Hydrogen atom transfer reactions of imido manganese(V) corroie: One reaction with two mechanistic pathways

Article abstract of DOI:10.1021/ja073027s

Hydrogen atom transfer (HAT) reactions of (tpfc)MnNTs have been investigated (tpfc = 5,10,-15-tris(pentafluorophenyl)corrole and Ts = p-toluenesulfonate). 9,10-Dihydroanthracene and 1,4-dihydrobenzene reduce (tpfc)MnNTs via HAT with second-order rate constants 0.16 ± 0.03 and 0.17 ± 0.01 M^-1 s^-1, respectively, at 22°C. The products are the respective arenes, TsNH₂ and (tpfc)Mn^III. Conversion of (tpfc)MnNTs to (tpfc)Mn by reaction with dihydroanthracene exhibits isosbestic behavior, and formation of 9,9′,10,10′- tetrahydrobianthracene is not observed, suggesting that the intermediate anthracene radical rebounds in a second fast step without accumulation of a Mn^IV intermediate. The imido complex (tpfc)-Mn^VNTs abstracts a hydrogen atom from phenols as well. For example, 2,6-di-tert-butyl phenol is oxidized to the corresponding phenoxyl radical with a second-order rate constant of 0.32 ± 0.02 M^-1 s^-1 at 22°C. The other products from imido manganese(V) are TsNH₂ and the trivalent manganese corrole. Unlike reaction with dihydroarenes, when phenols are used isosbestic behavior is not observed, and formation of (tpfc)-Mn ^IV(NHTs) is confirmed by EPR spectroscopy. A Hammett plot for various p-substituted 2,6-di-tert-butyl phenols yields a V-shaped dependence on σ, with electron-donating substituents exhibiting the expected negative ρ while electron-withdrawing substituents fall above the linear fit (i.e., positive ρ). Similarly, a bond dissociation enthalpy (BDE) correlation places electron-withdrawing substituents above the well-defined negative slope found for the electron-donating substituents. Thus two mechanisms are established for HAT reactions in this system, namely, concerted proton - electron transfer and proton-gated electron transfer in which proton transfer is followed by electron transfer.

Full text of DOI:10.1021/ja073027s

Published on Web 08/25/2007
Hydrogen Atom Transfer Reactions of Imido Manganese(V)
Corrole: One Reaction with Two Mechanistic Pathways
Michael J. Zdilla, Jennifer L. Dexheimer, and Mahdi M. Abu-Omar*
Contribution from the Brown Laboratory, Department of Chemistry, Purdue UniVersity,
560 OVal DriVe, West Lafayette, Indiana 47907
Received April 30, 2007; E-mail: mabuomar@purdue.edu
Abstract: Hydrogen atom transfer (HAT) reactions of (tpfc)MnNTs have been investigated (tpfc ) 5,10,-
15-tris(pentafluorophenyl)corrole and Ts ) p-toluenesulfonate). 9,10-Dihydroanthracene and 1,4-dihy-
drobenzene reduce (tpfc)MnNTs via HAT with second-order rate constants 0.16 ( 0.03 and 0.17 ( 0.01
M-1 s-1, respectively, at 22 °C. The products are the respective arenes, TsNH2 and (tpfc)MnIII. Conversion
of (tpfc)MnNTs to (tpfc)Mn by reaction with dihydroanthracene exhibits isosbestic behavior, and formation
of 9,9′,10,10′-tetrahydrobianthracene is not observed, suggesting that the intermediate anthracene radical
rebounds in a second fast step without accumulation of a MnIV intermediate. The imido complex (tpfc)-
MnVNTs abstracts a hydrogen atom from phenols as well. For example, 2,6-di-tert-butyl phenol is oxidized
to the corresponding phenoxyl radical with a second-order rate constant of 0.32 ( 0.02 M-1 s-1 at 22 °C.
The other products from imido manganese(V) are TsNH2 and the trivalent manganese corrole. Unlike reaction
with dihydroarenes, when phenols are used isosbestic behavior is not observed, and formation of (tpfc)-
MnIV(NHTs) is confirmed by EPR spectroscopy. A Hammett plot for various p-substituted 2,6-di-tert-butyl
phenols yields a V-shaped dependence on σ, with electron-donating substituents exhibiting the expected
negative F while electron-withdrawing substituents fall above the linear fit (i.e., positive F). Similarly, a bond
dissociation enthalpy (BDE) correlation places electron-withdrawing substituents above the well-defined
negative slope found for the electron-donating substituents. Thus two mechanisms are established for
HAT reactions in this system, namely, concerted proton-electron transfer and proton-gated electron transfer
in which proton transfer is followed by electron transfer.
or peroxides.3,4 Traditionally, compound I has been assumed
to be the active oxidant in various reaction types that have been
Introduction
Metal mediated hydrogen atom transfer (HAT) reactions
represent the initial step in many industrially and biologically
important processes.1 Industrially, the Amoco MC oxygenation
process as practiced by BP hinges on metal-mediated HAT in
the conversion of ∼3 billion pounds of p-xylene per year to
terephthalic acid (PTA), which is used in the synthesis of
polyethylene terephthalate (PET).2 In biology, oxygen insertion
reactions facilitated by biological oxygenases such as cyto-
chrome P450 rely on an initial HAT step according to the radical
rebound mechanism.3 In these systems, the active oxidant,
mechanism of HAT, and subsequent fate of the resulting organic
radical and metal intermediates are of fundamental interest to
synthetic and biological chemists.
observed for P450, such as oxygen insertion, epoxidation,
dealkylation, and deamination. However, recent results from
enzyme and model studies have pointed to the possibility of
multiple oxidative species (Scheme 1).4-7
Enzymatically, support for the multiple oxidant hypothesis
has been found in drastically varied product types and distribu-
tions resulting from protein mutations.5 For model complexes,
(4) (a) Lee, K. A.; Nam, W. J. Am. Chem. Soc. 1997, 119, 1916. (b) Primus,
J.-L.; Boersma, M. G.; Mandon, D.; Boeren, S.; Veeger, C.; Weiss, R.;
Rietjens, I. M. C. M. J. Biol. Inorg. Chem. 1999, 4, 274. (c) Lee, Y. J.;
Goh, Y. M.; Han, S.-Y.; Kim, C.; Nam, W. Chem. Lett. 1998, 837. (d)
Kamaraj, K.; Bandyopadhyay, D. J. Am. Chem. Soc. 1997, 119, 8099. (e)
Pratt, J. M.; Ridd, T. I.; King, L. J. J. Chem. Soc., Chem. Commun. 1995,
2297. (f) Machii, K.; Watanabe, Y.; Morishima, I. J. Am. Chem. Soc. 1995,
117, 6691. (g) Watanabe, Y.; Yamaguchi, K.; Morishima, I.; Takehira, K.;
Shimizu, M.; Hayakawa, T.; Orita, H. Inorg. Chem. 1991, 30, 2581.
(5) (a) Coon, M. J.; Vaz, A. D. N.; McGinnity, D. F.; Peng, H. M. Drug Metab.
Dispos. 1998, 26, 1190. (b) Vaz, A. D. N.; McGinnity, D. F.; Coon, M. J.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 3555. (c) Toy, P. H.; Newcomb,
M.; Coon, M. J.; Vaz, A. D. N. J. Am. Chem. Soc. 1998, 120, 9718. (d)
Vaz, A. D. N.; Pernecky, S. J.; Raner, G. M.; Coon, M. J. Proc. Natl.
Acad. Sci. U.S.A. 1996, 93, 4644. (e) Toy, P. H.; Dhanabalasingam, B.;
Newcomb, M.; Hanna, I. H.; Hollenberg, P. F. J. Org. Chem. 1997, 62,
9114. (f) Pratt, J. M.; Ridd, T. I.; King, L. J. J. Chem. Soc., Chem. Commun.
1995, 2297.
For years, the consensus mechanism for cytochrome P450
and its model complexes has been the formation of a terminal
oxoiron(IV) porphyrin radical cation (compound I) from the
reaction of reduced Fe-poryphyrin with an oxidant such as O2
(1) (a) Mayer, J. Acc. Chem. Res. 1998, 31, 441. (b) Meunier, B.; de Visser,
S. P.; Shaik, S. Chem. ReV. 2004, 104, 3947.
(2) (a) Partenheimer, W. Catal. Today 1995, 23, 69. (b) Koshino, N.; Saha,
B.; Espenson, J. H. J. Org. Chem. 2003, 68, 9364. (c) Saha, B.; Koshino,
N.; Espenson, J. H. J. Phys. Chem. A 2004, 108, 425.
(3) Groves, J. T.; Han, Y. Z. In Cytochrome P450: Structure Mechanisms
and Biochemistry; Ortiz de Monellano, P. R., Ed.; Plenum: New York,
1995; Chapter 1, pp 3-48.
(6) (a) Nam, W.; Lim, M. H.; Moon, S. K.; Kim, C. J. Am. Chem. Soc. 2000,
122, 10805. (b) Collman, J. P.; Zeng, L.; Decre´au, R. A. Chem. Commun.
2003, 2974.
(7) Song, W. J.; Ryu, Y. O.; Song, R.; Nam, W. J. Biol. Inorg. Chem. 2005,
10, 294.
9
10.1021/ja073027s CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 11505-11511
11505

A R T I C L E S
Zdilla et al.
observed by Warren and co-workers using nickel imidos16 and
by Theopold and co-workers with cobalt imidos.17 In an
interesting twist, an example of the reverse reaction has recently
been reported by Smith and co-workers, in which the synthesis
of a cobalt imido species is achieved by reaction of cobalt amido
with a phenoxyl radical.18 Also worthy of mention is the recent
work of Peters and Chirik, who have activated H2 with iron
imido compounds.19
Scheme 1. Two Possible Pathways for Cytochrome P450
Oxidations: Rebound Mechanism through Compound I (Pathway
1) and Reaction of Substrate with Oxidant Adduct (Pathway 2)
Herein, we report on hydrogen atom transfer reactions of an
imidomanganese(V) corrole complex. The reaction occurs via
two subsequent HAT reactions, the second of which is fast. We
have found that the initial rate-determining HAT reaction can
occur by two different pathways: a concerted proton-electron
transfer or a proton transfer followed by electron transfer
(PT-ET).
differences in the identity of oxidant,4f,4e,6 solvent,4f,6aand
coordination sphere7 have been shown to affect reactivity. In
the case of iron porphyrins, recent results by Collman, Nam,
and Que have pointed to an Fe(III)-oxidant adduct as the active
species.8 Furthermore, Goldberg and co-workers have found
compelling evidence that the oxidant is an adduct of MnVdO
and PhIdO in expoxidations and sulfoxidations catalyzed by
manganese corrolazine.9
Results and Discussion
Reactivity and Kinetics of (tpfc)MnNTs toward HAT
Substrates. The terminal tosylimido complex of manganese
corrole is formed by the reaction of (tpfc)Mn (1) with ArINTs
(Ar ) 2-tert-butylsulfonylbenzene), eq 1. In addition to its role
as an aziridination catalyst, 2 reacts via HAT with solvents that
possess a thermodynamically suitable hydrogen atom, such as
toluene.10 To further explore this reactivity of 2, we examined
its reduction by hydrogen atom donor substrates such as
dihydroarenes and 2,6-di-tert-butyl phenols, eqs 2 and 3.
Our group has recently reported on the analogous aziridination
reaction by a manganese corrole and found the active oxidant
to be an adduct of the imidomanganese(V) and the iodinane
ArIdNTs (ArI ) 2-t-butylsulfonyliodobenzene). Concurrently,
we discovered a competing deactivation pathway by which
(tpfc)MnVNTs (tpfc ) 5,10,15-tetrakis(pentafluorophenyl)-
corrole) is reduced via hydrogen atom abstraction from the
solvent toluene to give TsNH2 and (tpfc)MnIII.10 Mayer has
pioneered the study of HAT reactions by metal-oxo complexes
and has also advanced the analogy of metal-based HAT reactions
to their organic radical counterparts.1,11,12 As for porphyrinoid
complexes, the work of Goldberg on HAT reactions from
phenols to oxomanganese(V) corrolazine is worthy of special
mention.13 Several other reports on hydrogen atom transfer to
terminal imido complexes have been reported recently. Borovik
et al. have studied HAT to a putative imidoiron(IV) species to
give an isolable amidoiron(III), which is stabilized by pendant
hydrogen bonds.14 The second by Holland and co-workers
details the reaction chemistry of an iron(III)-imido stabilized
by a diketiminate ligand, which abstracts a hydrogen atom
intramolecularly from the ketiminate ligand to generate an Fe-
(III) amido with formal oxidation of the ligand.15 HAT has been
Both reactions proceed to completion, and all products as
shown have been identified and quantified to greater than 90%
yield. Product 1 is detected and quantified by absorption
spectroscopy. Diamagnetic organic products, anthracene, and
TsNH2 were detected and quantified by 1H NMR and gas
chromatography. Phenoxyl radical was detected by EPR spec-
troscopy. Quantification of phenoxyl radical was done indirectly
by the reaction of 2 with 2-tert-butyl-4-methylphenol, which
couples to give the biaryl product (Scheme 2).13,20
Under conditions of excess phenol or dihydroarene, the HAT
reaction follows pseudo-first-order kinetics and affords kψ that
exhibits first-order dependence on the excess substrate concen-
tration (dihydroarene or phenol). Typical kinetic data are
(8) (a) Nam, W.; Choi, S. K.; Lim, M. H.; Rohde, J.-U.; Kim, I.; Kim, J.;
Kim, C.; Que, L., Jr. Angew. Chem., Int. Ed. 2003, 42, 109. (b) Collman,
J. P.; Chien, A. S.; Eberspacher, T. A.; Brauman, J. I. J. Am. Chem. Soc.
2000, 122, 11098. (c) Nam, W.; Lim, M. H.; Lee, H. J.; Kim, C. J. Am.
Chem. Soc. 2000, 122, 6641.
(9) Wang, S. H.; Mandimutsira, B. S.; Todd, R.; Ramdhanie, B.; Fox, J. P.;
Goldberg, D. P. J. Am. Chem. Soc. 2004, 126, 18.
(10) Zdilla, M. J.; Abu-Omar, M. M. J. Am. Chem. Soc. 2006, 128, 16971.
(11) (a) Tahmassebi, S. K.; McNeil, W. S.; Mayer, J. M. Organometallics 1997,
16, 5342-5353. (b) Wang, K.; Mayer, J. M. J. Org. Chem. 1997, 62, 4248.
(c) Cook, G. K.; Mayer, J. M. J. Am. Chem. Soc. 1995, 117, 7139. (d)
Gardner, K. A.; Mayer, J. M. Science 1995, 269, 1849. (e) Cook, G. K.;
Mayer, J. M. J. Am. Chem. Soc. 1994, 116, 1855. (f) Cook, G. K.; Mayer,
J. M. J. Am. Chem. Soc. 1994, 116, 8859. (g) Conry, R. R.; Mayer, J. M.
Organometallics 1993, 12, 3179. (h) Conry, R. R.; Mayer, J. M. Organo-
metallics 1991, 10, 3160. (i) Conry, R. R.; Mayer, J. M. Inorg. Chem.
1990, 29, 4862.
(12) Rhile, I. J.; Markle, T. F.; Nagao, H.; DiPasquale, A. G.; Lam, O. P.;
Lockwood, M. A.; Rotter, K.; Mayer, J. M. J. Am. Chem. Soc. 2006, 128,
6075.
(13) Lansky, D. E.; Goldberg, D. P. Inorg. Chem. 2006, 45, 5119.
(14) Lucas, R. L.; Powell, D. R.; Borovik, A. S. J. Am. Chem. Soc. 2005, 127,
11596.
(15) Eckert, N. A.; Vaddadi, S.; Stoian, S.; Lachicotte, R. J.; Cundari, T. R.;
Holland, P. L. Angew. Chem., Int. Ed. 2006, 45, 6868.
(16) Kogut, E.; Wiencko, H. L.; Zhang, L.; Cordeau, D. E.; Warren, T. H. J.
Am. Chem. Soc. 2005, 127, 11249.
(17) (a) Shay, D. T.; Yap, G. P. A.; Zakharov, L. N.; Rheingold, A. L.; Theopold,
K. H. Angew. Chem., Int. Ed. 2005, 44, 1508. (b) Shay, D. T.; Yap, G. P.
A.; Zakharov, L. N.; Rheingold, A. L.; Theopold, K. H. Angew. Chem.,
Int. Ed. 2006, 45, 7870.
(18) Cowley, R. E.; Bontchev, R. P.; Sorrell, J.; Sarracino, O.; Feng, Y.; Wang,
H.; Smith, J. M. J. Am. Chem. Soc. 2007, 129, 2424.
(19) (a) Brown, S. D.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 4538. (b)
Bart, S. C.; Lobkovsky, E.; Bill, Eckhard; Chirik, P. J. J. Am. Chem. Soc.
2006, 128, 5302.
(20) Liguori, L.; Bjørsvik, H.-R.; Fontana, F.; Bosco, D.; Galimberti, L; Minisci,
F. J. Org. Chem. 1999, 64, 8812.
9
11506 J. AM. CHEM. SOC. VOL. 129, NO. 37, 2007

Hydrogen Atom Transfer Reactions
A R T I C L E S
Figure 2. Reaction of 2 and DHA monitored by absorption spectroscopy
in benzene: [2] ) 4.8 × 10-5 M; [dihydroanthracene] ) 1.67 × 10-3 M;
T ) 22 °C; scan interval ) 2 min.
Figure 1. Kinetic data for abstraction of hydrogen atoms from excess
dihydroanthracene by compound 2 in benzene solvent at 22 °C. The top
panel shows (s) data, (---) fit. [2]0 ) 5 × 10-5 M; [dihydroanthracene] )
3.5 × 10-3 M; kψ from fit ) 7.12 ( 0.02 × 10-4 s-1; R ) 0.9991. The
bottom panel shows a plot of kψ vs [dihydroanthracene]. Data designated
by points; fit designated by line. k from fit ) 0.175 ( 0.014 M-1 s-1; R )
0.986.
Figure 3. (a) EPR spectrum of amido intermediate (tpfc)MnIV(NHTs) (3)
in toluene glass at 5 K with signals at g ) 4 and g| ) 2 (only the g
region is simulated because of the g| being obscured by other g ) 2 signals
and by subtraction of a g ) 2 baseline signal in the instrument). (b) EPR
spectrum of [(tpfc)MnIV + (5) in toluene glass at 5 K with gx ) 4.57, gy )
]
3.60, and gz ) 1.995. Experimental data is shown in green with spectral
simulation in blue. Other signals are Mn(III) at g ) 8, phenoxyl radical at
g ) 2.003, and a trace amount of S ) 1/2 Mn dimer resulting from hydrolysis
by trace H2O (16-line signal at g ) 2). Modulation frequency ) 100 kHz;
Modulation amplitude ) 7.182 G; microwave frequency ) 9.44 GHz;
microwave power ) 10 mW.
Scheme 2. Aryl Coupling of 6-Unsubstituted Phenoxyl Radicals
Table 1. Rate Constants, KIE, and Activation Parameters for
Hydrogen Atom Abstraction by Compound 2
k
kH/kD
∆
H‡
∆
S‡
1
1
a
1 a
substrate
M-1s-
(KIE)a kJ mol-
J mol-1 K-
0.17 ( 0.01 3.5 nd
nd
0.16 ( 0.03 nd
nd
nd
displayed for the reaction of 2 with dihydroanthracene (DHA)
in Figure 1. A comment regarding the y-intercept of kψ versus
[substrate] plots is in order. Most investigated substrates when
dried thoroughly give negligible intercepts within statistical
error. 3,5-Di-t-butyl-4-hydroxyacetophenone and a couple of the
deuterated phenols are the exception. They must contain either
residual moisture or a small impurity that was not totally
removed in the purification process, which gives rise to an
appreciable y-intercept. Nevertheless, the kinetics are reproduc-
ible and the contribution from the y-intercept amounts to no
more than 20% of the overall rate. Therefore, the rate law in eq
4 can be simplified to an overall second-order equation.
X ) OMe 25 ( 1
4.3 32.3(5) -106(2)
X ) tBu 1.15 ( 0.06 5.1 nd
nd
X ) CH3 1.9 ( 0.1
nd
39(2)
40(4)
nd
-110(7)
X ) H
0.32 ( 0.02 nd
-114(14)
X ) Ac
0.31 ( 0.02 nd
nd
X ) CHO 0.59 ( 0.05 nd
X ) CN 1.5 ( 0.1
42(1)
3.0 32(4)
-114(4)
-157(15)
a nd ) not determined.
limiting hydrogen atom transfer mechanism.1 Rate constants,
KIE values, and activation parameters are summarized in
Table 1.
Intermediacy of Mn(IV) Amido Complex. Given that the
tosylimido complex is converted completely to the amine
-d[2]/dt ) (k[substrate] + b)[2] = k[2][substrate] (4)
A significant deuterium kinetic isotope effect (KIE) was
observed for both dihydroanthracene (DHA) as well as 2,4,6-
tri-tert-butylphenol. A hydrogen KIE is consistent with a rate-
9
J. AM. CHEM. SOC. VOL. 129, NO. 37, 2007 11507

A R T I C L E S
Zdilla et al.
Scheme 3. Reactivity of 2 with Different Substrates
product TsNH2, reactions 2 and 3 undergo two hydrogen
abstraction steps, passing through a putative manganese(IV)
amido complex, (tpfc)MnIV(NHTs) (3). However, the reaction
of 2 with DHA shows tight isosbestic conversion between 2
and 1 (Figure 2). This observation implies no significant
intermediate buildup, and the second hydrogen atom abstraction
must be fast. A fast second HAT step is consistent with the
single-phase kinetic profiles observed for DHA and first-order
dependence on substrate. Attempts to detect a “short-lived” Mn-
(IV) amido intermediate by EPR spectroscopy were unsuccess-
ful.21 Additional evidence for this fast second step is the absence
of coupled organic product 9,9′,10,10′-tetrahydro-9,9′-bian-
thracene, which suggests that the initially formed radical does
not escape, accumulate, and couple (Scheme 3).22 Rather, the
carbon radical rebounds with a second HAT reaction to give
anthracene exclusively. In comparison, in a report by Borovik14
the reaction of a putative iron(IV) imido with DHA was found
to terminate after a single HAT event to give iron(III) amido
and DHA•, which couples to afford the tetrahydrobianthracene.
at g ) 8, and that of trace manganese dimer at g ) 2.05,10,23
the prominent axial signal of Mn(IV) with g at 4 and a partially
buried g| near g ) 2 are apparent. These features are comparable
to axial spectra of previously reported Mn(IV) porphyrins.24 The
concentration of 3 was determined to be ca. 10 µM by double
integration and by comparison to the intensity of the comparable
3
S ) /2 system (tpfc)MnIVBr.25 Thus, our system provides a
rare example of a two-step oxidation rebound mechanism where
the metal intermediate can be observed spectroscopically by the
proper selection of a substrate that cannot rebound, namely
phenol. These mechanistic findings are summarized in Scheme 3.
Linear Free Energy Relationships (LFER): Evidence for
Multiple Mechanistic Pathways. To gather further insight for
the HAT mechanism, we constructed a Hammett plot using
substituted phenol substrates. All substituted phenols studied
give rise to the same rate law (eq 4) and reaction products,
namely 1, TsNH2, and the corresponding phenoxyl radical. In
performing the Hammett LFER, we expected to find a general
negative F similar to that observed by Goldberg for the HAT
reaction of phenol substrates with oxomanganese(V) corrola-
zines.13 What we observed instead was a V-shaped Hammett
plot as illustrated in Figure 4a. While electron-donating sub-
stituents show the expected negative slope (reaction constant,
F), electron-withdrawing substituents appear significantly above
the established trend (positive F). A Hammett plot such as this
is indicative of a change in mechanism on either side of the
vertex. To support the notion of two mechanistic possibilities,
we also plotted the log k versus bond dissociation enthalpy
(BDE)26 (Figure 4b). Again, electron-donating substitutents
show a negative slope comparable to that seen by Goldberg,13
but the electron-withdrawing acetyl and cyano substituents break
the correlation. A scatter plot of the second-order rate constants
versus pKa values as determined by Cohen and Jones27 does
not give a clear correlation, but demonstrates that the electron-
In the case of HAT from phenol substrates, UV-vis spectra
do not show isosbestic conversion between 2 and 1 (Supporting
Information). This result suggests that there is buildup of a third
species in the reaction with phenols, eq 3. After careful
consideration, we realized that this intermediate buildup can
be easily reconciled. Phenol substrates, unlike dihydroarenes,
have only one available hydrogen atom for abstraction. Even if
the second abstraction is faster than the first, the resulting
phenoxyl radical, which may stabilize 3 by hydrogen bonding,
must first diffuse into solution to allow a second phenol
molecule to approach the putative manganese(IV) amido
complex 3 for a second HAT reaction. Attempts to trap the
presumed intermediate (tpfc)MnIV(NHTs) (3) by addition of
substoichiometric amounts of phenol to 2 were unsuccessful,
resulting in complete conversion to 1, presumably by a
subsequent HAT from solvent under limiting phenol conditions.
However, the Mn(IV) intermediate 3 was successfully trapped
by freeze-quenching the reaction with phenol substrates and was
observed by EPR spectroscopy (Figure 3). In addition to the
phenoxyl radical signal at g ) 2.003, the broad Mn(III) signal
(23) The 16-line signal at g ) 2.05 is due a small amount of a S ) 1/2 manganese
µ-oxo dimer resulting from minor hydrolysis of terminal imido.10
(24) Camenzind, M. J.; Hollander, F. J.; Hill, C. L. Inorg. Chem. 1983, 22,
3776.
(25) Golubkov, G.; Bendix, J.; Gray, H. B.; Mahammed, A.; Goldberg, I.;
DiBilio, A. J.; Gross, Z. Angew. Chem., Int. Ed. 2001, 40, 2132.
t
(26) (a) For phenols with X ) OMe, Bu, Me, and H, the BDE values as
measured by Pedulli were used: Lucarini, M.; Pedrielli, P.; Pedulli, G. F.;
Cabiddu, S.; Fattuoni, C. J. Org. Chem. 1996, 61, 9259. (b) For X ) CN,
Ac the BDE values were approximated using the ∆BDEs reported by
Bordwell: Bordwell, F. G.; Cheng, J.-P. J. Am. Chem. Soc. 1991, 113,
1736. While these BDE values are not exact, 18 cases were examined by
Pedulli (reference 26a) and were shown to be highly accurate with one
exception for 4-methoxytetramethylphenol, which was attributed to steric
reorientation of the oxygen lone pair orbital. These approximations are
therefore sufficient to establish BDE trends.
(21) After addition of dihydroanthracene to a solution of 2, the reaction was
quenched by freezing in liquid nitrogen and analyzed by EPR spectroscopy
at 5 K. The tube was repeatedly thawed to allow continuation of the reaction
and refrozen for analysis. An axial MnIV signal was never observed.
(22) While it is possible that the weak DHA• C-H bond is the sole reason for
the absence of coupled product, the scenario is unlikely given that the radical
was seen by Borovik14 to escape the solvent cage and couple, even in the
more viscous dimethylacetamide solvent. The likelihood of benzene solvent
cage completely preventing any radical coupling is quite small.
(27) Cohen, L. A.; Jones, W. M. J. Am. Chem. Soc. 1963, 81, 3397.
9
11508 J. AM. CHEM. SOC. VOL. 129, NO. 37, 2007

Hydrogen Atom Transfer Reactions
A R T I C L E S
Figure 4. Linear free energy correlations for HAT reactions. (a) Hammett substituent plot for reaction of 2 with 2,6-di-tert-butyl phenols. For electron-
donating substituents (left), F ) -2.4 ( 0.1. (b) BDE correlation with log k; slope ) -0.42 ( 0.02. (c) Scatter plot of log k vs phenol pKa.
Scheme 4. Mechanistic Possibilities for Positive Hammett
Reaction Constant in Hydrogen Atom Transfer
withdrawing substituents are employed, the acidity of the
phenolic proton increases. It is therefore reasonable that in
moving left to right on the Hammett plot a point is encountered
where protonation of the terminal imido ligand becomes feasible
and more favored than concerted proton-electron transfer. The
second proposed mechanism (B) takes into consideration the
increasing coordinating ability of the electron-withdrawing
substituents. In this remote attack mechanism,29 the phenol
ligates to the metal center via the p-substituent, allowing for
inner-sphere (through-bond) electron transfer, followed by (most
likely fast) proton transfer from the protonated phenoxyl radical.
The reaction rate via pathway B relies on the equilibrium
constant K, hence, the ligand strength of the substituent group.
Thus, cyano would be expected to be faster than aldehyde.
The second mechanism (B) can be ruled out on the basis of
the presence of a primary hydrogen KIE for deutero versus
protio 2,6-di-t-butyl-4-cyanophenol (Table 1). Since neither the
equilibrium constant K, nor the rate-limiting electron transfer
withdrawing substitutents lie in a very different pKa regime when
compared to the electron donating groups (Figure 4c).
Further confirmation of the difference in mechanism was
sought in the determination of activation parameters for
substrates on either side of the vertex (Table 1). However, as
the entropy of activation increases for the series, the enthalpy
of activation does not change much. The isokinetic plot of ∆H‡
versus ∆S‡ for phenol substrates (Supporting Information) does
not show a correlation. All substrates have similar activation
parameters within error with the exception of the Hammett
extremes (i.e., X ) OMe, CN).
In considering alternative mechanisms for phenols with
electron-withdrawing groups, the following requisites must be
satisfied: First, the proposed mechanism must be expected to
result in a positive Hammett reaction constant F. If this condition
is not satisfied, the proposed mechanism would also have to be
in effect for electron-donating substituents and there would be
no vertex in the Hammett plot. Second, the end products and
intermediate must be identical to the concerted proton-electron
transfer mechanism, namely, phenoxyl radical 1 and inter-
mediate 3.
involve the hydrogen atom, mechanism B cannot result in a
KIE. Thus, mechanism A, PT-ET is the most likely alternate
mechanism for the reaction. Additional support for this mech-
anism was found in the feasible protonation of compound 2.
Mechanism A suggests that 2 should be a Brønsted base capable
of accepting a proton. Addition of anhydrous trifluoroacetic acid
to 2 (eq 5) results in a new species, which is identified by mass
spectrometry as the protonated imidomanganese(V), (tpfc)Mn-
(NHTs)+ (4, m/z )1017.53, calcd ) 1018).
(tpfc)MnVNTs (2) + CF3COOH f [(tpfc)MnV(NHTs)]+
[CF3COO]-(4[CF3OO]-) (5)
This species is short-lived, and is only detected within the
few seconds after it is generated, as it rapidly picks up a
hydrogen atom, probably from the solvent, to give TsNH2 and
[(tpfc)MnIV]+ (5). The higher reactivity of the protonated imide
complex 4 in comparison to 2 is consistent with our mechanistic
proposal (Scheme 4A).
[H]
[(tpfc)MnV(NHTs)]+ (4) 8 [(tpfc)MnIV]+ (5) + TsNH2 (6)
We have considered two mechanisms that satisfy these
conditions. These are shown in Scheme 4. The first is proton
transfer followed by electron transfer (A), commonly referred
to as proton-gated electron transfer (PT-ET).28 As more
1
The appearance of TsNH2 is confirmed by H NMR (50%
yield), and 5 is characterized by EPR, UV-vis, and mass
(28) Chen, K.; Hirst, J.; Camba, R.; Bonagura, C. A.; Stout, C. D.; Burgess, B.
(29) (a) Haim, A. Acc. Chem. Res. 1975, 8, 264. (b) Fraser, R. T. M.; Taube,
H. J. Am. Chem. Soc. 1961, 83, 2239.
K.; Armstrong, F. A. Nature 2000, 405, 814.
9
J. AM. CHEM. SOC. VOL. 129, NO. 37, 2007 11509

A R T I C L E S
Zdilla et al.
sodium-benzophenone ketyl. Solvents were degassed and stored over
activated 4-Å molecular sieves in a glovebox for 24 h prior to use.
NMR solvents and ethanol-d1 were purchased from Cambridge Isotope
Laboratories. NMR solvents were stored over activated 4-Å molecular
sieves for 24 h prior to use. 2,6-Di-tert-butyl-4-X-phenol compounds
were purchased from Aldrich, TCI, Alfa Aesar, or Acros. 9,10-
Dihydroanthracene and 1,4-cyclohexadiene (dihydrobenzene) were
purchased from Aldrich. Sodium metal was obtained from Fluka. All
purchased chemicals were used as obtained from the manufacturers
unless specified otherwise. (tpfc)Mn (1) was prepared according to a
previous report.10 2-(t-Butylsulfonyl)(p-toluene-sulfonyliminoiodo)-
benzene was prepared by the method of Protasiewicz.30 9,10-Dihy-
droanthracene-d4 was prepared by the method of Mayr.31 Lithium 2,6-
di-tert-butyl-4-cyanophenoxide was prepared according to the method
of Henderson et al.32
Figure 5. Comparative absorption spectra of compound 5 (solid line) and
(tpfc)MnIVBr (dashed line) in benzene.
spectrometry. The mass spectrum exhibits an overwhelming
UV-vis spectra were recorded on a Shimadzu UV-2501PC scanning
spectrophotometer. NMR spectra were obtained on Inova/Varian 300
MHz spectrometer. Gas chromatography was carried out on an Agilent
Technologies 6890N Network GC system with a J&W Scientific DB-5
capilary column. Oven temperature was ramped from 70 to 150 °C at
25 °/min, then from 150 to 230 at 15 °/min in constant flow mode at
2.5 mL/min. Kinetics experiments were performed using an Applied
Photophysics SX.18MV stopped-flow analyzer or a Shimadzu UV-
2501PC scanning spectrophotometer in 3 mL quartz cells with 1 cm
path length. EPR spectra were recorded on a Bruker ESP 300E EPR
spectrometer equipped with an hp 5350B microwave frequency counter,
an Oxford ITC4 temperature controller, and a VC40 gas flow controller
(for liquid He) and a Eurotherm temperature control unit (liquid N2).
Data fitting was carried out using KaleidaGraph. EPR spectra were
simulated using the Easyspin33 program for MATLAB.
signal at m/z ) 848, corresponding to 5 (see Supporting
Information). The EPR spectrum exhibits an S ) /2 rhombic
Mn(IV) signal that is different from that of 3 (Figure 3b). Last,
3
the absorption spectrum bears a striking resemblance to the
previously reported Mn(IV) corrole (tpfc)MnBr (Figure 5).25
5
reacts readily with phenoxides to yield 1 and phenoxyl radical
according to absorption and EPR spectroscopies, respectively.
[(tpfc)MnIV (5)]+ + ArO- f (tpfc)MnIII (1) + ArO•
(7)
Only 1 equiv of trifluoroacetic acid is needed to convert
nearly all of 2 to 5, with very little spectral change (<5%)
observed with additional equivalents of trifluoroacetic acid. This
result demonstrates the capability of 2 as a proton acceptor, a
result consistent with our previous report that the imido ligand
of 2 is bent with a nitrogen lone pair and MndNTs double-
bond character.10 As expected, the resulting Mn(V) amido (5)
is highly reactive toward reduction chemistry, which further
validates the likelihood of the PT-ET mechanism for electron-
poor phenols.
Phenols were dried by dissolving in a minimal amount of dry solvent
t
(hexanes for X ) H, Me, Bu; CH2Cl2 for X ) CHO, Ac, OMe) and
storing over activated molecular sieves for 24 h. After removing sieves,
the compounds were recrystalized from their respective solvents
(volume reduction for X ) H, Me, tBu; and by addition of hexanes for
X ) CHO, Ac, OMe). Solids were filtered and rinsed with a minimum
of cold n-pentane and dried in vacuo.
Synthesis of 4-X-2,6-t-Bu2C6H2OD. Deuterated phenols were
prepared by an adaptation of the method of Mayr31 for the deuteration
of 9,10-dihydroanthracene. Under an inert atmosphere, Na metal (1 g,
44 mmol) was dissolved in 20 mL of ethanol-d. To this was added
4-X-2,6-t-Bu2C6H2OH (ca. 6-9 mmol). After 10 h of stirring, the
solvent volume was reduced in vacuo to 8 mL, and 20 mL of D2O was
added yielding a light-yellow precipitate. The solid was cooled to 0
°C, filtered, and dried under vacuum. Degree of deuteration was
determined by the integration of the phenol proton peak (300 MHz 1H
NMR, C6D6). Degree of deuteration: X ) MeO at 4.5 ppm, 88%
deuteration, 94% yield; X ) CN at 5.2 ppm, 85% deuteration, 40%
Conclusions
HAT reactions with terminal tosylimidomanganese(V) corrole
proceed in a two-step mechanism for which the rate-determining
step is the first hydrogen atom transfer resulting in an intermedi-
ate tosylamidomanganese(IV) complex, which can be detected
by EPR spectroscopy if the hydrogen-donating substrate is a
non-rebounding substrate such as 2,6-di-t-butylphenol. In the
case of dihydroarenes, the arene radical rebounds with Mn(IV)
to give arene and Mn(III) at a faster rate than diffusing into
solution to react with another equivalent of radical or Mn(V)
imido. For reactions with 4-substituted 2,6-di-tert-butylphenols,
the rate-determining initial step can proceed by two mechanisms.
In the case of electron-donating substituents, the reaction
proceeds through concerted proton-electron transfer, while in
the case of electron-withdrawing substituents, the reaction
proceeds via PT-ET. In these examples the substrate type
(namely phenol) is the same, yet the reaction pathways are
dependent upon the acidity of the phenolic proton. These
findings demonstrate that the substrate pKa is an important
consideration not only for the rate of metal-mediated HAT
reactions, but for the molecular mechanism as well.
t
yield; X ) Bu at 3.9 ppm, >99% deuteration, 97% yield.
Kinetic Studies. Reaction of (tpfc)MnVNTs with Substrates.
General Procedures. Kinetic traces for the reaction between (tpfc)-
MnVNTs and phenol or dihydroarene substrates were obtained under
pseudo-first-order conditions with excess substrate. Reactions were
monitored by UV-vis spectroscopy or a stopped-flow analyzer by
following the increase in the absorbance of the Q-band peak for (tpfc)-
MnIII at 600 nm. Stock solutions were prepared and stored under an
inert atmosphere with benzene as the solvent. Prior to each kinetic run,
fresh solutions of (tpfc)MnNTs, 2, (3-4 × 10-5 M) were prepared by
reacting 50 µM of (tpfc)MnIII and limiting amount (3-4 × 10-5 M) of
2-(t-butylsulfonyl)(p-toluene-sulfonyliminoiodo)benzene for 5 min.
Different amounts of a stock solution of the phenol substrate were added
(30) Macikenas, D.; Skrzypczak-Jankun, E.; Prostasiewicz, J. D. J. Am. Chem.
Soc. 1999, 121, 7164.
Experimental
(31) Mayr, H.; Lang, G.; Ofial, A. R. J. Am. Chem. Soc. 2002, 124, 4076.
(32) MacDougall, D. J.; Noll, B. C.; Kennedy, A. R.; Henderson, K. W. Dalton
Trans. 2006, 1875.
General. All operations were carried out under inert atmosphere
using standard glovebox and Schlenk line techniques except where
otherwise noted. Benzene, toluene, and n-pentane were distilled from
(33) Stoll, S.; Schweiger, A. J. Magn. Reson. 2006, 178, 42.
9
11510 J. AM. CHEM. SOC. VOL. 129, NO. 37, 2007

Hydrogen Atom Transfer Reactions
A R T I C L E S
to initiate the reaction. Four to six different concentrations were used
for each substrate, and pseudo-first-order rate constants (kψ) were
determined at each concentration. Plots of kψ versus [substrate] yielded
the second-order rate constants k. Metal syringes were avoided for all
phenol kinetics because of our observation that contact with metal
surfaces enhances reaction rates for some phenols.
anhydrous trifluoroacetic acid was added (no color change). This
solution was added to 1 mg (2 mmol) of 2-(t-butylsulfonyl)(p-toluene-
sulfonyliminoiodo)benzene to afford a color change from green to
green-yellow. This solution was analyzed immediately by ESI mass
spectrometry: observed m/z ) 1017.53; theoretical for (tpfc)Mn-
(NHTs)+ m/z ) 1018.
Reaction of (tpfc)MnVNTs with 4-X-2,6-t-Bu2C6H2OH (X ) tBu,
OCH3, Ac, or CHO, or CHO) and 4-t-Bu-2,6-t-Bu2C6H2OD. Kinetic
experiments were carried out according to the general procedure using
UV-vis spectroscopy (stop flow for OCH3). (tpfc)MnVNTs (ca. 2-4
× 10-5 M) reacted with varied concentrations of phenol substrate (2
× 10-3 to 2 × 10-2 M).
Generation and Detection of [(tpfc)MnIV]+(5) by Absorption
Spectroscopy. To 3 mL of a 3.3 × 10-5 M solution of (tpfc)Mn (1),
59 µL of a 1.7 × 10-3 M stock solution of 2-(t-butylsulfonyl)(p-toluene-
sulfonyliminoiodo)benzene was added to generate 2. This solution was
titrated by µL additions of 0.15 M anhydrous trifluoroacetic acid in
benzene to give the spectrum of 5.
Generation and Detection of [(tpfc)MnIV]+(5) by EPR Spectros-
copy. To 1 mL of a 1 mM solution of 1 in toluene, 2 equiv of solid
2-(t-butylsulfonyl)(p-toluene-sulfonyliminoiodo)benzene was added, and
the solution was stirred until red-brown (ca. 1 min). A 0.25 mL aliquot
of this solution was added to an EPR tube containing 20 µL of
trifluoroacetic acid/benzene 1:1000 (1 equiv) to afford a color change
to brown/green. The EPR tube was frozen in liquid nitrogen and
analyzed by EPR.
Reaction of [(tpfc)MnIV]+ (5) with Phenoxide. An amount of 3
mL of a 5 × 10-5 M solution of 1 was stirred with 2 mg of PhINTs
until brown/orange in color. This solution was decanted to a new vial
and 12 µL of benzene/trifluoroacetic acid 1:1000 was added to give a
color change to green/brown. This was added to a new vial containing
1 mg of lithium 2,6-di-tert-butyl-4-cyanophenoxide32 to afford a color
change to green. Analysis by absorption spectroscopy identifies the
manganese product as 1. The experiment was repeated in toluene and
analyzed by EPR spectroscopy to confirm the appearance of phenoxyl
radical, and the disappearance of 5.
Decomposition of [(tpfc)MnV(NHTs)]+ (4): Detection of TsNH2
by NMR Spectroscopy. A 2.0 mM solution of (tpfc)MnVNTs (2) was
generated by combining 1.0 mg (1.2 µmol) of (tpfc)Mn (1) with 0.5
mg (1.0 µmol) of 2-(t-butylsulfonyl)(p-toluenesulfonyliminoiodo)-
benzene in 0.50 mL of C6D6. A 0.10 M solution (10 µL) of CF3COOH
(1.0 µmol) in C6D6 was added. TsNH2 was detected by 1H NMR (δ )
4.74, s, 2H) and quantified by integration against ethyl acetate internal
standard from 1‚(EtOAc)2 (δ ) 1.88, s, 3H, CH3).
Reaction of (tpfc)MnVNTs with 2,4,6-t-Bu2C6H2OD/2,6-t-
Bu2C6H2OH. Kinetic experiments were carried out according to the
general procedure using UV-vis spectroscopy. (tpfc)MnVNTs (ca. 2
× 10-5 M) reacted with varied concentrations of phenol substrate (3.8-
38 mM).
Reaction of (tpfc)MnVNTs with 4-CN-2,6-t-Bu2C6H2OH. Kinetic
experiments were carried out according to the general procedure using
UV-vis spectroscopy. (tpfc)MnVNTs (ca. 2 × 10-5 M) reacted with
varied concentrations of phenol substrate (0.46-13 mM).
Reaction of (tpfc)MnVNTs with 9,10-Dihydroanthracene and 1,4-
Cyclohexadiene. Kinetic experiments were carried out according to
the general procedure using UV-vis spectroscopy. (tpfc)MnVNTs (ca.
5 × 10-5 M) reacted with varied concentrations of 9,10-dihydroan-
thracene (6.1 × 10-4 to 5.6 × 10-3 M) or 1,4-cyclohexadiene (4.1 ×
10-3 to 3.8 × 10-2 M).
Activation Parameters Studies. Kinetic traces were obtained by
monitoring absorbance at 600 nm at varied temperatures for the reaction
of (tpfc)MnVNTs with excess phenol under pseudo-first-order conditions
in accordance with the general procedure described above. Reactions
were performed at four different temperatures (7-60 °C), and second-
order rate constants (k) were determined at each temperature. The Eyring
plot of ln(k/T) versus 1/T yielded ∆H‡ and ∆S‡ values.
Detection of (tpfc)MnIV(NHTs) Intermediate (3) by EPR Spec-
troscopy. (tpfc)MnVNTs (2) was generated by the addition of 1 mg of
2-(t-butylsulfonyl)(p-toluene-sulfonyliminoiodo)benzene to a solution
of 1 mg of (tpfc)Mn (1) in 1 mL of toluene. To initiate the HAT
reaction, 1.25 µL of a 0.1 M stock solution of 2,6-di-tert-butylphenol
was added to 0.25 mL of solution of 2. An immediate color change to
green ensued, and the reaction was quickly quenched by immersion
into liquid N2. The sample was analyzed by EPR at 5 K and showed
the characteristic S ) 3/2 spectrum for 3. Modulation frequency ) 100
kHz; modulation amplitude ) 7.182 G; microwave frequency ) 9.44
GHz; microwave power ) 10 mW.
Acknowledgment. This work was supported by a grant from
the National Science Foundation (Grant No. CHE-0502391).
Supporting Information Available: Kinetics plots of kψ
versus concentration of substrate, EPR spectra of product organic
radicals, data on determination of activation parameters, isoki-
netic plot, mass spectra for 4 and 5, and absorption spectra for
reaction of 2 with phenols. This material is available free of
charge via the Internet at http://pubs.acs.org.
An identical test for 3 using dihydroanthracene in place of 2,6-di-
tert-butyl phenol did not yield the signature for 3.
Detection of [(tpfc)Mn(NHTs)]+(4) by Mass Spectrometry. To a
solution of 1 mg (tpfc)Mn (1, 1 µmol) in 0.25 mL of benzene, 1 µL of
JA073027S
9
J. AM. CHEM. SOC. VOL. 129, NO. 37, 2007 11511