hydroxylation
of cyclohexane mediated by RuII(TPFPP)(CO). In a similar
reaction, benzene was oxidized to benzoquinone (Table 1).
Maximum turnover numbers that we obtained for the
hydroxylation of cyclohexane were compared with carbon-
ylruthenium(II) tetramesitylylporphyrin [RuII(TMP)(CO)] and
two pentafluorophenyl ruthenium porphyrin catalysts, RuII-
(TPFPP)(CO) and the â-pyrrole octabrominated analogue
RuII(TPFPPBr8)(CO). Ultimately, the ruthenium porphyrins
were completely degraded, as evidenced by the decrease of
their Soret absorbances in the UV-vis spectra and a loss of
catalytic activity. Catalyst stability dramatically improved
with increased halogenation of the porphyrin macrocycles.
Polyhalogenated ruthenium porphyrins are known to have
enhanced stability toward oxidative degradation.31,32 In
typical runs, RuII(TMP)(CO) gave 300 TO, and RuII(TPFPP)-
(CO) afforded 2500 TO. The greatest number of turnovers
for cyclohexane oxidation, nearly 6000, was achieved with
completely halogenated RuII(TPFPPBr8)(CO).
Here, we report our full findings on the mechanism of
hydrocarbon hydroxylation mediated by polyhalogenated
ruthenium porphyrin catalysts with 2,6-dichloropyridine
N-oxide as the oxidant. Remarkable rates, selectivities, and
efficiencies have been achieved for the catalytic oxygenations
with polyhalogenated ruthenium porphyrins. We show that
the Ru(II), Ru(IV), and Ru(VI) porphyrin species observed
during catalysis act as a reservoir of catalyst precursors, each
with marginal activity. The results support a Ru(III)-
oxoRu(V) cycle for the “fast catalysis” as we have previously
suggested.27,28
(13) Zhang, R.; Yu, W. Y.; Sun, H. Z.; Liu, W. S.; Che, C. M. Chem.s
Eur. J. 2002, 8, 2495-2507.
(14) Bonchio, M.; Scorrano, G.; Toniolo, P.; Proust, A.; Artero, V.; Conte,
V. AdV. Synth. Catal. 2002, 344, 841-844.
(15) Zhang, R.; Yu, W. Y.; Wong, K. Y.; Che, C. M. J. Org. Chem. 2001,
66, 8145-8153.
(16) Nakata, K.; Takeda, T.; Mihara, J.; Hamada, T.; Irie, R.; Katsuki, T.
Chem.sEur. J. 2001, 7, 3776-3782.
(17) Ogliaro, F.; de Visser, S. P.; Groves, J. T.; Shaik, S. Angew. Chem.,
Int. Ed. 2001, 40, 2874-2878.
(18) Groves, J. T.; Quinn, R. J. Am. Chem. Soc. 1985, 107, 5790-5792.
(19) Groves, J. T.; Roman, J. S. J. Am. Chem. Soc. 1995, 117, 5594-
5595.
(20) Tanaka, H.; Hashimoto, K.; Suzuki, K.; Kitaichi, Y.; Sato, M.; Ikeno,
T.; Yamada, T. Bull. Chem. Soc. Jpn. 2004, 77, 1905-1914.
(21) Hashimoto, K.; Tanaka, H.; Ikeno, T.; Yamada, T. Chem. Lett. 2002,
582-583.
(22) Ben-Daniel, R.; Weiner, L.; Neumann, R. J. Am. Chem. Soc. 2002,
124, 8788-8789.
(23) Yamada, T.; Hashimoto, K.; Kitaichi, Y.; Suzuki, K.; Ikeno, T. Chem.
Lett. 2001, 268-269.
Hydroxylations with polyhalogenated ruthenium porphy-
rins were characterized by extraordinary selectivity. For
example, the 3°/2° preference on a per hydrogen basis for
the RuII(TPFPP)(CO)-catalyzed oxidation of adamantane was
nearly 200. By comparison, tertiary versus secondary C-H
selectivities are reported to be 11-48 for iron porphyrin
(24) Ohtake, H.; Higuchi, T.; Hirobe, M. Heterocycles 1995, 40, 867-
903.
(25) Higuchi, T.; Hirobe, M. J. Mol. Catal. A 1996, 113, 403-422.
(26) (a) Shingaki, T.; Miura, K.; Higuchi, T.; Hirobe, M.; Nagano, T. Chem.
Commun. 1997, 861-862. (b) Higuchi, T.; Ohtake, H.; Hirobe, M.
Tetrahedron Lett. 1989, 30, 6545-6548.
(27) Groves, J. T.; Bonchio, M.; Carofiglio, T.; Shalyaev, K. J. Am. Chem.
Soc. 1996, 118, 8961-8962.
(28) Groves, J. T.; Shalyaev, K. V.; Bonchio, M.; Carofiglio, T. Stud. Surf.
Sci. Catal. 1997, 110, 865-872.
(31) Traylor, P. S.; Dolphin, D.; Traylor, T. G. J. Chem. Soc., Chem.
(29) Zhang, J. L.; Che, C. M. Chem.sEur. J. 2005, 11, 3899-3914.
(30) Ogawa, S.; Iida, T.; Goto, T.; Mano, N.; Goto, J.; Nambara, T. Org.
Biomol. Chem. 2004, 2, 1013-1018.
Commun. 1984, 279-280.
(32) Wijesekera, T. P.; Lyons, J. E.; Ellis, P. E. Catal. Lett. 1996, 36, 69-
73.
4770 Inorganic Chemistry, Vol. 45, No. 12, 2006

Fast Catalytic Hydroxylation of Hydrocarbons
Table 1. Oxidations with 2,6-Dichloropyridine N-Oxide Catalyzed by Ru(TPFPP) Complexesa
number
1
substrate, M
cis-decalin, 0.02
N-oxide, M
time, min
25
product, (yieldb, %)
TOc (max. rate, TO/min)
0.02
cis-9-decalol (80)
320 (64)
34
352
12070 (800)
1180
87
cis-decalin-9,10-diol (8.4)
2,3,3-trimethyl-2-butanol (88)
1-adamantanol (61)
adamantane-1,3-diol (6.0)
2-adamantanol (0.59)
2-adamantanone (0.97)
cyclohexanol (49)
cyclohexanone (35)
cyclohexanol (18)
cyclohexanone (55)
2
3
2,2,3-trimethylbutane, 0.04
adamantane, 0.2d
0.02
0.2
100
120
142
4
5
cyclohexane, 2.0d
0.05
0.05
100
1080 (30)e
376
cyclohexane, 2.0d,f
(CH2Cl2/CCl3Br 80:20%)
20 h
154e
232
bromocyclohexane (0.9)
benzoic acid (61)
7.5
82
6
toluene, 0.2
0.02
90
benzyl alcohol (9.3)
37
benzaldehyde (7.2)
15
benzyl chloride (1.6)
1,4-benzoquinone (40)
butyraldehyde (80)
6
7
8
benzene, 2.0
dibutyl ether, 0.2
0.02
0.02
12 h
180
53e
320
1-butanol (92)
368
9
cyclohexyl acetate, 0.1
0.05
250
3-acetoxycyclohexanone (33)
4-acetoxycyclohexanone (42)
cyclohexanone (1.5)
3-acetoxycyclopentanone (61)
cyclohexanone (85)
165
210
15
158
10
11
12
cyclopentyl acetate, 0.1
cyclohexanol, 0.04g
cyclobutanol, 0.04
0.05
0.02
0.02
200
240
14 h
268 (1.2)e
76e
cyclobutanone (60)
a Catalyst, RuII(TPFPP)(CO); [cat.] ) 5 × 10-5 M; CH2Cl2; 65 °C. b Percent yield based on oxidant consumed. c Turnover number, TO ) [product]/
[catalyst]. d [RuII(TPFPP)(CO)] ) 1 × 10-5 M. e The reaction was stopped before consumption of the N-oxide. f Reaction was carried out at 25 °C. g Catalyst,
RuVI(TPFPP)(O)2; [cat.] ) 5 × 10-5 M; 40 °C.
catalyzed hydroxylations,33 3-10 for manganese porphyrin
mediated oxidations,34 10 for reactions of the tert-butoxy
radical,35 and about 2 for the oxidation of alkanes by
hydroxyl radicals.36 The oxidation of cis-decalin was com-
pletely stereospecific with cis-9-decalol and cis-decalin-9,-
10-diol produced in high yields (90%; Table 1). No trans-
9-decalol was detected. Likewise, trans-decalin gave trans-
9-decalol and no cis-9-decalol, as we have reported.27 Similar
selectivity has been reported by Hirobe et al. in a related
system24 and recently by Bonchio et al. for ruthenium
oxometalates.14
Remarkably, the oxidation of cyclohexyl acetate with RuII-
(TPFPP)(CO)/2,6-dichloropyridine N-oxide gave a mixture
of 3- and 4-acetoxycyclohexanones (75% combined yield
based on the oxidant consumed). Only a trace of cyclohex-
anone was produced, and no 2-acetoxycyclohexanone was
detected. A single ketone, 3-acetoxycyclopentanone, was
formed in cyclopentyl acetate oxidation. For both substrates,
the acetyl group provided complete protection against
oxidation of the C-H bonds in the R and â positions relative
to the acetate group.
Catalytic oxygenation of dibutyl ether resulted in clean
cleavage to give 1-butanol and butyraldehyde in high yields.
Cyclohexanol was selectively oxidized to cyclohexanone.
Reaction Kinetics. These catalytic oxygenation reactions
have been found to follow an unusual time course. Typical
kinetic profiles for adamantane hydroxylation with 2,6-
Figure 1. Kinetics of adamantane hydroxylation with 2,6-dichloropyridine
N-oxide catalyzed by RuVI(TPFPP)(O)2 and RuII(TPFPP)(CO). [adamantane]
) 0.02 M, [RuPor] ) 50 µM, [2,6-dichloropyridine N-oxide] ) 0.02 M;
40 °C.
dichloropyridine N-oxide catalyzed by RuII(TPFPP)(CO) or
RuVI(TPFPP)(O)2 showed an induction period followed by
a linear, apparently zero-order phase, as we have reported
(Figure 1).27 This linear phase of the product appearance
profile extended until ca. 80% of the oxidant and substrate
had been consumed. Significantly, the starting catalyst, RuII-
(TPFPP)(CO) or RuVI(TPFPP)(O)2, remained the major
ruthenium porphyrin species present during ca. 400 turnovers
of adamantane hydroxylation, as revealed by the UV-vis
monitoring of the two reactions during turnover.
(33) Groves, J. T.; Nemo, T. E. J. Am. Chem. Soc. 1983, 105, 6243-6248.
(34) Hoffmann, P.; Robert, A.; Meunier, B. Bull. Soc. Chim. Fr. 1992,
129, 85-97.
(35) Barton, D. H. R.; Hu, B.; Taylor, D. K.; Wahl, R. U. R. J. Chem.
Soc., Perkin Trans. 2 1996, 1031-1041.
The catalytic hydroxylation of cyclohexane, which is a
less active substrate than adamantane, also proceeded with
(36) Russell, G. A. In Free Radicals; Kochi, J., Ed.; Wiley-Interscience:
New York, 1973; Vol. II, pp 689-690.
Inorganic Chemistry, Vol. 45, No. 12, 2006 4771

Wang et al.
Figure 2. Kinetics of cyclohexane hydroxylation with 2,6-dichloropyridine
N-oxide catalyzed by RuII(TPFPP)(CO). [cyclohexane] ) 2.0 M, [RuII-
(TPFPP)(CO)] ) 10 µM, [2,6-dichloropyridine N-oxide] ) 0.05 M; CH2-
Cl2; 65 °C.
Figure 3. Effect of dioxygen on the induction period of adamantane
hydroxylation with 2,6-dichloropyridine N-oxide catalyzed by RuVI(TPFPP)-
(O)2. [adamantane] ) 0.02 M, [RuVI(TPFPP)(O)2] ) 50 µM, [2,6-
dichloropyridine N-oxide] ) 0.02 M; 40 °C.
a fast initial linear phase (30 TO/min maximum rate at 65
°C) but eventually slowed because of the oxidative destruc-
tion of the porphyrin catalyst (bleaching) as revealed by the
decrease in the UV-vis absorbance (λmax ) 402 nm) in the
Soret region of RuII(TPFPP)(CO) (Figure 2). Both cyclo-
hexanol and cyclohexanone were formed (Table 1, entries 4
and 5). The relative reactivity of cyclohexane and cyclo-
hexanol, determined from the kinetics of the product evolu-
tion, remained constant throughout the entire course of the
reaction (kcyclohexanol/kcyclohexane ) 123 ( 5, see Experimental
Section). This observation is consistent with a sequential
oxidation to cyclohexanol and then cyclohexanone.
The induction period for the catalytic oxidations mediated
by RuII(TPFPP)(CO) and RuVI(TPFPP)(O)2 was shortened
with an increase in the reaction temperature and by exposure
to visible light.27 Interestingly, there was a distinct effect of
air on the kinetic profiles of the product evolution. The
kinetics of adamantane hydroxylation with RuVI(TPFPP)-
(O)2/2,6-dichloropyridine N-oxide in methylene chloride
showed no detectable induction period when air was not
strictly excluded from the reaction medium (Figure 3). The
presence of air in the reaction system slightly increased the
maximum rates of adamantane hydroxylation, measured as
the slope of the linear phase of kinetic traces (Table 2, entries
8-11; also see Figure 3).
The rates of catalytic hydroxylations were highly temper-
ature-dependent. A linear Eyring plot was obtained for the
rate of adamantane hydroxylation with a RuVI(TPFPP)(O)2
catalyst (Figure 4) with an apparent enthalpy of activation
∆Hq ) 19 kcal/mol (R ) 0.994).
The Role of RuVI(TPFPP)(O)2 in the Catalysis of
Alkane Hydroxylation. The similarity of the turnover rates
and small changes in the visible spectra during reactions with
RuII(TPFPP)(CO) and RuVI(TPFPP)(O)2 indicate that neither
RuII(TPFPP)(CO) nor RuVI(TPFPP)(O)2 lie on the “fast”
catalytic oxidation pathway. Obviously, RuII(TPFPP)(CO)
cannot be directly involved in the catalytic cycle and would
need to be activated in some way. The fact that the trans-
dioxoRu(VI) trace for adamantane hydroxylation has an
induction period also argues against trans-dioxoRu(VI) being
the only active species. The induction period implies that,
as in the case of RuII(TPFPP)(CO), the trans-dioxoRu(VI)
catalyst also undergoes activation before the maximum
turnover rates can be reached.
We have quantitatively evaluated the role of trans-
dioxoRu(VI) porphyrins in the fast catalysis process by
examining the stoichiometric reactions of these species. We
found that one equivalent of RuVI(TPFPP)(O)2 oxidized the
tertiary C-H positions of cis-decalin to produce one
equivalent of cis-9-decalol in nearly quantitative yield (99%;
Scheme 2). The reaction was completely stereospecific, and
no trans-9-decalol or secondary C-H bond oxidation
products were detected. The porphyrin product, which was
isolated from the reaction mixture in high yield, had a
diamagnetic NMR resonance at δ 8.62 for the pyrrole protons
and displayed an intense cluster of peaks at m/z ) 2164 in
the FAB-MS spectrum, which corresponds to the mass of
a dimeric [Ru(TPFPP)]2O fragment. Considered together with
We found that acids suppressed the rates of oxidations
mediated by Ru(TPFPP) complexes. When anhydrous HCl
was introduced to the reaction mixture of adamantane
hydroxylation in the presence of RuVI(TPFPP)(O)2, the rate
of the catalytic oxidation decreased 8-fold (Table 2, entries
11 and 12). Trifluoroacetic acid produced a similar inhibiting
effect. By contrast, the addition of mineral acids such as HCl
and HBr was essential for the hydroxylation of alkanes in
the catalytic system of Hirobe et al. in which Ru(TPP) and
Ru(TMP) catalysts were used.24,25
4772 Inorganic Chemistry, Vol. 45, No. 12, 2006

Fast Catalytic Hydroxylation of Hydrocarbons
Table 2. Hydroxylation of Adamantane with 2,6-Dichloropyridine N-Oxide Catalyzed by Ru(TPFPP) Complexesa
number
catalyst
solvent
time, min
T, °C
atmosphereb
TON
max. rate, TO/min
1
2
3
4
5
6
7
8
9
RuIV(TPFPP)Cl2 + Zn(Hg)
RuIV(TPFPP)Cl2
RuIV(TPFPP)Cl2
RuIV(TPFPP)Cl2
RuIV(TPFPP)Cl2
RuII(TPFPP)(CO)
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
CH2Cl2
CH2Cl2
CH2Cl2
PhCF3
PhCF3
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
C6H6
50
208
50
50
45
753
190
180
150
200
270
200
25
25
25
25
25
25
25
40
40
40
40
40
Ar
Ar
air
Ar
air
air
air
Ar
air
Ar
air
air
380
336
309
105
112
275
74
395
395
209
359
57
31.4
6.3
11.5
4.6
5.1
0.43
0.39
3.6
4.1
1.4
10
11
12
C6H6
1.9
0.25
C6H6,
HClc
13
RuII(TPFPP)(CO)
CH2Cl2
20
65
air
364
72
a [Adamantane] ) [2,6-dichloropyridine N-oxide] ) 0.02 M; [catalyst] ) 5 × 10-5 M. Yields of 1-adamantanol and 1,3-adamantanediol were 91-97%
based on oxidant consumed. b Reaction conditions: Ar, reaction was conducted under argon. c A total of 20 µL of saturated HCl (anh.) solution in benzene
was added to a 2 mL sample.
trans-DioxoRu(VI) complexes have been shown to be the
active species in epoxidations catalyzed by ruthenium
complexes.18,42 Che et al.37 have demonstrated that, in
addition to being reactive toward alkenes, trans-dioxoRu-
(VI) porphyrins, RuVI(TPP)(O)2 and RuVI(OEP)(O)2, were
selective oxidants for the tertiary C-H bonds of alkanes.
However, the stoichiometric reactions of these ruthenium
porphyrin complexes with alkanes proceeded with low
efficiency because the rate of oxidation was too slow to
compete with the self-degradation of the ruthenium com-
plexes. For example, only ∼20% yields of 1-adamantanol
were obtained in the oxidation of adamantane.
A comparison of the rates of the stoichiometric hydro-
carbon oxidations by RuVI(TPFPP)(O)2 with those of the
catalytic oxidations with 2,6-dichloropyridine N-oxide showed
that the catalytic reactions were too fast to be due to the
reaction of RuVI(TPFPP)(O)2. The rates of the stoichiometric
Figure 4. Eyring plot for adamantane hydroxylation with 2,6-dichloro-
reactions of RuVI(TPFPP)(O)2 with several substrates were
pyridine N-oxide catalyzed by RuVI(TPFPP)(O)2. [adamantane] ) 0.02 M,
measured in two ways: by monitoring the disappearance of
UV-vis absorbance of the Ru(VI) porphyrin or by following
the intensity of the pyrrole proton resonance of the trans-
[RuVI(TPFPP)(O)2] ) 50 µM, [2,6-dichloropyridine N-oxide] ) 0.02 M;
CH2Cl2; 40 °C.
Scheme 2. Stoichiometric Hydroxylation of cis-Decalin with
1
RuVI(TPFPP)(O)2
dioxoRu(VI) complex at δ 9.26 in H NMR spectra. The
reactions were carried out under pseudo-first-order conditions
with excess substrate relative to the ruthenium porphyrin.
The second-order rate constants of the stoichiometric oxida-
tions (kst) determined from a single-exponential fit of the
data to eq 1 are shown in Table 3.
Rate ) -d[RuVI(TPFPP)(O)2]/dt )
kst[substrate][RuVI(TPFPP)(O)2] (1)
the stoichiometry of cis-decalin oxidation, these data support
the assignment of this diamagnetic porphyrin as a µ-oxoru-
thenium(IV) dimer.37-41 We found that such a µ-oxo dimer
could also be produced by the slow (72 h) decomposition of
RuVI(TPFPP)(O)2 in methylene chloride in the absence of a
substrate.
Turnover rates for the catalytic reactions were measured
as the slope of the linear phase of kinetic traces of product
evolution in hydrocarbon oxidations with 2,6-dichloropyri-
dine N-oxide as an oxygen donor in the presence of RuVI-
(TPFPP)(O)2. We estimated the catalytic rate constants (kcat)
assuming that the trans-dioxoRu(VI) porphyrin is the sole
active catalytic species, and it is the reaction of this dioxoRu-
(VI) species with the substrate that determines the turnover
(37) Ho, C.; Leung, W. H.; Che, C. M. J. Chem. Soc., Dalton Trans. 1991,
2933-2939.
(38) Collman, J. P.; Barnes, C. E.; Brothers, P. J.; Collins, T. J.; Ozawa,
T.; Gallucci, J. C.; Ibers, J. A. J. Am. Chem. Soc. 1984, 106, 5151-
5163.
(39) Leung, W. H.; Che, C. M. J. Am. Chem. Soc. 1989, 111, 8812-8818.
(40) Masuda, H.; Taga, T.; Osaki, K.; Sugimoto, H.; Mori, M.; Ogoshi, H.
J. Am. Chem. Soc. 1981, 103, 2199-2203.
(41) Sugimoto, H.; Higashi, T.; Mori, M.; Nagano, M.; Yoshida, Z.; Ogoshi,
H. Bull. Chem. Soc. Jpn. 1982, 55, 822-828.
(42) Rajapakse, N.; James, B. R.; Dolphin, D. Catal. Lett. 1989, 2, 219-
225.
Inorganic Chemistry, Vol. 45, No. 12, 2006 4773

Wang et al.
Table 3. Comparison of Actual Rate Constants for Stoichiometric RuVI(TPFPP)(O)2 Reactions with Minimum, Hypothetical Rate Constants for
Catalytic Reactions of RuVI(TPFPP)(O)2
stoichiometric oxidation
catalytic oxidation
cat,a M-1 s-1, (R)
major resting state
of the catalystb
a
substrate
T, °C
kst, M-1 s-1, (R)
k
k
cat/kst
cis-decalin
25
2.3 × 10-3
1.9 × 10-2 [0.38]
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
N/D
8.3
[165]
9.7
(0.978)
(0.990)
adamantane
benzyl alcohol
cyclohexanol
styrene
40
40
40
25
3.1 × 10-1
(0.998)
3.0 [60]
(0.999)
2.8 [56]
[194]
46.6
[933]
330
6.0 × 10-2
(0.999)
(0.989)
1.5 × 10-3
(0.999)
5.0 × 10-1 [10]
(0.999)
[6.6 × 103]
4.3 × 10-1
1.25 [25]
(0.996)
RuIV(TPFPP)(O)
2.9
(0.999)
[58]
a kcat, conservative, hypothetical estimates calculated assuming that RuVI(TPFPP)(O)2 is the only active oxidant and its concentration is equal to the initial
1
concentration during the catalytic reactions. The values in brackets are catalytic rate constants assuming 5% active species. b Determined by H NMR or
UV-vis spectroscopy.
Scheme 3. Pathways Leading to Ru(III) Porphyrins from
RuII(TPFPP)(CO) and RuIV(TPFPP)Cl2 (OD ) 2,6-Dichloropyridine-N-
oxide)
rate (eq 2). The data show that the turnover rate is much too
fast for this hypothetical situation to be true, and accordingly,
RuVI(TPFPP)(O)2 cannot be the active catalytic oxidant.
Rate ) -d[substrate]/dt )
kcat[substrate][RuVI(TPFPP)(O)2]
[RuVI(TPFPP)(O)2] ) [RuVI(TPFPP)(O)2]0
(2)
Comparison of the catalytic and stoichiometric rate
constants for the series of substrates in Table 3 shows that
the catalytic rate constants (kcat) are all significantly larger
than the corresponding rate constants for the stoichiometric
reactions of RuVI(TPFPP)(O)2. For example, the rate constant
for the stoichiometric reaction between RuVI(TPFPP)(O)2 and
cis-decalin (kst) was determined to be 2.3 × 10-3 M-1 s-1
oxidative catalysis were obtained from observing the activa-
tion of the catalyst by redox processes. As mentioned above,
oxygen eliminated the induction period observed when
methylene chloride was the reaction solvent. Significantly,
we have found that oxygen had no obvious effect on the
rate of reaction with trifluorotoluene as the solvent. These
observations and the slow rate of adamantane hydroxylation
in the presence of cyclohexene (a stabilizer for methylene
chloride)43 are consistent with a component to the activation
of RuII(TPFPP)(CO) that involves autoxidation of the
solVent.44
The activation of RuII(TPFPP)(CO) that is observed by
one-electron oxidation supports this hypothesis. We have
previously reported the oxidation of RuII(TPFPP)(CO) with
Fe(ClO4)3 to the corresponding porphyrin cation radical, RuII-
(TPFPP+•)(CO) (λmax ) 635 nm).27 The addition of lutidine-
N-oxide to RuII(TPFPP+•)(CO) gave EPR signals character-
istic of Ru(III). The hydroxylation of adamantane in methylene
chloride with RuII(TPFPP)(CO) pretreated with Fe(ClO4)3
proceeded without an induction period (Scheme 3).
1
(25 °C). H NMR monitoring of the catalytic oxidation of
cis-decalin revealed that RuVI(TPFPP)(O)2 remained the
major porphyrin species during the reaction. If one assumes
that trans-dioxoRu(VI) is the only active species under the
catalytic conditions, the kinetics of the catalytic hydroxylation
of cis-decalin in the presence of 2,6-dichloropyridine N-oxide
would give kcat ) 1.9 × 10-2 M-1 s-1 (25 °C), which is
more than 8 times greater than kst. Such a substantial
difference in the rate of oxidation under the stoichiometric
and catalytic conditions indicates that the rate of catalytic
oxidation cannot be explained if the trans-dioxoRu(VI)
porphyrin is the only active oxidant. Some other ruthenium
species with significantly higher activity than that of RuVI-
(TPFPP)(O)2 must be involved in ruthenium porphyrin
catalysis with 2,6-dichloropyridine N-oxide as an oxygen
donor.
We found that RuIV(TPFPP)Cl2 is a faster catalyst than
RuII(TPFPP)(CO) or RuVI(TPFPP)(O)2 and that it can be
further activated by methylene chloride/air. The hydroxyla-
tion of adamantane in methylene chloride catalyzed by RuIV-
It is important to emphasize that the catalytic rate constants
(kcat) determined above represent very conservative estimates
of the true catalytic rate constants, because the fraction of
ruthenium porphyrin directly involved in the “fast” catalytic
cycle is certainly much smaller than the initial concentration
of the ruthenium porphyrin that was used for the calculation
of kcat. Indeed, the very modest changes in the visible spectra
of the catalysts under turnover conditions indicate that the
active species could never amount to more than ca. 5% of
the catalyst inventory.
(43) Commercial methylene chloride contains 40-100 ppm (∼1 mM) of
cyclohexene or amylene as stabilizing agents against autoxidation. We
found that the presence of even trace amounts of cyclohexene in the
solvent caused prolonged induction periods and inhibited the catalytic
hydroxylations of alkanes. Maximum turnover rates for cyclohexene
epoxidation were 3 times slower than that of adamantane under
comparable conditions even though cyclohexene was found to be 100
times intrinsically more reactive than adamantane in an intermolecular
competition. The situation in benzene would appear to require some
autoxidation of the substrate for catalyst activation.
Redox Activation of RuII(TPFPP)(CO) and RuIV(TPF-
PP)Cl2. Important clues regarding the mechanism of this fast
(44) Yin, C.-X.; Sasaki, Y.; Finke, R. G. Inorg. Chem. 2005, 44, 8521-
8530.
4774 Inorganic Chemistry, Vol. 45, No. 12, 2006

Fast Catalytic Hydroxylation of Hydrocarbons
Figure 6. Kinetics of RuIII(TPFPP)(OEt)(EtOH)-mediated adamantane
hydroxylation with 2,6-dichloropyridine N-oxide followed with GC and
UV-vis. Adamantane, 0.02 M; RuIII(TPFPP)(OEt)(EtOH), 7.5 µM; 2,6-
dichloropyridine N-oxide, 0.01 M; CH2Cl2; 25 °C.
Figure 5. X-band EPR characterization of RuIII(TPFPP) generated by the
reduction of RuIV(TPFPP)(Cl)2 with Zn(Hg). T ) 77 K. (a) RuIII(TPFPP)-
(OEt)(EtOH) in ethanol glass. (b) RuIV(TPFPP)Cl2 reacted with Zn(Hg) in
CD2Cl2 and quenched with liquid N2. (c) After the addition of 10 µL of
ethanol into 0.5 mL of solution b and quenched with liquid N2.
3). No EPR signals were observed when ethanol was added
to RuIV(TPFPP)Cl2 even after 3 h at 25 °C.
(TPFPP)Cl2 proceeded at a maximum rate of 6.3 TO/min
under argon and 11.5 TO/min under air at 25 °C (Table 2,
entries 2 and 3). Both rates were faster than the maximum
turnover rates for RuVI(TPFPP)(O)2- and RuII(TPFPP)(CO)-
catalyzed reactions under identical conditions (0.39 and 0.43
TO/min, respectively, see Table 2, entries 6 and 7). On the
other hand, the RuIV(TPFPP)Cl2-mediated oxidation pro-
ceeded much slower in R,R,R-trifluorotoluene with a rate of
5.1 TO/min under air and 4.6 TO/min under argon (Table
2, entries 4 and 5). These observations strongly suggest that
methylene chloride is involved in the activation of the
catalyst.
Interestingly, when authentic RuIII(TPFPP)(OEt) was used
as the catalyst, the maximum turnover rate of adamantane
oxidation at 25 °C at 7.5 µM catalyst was found to proceed
at 28.8 TO/min, several times faster than the reactions
mediated by RuII(TPFPP)(CO) or RuVI(TPFPP)(O)2. Soon
after the reaction started, the RuIII species was gradually
converted to a dioxo RuVI species, and the turnover rates in
this process decreased concomitantly (Figure 6). This
decrease in rate, accompanied by the conversion of RuIII to
RuVI, speaks against RuVI(TPFPP)(O)2 as the active oxidant.
The kinetic and spectroscopic results described above have
ruled out both RuVI(TPFPP)(O)2 and RuII(TPFPP)(CO) as
the active species in the “fast catalysis” regime for these
hydroxylations and epoxidation reactions. Activation of the
catalyst either by the oxidation of RuII(TPFPP)(CO) with
Fe(ClO4)3 or the reduction of RuIV(TPFPP)Cl2 to Ru(III) with
a zinc amalgam or air/methylene chloride strongly supports
the formation of Ru(III) in the catalyst activation process.
Proposed Catalytic Cycle. A mechanism that is consistent
with the data described above is presented in Scheme 4. To
explain the oxidation of hydrocarbons with the highest
turnover rates and efficiencies, we propose Ru(III) and
oxoRu(V) porphyrin species as the key intermediates. In this
scenario, one-electron redox processes involving the inven-
tory of Ru(II), Ru(IV), or Ru(VI) porphyrin species observed
during catalysis would produce a low steady-state concentra-
tion of Ru(III) that is oxidized to oxoRu(V) by dichloropyr-
idine-N-oxide. This interpretation differs from that of Hirobe
et al., who have suggested that the dichloroRu(IV) porphyrin
is a precursor of chloro,oxoRu(VI) cationic species,24 and
that of Gross et al.,47 who have suggested an adduct of the
oxoRu(IV) with the pyridine-N-oxide.
The addition of a zinc amalgam to the reaction mixture
of a RuIV(TPFPP)Cl2-catalyzed oxidation of adamantane led
to a sustained initial reaction rate 5 times faster than that
observed with RuIV(TPFPP)Cl2 alone (31.4 TO/min vs 6.3
TO/min; see entries 1 and 2 in Table 2). Other additives
such as AgBF4, AgPF6, trifluoroacetic acid, ferrocene,
methanol, Bu4NCl, and Me4NOH slowed the catalytic
oxidations. This activation by a zinc amalgam is consistent
with a one-electron reduction of the Ru(IV) porphyrin to give
a Ru(III) species in situ.
Indeed, proof of the formation of Ru(III) from RuIV-
(TPFPP)Cl2 by zinc amalgam reduction was obtained by EPR
spectroscopy. Thus, the reaction of RuIV(TPFPP)Cl2 with a
zinc amalgam in CH2Cl2 at 25 °C for 20 min caused a shift
in the visible spectrum in the Q-band region from 506 to
520 nm. This solution was EPR-silent (Figure 5, trace b),
and the NMR absorbance for the pyrrole protons of RuIV-
(TPFPP)Cl2 (δ -53) was no longer apparent. The addition
of ethanol at 25 °C caused the immediate appearance of a
UV spectrum characteristic of an ethoxyRu(III) com-
plex.38,45,46 An NMR resonance at δ -15 and strong, sharp
EPR signals (g ) 2.53, 2.12, and 1.89) (Figure 5, trace c)
observed in this solution at 77 K provide conclusive evidence
for the presence of a Ru(III) complex in this sample (Scheme
The “fast” catalytic pathway can be conveniently described
using the terminology of chain-reaction kinetics, which
includes initiation, propagation, and termination steps. The
presence of an induction period in the kinetic profiles of the
product evolution of the catalytic oxidations with Ru-
(45) Collman, J. P.; Rose, E.; Venburg, G. D. J. Chem. Soc. 1994, 11-12.
(46) James, B. R.; Dolphin, D.; Leung, T. W.; Einstein, F. W. B.; Willis,
A. C. Can. J. Chem. 1984, 62, 1238-1245.
(47) Gross, Z.; Ini, S. Inorg. Chem. 1999, 38, 1446-1449.
Inorganic Chemistry, Vol. 45, No. 12, 2006 4775

Wang et al.
Scheme 4. Mechanisms of Hydrocarbon Oxygenation by Ru(TPFPP)
Figure 7. Energy diagram consistent with a large deuterium isotope effect
in the competitive hydroxylation of adamantane and adamantane-d16 and a
similar turnover observed in separate hydroxylations of adamantane and
adamantane-d16. The axial ligand, OX, is the pyridine N-oxide.
In the termination step, the Ru(III) and oxoRu(V) por-
phyrin species either return to the resting pool of the catalyst
or the porphyrin macrocycles get oxidatively destroyed. The
maximum number of catalytic turnovers obtained was found
to increase only slightly as the concentration of the catalyst
was decreased, suggesting an intramolecular self-oxidation
as opposed to intermolecular deactivation pathways such as
bimolecular oxidative cleavage of the porphyrin macrocycle
and formation of inactive µ-oxo ruthenium porphyrin
dimers.38,49 The kinetics of oxidation of relatively unreactive
substrates, such as cyclohexane, were affected by the
oxidative bleaching of the porphyrin to a greater extent
because the reactive oxoRu(V) intermediate would be longer-
lived and more susceptible to decomposition. This need for
“substrate rescue” of the catalyst is why such high concen-
trations of cyclohexane (2 M) were required to maintain the
fast catalysis. Indeed, in the absence of a substrate, the
ruthenium porphyrin was bleached completely in only a few
minutes.
(TPFPP)/2,6-dichloropyridine N-oxide is an indication of an
initiation process. During the initiation phase, a buildup of
the metastable and highly catalytically active Ru(III) por-
phyrin would occur from the resting pool of ruthenium
porphyrin catalyst represented by RuVI(TPFPP)(O)2, RuIV-
(TPFPP)(O), RuIV(TPFPP)Cl2, or RuII(TPFPP)(CO). During
the propagation phase, the hydrocarbon substrates would be
efficiently oxidized within the “fast” catalytic cycle with the
oxoRu(V) porphyrin species formed via oxygen transfer from
the pyridine N-oxide ligated to the Ru(III) precursor. Such
a steady-state regime is uniquely consistent with the appar-
ently zero-order kinetics observed for the product evolution.
It was observed that the maximum turnover rates obtained
for the oxygenation of adamantane (at 72 TO/min) and trans-
â-methylstyrene (at 45 TO/min)48 differed only by a factor
of 1.5. However, in the competitive oxidation of these two
substrates, trans-â-methylstyrene was 50 times more reactive
than adamantane under the same conditions. This observation
provides support for the argument that the hydrocarbon
substrates are discriminated after the rate-limiting step, and
it is the formation of the oxoruthenium(V) intermediate from
a Ru(III)-O-pyridine precursor that is rate-limiting (Figure
7). Furthermore, a common actiVe intermediate is indicated
for both the alkane hydroxylation and epoxidation of olefins
in the Ru(TPFPP)/2,6-dichloropyridine N-oxide catalytic
system.
The even-number oxidation states of ruthenium, oxoRu-
(IV) and trans-dioxoRu(VI) porphyrins, are known to be
involved in a “slow” oxidation pathway. We have previously
described a catalytic cycle, mediated by Ru(II), oxoRu(IV),
and trans-dioxoRu(VI) porphyrins, for the aerobic epoxida-
tion of olefins.18,50,51 The oxidation of the substrate with RuVI-
(TPFPP)(O)2 would result in the formation of RuIV(TPFPP)-
(O). RuIV(TPFPP)(O) can either be directly reoxidized to a
trans-dioxoRu(VI)porphyrin by 2,6-dichloropyridine N-oxide
or undergo bimolecular disproportionation to give trans-
dioxoRu(VI) and Ru(II) porphyrin complexes with subse-
(49) James, B. R.; Shilov, A. E. Fundamental Research in Homogeneous
Catalysis; Gordon & Breach: New York, 1986; p 309.
(50) Groves, J. T.; Shalyaev, K.; Lee, J.; Kadish, K. M.; Smith, K. M.;
Guilard, R. The Porphyrin Handbook; Academic Press: San Diego,
2000; Vol. 4, pp 17-40.
(48) Wang, C.; Groves, J. T. Unpublished results.
(51) Quici, S.; Banfi, S.; Pozzi, G. Gazz. Chim. Ital. 1993, 123, 597-612.
4776 Inorganic Chemistry, Vol. 45, No. 12, 2006

Fast Catalytic Hydroxylation of Hydrocarbons
Table 4. Deuterium Isotope Effectsa
catalyst
oxidant
substrate
solvent
T, °C
DIE
RuVI(TPFPP)(O)2
[RuVL4(O)]2+
pyCl2 N-oxide
cyclohexane
cyclohexane
adamantane
adamantane
cis-decalin
cis-decalin
cis-decalin
cyclohexanol
propan-2-ol
propan-2-ol
toluene
CH2Cl2
CH3CN
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
H2O
25
17
25
65
40
25
25
40
25
25
20
45
4.5
5.3 ( 0.636
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
RuVI(TPFPP)(O)2
[RuVL4(O)]2+
pyCl2 N-oxide
pyCl2 N-oxide
pyCl2 N-oxide
pyCl2 N-oxide
5.2
4.2
5.5
6.4
6.5
pyCl2 N-oxide
5.4
5.3 ( 0.536
18 ( 337
9.738
639
[RuIV(trpy)(bpy)O]2+
[MnO4]-
H2O
H2O
toluene
[MnO4]-
toluene
a HL4 ) [2-hydroxy-2-(2-pyridyl)ethyl]bis[2-(2-pyridyl)ethyl] amine].
quent reoxidation of the Ru(II) porphyrin to RuIV(TPFPP)(O).
The contribution of this “slow” catalytic cycle toward the
oxygenation of alkanes is small, given the slow stoichio-
metric rate constants reported here, but it cannot be entirely
excluded, especially for the substrates such as olefins, which
have relatively high intrinsic activity toward RuVI(TPFPP)-
(O)2.
Nature of the Active Oxidant Species. Several mecha-
nistic probes have been employed to elucidate the nature of
the proposed oxoRu(V) intermediate in these catalytic
oxygenations.
Deuterium Isotope Effect Studies. The competitive oxida-
tion of a 1:1 mixture of adamantane and adamantane-d16 with
2,6-dichloropyridine-N-oxide catalyzed by RuVI(TPFPP)(O)2
showed a kinetic isotope effect, kH/kD ) 4.8 at 40 °C.
However, the deuterated and undeuterated substrates dis-
played virtually no difference in the turnoVer rates of
hydroxylation in separate reactions (kH/kD ) 1.2). We have
previously reported similar results obtained with a RuII-
(TPFPP)(CO) catalyst.27 Likewise, the catalytic oxidations
of cis-decalin and cis-decalin-d18 with RuVI(TPFPP)(O)2 used
as catalyst proceeded with the same turnover rate, although
kH/kD ) 5.5 at 40 °C was measured in a competitive
oxidation. Similar to the adamantane/trans-â-methylstyrene
case discussed above, these deuterium isotope effect data
are consistent with a process in which the active oxidant
species, an oxoRu(V) porphyrin, is formed in the rate-
determining step and differentiation of substrates occurs after
the rate-determining step on the reaction coordinate (Figure
7). The deuterium isotope effect for the competitive, stoi-
chiometric oxidation of cis-decalin and cis-decalin-d18 with
RuVI(TPFPP)(O)2 was 6.5 at 25 °C. The catalytic hydroxy-
lation of cis-decalins with 2,6-dichloropyridine N-oxide
mediated by a trans-dioxoRu(VI) complex gave a very
similar value, kH/kD ) 6.4.
pecially, oxoruthenium(V) species are rare.5,52,53 One of the
few known examples of an oxoRu(V) complex is [RuVL-
(O)]2+ {HL ) [2-hydroxy-2-(2-pyridyl)ethyl]bis[2-(2-py-
ridyl)ethyl] amine]}, reported by Che et al.54-56 Several
characteristics of this complex are strikingly reminiscent of
the reactivity features of the oxoRu(V) active species that
we propose.57-59 [RuVL(O)]2+ is one of the fastest and most
reactive oxoruthenium species toward stoichiometric C-H
bond oxidation, capable of oxidizing cyclohexane under mild
conditions similar to the behavior seen for the proposed
oxoRu(V) porphyrin. Further, the kinetic isotope effect (kH/
kD ) 5.3) for cyclohexane oxidation with [RuVL(O)]2+ is
similar to the values we find for adamantane hydroxylation
(kH/kD ) 5.2) and cyclohexane hydroxylation (kH/kD ) 4.5)
with a Ru(TPFPP)/2,6-dichloropyridine N-oxide system.
Electronic Effects. The electronic features of the catalytic
oxygenations have been probed through studies of the relative
reactivity of substituted toluenes. A typical Hammett treat-
ment of data for the hydroxylation of para-substituted
toluenes in the Ru(TPFPP)/2,6-dichloropyridine N-oxide
system produced an extraordinarily negative F+ ) -2.0 at
40 °C (Figure 8). This value is much greater than that for
the relative rates of the competitive oxidation of toluenes
by an oxoFe(IV) porphyrin radical cation, FeIV(TPP+•)(O),
the active oxidant in the FeIII(TPP)Cl/PhIO system (F+
)
-0.83).60 For typical radical reactions such as hydrogen-
atom abstraction from XC6H4CH3 by t-butoxy radicals and
bromine atoms, F+ values of -0.4 and -1.4, respectively,
have been reported.36 The large negative value of F+ for the
ruthenium-catalyzed process indicates an unusually high
charge separation and electrophilic nature of the oxygen
transfer step in the transition state of the “fast” ruthenium
(52) Dengel, A. C.; Griffith, W. P. Inorg. Chem. 1991, 30, 869-871.
(53) Fackler, N. L. P.; Zhang, S. S.; O’Halloran, T. V. J. Am. Chem. Soc.
1996, 118, 481-482.
(54) Che, C. M.; Wong, K. Y.; Mak, T. C. W. J. Chem. Soc. 1985, 988-
990.
Larger deuterium isotope effects than these values have
been reported for the competitive oxidation of cyclohexanes
with electron-deficient iron porphyrins, FeIII(TDCPP)/PhIO
(kH/kD ) 9 ( 3) and FeIII(TMPBr8)/PhIO (kH/kD ) 7.7 (
0.6).34 Smaller deuterium isotope effects have been measured
for the manganese porphyrin catalyzed hydroxylation of
cyclohexane with MnIII(TDCPP)/PhIO (kH/kD ) 2.6 ( 0.4;
Table 4).34
(55) Che, C. M.; Yam, V. W. W.; Mak, T. C. W. J. Am. Chem. Soc. 1990,
112, 2284-2291.
(56) Che, C. M.; Ho, C.; Lau, T. C. J. Chem. Soc., Dalton Trans. 1991,
1259-1263.
(57) Thompson, M. S.; Meyer, T. J. J. Am. Chem. Soc. 1982, 104, 5070-
5076.
(58) Gardner, K. A.; Mayer, J. M. Science 1995, 269, 1849-1851.
(59) Gardner, K. A.; Kuehnert, L. L.; Mayer, J. M. Inorg. Chem. 1997,
36, 2069-2078.
(60) Inchley, P.; Lindsay-Smith, J. R.; Lower, R. J. New J. Chem. 1989,
13, 669-676.
Reliably characterized ruthenium(V) complexes and, es-
Inorganic Chemistry, Vol. 45, No. 12, 2006 4777

Wang et al.
are initiated by hydrogen abstraction.62,63 In these cases, the
carbon radicals are long-lived enough to dimerize and
disproportionate via intermolecular reactions. By contrast,
the oxoRu(V) species proposed here is behaving very
differently, with no evidence of radical species that live long
enough to be trapped. Thus, if these reactions do proceed
via the prototypical oxygen rebound pathway, the intermedi-
ate geminate radical pair [Ru(IV)-OH •R] must collapse in
the solvent cage with very high efficiency.
18O exchange. 18O incorporation in the oxidation products
for hydrocarbon oxygenations mediated by metalloporphyrins
has been widely used to implicate metal-oxo species as the
oxygen transfer agents.64-68 The hydroxylation of adamantane
catalyzed by RuVI(TPFPP)(O)2 in the presence of H218O
(92%) led to an 18% incorporation of the 18O label from the
moist CH2Cl2 into the product, 1-adamantanol. By contrast,
3.4% 18O incorporation into 1-adamantanol was observed
with RuII(TPFPP)(CO) as the catalyst. This result supports
the hypothesis that the pyridine N-oxide reacts to form an
oxoruthenium porphyrin species, such as oxoRu(V), that is
capable of exchange with water. The relatively low degree
of 18O exchange into the product, especially when RuII-
(TPFPP)(CO) was used as the catalyst, is consistent with a
short lifetime of the highly active oxidant species.69 As
discussed above, the oxidations mediated by RuII(TPFPP)-
(CO) and RuVI(TPFPP)(O)2 have similar turnover rates (see
Figure 8. Hammett correlation of log krel for RuII(TPFPP)(CO)-catalyzed
oxidations of para-substituted toluenes by 2,6-dichloropyridine N-oxide.
Pairs of competing substrates (0.02 M) were oxidized with 2,6-dichloro-
pyridine N-oxide (0.02 M) in the presence of RuII(TPFPP)(CO) (50 µM) in
CH2Cl2 at 40 °C. The consumption of the substrates was followed by GC
using 1,2-dichlorobenzene (0.01 M) as an internal standard.
catalysis compared to the characteristics of other metallopor-
phyrin systems.
Oxidation in the Presence of CCl3Br. The catalytic
oxidation of cyclohexane in 20% CCl3Br/80% CH2Cl2 gave
cyclohexanol (18% based on the oxidant) and cyclohexanone
(55%), with only a small amount of bromocyclohexane
(0.9%). The formation of brominated products in the presence
of CCl3Br has been used as evidence for the involvement of
alkyl radical intermediates, because it is known that bro-
motrichloromethane reacts rapidly and efficiently with alkyl
radicals to form alkyl bromides.61 For example, the oxidation
of cycloheptane by iodosylbenzene and FeIII(TPP)Cl in the
presence of bromotrichloromethane produced 24% cyclo-
heptanol and an 18% yield of cycloheptyl bromide.33
Lindsay-Smith et al.60 reported that for FeIII(TPP)Cl/PhIO
oxidation approximately half of the starting material was
diverted to bromocyclohexane, and in the case of the MnIII-
(TPP)Cl-catalyzed reaction, bromotrichloromethane trapped
90% of the reacting cyclohexane.
Table 2) but with RuVI(TPFPP)(O)2 having much higher 18
O
incorporation. The majority of the 18O incorporation in the
hydroxylation of adamantane mediated by RuVI(TPFPP)(O)2
should be due to the “slow” catalytic cycle [dioxoRu(VI) to
oxoRu(IV)]. It is also known that the oxoRu(IV) complexes
undergo rapid oxygen exchange with water.70
ReactiVity and SelectiVity. The high 3°/2° selectivity
observed in the hydroxylation of adamantane is in contrast
with the far less selective hydrocarbon oxidations catalyzed
by iron and manganese porphyrins.33,34 The possible explana-
tion for such a difference in selectivity can be derived from
a Hammond postulate reasoning. The oxoRu(V) species is
expected to be lower in free energy than the corresponding
oxoMn(V)71 and oxoFe(IV)(porphyrins+•).72,73 Accordingly,
the reactive oxoRu(V) intermediate would interact with a
The results presented here indicate that long-lived alkyl
radicals are not formed during hydrocarbon oxygenations
with the Ru(TPFPP)/2,6-dichloropyridine N-oxide catalytic
system. First, the amount of cyclohexyl bromide formed in
the oxidation of cyclohexane in the presence of bromotri-
chloromethane is negligible. Second, the hydroxylation of
cis-decalin gave the retention products, cis-9-decalol and cis-
decalin-9,10-diol, exclusively. The cis-9-decalyl radical is
known to isomerize to the trans conformation with a rate in
excess of 108 s-1.61b Finally, the oxidation of cyclobutanol
did not produce the ring-opening products characteristic of
a one-electron oxidation step. Mayer et al. have reported
recently that C-H bond oxidations by oxoRu(IV) complexes
(62) Bryant, J. R.; Mayer, J. M. J. Am. Chem. Soc. 2003, 125, 10351-
10361.
(63) Bryant, J. R.; Matsuo, T.; Mayer, J. M. Inorg. Chem. 2004, 43, 1587-
1592.
(64) Groves, J. T.; Stern, M. K. J. Am. Chem. Soc. 1988, 110, 8628-
8638.
(65) Groves, J. T.; Stern, M. K. J. Am. Chem. Soc. 1987, 109, 3812-
3814.
(66) Groves, J. T.; Lee, J. B.; Marla, S. S. J. Am. Chem. Soc. 1997, 119,
6269-6273.
(67) Pitie, M.; Bernadou, J.; Meunier, B. J. Am. Chem. Soc. 1995, 117,
2935-2936.
(68) Bernadou, J.; Fabiano, A. S.; Robert, A.; Meunier, B. J. Am. Chem.
Soc. 1994, 116, 9375-9376.
(69) Nam, W. W.; Valentine, J. S. J. Am. Chem. Soc. 1993, 115, 1772-
1778.
(70) Ahn, K. H. Ph.D. Thesis, Princeton University, Princeton, New Jersey,
1988.
(71) Jin, N.; Bourassa, J. L.; Tizio, S. C.; Groves, J. T. Angew. Chem., Int.
Ed. 2000, 39, 3849-3851.
(72) Holm, R. H. Chem. ReV. 1987, 87, 1401-1449.
(73) Holm, R. H.; Donahue, J. P. Polyhedron 1993, 12, 571-589.
(61) (a) Huyser, E. S.; Burham, R. L.; Schimke, H. J. Org. Chem. 1963,
28, 2141. (b) Hummel, A.; Deleng, H. C.; Luthjens, L. H. Radiat.
Phys. Chem. 1995, 45, 745-748 and references therein.
4778 Inorganic Chemistry, Vol. 45, No. 12, 2006

Fast Catalytic Hydroxylation of Hydrocarbons
hydrocarbon substrate leading to a comparatively late,
product-like transition state while the more reactive oxoman-
ganese and oxoiron species react via early transition states
for C-H bond cleavage, resulting in a lower selectivity and
stereospecificity.
We propose a mechanism for the “fast” cycle mediated
by highly active Ru(III) and oxoRu(V) species. A second,
“slow” catalytic pathway involving trans-dioxoRu(VI) and
oxoRu(IV) porphyrin complexes proceeds simultaneously.
Experimental Section
We suggest that the apparent concertedness of the oxy-
genation reaction with this ruthenium system derives from
both this late transition state and the strong preference of
Ru to remain low-spin. Thus, a doublet d3 ground state for
oxoRu(V) could pass adiabatically to a doublet d5 Ru(III)
complex. The lack of a spin barrier during the transformation
between (dxy)2(dxz)1 and (dxy)2(dxz)2(dyz)1 electron configura-
tions of ruthenium porphyrin in the oxoRu(V)-Ru(III)
catalytic cycle might be a key to understanding the faster
rates of hydrocarbon oxidation with this metastable redox
couple compared to the rates of oxoRu(IV)/trans-dioxo-
Ru(VI)-mediated reactions.17,74-76 For example, we have
shown that low-spin, d2 oxoMn(V) porphyrin complexes do
display large kinetic differences in reactivity, which we have
attributed to the need to promote an electron from dxy to
produce high-spin Mn(IV) or Mn(III) states.71,77
General. Analytical gas chromatography was performed on
Hewlett-Packard 5890 gas chromatographs equipped with flame
ionization detectors. SPB-20 and Stabilwax capillary columns were
used. Mass spectra were taken on a Kratos MS50 RFA mass
spectrometer with a FAB source. 3-Nitrobenzyl alcohol was used
for the matrix. GC-MS data were obtained on a HP 5890A gas
spectrometer equipped with an SPB-20 column and a HP 5970
series mass-selective detector. 1H NMR spectra were taken on
Varian INOVA 500 (500 MHz), JEOL GSX-270 (270 MHz), and
Bruker WM 250 (250 MHz) spectrometers. Chemical shift values
are reported relative to residual solvent resonances (δ 7.26, 5.32,
and 1.11 for CHCl3, CHDCl2, and CHD2CD2OD, respectively). EPR
spectra were recorded at 77 K on a Bruker ESP 300 X-band
spectrometer using a low-temperature Dewar. UV-vis spectra were
measured on a HP 8452A diode array spectrophotometer. IR spectra
were taken on a Nicolet 5DXB FT-IR spectrometer.
Materials. Methylene chloride was purified to remove olefins
by washing it with concentrated sulfuric acid, then aqueous sodium
carbonate and water, followed by drying it over anhydrous CaCl2
and distillation from CaH2. Benzene (thiophene free grade) was
distilled over sodium. Deuterated solvents, CD2Cl2, CDCl3, and
ethanol-d6, were purchased from Aldrich or Cambridge Isotope
Laboratories. CDCl3 and CD2Cl2were purified by passage through
a short column packed with basic alumina. cis-Decalin was purified
by passage through basic alumina. Ru3(CO)12 purchased from
Aldrich was purified by dissolution in hot toluene, filtration, and
removal of the solvent. Free base 5,10,15,20-tetrakis(pentafluo-
rophenyl)porphyrin, H2TPFPP, was obtained from Mid-Century
Chemicals and used without further purification. Zinc powder (20
mesh) was purchased from EM Science. A 200-400 mesh, 60 Å
silica gel (Aldrich) was used for chromatographic separations. Basic
alumina, Brockmann activity grade I (J. T. Baker), was treated with
water (3 mL per 100 g of alumina) before use. All other reagents
were purchased from Aldrich and used without further purification
unless otherwise noted. Cyclopentyl acetate was prepared from
cyclopentanol and acetic anhydride in the presence of ZnCl2.
Iodosylbenzene was prepared by the hydrolysis of iodobenzene
diacetate.78 RuII(TMP)(CO)79 and RuII(TPFPPBr8)(CO)80,81 were
synthesized according to published procedures.
Summary and Conclusions
We have shown that ruthenium complexes of 5,10,15,20-
tetrakis(pentafluorophenyl)porphyrin are efficient catalysts
for the oxygenation of a variety of substrates with 2,6-
dichloropyridine N-oxide used as the oxidant. High rates and
yields and large turnover numbers have been obtained with
these catalysts under mild conditions in a nonacidic reaction
medium.
The carbonylRu(II) and trans-dioxoRu(VI) porphyrin
complexes have been excluded as the active species in this
fast catalytic system, on the basis of comparative kinetic
evidence. While RuVI(TPFPP)(O)2 was found to be an
effective oxidant, the rates of the stoichiometric oxidations
were much too slow to explain the catalysis.
Free-radical processes of substrate oxygenation have been
ruled out by the observed stereospecificity of the hydroxyl-
ation of decalin and the lack of brominated products with
added BrCCl3.
Catalyst activation was observed upon the oxidation of
RuII(TPFPP)(CO) with Fe(ClO4)3 to the corresponding por-
phyrin cation radical, RuII(TPFPP+•)(CO), or upon the
reduction of RuIV(TPFPP)Cl2 with a zinc amalgam. A Ru(III)
porphyrin species was shown to be formed under these
conditions.
Carbonyl(5,10,15,20-tetrakis(pentafluorophenyl)porphyri-
nato) Ruthenium(II) [RuII(TPFPP)(CO)].82 H2TPFPP (200 mg)
and 100 mg of Ru3(CO)12 were refluxed in 30 mL of 1,2-
dichlorobenzene under argon. After 3 h, another 100 mg portion
of Ru3(CO)12 was added, and the reaction continued for 12 h. The
solvent was removed under a vacuum, and the product was
chromatographed with toluene on a silica gel column (Silica gel,
Aldrich, 60-200 mesh, 5 × 20 cm). A bright orange band (Rf )
RuIII(TPFPP)(OEt)-mediated adamantane oxidation was
found to proceed at a rapid initial rate, and then, gradually
slowed as RuIII(TPFPP) was converted to RuVI(TPFPP)(O)2.
The observation confirms that RuVI(TPFPP)(O)2 cannot be
the active oxidant and that the RuIII porphyrin is a precursor
to the active oxidation species.
(78) Lucas, H. J.; Kennedy, E. R. Organic Structure; Wiley: New York,
1955; Vol. 3.
(79) Groves, J. T.; Quinn, R. Inorg. Chem. 1984, 23, 3844-3846.
(80) Birnbaum, E. R.; Schaefer, W. P.; Labinger, J. A.; Bercaw, J. E.; Gray,
H. B. Inorg. Chem. 1995, 34, 1751-1755.
(74) Shaik, S.; Filatov, M.; Schroder, D.; Schwarz, H. Chem.sEur. J. 1998,
4, 193-199.
(75) Hirao, H.; Kumar, D.; Thiel, W.; Shaik, S. J. Am. Chem. Soc. 2005,
127, 13007-13018.
(81) Mandon, D.; Ochsenbein, P.; Fischer, J.; Weiss, R.; Jayaraj, K.; Austin,
R. N.; Gold, A.; White, P. S.; Brigaud, O.; Battioni, P.; Mansuy, D.
Inorg. Chem. 1992, 31, 2044-2049.
(76) Sharma, P. K.; de Visser, S. P.; Ogliaro, F.; Shaik, S. J. Am. Chem.
Soc. 2003, 125, 2291-2300.
(77) Jin, N.; Groves, J. T. J. Am. Chem. Soc. 1999, 121, 2923-2924.
(82) Murahashi, S. I.; Naota, T.; Komiya, N. Tetrahedron Lett. 1995, 36,
8059-8062.
Inorganic Chemistry, Vol. 45, No. 12, 2006 4779

Wang et al.
0.39) was collected. Removal of the solvent yielded 165 mg (73%)
of the product. 1H NMR (CD2Cl2, 270 MHz): Hâ δ 8.76. UV-vis
(CH2Cl2): λmax (log ꢀ) 402 (5.45), 524 (4.29), 554 (4.00) nm.
trans-Dioxo(5,10,15,20-tetrakis(pentafluorophenyl)por-
phyrinato) Ruthenium(VI) [RuVI(TPFPP)(O)2]. The solution of
27 mg of m-CPBA (57-87%) in 1.5 mL of CH2Cl2 was added to
40 mg (36.3 µmol) of RuII(TPFPP)(CO) dissolved in 5 mL of
CH2Cl2. The reaction mixture was stirred for 2.5 min at room
temperature and then applied to a jacketed silica gel column (1 ×
10 cm, Silica Gel) cooled with a dry ice/acetone mixture. A fast-
moving brown band was collected; the solvent was removed on a
rotary evaporator, and the brown solid was dried under a vacuum
unreacted 2,6-dichloropyridine. The second slow-moving band was
collected, and 1.72 g (78%) of the product was obtained. 1H NMR
(CD2Cl2, 270 MHz): δ 7.10 (t, J ) 8.6, 1 H), 7.46 (d, J ) 8.6, 2
H).
General Procedure for Catalytic Oxidations. Most hydrocar-
bon oxidations were carried out in threaded 3.7 mL vials with an
open-top closure and a PTFE-faced silicone septum. The measure-
ments of volumes were done with gastight Hamilton syringes. The
conditions for a typical reaction were as follows: the substrate (200
µL of 0.2 M in CH2Cl2), 2,6-dichloropyridine N-oxide (200 µL of
0.2 M in CH2Cl2), and 1,2-dichlorobenzene (100 µL of 0.2 M in
CH2Cl2) were added to 1.4 mL of methylene chloride in a vial.
The reaction solution was purged with argon for 5 min with a needle
pierced through the septum. The vial was wrapped in aluminum
foil to protect it from exposure to light and preheated in a thermostat
for 10 min to the desired temperature. The reaction was initiated
by the injection of 100 µL of 1 mM ruthenium porphyrin stock
solution in CH2Cl2 into the vial. The porphyrin catalyst stock
solution was also purged with argon for 5 min before use. A gas
chromatographic analysis was performed on aliquots withdrawn
directly from the reaction vessel. The yields and turnover numbers
were determined against 1,2-dichlorobenzene used as an internal
standard.
Maximum turnover numbers obtained for the hydroxylation of
cyclohexane with 2,6-dichloropyridine N-oxide with several ruthe-
nium porphyrin catalysts (10 µM) were compared {total TON )
([cyclohexanol] + 2[cyclohexanone])/[Ru catalyst]0. [Cyclohexane]
) 2.0 M, [2,6-dichloropyridine N-oxide] ) 0.05 M, CH2Cl2, 65
°C}. Under these conditions, the reaction catalyzed by RuII(TMP)-
(CO) and RuII(TPFPPBr8)(CO) gave 300 TO, RuII(TPFPP)(CO)
afforded 2500 TO, and RuII(TPFPPBr8)(CO) gave 5800 TO. RuII-
(TPFPPBr8)(CO) was recharged with additional 0.05 M 2,6-
dichloropyridine N-oxide.
For the reactions carried out in the presence of a zinc amalgam,
100 mg of the zinc amalgam and a small stirrer were introduced
into the reaction vial. The reaction solution was rapidly stirred.
Competitive Oxidation of Hydrocarbons. Competition between
pairs of substrates in catalytic oxidations were performed to
determine the relative reactivities of different substrates as such as
well as to measure the intermolecular deuterium isotope effects and
obtain data for the Hammett correlations. In a typical experiment,
the following stock solutions prepared in methylene chloride were
added to 1.2 mL of CH2Cl2 in a 3 mL vial: substrate a (200 µL of
0.2 M), substrate b (200 µL of 0.2 M), 2,6-dichloropyridine N-oxide
(200 µL of 0.2 M), and 1,2-dichlorobenzene (100 µL of 0.2 M).
This mixture was purged with argon for 5 min and, after that,
analyzed by gas chromatography at least three times to determine
the exact initial concentrations of the substrates relative to the 1,2-
dichlorobenzene internal standard. After that, the vial was preheated
in a thermostat for 10 min, and 100 µL of 1 mM ruthenium
porphyrin catalyst was injected. The final composition of the
reaction mixture was analyzed by gas chromatography. At least
three GC analyses were performed for each sample.
1
(30.6 mg, 76%). H NMR (CD2Cl2, 270 MHz): Hâ δ 9.26. UV-
vis (CH2Cl2): λmax (log ꢀ) 412 (5.35), 506 (4.18) nm. IR (KBr)
ν
RuO: 826 cm-1
.
Dichloro(5,10,15,20-tetrakis(pentafluorophenyl)por-
phyrinato) Ruthenium(IV) [RuIV(TPFPP)Cl2]. RuIV(TPFPP)Cl2
was prepared both by refluxing RuII(TPFPP)(CO) in carbon
tetrachloride in air, in a procedure similar to that described by Gross
and Barzilay83 and by the reaction of RuVI(TPFPP)(O)2 with HCl.
Both methods produced substantial amounts of µ-oxo dimer
byproducts, [RuIV(TPFPP)X]2O, along with RuIV(TPFPP)Cl2.38,83
The dichloroRu(IV) porphyrin could be easily isolated by refluxing
the crude reaction mixture in a Soxhlet apparatus with diethyl ether,
because the µ-oxo dimers were soluble in ether and RuIV(TPFPP)-
Cl2 was not. RuIV(TPFPP)Cl2 was EPR-silent. The position of its
paramagnetically shifted pyrrole H (δ -53) in the 1H NMR
spectrum was close to those for the related complexes, RuIV(TMP)-
Cl2 (δ -55.0),70 RuIV(TDMPP)Cl2 (δ -55.4), and RuIV(TPP)Cl2
(δ -57.7).83
Method 1. RuII(TPFPP)(CO) (60 mg, 54.5 µmol) was dissolved
in 250 mL of carbon tetrachloride in a 500 mL round-bottom flask
equipped with a reflux condenser. The solution was refluxed for 4
h exposed to air. The color of the reaction mixture changed from
bright orange to green. The solvent was removed on a rotary
evaporator, and the solid was dried under a vacuum. The crude
product was placed in a Soxhlet apparatus and extracted with diethyl
ether for 8 h. A brown solid insoluble in ether was dried under a
vacuum (32.8 mg, 53%).
Method 2. Anhydrous HCl gas was bubbled through a solution
of 30 mg of RuVI(TPFPP)(O)2 in 50 mL of CH2Cl2 for 30 min.
After 12 h, the solvent was removed, and the brown solid was dried
under a vacuum. The product was purified as described in method
1 (12.5 mg, 40%). NMR (CD2Cl2, 270 MHz): Hâ δ -53. UV-vis
(CH2Cl2): λmax (log ꢀ) 406 (5.25), 506 (5.07) nm.
2,6-Dichloropyridine N-Oxide. To 15 mL of trifluoroacetic acid
was added 2.0 g (13.5 mmol) of 2,6-dichloropyridine in a 100 mL
round-bottom flask. A total of 3 mL of 30% hydrogen peroxide
was added dropwise, and the reaction was run at 75-80 °C for 5
h. Then, 1.5 mL of hydrogen peroxide was added, and the reaction
was carried out for another 5 h. After that, the reaction was diluted
with 20 mL of chloroform and quenched with 20 mL of a 5%
aqueous sodium metabisulfate solution. Solid K2CO3 was added
to neutralize the trifluoroacetic acid. The organic phase was
separated, washed with water until a neutral reaction of the aqueous
phase was established, and dried over anhydrous sodium sulfate.
The solvent was removed on a rotary evaporator, and the solid
product was purified by chromatography on a basic alumina column
(4 × 7 cm, basic alumina was partially deactivated with 3% water
by weight, eluent CH2Cl2). The first fast-moving band was
The relative reactivities of substrates a and b, krel ) ka/kb, were
calculated as follows:
krel
)
ln([substrate a]/[substrate a]0)/ln([substrate b]/[substrate b]0)
where [substrate a]0 and [substrate b]0 are the initial concentrations
of the two substrates.
The relative reactivity of cyclohexanol versus that of cyclohexane
was obtained from the kinetics of appearance of the products in
(83) Gross, Z.; Barzilay, C. M. J. Chem. Soc., Chem. Commun. 1995,
1287-1288.
4780 Inorganic Chemistry, Vol. 45, No. 12, 2006

Fast Catalytic Hydroxylation of Hydrocarbons
Table 5
In experiments on simultaneous monitoring of the product
evolution and the fate of the porphyrin catalyst, the reactions were
run directly in a 1-cm-path-length UV-vis cell with a threaded
neck closed with an open-top cap and a PTFE-faced silicone septum.
The reaction temperature was adjusted with a temperature control
unit connected to the cell compartment of a spectrophotometer. The
transformations of the catalyst were followed by UV-vis spec-
troscopy in the 450-650 nm range using a <430 nm cutoff filter
placed in the light beam of the spectrophotometer. Aliquots of 30
µL were periodically withdrawn directly from the UV-vis cell at
time intervals, frozen in liquid nitrogen, and later analyzed by GC
in order to follow the evolution of the products of oxidation.
Stoichiometric Reactions of RuVI(TPFPP)(O)2. A 2.2 mM
solution of RuVI(TPFPP)(O)2 and a 40 mM solution of cis-decalin
in CD2Cl2 were prepared in an inert-atmosphere glovebox. The
reaction was initiated by injecting 90 µL of a 40 mM solution of
cis-decalin into 0.80 mL of a 2.2 mM solution of RuVI(TPFPP)-
(O)2 in an NMR tube with a threaded neck covered with a screw
cap and a PTFE-faced silicone septum. The disappearance of the
â-pyrrole resonance of RuVI(TPFPP)(O)2 was monitored at 25 °C,
and the rate constant of the reaction was obtained from a single-
exponential fit of the data. Control experiments in the absence of
a substrate showed that catalyst decomposition was negligible in
these runs. A similar procedure was used to determine the rate
constant for the stoichiometric oxidation of styrene.
The reactions of the other substrates were studied by UV-vis
spectroscopy. A total of 2.85 mL of the substrate solution (21.1
mM for adamantane, 42.1 mM for benzyl alcohol, and 2 M for
cyclohexanol) was placed in a UV-vis cell and purged with argon
for at least 5 min. The cell was equilibrated for 10 min to the desired
temperature in the cell compartment of the spectrophotometer. A
<430 nm cutoff filter was placed in the light beam to decrease the
exposure of the sample to light. A 1.0 mM solution of RuVI(TPFPP)-
(O)2 was prepared under argon. After the injection of 150 mL of
RuVI(TPFPP)(O)2 into the cell, an increase in absorbance at 450
nm followed. The rise in absorbance at this wavelength was due to
the appearance of oxoRu(IV) porphyrin in the case of cyclohexanol
and benzyl alcohol and to the formation of a µ-oxoRu(IV) dimer
during adamantane oxidation. The rate constants (kst) were obtained
from a single-exponential fit of the data.
time,
min
cyclohexanol,
mM (TO)
cyclohexanone,
mM (TO)
k
cyclohexanol/kcyclohexane
5
10
20
30
40
1.92 (192)
3.46 (346)
6.12 (612)
7.63 (763)
8.91 (891)
10.53 (1053)
10.80 (1080)
11.98 (1198)
12.54 (1254)
12.33 (1233)
0.12 (12)
0.38 (38)
127.4
128.6
119.9
122.7
117.3
125.4
128.8
124.0
117.4
121.6
1.12 (112)
1.79 (179)
2.33 (233)
3.48 (348)
3.76 (376)
4.45 (445)
4.61 (461)
4.62 (462)
85
103
223
268
354
the catalytic hydroxylation of cyclohexane (2.0 M) with 2,6-
dichloropyridine N-oxide (0.05 M) catalyzed by 10 µM RuII(TPF-
PP)(CO) in CH2Cl2 at 65 °C following the general procedure. The
concentrations of the oxidation products, cyclohexanol and cyclo-
hexanone, were determined by gas chromatography at time intervals
during the course of the reaction. 1,2-Dichlorobenzene (0.01 M)
was used as an internal standard. Assuming that the same active
oxidant is involved in the oxidation of both cyclohexane and
cyclohexanol, the rate expressions can be written as follows:
d[cyclohexanol]/dt ≈ kcyclohexane[cyclohexane] f(X)
(I)
where kcyclohexane and kcyclohexanol are absolute rate constants of
d[cyclohexanone]/dt ) kcyclohexanol[cyclohexanol] f(X) (II)
oxidation of cyclohexane and cyclohexanol, respectively, and f(X)
is a substrate-independent component of the rate. The division of
eq I by eq II and rearrangement followed by integration will give
a time-independent expression:
kcyclohexanol/kcyclohexane
)
(2[cyclohexane][cyclohexanone])/[cyclohexanol]2 (III)
Because [cyclohexane]0 . [2,6-dichloropyridine N-oxide], we can
assume that [cyclohexane] ≈ [cyclohexane]0. Using formula III to
calculate the relative reactivity of cyclohexane and cyclohexanol,
kcyclohexanol/kcyclohexane was found to be 123 ( 5 during the whole
Catalytic Oxidation of cis-Decalin Followed by NMR. To 0.7
mL of a 2.86 mM solution of RuVI(TPFPP)(O)2 in an NMR tube
were injected 2,6-dichloropyridine N-oxide and cis-decalin (0.20
mL of 0.2 M). The course of the reaction was followed at 25 °C
by monitoring the disappearance of 2,6-dichloropyridine N-oxide
peaks and the appearance of the resonances of 2,6-dichloropyridine.
A turnover rate of 0.045 TO/min for cis-decalin oxidation was
determined from a linear fit of the data obtained during the first
250 min of the oxidation.
Formation of µ-Oxoruthenium(IV) Porphyrin Dimer [RuIV-
(TPFPP)(X)]2O upon Decomposition of RuVI(TPFPP)(O)2. Freshly
prepared RuVI(TPFPP)(O)2 (1 mg) was dissolved in 1 mL of CD2-
Cl2 and left at room temperature for 72 h. The reaction mixture
was analyzed by 1H NMR. 1H NMR (270 MHz): Hâ δ 8.62. UV-
vis (CH2Cl2): λmax (log ꢀ) 384 (5.28), 398 (5.21), 532 (4.42), 552
(4.38) nm. FAB-MS (3-nitrobenzyl alcohol as the matrix): cluster
centered at m/z 2164 for [Ru(TPFPP)]2O (observed relative
intensity, calculated relative intensity), 2159 (31, 27), 2160 (45,
32), 2161 (65, 56), 2162 (77, 72), 2163 (91, 89), 2164 (100, 100),
2165 (82, 80), 2166 (73, 74), 2167 (48, 45).
course of the oxidation (ca. 350 min). Data for these assays are
shown in Table 5.
Hydroxylation of Adamantane in the Presence of H218O. A
500 µL reaction mixture containing 20 µM RuVI(TPFPP)(O)2 or
RuII(TPFPP)(CO), 40 µM 2,6-dichloropyridine N-oxide, 20 mM
adamantane, and 10 µL H218O was stirred at 25 °C under argon.
Product development was monitored periodically by GC. Separate
GC-MS analyses were performed on two reaction mixtures taken
after 30% conversion to 1-adamantanol. To determine the 18O
content of the water, a 100 µL aliquot of each reaction mixture
was treated with 0.4 µL of triethylorthopropionate and the 18O
content of the resultant ethyl propionate as determined by GC-
MS. The 18O content of the oxidation product, 1-adamantanol, was
determined from the ratio of m/e, 152 and 154 after correction for
the 18O content. The 18O incorporation in the 1-adamantol obtained
from the oxidation catalyzed by RuVI(TPFPP)(O)2 was found to be
18%. If RuII(TPFPP)(CO) was used as the catalyst, the 18O
incorporation in 1-adamantanol was 3.42%.
Kinetic Measurements. Oxidations were carried out as described
in the general procedure. Aliquots of 30 µL were periodically
withdrawn from a reaction, placed in small vials, and immediately
frozen in liquid nitrogen. Later, the aliquots were defrosted and
analyzed by gas chromatography one at a time.
Reduction of Dichloro(5,10,15,20-tetrakis(pentafluorophenyl)-
porphyrinato) Ruthenium(IV) [RuIV(TPFPP)Cl2] with a Zinc
Amalgam. To 1 mL of 0.5 mM RuIV(TPFPP)Cl2 in CD2Cl2, 100
Inorganic Chemistry, Vol. 45, No. 12, 2006 4781

Wang et al.
mg of a freshly prepared zinc amalgam was added. The reaction
mixture was stirred at 25 °C for 20 min under argon. The 1H NMR
spectrum taken at this point showed no resonances in the upfield
region (δ < 0) of the spectrum. No signals were apparent in the
X-band EPR spectrum of this solution at 77 K. A 10 µL aliquot of
ethanol-d6 was added at 25 °C, and the NMR spectrum, taken after
15 min, displayed a typically broadened resonance at δ -15
characteristic of RuIII(TPFPP)(OEt)(HOEt). The EPR spectrum of
this solution at 77 K displayed a sharp, three-line signal at g )
2.53, 2.12, and 1.89 as expected for a low-spin d5 system. In a
control experiment, the treatment of RuIV(TPFPP)Cl2 with ethanol-
d6 under identical conditions but without a zinc amalgam treatment
showed only the δ -53 resonance of the starting Ru(IV) porphyrin.
The EPR spectrum was silent both 20 min and 3 h after the ethanol
treatment.
oxidation experiment was carried out in quartz UV cells with open-
top, screw-cap, and Teflon-coated rubber septa. The UV-vis spectra
were recorded in cycle mode with the temperature maintained at
25 °C. Reaction time was measured from the addition of the
substrate and the oxidant. At regular intervals, an aliquot (30 µL)
was withdrawn for GC analysis. The reaction mixture was prepared
with RuIII(TPFPP)(OEt)(EtOH) as a catalyst (7.5 µM), a substrate
(0.02 M), an oxidant (0.01 M), 1,2-dichlorobenzene as a GC internal
standard (0.01 M), and methylene chloride (2 mL) as the solvent.
Acknowledgment. The National Science Foundation
(CHE-316301) for support of this research, the Monsanto
Company for fellowship support to K.V.S., and Bayer, AG
for partial support of C.W. are gratefully acknowledged.
Adamantane Oxidation Catalyzed by RuIII(TPFPP)(OEt)-
(EtOH) Followed by GC and UV-vis Spectroscopy. The
IC0520566
4782 Inorganic Chemistry, Vol. 45, No. 12, 2006