Laser flash photolysis generation and kinetic studies of porphyrin-manganese-oxo intermediates. Rate constants for oxidations effected by porphyrin-MnV-oxo species and apparent disproportionation equilibrium constants for porphyrin-MnIV-oxo species

Article abstract of DOI:10.1021/ja045042s

Porphyrin-manganese(V)-oxo and porphyrin-manganese(IV)-oxo species were produced in organic solvents by laser flash photolysis (LFP) of the corresponding porphyrin-manganese(III) perchlorate and chlorate complexes, respectively, permitting direct kinetic studies. The porphyrin systems studied were 5,10,15,20-tetraphenylporphyrin (TPP), 5,10,15,20- tetrakis(pentafluorophenyl)porphyrin (TPFPP), and 5,10,15,20-tetrakis(4- methylpyridinium)porphyrin (TMPyP). The order of reactivity for (porphyrin)Mn^V(O) derivatives in self-decay reactions in acetonitrile and in oxidations of substrates was (TPFPP) > (TMPyP) > (TPP). Representative rate constants for reaction of (TPFPP)Mn^V(O) in acetonitrile are k = 6.1 ×10⁵ M^-1 s^-1 for cis-stilbene and k = 1.4 × 10⁵ M^-1 s^-1 for diphenylmethane, and the kinetic isotope effect in oxidation of ethylbenzene and ethylbenzene-d₁₀ is k_H/k_D = 2.3. Competitive oxidation reactions conducted under catalytic conditions display approximately the same relative rate constants as were found in the LFP studies of (porphyrin)Mn^V(O) derivatives. The apparent rate constants for reactions of (porphyrin)Mn^IV(O) species show inverted reactivity order with (TPFPP) < (TMPyP) < (TPP) in reactions with cis-stilbene, triphenylamine, and triphenylphosphine. The inverted reactivity results because (porphyrin)Mn^IV(O) disproportionates to (porphyrin)Mn^IIIX and (porphyrin)Mn^V(O), which is the primary oxidant, and the equilibrium constants for disproportionation of (porphyrin)Mn^IV(O) are in the order (TPFPP) < (TMPyP) < (TPP). The fast comproportionation reaction of (TPFPP)Mn^V(O) with (TPFPP)Mn^IIICl to give (TPFPP)Mn^IV-(O) (k = 5 × 10⁸ M^-1 s ^-1) and disproportionation reaction of (TPP)Mn^IV(O) to give (TPP)Mn^V(O) and (TPP)-Mn^IIIX (k ≈ 2.5 × 10⁹ M^-1 s^-1) were observed. The relative populations of (porphyrin)Mn^V(O) and (porphyrin)Mn^IV(O) were determined from the ratios of observed rate constants for self-decay reactions in acetonitrile and oxidation reactions of cis-stilbene by the two oxo derivatives, and apparent disproportionation equilibrium constants for the three systems in acetonitrile were estimated. A model for oxidations under catalytic conditions is presented.

Full text of DOI:10.1021/ja045042s

Published on Web 04/15/2005
Laser Flash Photolysis Generation and Kinetic Studies of
Porphyrin-Manganese-Oxo Intermediates. Rate Constants
for Oxidations Effected by Porphyrin-MnV-Oxo Species and
Apparent Disproportionation Equilibrium Constants for
Porphyrin-MnIV-Oxo Species
Rui Zhang, John H. Horner, and Martin Newcomb*
Contribution from the Department of Chemistry, UniVersity of Illinois at Chicago,
845 West Taylor Street, Chicago, Illinois 60607
Received August 17, 2004; E-mail: men@uic.edu
Abstract: Porphyrin-manganese(V)-oxo and porphyrin-manganese(IV)-oxo species were produced in
organic solvents by laser flash photolysis (LFP) of the corresponding porphyrin-manganese(III) perchlorate
and chlorate complexes, respectively, permitting direct kinetic studies. The porphyrin systems studied were
5,10,15,20-tetraphenylporphyrin (TPP), 5,10,15,20-tetrakis(pentafluorophenyl)porphyrin (TPFPP), and
5,10,15,20-tetrakis(4-methylpyridinium)porphyrin (TMPyP). The order of reactivity for (porphyrin)MnV(O)
derivatives in self-decay reactions in acetonitrile and in oxidations of substrates was (TPFPP) > (TMPyP)
> (TPP). Representative rate constants for reaction of (TPFPP)MnV(O) in acetonitrile are k ) 6.1 × 105
M-1 s-1 for cis-stilbene and k ) 1.4 × 105 M-1 s-1 for diphenylmethane, and the kinetic isotope effect in
oxidation of ethylbenzene and ethylbenzene-d10 is kH/kD ) 2.3. Competitive oxidation reactions conducted
under catalytic conditions display approximately the same relative rate constants as were found in the LFP
studies of (porphyrin)MnV(O) derivatives. The apparent rate constants for reactions of (porphyrin)MnIV(O)
species show inverted reactivity order with (TPFPP) < (TMPyP) < (TPP) in reactions with cis-stilbene,
triphenylamine, and triphenylphosphine. The inverted reactivity results because (porphyrin)MnIV(O)
disproportionates to (porphyrin)MnIIIX and (porphyrin)MnV(O), which is the primary oxidant, and the
equilibrium constants for disproportionation of (porphyrin)MnIV(O) are in the order (TPFPP) < (TMPyP) <
(TPP). The fast comproportionation reaction of (TPFPP)MnV(O) with (TPFPP)MnIIICl to give (TPFPP)MnIV-
(O) (k ) 5 × 108 M-1 s-1) and disproportionation reaction of (TPP)MnIV(O) to give (TPP)MnV(O) and (TPP)-
MnIIIX (k ≈ 2.5 × 109 M-1 s-1) were observed. The relative populations of (porphyrin)MnV(O) and
(porphyrin)MnIV(O) were determined from the ratios of observed rate constants for self-decay reactions in
acetonitrile and oxidation reactions of cis-stilbene by the two oxo derivatives, and apparent disproportionation
equilibrium constants for the three systems in acetonitrile were estimated. A model for oxidations under
catalytic conditions is presented.
High-valent transition metal-oxo intermediates are active
oxidizing species formed in metal-catalyzed oxidation reactions
in nature and in the laboratory.1 In typical synthetic applications,
a catalytic amount of porphyrin-, salen-, or other Schiff-base-
metal complex and a stoichiometric amount of sacrificial oxidant
(also called terminal oxidant) is employed. The low concentra-
tions and high reactivities of the metal-oxo transients result in
difficulties in physical studies that can be addressed in some
cases by rapid mixing techniques or by the production of low-
reactivity analogues.2 Nonetheless, most of the mechanistic
information for transients that are thought to be the active species
in catalytic oxidation processes has been inferred from product
studies.
Manganese-oxo intermediates are among the more reactive
transition metal derivatives. A variety of these species are
employed catalytically in synthesis,1 and nature uses Mn-oxo
species in the production of oxygen.3 Highly reactive porphy-
rin-manganese(V)-oxo derivatives,4,5developed as models for
cytochrome P450 enzymes,6 are putative intermediates in
catalytic processes that have been known for decades,7 but they
eluded synthesis until 1997 when Groves and co-workers8
reported the preparation of (TMPyP)MnV(O).9 As recently as
2003, only two additional examples of (porph)MnV(O) were
(2) Examples of the isolation and/or characterization of oxometalloporphyrin
complexes include the following. FeIV(O) radical cation species: Groves,
J. T.; Haushalter, R. C.; Nakamura, M.; Nemo, T. E.; Evans, B. J. J. Am.
Chem. Soc. 1981, 103, 2884-2886. FeIV(O) species: Groves, J. T.; Gross,
Z.; Stern, M. K. Inorg. Chem. 1994, 33, 5065-5072. CrV(O) species:
Groves, J. T.; Kruper, W. J. J. Am. Chem. Soc. 1979, 101, 7613-7614.
RuVI(O) species: Groves, J. T.; Quinn, R. Inorg. Chem. 1984, 23, 3844-
3846; Leung, W.-H.; Che, C.-M. J. Am. Chem. Soc. 1989, 111, 8812-
8818.
(1) Metalloprophyrins in Catalytic Oxidations; Sheldon, R. A., Ed.; Marcel
Dekker: New York, 1994. Metal-Oxo and Metal-Peroxo Species in
Catalytic Oxidations; Meunier, B., Ed.; Springer-Verlag: Berlin, 2000.
Meunier, B. Chem. ReV. 1992, 92, 1411-1456. Jacobsen, E. N. In
ComprehensiVe Organometallic Chemistry II; Wilkinson, G. W., Stone, F.
G. A., Abel, E. W., Hegedus, L. S., Eds.; Pergamon: New York, 1995;
Vol. 12, pp 1097-1135.
(3) Yachandra, V. K.; Sauer, K.; Klein, M. P. Chem. ReV. 1996, 96, 2927-
2950.
9
10.1021/ja045042s CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 6573-6582
6573

A R T I C L E S
Zhang et al.
reported,10-12 and these derivatives were produced in water,
which apparently stabilizes the intermediates. As one might
expect, kinetic information for reactions of (porph)MnV(O)
species is limited; Groves and co-workers measured the kinetics
of reactions of one species directly,8 and Bruice and co-workers
determined the rate constants for olefin epoxidations by another
species under catalytic turnover conditions.13
Table 1. UV-Visible Absorbances for Porphyrin-Manganese(III)
Complexesa
complex
Soret band (log
ꢀ
)
q-band (log ꢀ)
(TPFPP)Mn(Cl)
(TPFPP)Mn(Cl)b
(TMPyP)Mn(Cl)
(TPP)Mn(Cl)
(TPFPP)Mn(ClO4)
(TPFPP)Mn(ClO4)b
(TMPyP)Mn(ClO4)
(TPP)Mn(ClO4)
(TPFPP)Mn(ClO3)
(TPFPP)Mn(ClO3)b
(TPFPP)Mn(NO3)
(TMPyP)Mn(ClO3)
(TPP)Mn(ClO3)
359 (5.09), 470 (5.23)
364 (4.90), 475 (4.99)
370 (4.45), 470 (4.88)
373 (4.72), 475 (5.00)
375 (5.22), 474 (5.18)
377 (4.80), 475 (4.90)
386 (4.70), 475 (4.85)
390 (5.02), 486 (4.80)
370 (5.05), 460 (5.25)
368 (4.89), 470 (5.07)
370 (5.05), 464 (5.28)
377 (4.48), 465 (4.92)
384 (4.74), 473 (4.82)
571 (4.44)
573 (4.25)
580 (4.04)
583 (4.08), 617 (4.13)
557 (4.45)
562 (4.23)
563 (4.10)
572 (4.10), 606 (4.08)
557 (4.47)
566 (4.47)
558 (4.48)
In an extension of known photocatalytic reactions of por-
phyrin-manganese species.14 we used laser flash photolysis
(LFP) methods to produce in organic solvents porphyrin-MnV-
oxo species that are thought to be the active species in catalytic
oxidation reactions.15 The temporal resolution of LFP studies
is many orders of magnitude shorter than those of the fastest
mixing methods, and the approach permitted direct kinetic
studies of (porph)MnV(O) reactions with substrates. In the
present work, we detail the LFP studies of porphyrin-MnV-
oxo intermediates and also report LFP production and direct
kinetic studies of porphyrin-MnIV-oxo intermediates. The
reactions of (porph)MnV(O) are much faster than one would
deduce from rates of product formation under catalytic condi-
tions because comproportionation of (porph)MnV(O) with (por-
ph)MnIIIX limits the amount of active species in the catalytic
processes. A major pathway, perhaps the only pathway, for
oxidations by (porph)MnIV(O) intermediates involves the same
equilibrium, with disproportionation giving (porph)MnIIIX spe-
cies and (porph)MnV(O) intermediates that are the primary
oxidants. Apparent equilibrium constants for the disproportion-
ation reactions of three systems in acetonitrile were determined
from the kinetic results, and a mechanistic scheme for oxidations
by a commonly employed system was constructed. The mecha-
nistic conclusions have implications for studies of oxidations
by other transition-metal-oxo derivatives.
566 (4.10)
571 (4.13), 605 (4.08)
a λmax values in nm in acetonitrile solvent unless noted. b In C6H5CF3.
Results and Discussion
We studied manganese complexes of three porphyrins that
encompass the typical range of reactivities of these intermedi-
ates, 5,10,15,20-tetraphenylporphyrin (TPP), 5,10,15,20-tetrakis-
(4-methylpyridinium)porphyrin (TMPyP), and 5,10,15,20-
tetrakis(pentafluorophenyl)porphyrin (TPFPP).9 TPP- and
TPFPP-metal complexes are among the more widely studied.
The water-soluble TMPyP-Mn complexes are important be-
cause (TMPyP)MnV(O) was the first porphyrin-manganese-
(V)-oxo complex synthesized and characterized spectroscopi-
cally.8 The porphyrin ring is doubly deprotonated, with the result
that (porph)MnIII and (porph)MnV(O) species are cationic,
whereas (porph)MnIV(O) species are neutral. The consensus
view is that the reactivities of porphyrin-metal-oxo complexes
are related to electronic demand of the porphyrins, with the
electron-withdrawing TPFPP complexes being more highly
reactive.16
Photochemical Production of Manganese-Oxo Deriva-
tives. Reaction of (porph)MnIIICl complexes with Ag(ClO4),
with Ag(ClO3), or with Ag(NO3) gave the corresponding
porphyrin-manganese perchlorate, chlorate, or nitrate com-
plexes, respectively, that were characterized by their UV-visible
spectra. Table 1 lists the Soret and q-band absorbances of these
salts. Acetonitrile solutions were prepared, and other organic
solvents could be employed. Slight shifts in the UV-visible
absorbances were found when the solvent was changed.
Laser flash photolysis (LFP) of the perchlorate, chlorate, or
nitrate complexes gave manganese-oxo species that were
observed within microseconds, as discussed below. We super-
ficially studied the fast processes following photolyses of
(TPFPP)MnIII(ClO4) and (TPFPP)MnIII(ClO3) salts in CH3CN
to ensure that decays of excited states were not convoluted with
the kinetics of reactions of the oxo species discussed later.
Irradiation of these salts with 355 nm light resulted in
fluorescence from the singlet states that decayed with rate
constants of 2 × 108 s-1; Figure 1 shows the results from
(TPFPP)MnIII(ClO4). Short-lived triplet states also apparently
were formed as indicated by rapidly decaying small absorbances
(k ≈ 2 × 107 s-1) that were not fully characterized. The decays
of the excited states in these studies were at least 4 orders of
magnitude faster, and up to 8 orders of magnitude faster, than
the kinetics of the reactions of manganese-oxo species with
substrates.
(4) Stable non-porphyrin-manganese(V)-oxo species include bis-amido-bis-
alkoxo, tetraamido, corrole, and corrolazine complexes. For examples,
see: MacDonnell, F. M.; Fackler, N. L. P.; Stern, C.; O’Halloran, T. V. J.
Am. Chem. Soc. 1994, 116, 7431-7432; Miller, C. G.; Gordon-Wylie, S.
W.; Horwitz, C. P.; Strazisar, S. A.; Peraino, D. K.; Clark, G. R.; Weintraub,
S. T.; Collins, T. J. J. Am. Chem. Soc. 1998, 120, 11540-11541; Gross,
Z.; Golubkov, G.; Simkhovich, L. Angew. Chem., Int. Ed. Engl. 2000, 39,
4045-4047; Mandimutsira, B. S.; Ramdhanie, B.; Todd, R. C.; Wang, H.;
Zareba, A. A.; Czemuszewicz, R. S.; Goldberg, D. P. J. Am. Chem. Soc.
2002, 124, 15170-15171; Liu, H.-Y.; Lai, T.-S.; Yeung, L.-L.; Chang, C.
K. Org. Lett. 2003, 5, 617-620.
(5) For the identification of oxomanganese-salen complexes, see: Feichtinger,
D.; Plattner, D. A. Angew. Chem., Int. Ed. Engl. 1997, 36, 1718-1719.
(6) For leading references, see the following. Hill, C. L.; Schardt, B. C. J.
Am. Chem. Soc. 1980, 102, 6374-6375; Groves, J. T.; Kruper, W. J.;
Haushalter, R. C. J. Am. Chem. Soc. 1980, 102, 6375-6377; Meunier, B.;
Guilmet, E.; De Carvalho, M. E.; Poilblanc, R. J. Am. Chem. Soc. 1984,
106, 6668-6676; Collman, J. P.; Brauman, J. I.; Meunier, B.; Raybuck, S.
A.; Kodadek, T. Proc. Natl. Acad. Sci. U.S.A. 1984, 81, 3245-3248.
Collman, J. P.; Brauman, J. I.; Meunier, B.; Hayashi, T.; Kodadek, T.;
Raybuck, S. A. J. Am. Chem. Soc. 1985, 107, 2000-2005.
(7) Groves, J. T.; Watanabe, Y.; McMurry, T. J. J. Am. Chem. Soc. 1983,
105, 4489-4490.
(8) Groves, J. T.; Lee, J.; Marla, S. S. J. Am. Chem. Soc. 1997, 119, 6269-
6273.
(9) Abbreviations: mcpba, m-chloroperoxybenzoic acid; porph, porphyrinato;
TMP, 5,10,15,20-tetramesitylporphyrin or -porphyrinato; TMPyP, 5,10,15,20-
tetrakis(4-methylpyridinium)porphyrin or -porphyrinato; TPFPP, 5,10,15,20-
tetrakis(pentafluorophenyl)porphyrin or -porphyrinato; TPP, 5,10,15,20-
tetraphenylporphyrin or -porphyrinato.
(10) Jin, N.; Groves, J. T. J. Am. Chem. Soc. 1999, 121, 2923-2924.
(11) Nam, W.; Kim, I.; Lim, M. H.; Choi, H. J.; Lee, J. S.; Jang, H. G. Chem.
Eur. J. 2002, 8, 2067-2071.
(12) A stable dinuclear porphyrin-manganese(V)-oxo example was reported
recently. See: Shimazaki, Y.; Nagano, T.; Takesue, H.; Ye, B.-H.; Tani,
F.; Naruta, Y. Angew. Chem., Int. Ed. 2004, 43, 98-100.
(13) Lee, R. W.; Nakagaki, P. C.; Bruice, T. C. J. Am. Chem. Soc. 1989, 111,
1368-1372.
(14) Suslick, K. S.; Acholla, F. V.; Cook, B. R. J. Am. Chem. Soc. 1987, 109,
2818-2819. Suslick, K. S.; Watson, R. A. New J. Chem. 1992, 16, 633-
642. Hennig, H. Coord. Chem. ReV. 1999, 182, 101-123.
(16) Dolphin, D.; Traylor, T. G.; Xie, L. Y. Acc. Chem. Res. 1997, 30, 251-
(15) Zhang, R.; Newcomb, M. J. Am. Chem. Soc. 2003, 125, 12418-12419.
259.
9
6574 J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005

Porphyrin−Manganese−Oxo Intermediates
A R T I C L E S
Table 2. UV-Visible Absorbances for
Porphyrin-Manganese-Oxo Complexesa
Mn
−oxo
precursor salt
Soret band (log
ꢀ
)
q-band
% conversionb
(TPFPP)MnV(O)
(TMPyP)MnV(O)
(TPP)MnV(O)
(ClO4)
(ClO4)
(ClO4)
(ClO3)
(ClO3)
(NO3)
(ClO3)
(ClO3)
432 (5.40)
450 (4.94)
435 (nd)c
422 (5.27)
418 (5.00)
422 (5.27)
438 (4.88)
425 (4.80)
ndc
ndc
10
3
4
17
12
1
ndc
(TPFPP)MnIV(O)
(TPFPP)MnIV(O)d
(TPFPP)MnIV(O)
(TMPyP)MnIV(O)
(TPP)MnIV(O)
538
540
538
545
540
6
3
a λmax values in nm in CH3CN solution unless noted. b Percent conversion
is laser power dependent; (porph)MnV(O) species were produced with 100
mJ of power, whereas (porph)MnIV(O) species were produced with 30 mJ
of power. c Not determined. d In C6H5CF3.
Figure 1. Fluorescence emission spectrum from 355 nm irradiation of
(TPFPP)MnIII(ClO4) in CH3CN. The inset show the fluorescence emission
kinetic trace at 400 nm; the rate constant for decay is kobs ) 2 × 108 s-1
.
Figure 2. Decay spectra of porphyrin-manganese-oxo derivatives. (A)
(TPFPP)MnV(O) formed by 355 nm LFP of (TPFPP)MnIII(ClO4) in CH3-
CN decaying over 11 ms. (B) (TPFPP)MnIV(O) formed by 355 nm LFP of
the (TPFPP)MnIII(ClO3) in CH3CN decaying over 5 s. Dashed lines in both
spectra are drawn at 427 nm for calibration.
Figure 3. Log-log plot of absorbance at 432 nm vs laser power for
irradiation of (TPFPP)MnIII(ClO4) in CH3CN with 355 nm light. The slope
of the regression line shown is (0.98 ( 0.07).
Scheme 1
Photolysis of (porph)MnIII(ClO4) complexes with 355 nm
laser light resulted in heterolytic cleavage of an O-Cl bond to
give (porph)MnV(O) derivatives (Scheme 1). Figure 2A shows
a representative time-resolved decay spectrum of (TPFPP)-
MnV(O) in CH3CN, and spectra for the other (porph)MnV(O)
species are in Supporting Information. The λmax values for the
Soret bands of (porph)MnV(O) species are listed in Table 2.
For formation of (TPFPP)MnV(O), the yield was linearly related
to laser power (Figure 3), indicating that the reaction was a
one-photon process. The initially formed (porph)MnV(O) cation
presumably complexes with an anion (chlorate from the
photolysis reaction or perchlorate from residual precursor) at a
diffusion-controlled rate.
The spectral signatures of the MnV-oxo derivatives were
confirmed by the production of the same species in rapid mixing
experiments of the (porph)MnIIICl with mcpba. In acetonitrile
solution, it was possible to observe formation of (TMPyP)MnV-
(O), which had a λmax value for the Soret band that was the
same as that found in the LFP study. For the TPP system, this
method gave a mixture of (TPP)MnV(O) with the MnIV-oxo
analogue. Spectra are in Supporting Information.
Figure 4. Time-resolved spectrum over 1 s of (TPFPP)MnV(O) from
reaction of (TPFPP)MnIIICl with 3 equiv of mcpba in water/CH3CN (1:3,
v:v). The time resolution is obtained by subtracting spectra at time ) t
from the spectrum at time ) zero. In this representation, decaying peaks
have positive absorbances, whereas growing peaks have negative absor-
bances.
The more reactive (TPFPP)MnV(O) species could not be
detected in stopped-flow studies in CH3CN, as we previously
reported.15 In a water-CH3CN mixture (1:3, v:v), however, we
observed “instant” formation and then decay of (TPFPP)MnV-
(O) when the manganese(III) chloride was treated with mcpba
(Figure 4). In the water/CH3CN mixture, the Soret band for
(TPFPP)MnV(O) had λmax ) 430 nm; which is blue-shifted from
the value in found in the LFP experiment by 2 nm. It is
noteworthy that the rate of decay of (TPFPP)MnV(O) in the
water/CH3CN mixture was reduced by nearly 3 orders of
magnitude in comparison to the decay in CH3CN,17 indicating
a significant stabilization in the aqueous solution. The three
9
J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005 6575

A R T I C L E S
Scheme 2
Zhang et al.
relative yieldb
Table 3. Quantum Yields for Formation of
Porphyrin-Manganese-Oxo Complexesa
Mn
−oxo
precursor salt
quantum yield
(TPFPP)MnV(O)
(TMPyP)MnV(O)
(TPP)MnV(O)c
(ClO4)
(ClO4)
(ClO4)
(ClO3)
(NO3)
(ClO3)
(ClO3)
(NO3)
2 × 10-3
1.0
0.4
1.0
8.5
0.4
7
9 × 10-4
ca. 2 × 10-3
1.7 × 10-2
7 × 10-4
(TPFPP)MnIV(O)
(TPFPP)MnIV(O)
(TMPyP)MnIV(O)
(TPP)MnIV(O)
1.4 × 10-2
7 × 10-3
3.5
0.03
(TPP)MnIV(O)
7 × 10-5
examples of (porph)MnV(O) derivatives previously reported
from mixing studies also were in prepared in aqueous solu-
tions.8,10,11 Water might stabilize these intermediates by hydrogen-
bonding interactions with the oxo moiety and/or by providing
an hydroxide ligand for manganese, and the origin of this
stabilization deserves further study.
a Quantum yields in units of mol/einstein of 355 nm light in CH3CN
solutions. b Relative quantum yields. c Estimated value.
Photolysis of (porph)MnIII(NO3) complexes was reported by
Suslick and Watson to give (porph)MnIV(O) species by ho-
molytic cleavage of an O-N bond.18 Photolyses of nitrate
complexes also can be performed in LFP experiments, but we
found that photochemical cleavages of the chlorate complexes
were considerably more efficient than cleavages of nitrate
complexes. Thus, 355 nm laser light irradiation of the (porph)-
MnIII(ClO3) complexes in CH3CN gave (porph)MnIV(O) deriva-
tives (Scheme 2) that decayed slowly in comparison to the
analogous (porph)MnV(O) derivatives. Figure 2B shows the
spectrum of (TPFPP)MnIV(O), and other spectra are shown in
Supporting Information. The absorbances are listed in Table 2.
Again, the spectral assignments in the LFP experiments were
confirmed by production of the same MnIV-oxo derivatives by
rapid mixing methods.19 We generated these species by mixing
the (porph)MnIIICl salts with mcpba in the presence of tri-
phenylamine. The amine served as a reductant that rapidly
reduced (porph)MnV(O) derivatives to (porph)MnIV(O) species
and then less rapidly reduced the (porph)MnIV(O) species to
(porph)MnIII derivatives. The spectra of (proph)MnIV(O) from
the mixing experiments are in Supporting Information.
Quantum yields for production of the manganese-oxo species
in LFP experiments are listed in Table 3. The method employed
was the same as that described by Hoshino, Ford, and co-
workers.20,21 where the quantum yields were determined relative
to a quantum yield of 1.0 for formation of the benzophenone
triplet.21 One LFP value, that for formation of (TPP)MnIV(O)
from the nitrate salt, can be compared to a literature value.
Suslick and Watson18 reported a quantum yield for continuous
irradiation of (TPP)MnIII(NO3) of 1.6 × 10-4 mol/einstein at
350 nm, which is in reasonable agreement with the value found
in this work. The quantum yields for formation of (porph)MnIV-
Figure 5. (Left) Kinetic traces at λmax for the self-decay reaction of
(TPFPP)MnIV(O) in CH3CN on two time scales. (Right) Plots of log(kobs
)
for the fast processes in self-decay of (porph)MnIV(O) species vs log of the
concentration of oxo species. The slopes of the regression lines are 0.77
(TPP), 0.83 (TMPyP), and 0.83 (TPFPP).
(O) transients from the chlorate salts were more than 1 order
of magnitude larger than those for formation of the same species
from the nitrate salts, and formation of the (porph)MnIV(O)
derivatives from chlorate salts was more efficient than formation
of the (porph)MnV(O) intermediates from perchlorate salts.
Self-Decay and Disproportionation Reactions. Following
formation of (porph)MnV(O) and (porph)MnIV(O) species in
CH3CN solution, the spectra of the transients decayed in a
complex manner. The major reactions likely were oxidations
of the solvent, but oxidation of a porphyrin ring on another
molecule is possible, and we note the conversions to oxo species
were relatively low, resulting in unreacted (porph)MnIII salts at
concentrations greater than those of the oxo derivatives. It is
also possible that (porph)Mn-oxo species decayed in part by
formation of the µ-oxo dimer.22 From results discussed below,
it is clear that rapid disproportionation and comproportionation
reactions were involved in the kinetics of the decay processes
in the absence of reactive substrates. Despite the complexity,
the kinetics of both (porph)MnV(O) and (porph)MnIV(O) decay
could be solved reasonably well for biexponential pseudo-first-
order reactions with a fast and a slow component. Both of these
processes were several orders of magnitude slower than the
decays of the excited states.
(17) A similar phenomenon was previously reported for porphyrin-iron-oxo
species. See: Nam, W.; Park, S.-E.; Lim, I. K.; Lim, M. H.; Hong, J.;
Kim, J. J. Am. Chem. Soc. 2003, 125, 14674-14675.
For the (porph)MnIV(O) species, the slow component of decay
was much less rapid than the fast component. As shown in
Figure 5, the fast reaction occurs on the millisecond time scale,
which is too fast to observe in rapid mixing studies. Importantly,
the rates of self-decay for (porph)MnIV(O) were found to be
dependent on the concentration of reactive manganese transients.
In these studies, the concentration of precursor salt in the LFP
(18) Suslick, K. S.; Watson, R. A. Inorg. Chem. 1991, 30, 912-919.
(19) For the isolation and/or characterization of porphyrin-manganese(IV)-
oxo complexes, see: Schappacher, M.; Weiss, R. Inorg. Chem. 1987, 26,
1189-1190; Czernuszewicz, R. S.; Su, Y. O.; Stern, M. K.; Macor, K. A.;
Kim, D.; Groves, J. T.; Spiro, T. G. J. Am. Chem. Soc. 1988, 110, 4158-
4165; Groves, J. T.; Stern, M. K. J. Am. Chem. Soc. 1988, 110, 8628-
8638; Rodgers, K. R.; Goff, H. M. J. Am. Chem. Soc. 1988, 110, 7049-
7060; Ayougou, K.; Bill, E.; Charnick, J. M.; Garner, C. D.; Mandon, D.;
Trautwein, A. X.; Weiss, R.; Winkler, H. Angew. Chem., Int. Ed. Engl.
1995, 34, 343-346; Arasasingham, R. D.; He, G. X.; Bruice, T. C. J. Am.
Chem. Soc. 1993, 115, 7985-7991.
(20) Hoshino, M.; Arai, S.; Yamaji, M.; Hama, Y. J. Phys. Chem. 1986, 90,
2109-2111.
(22) Schardt, B. C.; Smegal, J. A.; Hollander, F. J.; Hill, C. L. J. Am. Chem.
Soc. 1982, 104, 3964-3972. Nolte, R. J. M.; Razenberg, J. A. S. J.;
Schuurman, R. J. Am. Chem. Soc. 1986, 108, 2751-2752.
(21) Hoshino, M.; Nagashima, Y.; Seki, H.; De Leo, M.; Ford, P. C. Inorg.
Chem. 1998, 37, 2464-2469.
9
6576 J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005

Porphyrin−Manganese−Oxo Intermediates
A R T I C L E S
Table 4. Rate Constants for Porphyrin-Manganese-Oxo
experiments was varied, and the concentrations of oxo species
were determined from their absorbances. The observed fast
decay pseudo-first-order rate constants are plotted against
transient concentration in the log-log plots in Figure 5. The
slopes of the plots were ca. 0.8, indicating complex stoichiom-
etry in the self-decay processes; slopes of 1.0 would be obtained
if the reactions were clean second-order processes with respect
to the oxo species. The slow component of decay also increased
in rate with increasing concentrations of reactive transient (data
not shown). The high order for transient concentration in the
decay reactions most likely reflects rapid disproportionation
reactions of (porph)MnIV(O) to give reactive (porph)MnV(O)
and (porph)MnIII species and subsequent reaction of the (porph)-
MnV(O) species with solvent. Disproportionation of (porph)-
MnIV(O) to produce highly reactive (porph)MnV(O) was sug-
gested years ago,23 and it is implicit in the oxidation model
developed by Bruice and co-workers.13 Additional evidence for
such a reaction pathway in the oxidations of reactive substrates
is presented below.
The apparent rapid disproportionation of (porph)MnIV(O)
species and the concomitant rapid comproportionation of
(porph)MnV(O) with (porph)MnIII species that it implies creates
a conundrum. The photochemical conversions of (porph)MnIII-
(ClO4) to (porph)MnV(O) were relatively inefficient processes,
but we did not observe fast reactions of (porph)MnV(O) with
residual (porph)MnIII(ClO4) to give (porph)MnIV(O) species. If
comproportionation was fast, the MnIV-oxo derivatives would
have been formed.
Experiments wherein (TPFPP)MnV(O) was generated in the
presence of (TPFPP)MnIIICl resolved the apparent conflict. In
these studies, a solution of (TPFPP)MnIII(ClO4) in PhCF3 was
mixed with a PhCF3 solution of (TPFPP)MnIIICl in a stopped-
flow mixing unit such that the final concentration of both species
was 1 × 10-5 M, and following a 1 s delay, the mixture was
irradiated with 355 nm laser light. Instead of a relatively
persistent spectrum for the (TPFPP)MnV(O) species, we ob-
served rapid growth of the (TPFPP)MnIV(O) species (λ ) 418
nm) and rapid decay of (TPFPP)MnIIICl (λ ) 475 nm). The
pseudo-first-order rate constants for both processes were kobs
≈ 5 × 103 s-1, which gives a second-order rate constant for
reaction of (TPFPP)MnV(O) with (TPFPP)MnIIICl of k ≈ 5 ×
108 M-1 s-1. This rate constant is similar to the second-order
rate constant reported by Bruice and co-workers for the
comproportionation reaction of (TMP)MnV(O) with its (TMP)-
MnIIIX precursor.13
Thus, (TPFPP)MnV(O) does react rapidly with the chloride
salt of its precursor, but the comproportionation reaction of the
MnV-oxo species with the perchlorate salt is slower. This
apparently reflects the fact that perchlorate is a weak-binding
ligand, whereas chloride is a tight-binding ligand. Accordingly,
the (TPFPP)MnIII(ClO4) complex should have more positive
charge on manganese in comparison to the chloride salt, and
electron transfer from the MnIII perchlorate complex would
increase that charge. Support for this conclusion is provided
by comparison of the reduction potentials of (porph)MnIII(Cl)
and (porph)MnIII(ClO4). The reduction potentials of (TPP)MnIII-
(Cl) and (TPP)MnIII(ClO4) in CH3CN were reported by Kelly
and Kadish to be -0.23 and -0.19 V versus SCE, respec-
tively,24 and we found that the reduction potentials of (TPFPP)-
Reactionsa
porphyrin
substrate
(porph)MnV(O)
(porph)MnIV(O)b
ratioc
TPFPP
self-decay
self-decayd
cis-stilbene
Ph2CH2
Ph2CH2
PhEt
PhEt-d10
Ph3N
Ph3P
self-decay
cis-stilbene
Ph2CH2
Ph3N
(1.4 ( 0.1)E3
(1.6 ( 0.5)E2
(6.1 ( 0.3)E5
(1.37 ( 0.10)E5
(7.4 ( 0.6)E4
(1.28 ( 0.05)E5
(5.5 ( 0.5)E4
(6.0 ( 0.7)E-1
2,300
(2.4 ( 0.2)E2
2,500
d
(4.9 ( 0.5)E4
(1.5 ( 0.1)E4
(1.0 ( 0.15)
(6.2 ( 0.8)E2
TMPyP
TPP
(9.9 ( 1.6)E1
(4.3 ( 0.3)E4
(5.8 ( 0.1)E3
100
70
(1.3 ( 0.1)E5
(6.9 ( 0.4)E4
(1.4 ( 0.2)E1
(1.2 ( 0.1)E3
Ph3P
self-decay
cis-stilbene
Ph2CH2
Ph3N
(1.5 ( 0.1)E2
(1.1 ( 0.1)E4
<2 E3
11
9
(2.2 ( 0.3)E5
(1.1 ( 0.3)E5
Ph3P
a Second-order rate constants for reactions with substrates in CH3CN at
22 °C in units of M-1 s-1 unless noted. The self-decay rate constants are
pseudo-first-order rate constants for decay in units of s-1 for the major
decay process. b The values listed for (porph)MnIV(O) are apparent second-
order-rate constants; see text. c Ratio of rate constants for (porph)MnV(O)
to (porph)MnIV(O). d The solvent was PhCF3.
MnIII(Cl) and (TPFPP)MnIII(ClO4) in CH3CN are -0.01 and
+0.02 V versus SCE, respectively. The more positive potentials
for the perchlorate salts indicate that the manganese ions in these
salts bear more positive charge than the manganese ions in the
chloride salts.
We refer to the decay of (porph)Mn(O) species in the absence
of reactive substrates as “self-decay”. Table 4 contains a set of
self-decay rate constants for the major decay reactions deter-
mined when the concentration of precursor was 8 × 10-6
M
(perchlorate salts) or 1 × 10-5 M (chlorate salts). We note again
that the rates for self-decay of (porph)MnIV(O) species were
dependent on the concentrations of the oxo species. An
interesting result for (TPFPP)MnV(O) was the smaller rate
constant for self-decay in PhCF3 in comparison to that found
in CH3CN; we speculate that the major pathway for self-decay
was reaction with the solvent, and that CH3CN is more reactive
than PhCF3.
Kinetics of Porphyrin-Manganese(V)-Oxo Reactions.
Kinetic studies of reactions of (porph)MnV-oxo species with
substrates were performed with solutions containing the ap-
propriate precursor and varying concentrations of cis-stilbene
or arene substrates. At concentrations of up to 10-20 mM, the
substrates did not appear to react with (porph)MnIII(ClO4) for
up to an hour after mixing, as indicated by unchanging UV-
vis spectra. At higher concentrations of substrates, however,
the UV-visible spectrum of the perchlorate complex was altered
suggesting that the precursor reacted in unknown processes.
Therefore, studies of (porph)MnV(O) species were conducted
with <20 mM concentrations of substrates in most cases.
The kinetics of decay for the (porph)MnV(O) species in the
presence of substrates were simplified. The fast decay process
dominated (typically >90% of the total reaction), and this
process accelerated linearly as a function of substrate concentra-
tion (Figure 6). Second-order rate constants for reactions with
substrates were determined from eq 1, where kobs is the observed
(23) Groves, J. T.; Stern, M. K. J. Am. Chem. Soc. 1987, 109, 3812-3814.
(24) Kelly, S. L.; Kadish, K. M. Inorg. Chem. 1982, 21, 3631-3639.
9
J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005 6577

A R T I C L E S
Zhang et al.
Figure 7. (A) Kinetic traces at λmax for reactions of (TPFPP)MnIV(O) with
cis-stilbene (concentrations listed on graph) in CH3CN. (B) Observed
pseudo-first-order rate constants for decay of (TPFPP)MnIV(O) in the
presence of cis-stilbene.
Figure 6. Observed pseudo-first-order rate constants for reactions of
(TPFPP)MnV(O) with substrates in CH3CN.
Little kinetic data exists for comparison to the rate constants
for (porph)MnV(O) reactions determined in this work. Groves
and co-workers reported that (TMPyP)MnV(O) reacted with the
reactive substrate carbamazepine in water with a rate constant
of 6.5 × 105 M-1 s-1, which is consistent with the large rate
constants found in this work.8 Bruice and co-workers studied
epoxidation reactions of alkenes with the tetramesitylporphyrin
(TMP) system under catalytic conditions in water-saturated CH2-
Cl2 solutions.13 The rate constant for epoxidation of norbornene
fast rate constant, k0 is the intercept of the plot (background
reaction), kox is the second-order rate constant for oxidation of
substrate, and [sub] is the substrate concentration. Note that the
mechanistic details of the self-decay reactions are not important
because these processes were minor in comparison to reactions
with substrates. Second-order rate constants for reactions of
(porph)MnV(O) are listed in Table 4.
by (TMP)MnV(O) was about 2000 M-1 s-1 13
, which appears to
be consistent with our finding that (TPP)MnV(O) reacted with
cis-stilbene with a rate constant of 1.1 × 104 M-1 s-1. cis-
Stilbene is somewhat less reactive than norbornene in epoxi-
dation reactions,26 but the water in the system studied by Bruice
and co-workers should have stabilized the Mn-oxo species.
Kinetics of Porphyrin-Managanese(IV)-Oxo Reactions.
Kinetic studies of (porph)MnIV(O) included reactions with cis-
stilbene, triphenylamine, and triphenylphosphine. The chlorate
precursors were unstable in the presence of the highly reactive
substrates Ph3N and Ph3P, reacting in less than 1 min. Therefore,
a stopped-flow mixing unit was employed. Solutions of precur-
sor and substrate were mixed in the stopped-flow cell, and the
resulting mixture was irradiated by the laser after a delay of
less than 1 s.
The (porph)MnIV(O) species in the presence of substrates
decayed to give (porph)MnIII products, with no evidence for
formation of (porph)MnII species in any of our studies.27,28 The
observed rate constants for decay of (porph)MnIV(O) with the
substrates Ph3P, Ph3N, and cis-stilbene were fit reasonably well
by pseudo-first-order solutions. The rate constants increased as
a function of substrate concentration, and plots of kobs versus
substrate concentration were linear (Figure 7). Apparent second-
order rate constants for reactions of (porph)MnIV(O), solved by
eq 1, are collected in Table 4. As discussed below, however,
these values are not the actual rate constants for reactions of
the manganese(IV)-oxo species with substrates.
kobs ) k0 + kox[sub]
(1)
Each of the reactions of (porph)MnV(O) with the substrates
studied in this work appeared to be a two-electron process,
although the mechanistic details might be debated. The Soret
bands for (porph)MnV(O) decayed cleanly in the presence of
substrates with no indication of formation of (porph)MnIV(O).
This behavior requires that the velocities of comproportionation
reactions of (porph)MnV(O) with the (porph)MnIII products from
the oxidation of substrate were smaller than the velocities of
the reactions of (porph)MnV(O) with substrate. That conclusion
is consistent with the low concentrations of (porph)MnV(O)
formed in the photolysis reactions (<8 × 10-7 M) and the
second-order rate constant found for reaction of (TPFPP)MnV(O)
with (TPFPP)MnIIICl (k ≈ 5 × 108 M-1 s-1). At 50% conversion
of the (TPFPP)MnV(O) species, the pseudo-first-order rate
constant for its comproportionation reaction with the MnIII
products of the oxidation reaction would be <200 s-1, whereas
the observed pseudo-first-order rate constants for reactions with
substrates were >1000 s-1 for this system.
The second-order rate constants for (porph)MnV(O) reactions
with substrates follow the expected trends in reactivity.16 The
more electron-withdrawing TPFPP complex reacted faster with
a given substrate than the TMPyP complex, and the TPP
complex reacted least rapidly. Substrate reactivities also were
as expected, with cis-stilbene reacting faster than Ph2CH2 and
PhEt. The reaction of (TPFPP)MnV(O) with diphenylmethane
displayed an environment effect with reaction in the moderately
polar solvent CH3CN about twice as fast as reaction in the less
polar solvent PhCF3; this result indicates that the transition state
for the oxidation reaction is somewhat more polarized than the
reactant ground state. The rate constants for oxidation of PhEt
and PhEt-d10 by (TPFPP)MnV(O) give a kinetic isotope effect
of kH/kD ) 2.3, indicating that the transition state for the
oxidation is early, as would be expected for a highly reactive
oxidant.25
A striking feature of the apparent rate constants for reactions
of (porph)MnIV(O) is that they display an inverted reactivity
(25) A similar KIE (kH/kD ) 2.7) was reported for oxidation of PhEt and PhEt-
d10 in a competitive oxidation by (TMP)MnV(O). See: Campestrini, S.;
Cagnina, A. J. Mol. Catal. A 1999, 150, 77-86.
(26) Liu, C. J.; Yu, W. Y.; Che, C. M.; Yeung, C. H. J. Org. Chem. 1999, 64,
7365-7374.
(27) UV-visible spectra of authentic (porph)MnII species were obtained in
photoinduced electron-transfer reactions by excitation of benzophenone in
the presence of (porph)MnIIICl.28 No evidence for formation of these species
was found in the reactions of (porph)MnIV(O).
(28) Hoshino, M.; Shizuka, H. In Photoinduced Electron Transfer; Fox, M. A.,
Chanon, M., Eds.; Elsevier: Amsterdam, 1988; pp 313-371.
9
6578 J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005

Porphyrin−Manganese−Oxo Intermediates
A R T I C L E S
Scheme 4
(O) are “apparent” rate constants and not true rate constants
for reactions with the substrates.
Figure 8. (A) Time-resolved spectrum for self-decay of (TPP)MnIV(O) in
CH3CN over 5 s. (B) Time-resolved spectrum for decay of (TPP)MnIV(O)
in CH3CN containing 4 mM Ph3P over 20 ms; the inset shows kinetic traces
at 425 and 435 nm for the first 4 ms of the reaction.
Scheme 3 is complicated because the identity of (porph)-
MnIIIX that participates in the equilibrium is not known; multiple
counterions are present in the photolysis reactions, and X could
be hydroxide. Nonetheless, one can determine apparent equi-
librium constants that are mechanistically instructive and serve
as vehicles for calculating rate constants. We assume that the
self-decay reactions and reactions with cis-stilbene involved
oxidations by only the (porph)MnV(O) species. Thus, the
concentration of the (porph)MnV(O) species relative to the
(porph)MnIV(O) species is given by the inverse of the ratio of
rate constants listed in Table 4. The photolysis reactions
producing the (porph)MnIV(O) species in CH3CN were con-
ducted with 1 × 10-5 M chlorate salt, and they resulted in 17%,
6%, and 3% conversion to the oxo species for the TPFPP,
TMPyP, and TPP systems, respectively; therefore, for example,
the concentration of (TPFPP)MnV(O) in equilibrium with the
(TPFPP)MnIV(O) would be 7 × 10-10 M.
Scheme 3
order in comparison to those found for (porph)MnV(O) oxida-
tions as well as those expected based on the electron-demand
of the porphyrin ring.16 Thus, for each substrate studied, (TPP)-
MnIV(O) was apparently the most reactive oxidant, and (TPFPP)-
MnIV(O) was apparently the least reactive. To the best of our
knowledge, this surprising trend in reactivity has not been
observed previously, a fact that apparently is related to the
general lack of kinetic information available for transition-
metal-oxo reactions.
We further assume that the identity of X and the concentration
of HX are the same for all three porphyrin systems, and we
then calculate apparent equilibrium constants that ignore the
concentration of HX and X. The resulting equilibrium constants
Another unusual feature was observed in the behavior of the
Soret band for (TPP)MnIV(O) in various reactions. As shown
in Figure 8, the λmax value for the Soret band in this system
was different during the self-decay reaction and during the
oxidation of Ph3P. Moreover, the λmax value shifted during the
self-decay reaction. Both features suggest that multiple man-
ganese species were present at appreciable concentrations.
Remarkably, in the first millisecond of reaction of (TPP)MnIV-
(O) with Ph3P, signal growth was observed at λ ) 435 nm,
which is λmax for the (TPP)MnV(O) species. Similar signal
growth at λ ) 435 nm was observed when (TPP)MnIV(O)
reacted with other substrates, but this behavior was not observed
in the reactions of the other (porph)MnIV(O) derivatives.
Taken together, the absence of (porph)MnII products from
(porph)MnIV(O) reactions, the reactivity order for (porph)MnIV-
(O), and the unusual spectral behavior observed for reactions
of (TPP)MnIV(O) indicate that reactions of (porph)MnIV(O)
involved rapid disproportionation to give (porph)MnV(O) and
a (porph)MnIIIX species, and that (porph)MnV(O) was a major
oxidant in the reactions (Scheme 3). In fact, because the ratios
of rate constants for (porph)MnV(O) and (porph)MnIV(O) for a
given porphyrin were effectively the same in the self-decay and
cis-stilbene reactions (see Table 4), the (porph)MnV(O) species
appears to be the only active oxidant in these reactions. With
the more reactive substrates Ph3N and Ph3P, it is possible that
both (porph)MnV(O) and (porph)MnIV(O) acted as oxidants, but
the inverted reactivity patterns for the oxidants found with these
substrates suggest that (porph)MnV(O) was still a major oxidant.
Thus, the experimental rate constants for decay of (porph)MnIV-
are Kdis ) 0.002 for TPFPP, Kdis ) 0.2 for TMPyP, and Kdis
)
3 for TPP. The order of these apparent Kdis values illustrates
that the apparent rate constants for oxidations by (porph)MnIV-
(O) do follow a predictable trend. Specifically, increasing
electron demand in the porphyrin ring disfavors formation of
the active oxidant (porph)MnV(O) species, resulting in slower
apparent oxidations by (porph)MnIV(O) species that have more
electron-withdrawing porphyrin rings. The apparent Kdis for the
TPFPP system is by far the smallest value among the three
systems studied, which is consistent with the expected desta-
bilization of the cationic MnV-oxo species by the electron-
withdrawing porphyrin ring of (TPFPP).16
Disproportionation of (porph)MnIV(O) must be very fast if
the oxidation reactions are explained by Scheme 3. In fact, the
disproportionation reaction of (TPP)MnIV(O) apparently occurs
with nearly a diffusion-controlled rate constant. The growth of
the signal for the (TPP)MnV(O) species seen in reactions of
(TPP)MnIV(O) with substrates (cf. the reaction with Ph3P shown
in Figure 8B) had a pseudo-first-order rate constant of kobs ≈ 1
× 103 s-1. Given that the concentration of (TPP)MnIV(O) formed
in our reactions was ca. 3 × 10-7 M, this corresponds to a
second-order rate constant for disproportionation of k ≈ 3 ×
109 M-1 s-1. This value permits an estimation of the rate
constant for the disproportionation equilibrium for the (TPP)
system as shown in Scheme 4. Here we have used the apparent
K
dis value for the (TPP) system as a vehicle for calculating the
rate constant of the comproportionation reaction, and the result
is expected to be reasonably accurate due to cancellation of
errors. In a similar manner, the observed rate constant for
9
J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005 6579

A R T I C L E S
Zhang et al.
reaction of (TPFPP)MnV(O) with (TPFPP)MnIIICl permits a
calculation of the rate constants for the equilibration of the
(TPFPP) system, where we assume that the comproportionation
reaction has the same rate constant in CH3CN as found in PhCF3.
The comproportionation rate constants are quite similar for the
two systems and also similar to the value deduced by Bruice
and co-workers for the comproportionation of (TMP)MnV(O)
with (TMP)MnIIIX.13
Table 5. Relative Rate Constants for Porphyrin-Manganese-Oxo
Reactionsa
porphyrin
substratesb
oxidant
solvent
krel
TPFPP
stilbene/ Ph2CH2
LFP results
PhIO
PhIO
mcpba
mcpba
LFP results
PhIO
mcpba
LFP results
PhIO
mcpba
CH3CN
CH3CN
CDCl3
CH3CN
CDCl3
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
CH3CN
4.7
4.6
5.0
6.5
6.7
1.1
1.0
1.2
2.3
2.1
2.6
7.5
8.8
TPFPP
TPFPP
TMPyP
Ph2CH2/PhEt
The values in Scheme 4 apply to CH3CN solvent, which is
moderately polar.29 One expects that the equilibrium constants
for disproportionation of (porph)MnIV(O) will be smaller in less
polar solvents such as CH2Cl2, but the magnitudes of the rate
constants are expected to remain quite large. The kinetic values
illustrate that the velocities of disprotionation and compropor-
tionation reactions can readily exceed the velocities of substrate
oxidations, as we observed for the oxidations of substrates with
(TPP)MnIV(O).
PhEt/ PhEt-d10
stilbene/ Ph2CH2
LFP results
mcpba
a Relative rate constants from LFP results with (porph)MnV(O) and for
competitive oxidations with catalytic (porph)MnIIICl at ambient temperature.
b The more rapidly reacting substrate is listed first.
Comparisons to and Implications for Catalytic Oxidations.
The active (porph)MnV(O) oxidants produced in the LFP studies
are not identical to those formed under catalytic conditions with
a sacrificial oxidant such as mcpba or iodosobenzene. In the
LFP experiments with perchlorate salts, heterolytic cleavage of
the O-Cl bond in the ligand gives a (porph)MnV(O) species
with +1 charge. At the instant of formation, this cationic species
will be ligated by solvent, but the decay reactions are slow
enough such that the byproduct chlorate anion and perchlorate
from unreacted precursor can ligate the metal at a diffusion-
controlled rate to give (porph)MnV(O)(ClO3(4)). In catalytic
oxidations, on the other hand, (porph)MnIIICl typically is
allowed to react with a sacrificial oxidant, and because chloride
is a tight-binding ligand, the oxidant is likely (porph)MnV(O)-
(Cl). Counterion effects are known in porphyrin-metal oxo
chemistry,30 and it is possible that they will be displayed in
oxidations by manganese(V)-oxo species. The (porph)MnIV-
(O) species are neutral, and the same species will be formed in
the LFP and catalytic reactions.
Scheme 5
The agreement in the krel values from the LFP studies and
those determined from the consumption of the reagents under
catalytic oxidation conditions in CH3CN is good. There is some
evidence that catalytic oxidations with porphyrin-metal-oxo
species involve reactions of both the metal-oxo species and
complexes of the metal salt with the sacrificial oxidant,31 which
might explain the small differences in relative rate constants
for the LFP and catalytic reactions, or the differences might be
due to counterion effects.30 In any case, it is clear that the
reactivity patterns determined from the LFP studies of (TPFPP)-
MnV(O) were reproduced in the catalytic reactions. The similar-
ity in relative rate constants suggests that (TPFPP)MnV(O)
species were the active oxidants in the catalytic reactions, as
commonly assumed, and demonstrates that the LFP results can
be used in a predictive manner for estimating the relative
reactivities of substrates under catalytic conditions. Conversely,
the absolute rate constant for oxidation of substrate A could be
estimated from the results of a catalytic oxidation reaction where
substrates A and B competed and substrate B was one of those
calibrated in this work.
Practical (porph)MnV(O) oxidations are conducted under
catalytic conditions, of course, and the rate constants for
oxidation of substrates will be convoluted with the equilibrium
constant for disproportionation of (porph)MnIV(O) and the rate
constant for oxidation of the (porph)MnIII salt by the sacrificial
oxidant. Scheme 5 illustrates the mechanistic picture, where we
have used the kinetic values for the (TPFPP)MnV(O) oxidation
of cis-stilbene in CH3CN. The equilibrium between manganese
species can serve to sequester much of the potential catalyst as
unreactive (porph)MnIV(O), and an ironic conclusion from our
results is that attempts to accelerate oxidation reactions by using
One cannot readily evaluate the absolute rate constants for
reactions under catalytic oxidation conditions (see discussion
below), but a series of competitive catalytic oxidations showed
that the relative rate constants for oxidations found in the LFP
studies are similar to those in catalytic systems. Table 5 contains
the results of competition reactions where mcpba and io-
dosobenzene (PhIO) were employed as sacrificial oxidants with
catalytic (porph)MnIIICl. In these reactions, we used a limiting
amount of sacrificial oxidant and allowed the reactions to
proceed essentially to completion. The amounts of substrates
consumed were determined, as opposed to the amounts of
oxidation products formed; this method reduces potential
problems in product analysis due to secondary oxidations and
multiple oxidation products. A control reaction demonstrated
that mcpba in the absence of manganese salt did not epoxidize
cis-stilbene appreciably in the time frame of our studies (<5%
yield relative to the yield in the catalytic reaction). In practice,
the mcpba reactions were complete in about 10 min, whereas
the PhIO reactions required about 1 day.
(29) Reichardt, C. Chem. ReV. 1994, 94, 2319-2358.
(30) For leading references to counterion effects, see: Gross, Z.; Nimri, S. Inorg.
Chem. 1994, 33, 1731-1732; Nam, W.; Jin, S. W.; Lim, M. H.; Ryu, J.
Y.; Kim, C. Inorg. Chem. 2002, 41, 3647-3652; Lai, T.-S.; Lee, S. K. S.;
Yeung, L.-L.; Liu, H.-Y.; Williams, I. D.; Chang, C. K. Chem. Commun.
2003, 620-621.
(31) For example, see: Adam, W.; Roschmann, K. J.; Saha-Moller, C. R.;
Seebach, D. J. Am. Chem. Soc. 2002, 124, 5068-5073; Collman, J. P.;
Zeng, L.; Decreau, R. A. Chem. Commun. 2003, 2974-2975.
9
6580 J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005

Porphyrin−Manganese−Oxo Intermediates
A R T I C L E S
(TPFPP)MnIIIX catalysts might have just the opposite effect
because the (TPFPP)MnV(O) species is highly disfavored in the
equilibrium.
intermediates in organic solvents and direct kinetic studies of
their reactions with typical organic substrates. Apparent rate
constants for reactions of porphyrin-manganese(IV)-oxo
derivatives, produced photochemically from the corresponding
manganese(III) chlorate salts, indicate that these intermediates
react via disproportionation to give the (porph)MnV(O) species
that is a major, or possibly the only, active oxidant in the system.
The rate constant for disproportionation of (TPP)MnIV(O) and
the rate constant for comproportionation of (TPFPP)MnV(O) and
(TPFPP)MnIIICl found in this work confirm that, in catalytic
processes, equilibrations of the various manganese species
typically will be faster than oxidations of substrates. The
apparent equilibrium constants for the disproportionation reac-
tions of (porph)MnIV(O) are important in determining the
concentrations of active oxidant, and the values in CH3CN
determined here indicate that little of the highly reactive
(TPFPP)MnV(O) species will be present in a catalytic reaction.
In such a case, the observed rate constants for formation of
oxidized products can be much smaller than the true rate
constants for reactions of the active oxidant with substrates. Our
work provides fundamental kinetic information for reactions of
porphyrin-manganese-oxo intermediates when limited infor-
mation was previously available, and it implicates complex
reaction mechanisms with kinetically important disproportion-
ation equilibria. Additional kinetic information in regard to
concentration effects of the oxo species and solvent effects,
especially water effects, will be important for complete mecha-
nistic descriptions of manganese-oxo reactions.
An unknown rate constant in Scheme 5 is that for oxidation
of the (porph)MnIII salt by the sacrificial oxidant X-O, but that
value might be determined if a calibrated substrate such as cis-
stilbene was used. Alternatively, simulations of the entire
reaction scheme might be considered in order to obtain the rate
constants for all processes. Such a procedure was reported by
Bruice and co-workers,13 but the method is difficult and prone
to systematic errors, especially if the porphyrin complex is
unstable over the course of the reaction.
Although complicated by the equilibrium with the (porph)-
MnIV(O) species, the reaction sequence in Scheme 5 is a chain
reaction in terms of the MnV(O) and MnIII species, and therefore,
the velocity of the oxidation of MnIII species to MnV(O) species
will be equal to the velocity of the reaction of the MnV(O)
species with substrate. Therefore, the rate constant for the MnIII
oxidation reaction will appear in the rate law for formation of
oxidized substrate when that oxidation is slow enough such that
the MnV-oxo species does not accumulate substantially. This
is illustrated qualitatively in our catalytic studies. From the ratio
of rate constants in Table 5, it seems apparent that (TPFPP)-
MnV(O) was the major oxidant in reactions with both mcpba
and PhIO as the sacrificial oxidants, but oxidations using mcpba
were about 2 orders of magnitude faster than oxidations using
PhIO in terms of rates of product formation. This observation
indicates that mcpba oxidized the manganese(III) species about
3-4 orders of magnitude faster than did PhIO. The slow
oxidation by PhIO undoubtedly reflects the low solubility of
this oxidant in organic solvents, resulting in rate-limiting
dissolution of PhIO.32
In the absence of reactive substrates, the oxidation of (porph)-
MnIIIX might eventually drive the system to complete formation
of the (porph)MnV(O) species, even if it is not the major species
present under catalytic conditions. Alternatively, the oxidation
of solvent by (porph)MnV(O) might be fast enough to prevent
accumulation of the highly reactive transient, as we found for
the mcpba reaction with (TPFPP)MnIIICl in acetonitrile. In any
event, it is clear that spectroscopic information obtained under
a special set of conditions might not be useful for determining
the identity or concentration of the active catalyst under fast
turnover conditions. Even in so-called “stoichiometric” oxida-
tions, wherein one generates a (porph)Mn-oxo derivative and
then adds one equivalent of substrate, one does not know if the
observed oxo species was the actual oxidant.
The essence of the reaction sequence shown in Scheme 5
likely applies to many, if not most, catalytic oxidation processes
that involve high-valent transition metal-oxo derivatives.
Simply stated, detection of a particular high valent metal-oxo
derivative under selected conditions will not necessarily dem-
onstrate the identity of the active oxidant under either catalytic
turnover or “stoichiometric” conditions. This fact has not been
appreciated in some studies of transition metal-oxo intermedi-
ates.
Experimental Section
Materials. Acetonitrile was obtained from Fisher Scientific and
distilled over P2O5 prior to use. m-CPBA (77%) purchased from Aldrich
Chemical Co. was purified by crystallization in methylene chloride and
then dried in a vacuum. Iodosobenzene was purchased from the TCI
America Co. and used as obtained. All reactive substrates for LPF
kinetic studies and catalytic competition oxidations were the best
available purity from Aldrich Chemical Co. and were passed through
a dry column of active alumina (Grade I) before use. (TPFPP)MnIII-
(Cl) was prepared according to the literature procedure33 and purified
by chromatography on silica gel. (TPP)MnIII(Cl) and (TMPyP)MnIII-
(Cl) were obtained from Aldrich Chemical Co. and Mid-Century
Chemicals, respectively, and used without further purification. All
perchlorate salts, chlorate salts, and nitrate salts of manganese(III)
porphyrin complexes were prepared by stirring equimolar amounts of
(porph)MnIII(Cl) with AgClO4, AgClO3, and AgNO3, respectively. The
resulting mixtures were filtered and used for LFP studies immediately
after preparation.
Caution! Perchlorate salts of metal complexes are potentially
explosive and should be handled with care.
Instrumentation. UV-vis spectra were recorded on an Agilent 8453
spectrophotometer. Laser flash photolysis studies were conducted on
an LK-60 kinetic spectrometer (Applied Photophysics) at ambient
temperature (ca. 22 ( 2 °C). Stock solutions of porphyrin manganese-
(III) perchlorate, chlorate, or nitrate complexes with concentrations in
the range of 8-10 × 10-6 M were sparged with helium and then
allowed to flow through a 1 × 1 cm quartz cell. Samples were irradiated
with 355 nm light from a Nd:YAG laser (ca. 7 ns pulse). Data were
acquired and analyzed with the Applied Photophysics software.
Oversampling (64:1) was employed in all cases to improve the signal-
Conclusion
(32) The rate constants for formation of (TMPyP)MnV(O) from the MnIII
precursor in water were reported to be as follows: mcpba, 2.7 × 107 M-1
Laser flash photolysis production of porphyrin-manganese-
(V)-oxo species from the corresponding manganese(III) per-
chlorate salts has permitted detection of these highly reactive
s-1; HSO5 anion, 6.9 × 105 M-1 s-1; ClO anion, 6.3 × 105 M-1 s-1 8
.
(33) Adler, A. D.; Longo, F. R.; Kampas, F.; Kim, J. J. Inorg. Nucl. Chem.
1970, 32, 2443-2445.
9
J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005 6581

A R T I C L E S
Zhang et al.
to-noise. For reactions of (porph)MnIV(O) with substrates, an SC-18MV
stopped-flow mixing unit affixed to the kinetic spectrometer was
employed.
constants for reactions with substrates were determined from the
observed slow decay rate constants via eq 1. The results in Table 4 are
averages of 2-3 runs, and the errors in the rate constants are 1σ.
For generation of (porph)MnV(O) in mixing studies, the SC-18MV
stopped-flow unit was employed, and data was acquired in the
millisecond to second time ranges; solutions of (porph)MnIIICl com-
plexes were mixed with solutions containing excess mcpba (ca. 3 equiv).
For production of (porph)MnIV(O) intermediates in mixing studies, an
RX2000 rapid kinetics spectrometer accessory (Applied Photophysics)
coupled with the above UV spectrometer was employed with 400 ms
to seconds time scales; solutions of (porph)MnIIICl complexes were
treated with excess mcpba in the presence of the reductant Ph3N.
Quantum Yields were determined relative to the yields of a standard
by the general method of Hoshino.20 Solutions of the manganese-oxo
precursor with absorbances of 0.5 at 355 nm were irradiated with the
third harmonic of the Nd:YAG laser (355 nm) at 30 mJ of power per
pulse, and the absorbance increase at λmax of the Soret band for the
oxo transient was measured. We assume that no bleaching occurred at
λmax of the oxo product. The molar yields of oxo derivative were
determined using the extinction coefficients listed in Table 2. These
yields were compared to the yield of benzophenone triplet formed by
355 nm irradiation, where the standard solution had an absorbance of
0.5 at 355 nm. In this method, the quantum yield for excitation of
benzophenone is taken to be 1.0.21 The quantum yields for manganese-
oxo formation were calculated from the ratio of molar yields and ratio
of molar extinction coefficients for the precursor of interest and the
standard.
Cyclic Voltammetry was performed with a CHI 900 potentiostat
in a three-electrode, single-compartment cell using acetonitrile as solvent
(5 mL). The working electrode was a platinum disk, the counter
electrode was platinum wire, and the reference electrode was silver
wire. Potentials were internally referenced to the Fc/Fc+ couple and
converted to a saturated calomel electrode (SCE) reference. The
solutions were purged with nitrogen prior to use. The supporting
electrolyte was 0.1 M (Bu4N)(PF6) (TBAP, Fluka), which was
recrystallized twice from ethanol/water and dried under high vacuum.
Catalytic Competitive Oxidations. A solution containing equal
amounts of two substrates, e.g., cis-stilbene (0.5 mmol) and diphenyl-
methane (0.5 mmol), manganese(III) porphyrin catalyst (1 µmol), and
an internal standard of ethyl benzoate or bromobenzene (0.5 mmol)
was prepared (volume ) 5 mL). The standards were shown to be stable
to the oxidation conditions in control reactions. PhIO or mcbpa (0.2
mmol) was added, and the mixture was stirred under an inert atmosphere
at ambient temperature (ca. 22 °C) until the reaction was complete.
The amounts of substrates before and after the reactions were
determined by GC (FID, Carbowax or DB-5) for reactions in CH3CN
1
and by both GC and H NMR spectroscopy for reactions in CDCl3.
Relative rate constants for oxidations were determined based on the
average concentrations of the substrates, which introduces a minor error
(<5%).34 The values reported in Table 5 are the averages of 2 or 3
runs.
LFP Kinetics. The decay rates of (porph)MnV(O) intermediates in
the presence of reactive substrates were directly measured at λmax. The
kinetic profiles exhibited a double exponential process, and the fast
process represented >90% of the total decay. The fast process
accelerated as a function of substrate concentration, and the slow process
was not altered by the addition of substrates. The rate constants for
reactions with substrates were determined from the observed fast decay
rate constants via eq 1. For studies of (porph)MnIV(O), solutions of
precursor were mixed with solutions of substrate, and the resulting
mixture was irradiated by the laser after a delay of < 1 s. For (porph)-
MnIV(O) species, the slow decay reaction was the major process, and
this reaction accelerated as a function of substrate concentration; rate
Acknowledgment. This work was supported by a grant from
the National Institutes of Heath (GM-48722). We thank Ms. J.
Zhang for assistance with the CV measurements.
Supporting Information Available: UV-visible spectra of
porphyrin-manganese-oxo species. This material is available
free of charge via the Internet at http://pubs.acs.org.
JA045042S
(34) Newcomb, M. Tetrahedron 1993, 49, 1151-1176.
9
6582 J. AM. CHEM. SOC. VOL. 127, NO. 18, 2005