Oxidative reactivity difference among the metal oxo and metal hydroxo moieties: pH dependent hydrogen abstraction by a manganese(IV) complex having two hydroxide ligands

Article abstract of DOI:10.1021/ja804305x

Clarifying the difference in redox reactivity between the metal oxo and metal hydroxo moieties for the same redox active metal ion in identical structures and oxidation states, that is, Mⁿ⁺=O and M ⁿ⁺-OH, contributes to the understanding of nature's choice between them (Mⁿ⁺=O or Mⁿ⁺-OH) as key active intermediates in redox enzymes and electron transfer enzymes, and provides a basis for the design of synthetic oxidation catalysts. The newly synthesized manganese(IV) complex having two hydroxide ligands, [Mn(Me₂EBC)₂(OH) ₂](PF₆)₂, serves as the prototypic example to address this issue, by investigating the difference in the hydrogen abstracting abilities of the Mn^IV=O and Mn^IV-OH functional groups. Independent thermodynamic evaluations of the O-H bond dissociation energies (BDE_OH) for the corresponding reduction products, Mn^III-OH and Mn^III-OH₂, reveal very similar oxidizing power for Mn^IV=O and Mn^IV-OH (83 vs 84.3 kcal/mol). Experimental tests showed that hydrogen abstraction proceeds at reasonable rates for substrates having BDE_CH values less than 82 kcal/mol. That is, no detectable reaction occurred with diphenyl methane (BDE_CH = 82 kcal/mol) for both manganese(IV) species. However, kinetic measurements for hydrogen abstraction showed that at pH 13.4, the dominant species Mn(Me ₂EBC)₂(O)₂, having only Mn^IV=O groups, reacts more than 40 times faster than the Mn^IV-OH unit in Mn(Me₂EBC)₂(OH)₂²⁺, the dominant reactant at pH 4.0. The activation parameters for hydrogen abstraction from 9,10-dihydroanthracene were determined for both manganese(IV) moieties: over the temperature range 288-318 K for Mn^IV(OH)₂²⁺, ΔH^? = 13.1 ± 0.7 kcal/mol, and ΔS ^? = -35.0 ± 2.2 cal K^-1 mol^-1; and the temperature range 288-308 K for for Mn^IV(O)₂, ΔH^? = 12.1 ± 1.8 kcal/mol, and ΔS ^? = -30.3 ± 5.9 cal K^-1 mol^-1.

Full text of DOI:10.1021/ja804305x

Published on Web 11/11/2008
Oxidative Reactivity Difference among the Metal Oxo and
Metal Hydroxo Moieties: pH Dependent Hydrogen Abstraction
by a Manganese(IV) Complex Having Two Hydroxide Ligands
Guochuan Yin,† Andrew M. Danby,† David Kitko,‡ John D. Carter,‡
William M. Scheper,‡ and Daryle H. Busch*,†
Department of Chemistry, The UniVersity of Kansas, Lawrence, Kansas 66045, and The Procter
and Gamble Company, Cincinnati, Ohio 45202
Received June 6, 2008; E-mail: busch@ku.edu
Abstract: Clarifying the difference in redox reactivity between the metal oxo and metal hydroxo moieties
for the same redox active metal ion in identical structures and oxidation states, that is, Mn+dO and Mn+-OH,
contributes to the understanding of nature’s choice between them (Mn+dO or Mn+-OH) as key active
intermediates in redox enzymes and electron transfer enzymes, and provides a basis for the design of
synthetic oxidation catalysts. The newly synthesized manganese(IV) complex having two hydroxide ligands,
[Mn(Me2EBC)2(OH)2](PF6)2, serves as the prototypic example to address this issue, by investigating the
difference in the hydrogen abstracting abilities of the MnIVdO and MnIV-OH functional groups. Independent
thermodynamic evaluations of the O-H bond dissociation energies (BDEOH) for the corresponding reduction
products, MnIII-OH and MnIII-OH2, reveal very similar oxidizing power for MnIVdO and MnIV-OH (83 vs
84.3 kcal/mol). Experimental tests showed that hydrogen abstraction proceeds at reasonable rates for
substrates having BDECH values less than 82 kcal/mol. That is, no detectable reaction occurred with diphenyl
methane (BDECH ) 82 kcal/mol) for both manganese(IV) species. However, kinetic measurements for
hydrogen abstraction showed that at pH 13.4, the dominant species Mn(Me2EBC)2(O)2, having only MnIVdO
groups, reacts more than 40 times faster than the MnIV-OH unit in Mn(Me2EBC)2(OH)2 the dominant
2+
,
reactant at pH 4.0. The activation parameters for hydrogen abstraction from 9,10-dihydroanthracene were
determined for both manganese(IV) moieties: over the temperature range 288-318 K for MnIV(OH)2
,
2+
∆Hq ) 13.1 ( 0.7 kcal/mol, and ∆Sq ) -35.0 ( 2.2 cal K-1 mol-1; and the temperature range 288-308
K for for MnIV(O)2, ∆Hq ) 12.1 ( 1.8 kcal/mol, and ∆Sq ) -30.3 ( 5.9 cal K-1 mol-1
.
peroxidation catalysts;3 and (3) high valent metal hydroxo
moieties in lipoxygenases, and their models.4 Further, the high
valent metal oxo or hydroxo functional groups may also
participate in a series of electron transfer processes in biological
and chemical events.1 Because of their wide significance in
biological and chemical sciences, an immediate issue is
determining the reactivity differences among these related active
transition metal ion species and then understanding the roles
they can play in various oxidation processes. For example, in
the case of iron, the (L+•)FeIVdO moiety is a general active
intermediate in an oxidation process mediated by P450 enzymes
and by synthetic iron porphyrin and related compounds.2a,5 The
FeIII-OOH moiety has recently been proposed to serve as a
second active intermediate in certain P450 enzymes and in the
destruction of porphyrins in certain oxygenase enzymes.3,6 Also,
Introduction
Transition metal ions play a critical role in various oxidation
and electron transfer processes in nature and in industrial and
academic chemistry.1 Generally, there are three major categories
of redox metal ions when defined in terms of the active oxygen
species: (1) high valent metal oxo moieties in various redox
enzymes along with their synthetic counterparts, and lattice
oxygen in metal oxide catalysts;2 (2) metal hydroperoxide
complexes in certain redox enzymes and in transition metal
† The University of Kansas.
‡ The Procter and Gamble Company.
(1) (a) Frey, P. A.; Hegeman, A. D. Enzymatic Reaction Mechanisms;
Oxford University Press: New York, 2007. (b) Special topic issue:
Bioinorganic Enzymology. Chem. ReV. 1996, 96, 2237. (c) Meunier,
B., Ed. Biomimetic Oxidations Catalyzed by Transition Metal Com-
plexes; Imperial College Press: London, 2000. (d) Sigel, H., Sigel,
A., Eds. Electron Transfer Reactions in Metalloproteins. In Metal ions
in Biological Systems; Marcel Dekker: New York, 1991; Vol. 27.
(2) (a) Ortiz de Montellano, P. R., Ed. Cytochrome P450: Structure,
Mechanism, and Biochemistry; Plenum Press: New York, 1986. (b)
Stanbury, D. M. AdV. Inorg. Chem. 2003, 54, 351. (c) Holm, R. H.
Chem. ReV. 1987, 87, 1401. (d) Holm, R. H. Coord. Chem. ReV. 1990,
100, 183. (e) Kryatov, S. V.; Rybak-Akimova, E. V.; Schindler, S.
Chem. ReV. 2005, 105, 2175. (f) Yoshizawa, K. Acc. Chem. Res. 2006,
39, 375. (g) Shaik, S.; Hirao, H.; Kumar, D. Acc. Chem. Res. 2007,
40, 532. (h) Li, C. J. Catal. 2003, 216, 203. (i) Grasselli, R. K. Top.
Catal. 2002, 21, 79.
(3) (a) Vaz, D. N.; McGinnity, D. F.; Coon, M. J. Proc. Natl. Acad. Sci.
U.S.A. 1998, 95, 3555. (b) Newcomb, M.; Aebisher, D.; Shen, R.;
Chandrasena, R. E. P.; Hollenberg, P. F.; Coon, M. J. J. Am. Chem.
Soc. 2003, 125, 6064. (c) Toy, P. H.; Newcomb, M.; Coon, M. J.;
Vaz, A. D. N. J. Am. Chem. Soc. 1998, 120, 9718.
(4) (a) Samuelsson, B.; Dahlen, S. E.; Lindgren, J. A.; Rouzer, C. A.;
Serhan, C. N. Science 1987, 237, 1171. (b) Clark, K. B.; Culshaw,
P. N.; Griller, D.; Lossing, F. P.; Simoes, J. A. M.; Walton, J. C. J.
Org. Chem. 1991, 56, 5535. (c) Schilstra, M. J.; Veldink, G. A.;
Vliegenthart, J. F. G. Biochemistry 1994, 33, 3974. (d) Goldsmith,
C. R.; Jonas, R. T.; Stack, T. D. P. J. Am. Chem. Soc. 2002, 124, 83.
9
10.1021/ja804305x CCC: $40.75
2008 American Chemical Society
J. AM. CHEM. SOC. 2008, 130, 16245–16253 16245

A R T I C L E S
Yin et al.
in iron lipoxygenases, the FeIII-OH moiety has most recently
been suggested as the key active species for hydrogen abstraction
from 1,4-diene-containing fatty acids to form alkyl hydroper-
oxides.4 A similar issue occurs in the redox reactions involving
manganese. The MnVdO function is generally believed to serve
as the key active species in a series of oxygen transfer
process,5b,c,7 whereas a MnIII-OH moiety has recently been
assigned responsibility for hydrogen abstraction in manganese
lipoxygenases.8 In addition to these manganese reactants, a
manganese(III)/manganese(II) couple has been credited with
lignin degradation in manganese peroxidases, although no
evidence was provided to determine whether MnIIIdO or
MnIII-OH served as the active species.9 Clarifying the reactivity
differences among these active intermediates would provide a
basis for enhanced understanding of nature’s redox enzymes
and the knowledge for matching catalyst systems to targeted
oxidation processes. Unfortunately, prior to our preliminary
results10 and the detailed work presented here, there was no
reported example of precisely related MnIIIdO or MnIII-OH
moieties to address this broadly significant issue. The major
obstacle has been that it is difficult to produce an effective
platform, that is, an uncomplicated high valent metal complex
of manganese or iron within which the metal oxo and the metal
hydroxo groups could be alternatively generated and manipu-
lated in confidence, using a single central metal ion with
identical oxidation states and an otherwise identical coordination
sphere. The difficulty exists because such high oxidation state
iron and manganese ion moieties (MdO and M-OH) have great
tendencies to form µ-oxo bridged dimers and/or higher oligo-
mers. Following our preliminary publication,10 Fujii, et al.,
reported solution studies on the physical chemical property
differences between OdMnIV(salen), HO-MnIV(salen) and
H2O-MnIV(salen) using various spectroscopic techniques. How-
ever, the poor oxidizing capability of HO-MnIV(salen) and
H2O-MnIII(salen+·) prevented their investigation of the reactiv-
ity differences between the corresponding reaction centers,
Figure 1. The structure of MnII(Me2EBC)Cl2.
OdMnIV(salen) and HO-MnIV(salen), although OdMn(salen)
demonstrated its ability to abstract hydrogen from cyclohex-
ene.11
In these laboratories, a series of transition metal complexes
having rare abilities to host stable MnIVdO and MnIVOH
functional groups has been successfully developed with ultra
rigid ethylene cross-bridged macrocyclic ligands.12 Compared
with other ligands, these cross-bridged macrocycles produce
manganese(II) complexes having extreme kinetic stability in
both acidic and basic media, despite the well-known lability of
the high-spin d5 Mn2+ ion. The complexes have rich redox
chemistries and an industrial partner has demonstrated their
excellent catalytic oxidation activity in demanding applications,
for example in home care products.13 Specifically, the manga-
nese complex, Mn(Me2EBC)Cl2, in which Me2EBC is 4,11-
dimethyl-1,4,8,11-tetraazabicyclo[6.6.2]hexadecane (Figure 1),
demonstrated unexpected selectivity when epoxidizing olefins
with hydrogen peroxide. Epoxidations by transition metal ions
known for readily displaying a range of oxidation states, such
as manganese, have most often been attributed to the oxygen
rebound mechanism, in which oxygen atom transfer is ac-
companied by 2-electron reduction of the metal ion.5e
(5) (a) Selected examples: (a) Garrison, J. M. ; Bruice, T. C. J. Am. Chem.
Soc. 1989, 111, 191. (b) Adam, W.; Roschmann, K. J.; Saha-Mo¨ller,
C. R.; Seebach, D. J. Am. Chem. Soc. 2002, 124, 5068. (c) Finney,
N. S.; Pospisil, P. J.; Chang, S.; Palucki, M.; Konsler, R. G.; Hansen,
K. B.; Jacobsen, E. N. Angew. Chem., Int. Ed. Engl. 1997, 36, 1720.
(d) Collman, J. P.; Chien, A. S.; Eberspacher, T. A.; Brauman, J. I.
J. Am. Chem. Soc. 2000, 122, 11098. (e) Groves, J. T. J. Chem. Educ.
1985, 65, 928.
The capabilities of the high valent intermediates in these
complexes include 1-electron oxidations as well, often limiting
their selectivities. Detailed studies with the Mn(Me2EBC)Cl2
catalyst system found the origin of this selectivity to be
mechanistic with the epoxidation occurring by a Lewis acid
(6) (a) Bach, R. D.; Su, M. D.; Andres, J. L.; Schlegel, H. B. J. Am.
Chem. Soc. 1993, 115, 8763. (b) Nam, W.; Ho, R.; Valentine, J. S.
J. Am. Chem. Soc. 1991, 113, 7052. (c) Sam, J. W.; Tang, X. J.;
Peisach, J. J. Am. Chem. Soc. 1994, 116, 5250. (d) Ho, R. Y. N.;
Roelfes, G.; Feringa, B. L.; Que, L., Jr. J. Am. Chem. Soc. 1999, 121,
264. (e) Nam, W.; Lim, M. H.; Lee, H. J.; Kim, C. J. Am. Chem. Soc.
2000, 122, 6641. (f) Wadhwani, P.; Mukherjee, M.; Bandyopadhyay,
D. J. Am. Chem. Soc. 2001, 123, 12430. (g) Kim, C.; Chen, K.; Kim,
J.; Que L., Jr, J. Am. Chem. Soc. 1997, 119, 5964. (h) Avila, L.; Huang,
H.; Damaso, C. O.; Lu, S.; Moe¨nne-Loccoz, P.; Rivera, M. J. Am.
Chem. Soc. 2003, 125–4103.
(11) Kurahashi, T.; Kikuchi, A.; Tosha, T.; Shiro, Y.; Kitagawa, T.; Fujii,
H. Inorg. Chem. 2008, 47, 1674.
(12) (a) Hubin, T. J.; McCormick, J. M.; Collinson, S. R.; Buchalova, M.;
Perkins, C. M.; Alcock, N. W.; Kahol, P. K.; Raghunathan, A.; Busch,
D. H. J. Am. Chem. Soc. 2000, 122, 2512. (b) Hubin, T. J.;
McCormick, J. M.; Collinson, S. R.; Alcock, N. W.; Clase, H. J.;
Busch, D. H. Inorg. Chim. Acta 2003, 346, 76. (c) Collinson, S. R.;
Alcock, N. W.; Hubin, T. J.; Busch, D. H. J. Coord. Chem. 2001, 52,
317. (d) Collinson, S. R.; Alcock, N. W.; Raghunathan, A.; Kahol,
P. K.; Busch, D. H. Inorg. Chem. 2000, 39, 757. (e) Hubin, T. J;
McCormick, J. M.; Alcock, N. W.; Busch, D. H. Inorg. Chem. 2001,
40, 435.
(7) (a) Meunier, B.; Guilmet, E.; De Carvalho, M.; Poilblanc, R. J. Am.
Chem. Soc. 2000, 122, 2675. (b) Battioni, P.; Renaud, J. P.; Bartoli,
J. F.; Reina-Arilles, M.; Fort, M.; Mansuy, D. J. Am. Chem. Soc. 1988,
110, 8462. (c) Siegbahn, P. E. M.; Crabtree, R. H. J. Am. Chem. Soc.
1999, 121, 117. (d) Wessel, J.; Crabtree, R. H. J. Mol. Catal., A 1996,
113, 13.
(8) (a) Su, C.; Oliw, E, H J. Biol. Chem. 1998, 273, 13072. (b) Goldsmith,
C. R.; Cole, A. P.; Stack, T. D. P. J. Am. Chem. Soc. 2005, 127, 9904.
(9) (a) ten Have, R.; Teunissen, P. J. M. Chem. ReV. 2001, 101, 3397. (b)
Pecoraro, V. L., Ed. Manganese Redox Enzymes; VCH: New York,
1992.
(10) Yin, G.; Danby, A. M.; Kitko, D.; Carter, J. D.; Scheper, W. M.; Busch,
D. H. J. Am. Chem. Soc. 2007, 129, 1512.
(13) (a) Busch, D. H.; Collinson, S. R.; Hubin, T. J.; Labeque, R.; Williams,
B. K.; Johnston, J. P.; Kitko, D. J.; St. Laurent, J. C. T. R. B.; Perkins,
C. M. Bleach Compositions Containing Metal Bleach Catalyst for
Detergents. WO 98/39406. (b) Busch, D. H.; Collinson, S. R.; Hubin,
T. J. Catalysts and Methods for Catalytic Oxidation. WO98/39098,
Sept 11, 1998.
9
16246 J. AM. CHEM. SOC. VOL. 130, NO. 48, 2008

Oxidative Reactivity among Transition Metal Moieties
A R T I C L E S
at least 15 times faster than does the hydroxo species
Mn(Me2EBC)(OH)22+ in hydrogen abstraction processes.10 The
present studies address this issue in detail, showing that, for
this manganese complex, the MndO group in Mn(Me2EBC)(O)2
reacts more than 40 times faster than the corresponding
MnIV-OH group in Mn(Me2EBC)(OH)22+ in hydrogen abstrac-
tion reactions.
Experimental Section
MnII(Me2EBC)Cl2 (+99.9%) was generously supplied by the
Procter and Gamble Company. [MnIII(Me2EBC)Cl2]PF6 and
[MnIV(Me2EBC)(OH)2](PF6)2, were synthesized as reported.12e,16
Deuterated 9,10-dihydroanthracene (DHA-d4) was synthesized
according to the literature.4 Other reagents were obtained from
d
Aldrich and used as received except 1,4-cyclohexadiene which was
distilled before use.
Hydrogen Abstraction Reaction, Small Synthetic Scale. In a
typical reaction, 0.25 mmol of the manganese(IV) complex was
added to 0.5 mmol of 1,4-cyclohexadiene in 5 mL of acetone/water
(ratio 4:1) solvent in a controlled atmosphere wet box. The reaction
mixture was stirred until the purple color of the manganese(IV)
species turned to the red-brown color of the corresponding
manganese(III) species. Quantitative analysis of products was
performed by GC using an internal standard, and the identities of
the reaction products were confirmed by GC-MS.
Hydrogen Abstraction Reaction in Base, Analytical Scale. In
8 mL of acetone/water (ratio 4:1), 0.063 g (0.1 mmol) of
manganese(IV) complex was dissolved. Before adding 0.4 mL (2.4
mmol) of diphenylmethane (or some other substrate) to initiate the
reaction, the pH of the solution was adjusted to 8.4 by NaOH
addition, making MnIV(Me2EBC)(O)(OH)+ the dominant catalyst
species in the solution. After stirring overnight, the manganese(IV)
in the complex was reduced to manganese(III). In the case of
diphenylmethane, GC and GC-MS analysis of the reaction mixture
showed that no hydrogen abstraction product was formed. The
identical procedure was performed at the initial pH value of 13.4,
making MnIV(Me2EBC)(O)2 the dominant species and, for the
reaction with diphenylmethane as the substrate, the GC and
GC-MS analysis of the resulting reaction mixture also confirmed
that no hydrogen abstraction product was formed.
Rate Measurements. (a) In a typical kinetic experiment, a 20-
fold excess of substrate was used. For example, a 20-fold excess
of 1,4-cyclohexadiene (CHD, 40 mM) and 2 mM of the manga-
nese(IV) complex were dissolved in 10 mL of acetone/water (ratio
4:1, initial pH 6.1), at room temp. The disappearance of the
manganese(IV) species was monitored by UV-visible spectropho-
tometry and the pseudo-first-order rate constant was calculated.
(b) To determine the reactivity differences for three distinct catalyst
species, that is, MnIV(Me2EBC)(OH)22+, MnIV(Me2EBC)(O)(OH)+,
and MnIV(Me2EBC)(O)2, the oxidations of the selected substrates
(40 mM) were carried out with the freshly synthesized manga-
nese(IV) complex (4 mM) in acetone/water (ratio 4:1) at the selected
initial pH value, including pH values of 4.0, 8.4, and 13.4, at the
temperature of 288 K. The initial pH value was adjusted by NaOH
or HCl as needed. The disappearance of the manganese(IV) species
was monitored by UV-visible spectrophotometry and the pseudo-
first-order rate constant was calculated. The second order constant
was calculated from a set of the pseudo-first-order rate constants
with the different initial substrate concentrations, including 25, 30,
35, and 40 mM.
(c) To demonstrate the influence of the pH of the reaction
medium on the hydrogen abstraction rate, the hydrogen abstractions
from 9,10-dihydroanthracene by Mn(IV) complexes were performed
in solutions having each of the different initial pH values, and
monitored by UV-visible spectrophotometer. The initial pH values
were adjusted by the addition of HCl or NaOH as needed.
(d) To measure the activation parameters for hydrogen abstraction
with MnIV(Me2EBC)(OH)22+, and MnIV(Me2EBC)(O)2, 9,10-dihy-
2+
Figure 2. The structure of MnIV(Me2EBC)(OH)2
.
pathway.14 That mechanism has generally been found for the
early transition metal ions such as Ti(IV), Mo(IV), W(IV), and
Re(VII), not with the elements manganese and iron.15 Through
simple oxidation of Mn(Me2EBC)Cl2 by H2O2 in the presence
of NH4PF6, the first mononuclear manganese(IV) complex
having two hydroxide ligands, [Mn(Me2EBC)(OH)2](PF6)2, was
synthesized and thoroughly characterized (Figure 2).16 Because
of the two bulky methyl groups on the nonbridged nitrogen
atoms of the ligand, the dehydration of MnIV-(OH)2 unit to
form dimers or oligomers is prevented. The measured
deprotonation constants for this manganese(IV) complex to
generate MnIV(Me2EBC)(O)(OH)+ and MnIV(Me2EBC)(O)2
are 6.86 and 10, respectively (eq 1 and 2). The deprotonated
manganese(IV) species, including Mn(Me2EBC)(O)(OH)+
and Mn(Me2EBC)(O)2, slowly decompose (pH > 7), gradu-
ally degrading to the manganese(III) species in base with
88% yield.
Mn(Me2EBC)(OH)22+ r f Mn(Me2EBC)(O)(OH)+ + H+
pKa1 ) 6.86 (1)
Mn(Me2EBC)(O)(OH)+ r f Mn(Me2EBC)(O)2+H+
pKa2 ) 10 (2)
Since the two pKa values, 6.86 and 10, are within the easily
controllable pH range, this manganese(IV) complex provides a
good platform to address the main issue stated above, that is,
clarifying the reactivity difference between the MnIV-OH vs
MnIVdO reactive groups on the same transition metal ion with
identical ligands and oxidation states. Our preliminary results
showed that the oxo derivative, Mn(Me2EBC)(O)(OH)+, reacts
(14) (a) Yin, G.; Buchalova, M.; Danby, A. M.; Perkins, C. M.; David
Kitko, D.; Carter, J.; Scheper, W. M.; Busch, D. H. J. Am. Chem.
Soc. 2005, 127, 17170. (b) Yin, G.; Buchalova, M.; Danby, A. M.;
Perkins, C. M.; David Kitko, D.; Carter, J.; Scheper, W. M.; Busch,
D. H. Inorg. Chem. 2006, 45, 3467.
(15) (a) Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102, 5974.
(b) Chaumette, P.; Mimoun, H.; Saussine, L.; Fischer, J.; Mitschler,
A. J. Organomet. Chem. 1983, 250, 291. (c) Fujiwara, M.; Wessel,
H.; Park, H.; Roesky, H. W. Tetrahedron 2002, 58, 239. (d) Thiel,
W. R.; Priermeier, T. Angew. Chem., Int. Ed. Engl. 1995, 34, 1737.
(e) Herrmann, W. A.; Kratzer, R. M.; Ding, H.; Thiel, W.; Glas, H. J.
Organomet. Chem. 1998, 555, 293. (f) Xi, Z.; Zhou, N.; Sun, Y.; Lu,
K. Science 2001, 292, 1139.
(16) Yin, G.; McCormick, J. M.; Buchalova, M.; Danby, A. M.; Rodgers,
K.; Smith, K.; Perkins, C.; Kitko, D.; Carter, J.; Scheper, W. M.;
Busch, D. H. Inorg. Chem. 2006, 45, 8052.
9
J. AM. CHEM. SOC. VOL. 130, NO. 48, 2008 16247

A R T I C L E S
Yin et al.
Scheme 1
Scheme 2
Figure 3. Plot of log(k2corr) vs BDECH for hydrogen abstraction.
abstracting abilities of MnIV-OH and MnIVdO are very similar
despite the difference in charges on the two manganese(IV)
complexes, that is, +2 for MnIV(Me2EBC)(OH)22+ vs +1 for
MnIV(Me2EBC)(O)(OH)+. The effect of the difference in ionic
charge is largely compensated by the change in the nature of
the ionizing moiety (2+ for MnIV-(OH)2 versus 1+ for
MnIV-(O)(OH). In the calculation this is reflected by the
similarities in the two pKa values, 5.87 and 6.86, respectively.12e,16
Interestingly, the calculated BDEOH values (83.0 and 84.3
kcal/mol) for water and hydroxide ligated to this manganese
ion are far below the value for bulk water (119 kcal/mol),
differing by well over 30 kcal/mol. This decrease in the
activation energy for the OH bond is a simple example of
water activation by ligation to a monomeric manganese ion
center, reflecting the expectations for water associated with
the tetramanganese center in photosystem II.20 A second
consideration also comes from these simple results. The BDE
calculations reported here show that the accumulation of a
charge of one positive unit on a monomeric manganese
complex need not necessarily change the energy barrier
substantially (activation energy proportional to 83.0 kcal/
mol for [MnIII(Me2EBC)(OH)(H2O)]2+ vs 84.3 kcal/mol for
[MnIII(Me2EBC)(OH)2]+). This may be pertinent to the view
that separate proton and electron transfer events that result
in one unit of charge accumulation may be energetically
feasible in processes like water oxidation.20c,21
droanthracene (0.04 M) was chosen as substrate, and kinetic studies
were performed at initial pH values of 4.0 and 13.4 with the
manganese(IV) complex (4 mM) in acetone/water (ratio 4:1), at
the temperatures of 288, 298, 303, 308, and 318 K.
Results and Discussions
Thermodynamic Evaluation of the Oxidizing Power of
Manganese(IV) Oxo and Hydroxo Groups. In our earlier report,
the Bordwell/Mayer analysis17 was applied to calculate the
BDEOH values for the OH bonds formed when
2+
MnIV(Me2EBC)(O)(OH)+ or MnIV(Me2EBC)(OH)2 abstracts
a hydrogen atom from a substrate.10 The corresponding ther-
mochemical cycles are shown in Schemes 1 and 2. The method
has been applied by others to certain oxygen evolution models
for photosystem II, and to evaluate hydrogen abstraction abilities
of high oxidation state transition metal complexes.10,18,19 The
BDEOH values of the reduced forms of these oxidizing agents
correlate with activation energies for substrate oxidations and,
in that unusual Polanyi correlation, can serve as measures of
the oxidizing power of the reagent in that category of reactions.
The BDEOH of the water molecule ligated to the Mn3+ species,
MnIII(Me2EBC)(OH)(H2O)2+, BDEOH ) 83.0 kcal/mol (Scheme
1) is taken to be a measure of the oxidizing power of the
MnIV-OH group in MnIV(Me2EBC)(OH)22+. In the same way,
the BDEOH value for hydroxide ligated to the Mn3+ species in
Experimental Test for the Oxidizing Power of the
+
MnIII(Me2EBC)(OH)2 provided the corresponding parameter
Manganese(IV) Complex. Our earlier study reported oxidation
by [MnIV(Me2EBC)(OH)2](PF6)2 of substrates of known BDECH
,
for the oxidizing power of MnIVdO group in
MnIV(Me2EBC)(O)(OH)+: BDEOH ) 84.3 kcal/mol. It is sig-
nificant that the two bond dissociation energies are similar in
magnitude (83 and 84.3 kcal/mol), indicating that the hydrogen
giving the Polanyi correlation presented graphically in Figure
3.10 Rate constants were determined in 4:1 acetone/water
solutions spectrophotometrically as the purple Mn(IV) changed
(17) (a) Bordwell, F. G.; Cheng, J. P.; Harrelson, J. A. J. Am. Chem. Soc.
1988, 110, 1229. (b) Mayer, J. M. Acc. Chem. Res. 1998, 37, 441. (c)
Cook, G. K.; Mayer, J. M. J. Am. Chem. Soc. 1994, 116, 1855.
(18) (a) Baldwin, M. J.; Percoraro, V. L. J. Am. Chem. Soc. 1996, 118,
11325. (b) Caudle, M. T.; Percoraro, V. L. J. Am. Chem. Soc. 1997,
119, 3415.
(20) (a) Carrell, T. G.; Bourles, E.; Lin, M.; Dismukes, G. C. Inorg. Chem.
2003, 42, 2849. (b) Tommos, C.; Babcock, G. T. Acc. Chem. Res.
1998, 31, 18. (c) Vrettos, J. S.; Brudvig, G. W. Compr. Coord. Chem.,
II 2004, 8, 507. (d) Pecoraro, V. L.; Baldwin, M. J.; Caudle, M. T.;
Hsieh, W.; Law, N. A. Pure Appl. Chem. 1998, 70, 925.
(21) Schlodder, E.; Witt, H. T. J. Biol. Chem. 1999, 274, 30387.
(22) Rush, J. D.; Bielski, B. H. J Inorg. Chem. 1995, 34, 5832.
(23) Gupta, R.; Borovik, A. S. J. Am. Chem. Soc. 2003, 125, 13234.
(24) Moyer, B. A.; Meyer, T. J. Inorg. Chem. 1981, 20, 436.
(19) (a) Gardner, K. A.; Mayer, J. M. Science 1995, 269, 1849. (b) Mayer,
J. M. Acc. Chem. Res. 1998, 31, 441. (c) Feng, Y.; Gunnoe, T. B.;
Grimes, T. V.; Cundari, T. R. Organometallics 2006, 25, 5456.
9
16248 J. AM. CHEM. SOC. VOL. 130, NO. 48, 2008

Oxidative Reactivity among Transition Metal Moieties
A R T I C L E S
Table 1. pKa Values in the Reported Monomeric Redox Metal
Table 2. The Second Order Rate Constants for Hydrogen
2+
MnIV(Me2EBC)(O)2 at pH 4.0 and pH 13.4a
Complexes
Abstraction from Substrates with MnIV(Me2EBC)(OH)2 and
complexa
pKa1
pKa2
medium
ref
BDECH
kcal/mol
k2OH at pH
4.0 M-1 s-1
k2OXO at pH
13.4 M-1 s-1
-
substrate
k2OXO/k2OH
HMnO4
7.4
28.3
25.0
aqueous
DMSO
DMSO
aqueous
aqueous
acetonitrile
DMSO
22
23
23
24
16
25
26a
[MnIIIH31(OH)]-
[FeIIIH31(OH)]2-
xanthene
1,4-cyclohexadiene
9,10-dihydroxyanthracene 78
fluorene 80
75.5 0.00107
0.048
0.0159
0.01496
44
44
43
36
3.64 × 10-4
[(bpy)2(py)RuIVOH2]3+
[MnIV(Me2EBC)(OH)2]2+
[MnII(PY5)(OH2)]2+
[FeII(PY5)(OH2)]2+
0.85
6.86
13
8.1
76
3.52 × 10-4
10
∼2.5 × 10-4 0.00912
a Reaction conditions: solvent, acetone/water (ratio 4:1), initial
concentration of substrate, 40 mM, initial concentration of
manganese(IV) complex, 4 mM, at temperature 288 K.
a H31: tris[(N′-tert-butylureaylato)-N-ethyl)]aminato. PY5: 2,6-bis(bis-
(2-pyridyl)methoxymethane)pyridine.
into the red-brown of Mn(III), revealing the logarithmic
dependence on the strength of the targeted CH bond, as
expected. The rates varied from 1.8 × 10-3 M-1 s-1 for xanthene
with a BDE of 75.5 kcal/mol to 2.6 × 10-5 M-1 s-1 for fluorene
with a BDE of 80 kcal/mol. Diphenylmethane with a BDE of
82 kcal/mol showed zero reaction after over 19 days in stirred
solution. The observation of the termination of the hydrogen
abstraction reaction is rare,19c presumably because unfavorable
1-electron processes (e.g., high energy hydrogen abstractions)
are often driven by irreversible follow-up reactions. In addition,
a deuterated analogue of 9,10-dihydroanthracene (DHA-d4)
displays a significant rate decrease relative to the normal 9,10-
dihydroanthracene (DHA-h4). The ratio of the rate constants is
3.27 at 297 K, implying that the hydrogen abstraction by
MnIII(Me2EBC)(OH)2+ is the rate determining step (kobs(DHA-
d4) ) 7.79 × 10-6 s-1; kobs(DHA-h4) ) 2.55 × 10-5 s-1).
Kinetic Rate Constants for MnIV-OH in MnIV(Me2EBC)-
(OH)22+ and MnIVdO in MnIV(Me2EBC)(O)2. Conveniently, the
that the concentration of MnIV(Me2EBC)(O)(OH)+ is at the
97-98% level. Further, about 2-3% of the manganese is
estimated to be present as either the dihydroxo complex,
MnIV(Me2EBC)(OH)22+, or the dioxo complex, MnIV(Me2EBC)-
(O)2. Accordingly, preliminary rate studies were conducted at
pH values of 4.0 and 8.4 to demonstrate the difference in
reactivities of the two catalyst forms and HCl or NaOH was
used to adjust the initial pH. Buffers were not used in these
preliminary studies because of the complications they introduce.
Specifically, phosphate and carbonate buffers complex with the
metal ions, displacing OH groups, while some organic buffers
can also coordinate to the metal ion and some are subject to
oxidation. Since the manganese(IV) complexes decompose
slowly in the acidic solution and are quite unstable in basic
solution, the calculated rate constants have been corrected for
the separately determined catalyst decomposition rates in the
absence of the substrate. The hydrogen abstraction reaction using
9,10-dihydroanthracene (DHA) as the substrate proceeds much
faster in base than in acid. The pseudo-first-order rate constant
at pH ) 8.4 is k1 ) 6.64 × 10-4 s-1, at 297 K, while that in
acidic solution (pH ) 4.0) is k1 ) 4.39 × 10-5 s-1, also at 297
K. These results demonstrate the fact that the MnIVdO moiety
performs hydrogen abstraction more rapidly than does
MnIV-OH, even though there are two OH groups in the
dihydroxo complex and only one oxo group in the oxo-hydroxy
compound.10 These rates differ by an average factor of 15 and
this directs attention toward the relative intrinsic efficacies of
oxidizing species differing only in the level of their protonation
in such fundamental chemical processes as hydrogen abstraction.
Accepting the result that hydrogen abstraction from 9,10-
dihydroanthracene by MnIV(Me2EBC)(O)(OH)+ proceeds more
than 15 times faster than that by MnIV(Me2EBC)(OH)22+, it is
to be expected that, the reaction rate with a catalyst having two
MnIVdO units (that is MnIV(dO)2) would be much faster than
the rate for a catalyst having one MnIV-OH and one MnIVdO
unit (i.e., MnIV(O)(OH)). The hydrogen abstraction rate for
MnIV(Me2EBC)(O)(OH)+ should be a combination of the rates
of both an MnIVdO and an MnIV-OH group. To explore the
reactivity difference between MnIVdO and MnIV-OH groups
further, hydrogen abstraction experiments were conducted at
the initial pH values of 4.0 and 13.4, respectively, and the second
order rate constants are listed in Table 2. As stated earlier, at
two
deprotonation
constants
for
[MnIV(Me2EBC)-
(OH)2](PF6)2, pKa1 ) 6.86 and pKa2 ) 10, fall in the easily
controllable pH range, and this facilitates the generation of
activated catalyst forms containing only MnIVdO and MnIV-OH
moieties, respectively. The data in Table 1 suggest the rarity of
this useful set of pKa values. Importantly, this manganese(IV)
complex is sufficiently stable in water to permit useful physi-
cochemical measurements, including acid-base titrations, and
electrochemical and kinetic measurements, providing a unique
opportunity to address the open issue about the reactivity
differences between metal oxo and metal hydroxo oxidants,
when all else is constant. This discussion concerns the difference
in hydrogen abstracting abilities of MnIV-OH and MnIVdO
groups.
In preliminary work, pH values of 4.0 and 8.4 were chosen
to reveal the relative reactivities of two distinct catalyst species
capable of performing hydrogen abstractions, MnIV(Me2EBC)-
2+
(OH)2 and MnIV(Me2EBC)(O)OH+. These pH values were
chosen on the basis of a well-defined pKa value of 6.86 relating
the two species, whereas only approximate values of ∼2 and
∼10 were available for the equilibria under more acidic or more
basic conditions. For pKa ≈ 2, the third proton is added to
2+
MnIV(Me2EBC)(OH)2 converting one bound OH to a water
molecule and the pKa ≈ 10 involves removing the second proton
from MnIV(Me2EBC)(OH)2 to form the dioxo complex,
the pH of 4.0, the MnIV(Me2EBC)(OH)2 moiety is almost
2+
2+
MnIV(Me2EBC)(O)2. Although the calculations are rendered
highly approximate because the oxidations were studied in 4:1
acetone/water in order to reliably dissolve the substrates, the
range of acidities should be reasonably represented. For water
at a pH of 4, MnIV(Me2EBC)(OH)22+ is expected to be present
at g99%, and the concentrations of both MnIV(Me2EBC)-
(OH2)OH3+ and MnIV(Me2EBC)(O)OH+ are estimated to be less
than 1%. Similarly, calculations in basic media pH 8.4 suggest
certainly the dominant species and is estimated to be present at
g99%. Correspondingly, since the second pKa, which indicates
removal of the last proton from MnIV(Me2EBC)(O)(OH)+ to
form MnIV(Me2EBC)(O)2, is 10, at pH 13.4, the fully deproto-
nated species, MnIV(Me2EBC)(O)2, is confidently estimated as
g99%. Therefore, the reaction rates presented in Table 2 should
clearly reveal the difference in hydrogen abstracting abilities
between the two active manganese(IV) species, that is, MnIVdO
9
J. AM. CHEM. SOC. VOL. 130, NO. 48, 2008 16249

A R T I C L E S
Yin et al.
vs MnIV-OH, providing the first direct measure of this important
difference in reactivities. At 288 K, as the BDECH value of the
substrates increases from 75.5 to 80 kcal/mol, the second order
rate constants drop from 0.00107 M-1 s-1 to ∼2.5 × 10-4 M-1
s-1 for MnIV-OH in MnIV(Me2EBC)(OH)22+, and for Mn+dO in
MnIV(Me2EBC)(O)2 it drops from 0.048 to 0.00912 M-1 s-1. Also,
as shown in Table 2, the ratio of the more rapid MnIVdO rate to
that of the slower MnIV-OH ranges from 36 to 44. From this result,
one might suspect that the relative reactivities of these two
functional groups (M-OH and MdO), as measured by rates of
reaction, may vary by between 1 and 2 orders of magnitude, but,
probably, not by a great deal more.
It is interesting to compare the results of these simple rate
measurements with the calculated hydrogen abstracting abilities
of the two complexes, MnIV(Me2EBC)(O)(OH)+ and
MnIV(Me2EBC)(OH)22+, recalling that they operate at BDEOH
strengths of 84.3 and 83 kcal/mol, respectively. Viewing this
small difference in enthalpies (1.3 kcal) as if it were a difference
in the free energy of activation, the difference in rates of
hydrogen abstraction between the two functional groups should
only be a factor of a little less than 10 at room temperature.
The corresponding variation in rates was found to be 15-fold
for the complexes whose physical constants were used for the
calculations of BDEOH values. We suggest that the agreement
is surprisingly good.
Figure 4. Eyring plot for hydrogen abstraction from 9,10-dihydroanthracene
with MnIV(Me2EBC)(O)2 at 288 K, initial pH ) 13.4.
Since the redox potential of the MnIV(Me2EBC)(O)(OH)+/
MnIII(Me2EBC)(O)(OH) couple is not available, the calculated
BDEOH value of the O-H bond associated with the
MnIV(Me2EBC)(O)2 oxidant was not accessible. However,
experimentally, the rate ratio for hydrogen abstraction between
MnIVdO and MnIV-OH, k2OXO/k2OH, is ∼43 for 9,10-dihy-
droanthracene as the substrate, a number that indicates a
difference of ∼2.3 kcal/mol between the activation free energies
for hydrogen abstraction, suggesting that the BDEOH associated
with the reaction of MnIV(Me2EBC)(O)2 might be 85 or 86 kcal/
mol. In seeking to observe the expected greater oxidizing
capability of this source of MnIVdO groups, attempts were made
to abstract hydrogen from diphenyl methane (BDECH value of
82 kcal/mol) at pH 8.4, where MnIV(Me2EBC)(O)(OH)+ is the
dominant form of the reagent and at pH 13.4, where
MnIV(Me2EBC)(O)2 dominates, in acetone/water(ratio 4:1) at
room temperature. No hydrogen abstraction reaction was
observed in either case after the reaction ran overnight, with
stirring. The catalyst was completely degraded to the previously
studied manganese(III) derivatives in this reaction time, as
expected, because of the previously reported instability of the
catalyst in media at high pH.16 At this point it must be concluded
that the substantial errors and approximations in both the
thermodynamic model and the experiments reported here are
too great to characterize the two MnIV(OH)2 derived functional
groups further with the available information.
Figure 5. Eyring plot for hydrogen abstraction from 9,10-dihydroanthracene
2+
with LMnIV(OH)2 at 288 K, initial pH ) 4.0.
is also in agreement with hydrogen abstraction from substrate
by MnIV-OH in the rate determining step.
Activation Parameters for MnIV-OH and MnIVdO. The data
presented above have revealed that the kinetic rates between
MnIV-OH and MnIVdO units are distinctly different, although the
thermodynamic oxidizing power of MnIV-OH and MnIVdO units
are very similar as calculated (83.0 vs 84.3 kcal/mol). To further
examine the reactivity differences between the MnIV-OH and
MnIVdO groups, the enthalpy (∆Hq) and entropy (∆Sq) of
activation have been determined for hydrogen abstraction from
9,10-dihydroanthracene by MnIV(Me2EBC)(OH)22+ at the initial
pH of 4.0 and by MnIV(Me2EBC)(O)2 at the initial pH of 13.4,
over the temperature range of 288-318 K, (Figures 4 and 5).
Because of the solubility limitations of 9,10-dihydroanthracene in
acetone/water (ratio 4/1), the rapid degradation of the manga-
nese(IV) species to the manganese(III) derivative16 in very strong
base, and the rapid hydrogen abstraction rate of the Mn(IV) moiety,
the measurements of the activation parameters for MnIVdO in base
were limited to the relatively narrow temperature range from 288
to 308 K. Also, because of the same low solubility of 9,10-
dihydroanthracene, the slow hydrogen abstraction rate of MnIV-OH
in acid, and the relatively low boiling point of acetone (56 °C),
the measurements of the activation parameters for MnIV-OH were
limited to the temperature range from 288 to 318 K. The data are
summarized in Table 3 along with the literature data for other
In addition, the kinetic isotope effect for hydrogen abstraction
from 9,10-dihydroanthracene (DHA-h4) and deuterated 9,10-
dihydroanthracene (DHA-d4) with MnIV(Me2EBC)(O)2 at pH
13.4 supports hydrogen abstraction by MnIVdO as the rate
determining step with a KIE value of 3.78 at 288 K (the pseudo-
first-order rate constant for DHA-h4, k1 ) 9.37 × 10-4 s-1, and
for DHA-d4, k1 ) 2.48 × 10-4 s-1). The hydrogen abstraction
from DHA-d4 with MnIV(Me2EBC)(OH)2 at pH 4.0 was too
2+
slow at 288 K and was lost in the noise of the blank experiment,
thus the accurate KIE value is not available, however, the KIE
value of 3.27 at 297 K from the neutral acetone/water mixture
9
16250 J. AM. CHEM. SOC. VOL. 130, NO. 48, 2008

Oxidative Reactivity among Transition Metal Moieties
A R T I C L E S
A r f B + H+
B r f C + H+
pKa1 ) 6.86
pKa2 ) 10
(10)
(11)
(12)
appropriate monomeric metal complexes. As shown in Table 3,
the ∆Hq and ∆Sq values for the two manganese(IV) complexes
studied here fall in the range of other reported catalytic metal
complexes.
Superficially,
the
∆Hq
values
for
[A] + [B] + [C] ) [MnIV]
MnIV(Me2EBC)(OH)22+ and MnIV(Me2EBC)(O)2 are close in value,
13.1 ( 0.7 vs 12.1 ( 1.8 kcal/mol, whereas the ∆Sq values appear
different, -35.0 ( 2.2 vs -30.3 ( 5.9 cal K-1 mol-1. However,
the uncertainties of these thermal parameters are relatively large
and the relative contributions of the two terms to the free energies
of activation are comparable. A sizable ∆Sq difference could
indicate that the transition state for hydrogen abstraction from
MnIV-(OH)22+ requires a more ordered or rigid state, relative to
the ground state, than that involving MnIVdO. This could derive
from the necessity to rearrange and fix the positions of the protons
in the OH groups in the transition state. A second possibility may
involve more extensive ground-state solvation via hydrogen bond-
ing for the MnIV(OH)2 functional group. The overall perspective
is that the MnIV(dO)2 functional group undergoes reaction with
less constraint than is the case for the corresponding hydroxo
derivative. The calculated activation free energy (∆Gq) for hydrogen
abstraction from 9,10-dihydroanthracene by MnIVdO of
MnIV(Me2EBC)(O)2 at 298 K is +21.1 kcal/mol while for the
MnIV-OH unit in MnIV(Me2EBC)(OH)22+, it is +23.5 kcal/mol.
The difference between the two activation energies is 2.4 kcal/
mol, suggesting a 58 fold reaction rate ratio to which the activation
enthalpy contributes an approximate 5.4-fold rate ratio while
activation entropy contributes a ∼10.6-fold rate ratio. With an
uncertainty of at least a kilocalorie, this result includes the range
of experimental values observed for the rate parameters; that is, a
reaction rate ratio of 44 comes from a ∆ (∆Gq) value of 2.2 kcal/
mol.
Hydrogen abstraction from 9,10-dihydroanthracene with the
three active manganese(IV) species can be expressed by eq 7-9.
Here S is substrate, P is the product, and A, B, and C represent
MnIV(Me2EBC)(OH)2
,
MnIV(Me2EBC)(O)(OH)+,
and
2+
MnIV(Me2EBC)(O)2, respectively. The corresponding second
order rate constants of eqs 7-9 are k1, k2, and k3, and the link
connecting complexes A, B, and C is provided by the two
deprotonation constants, pKa1 and pKa2 (eqs 10 and 11). Under
the reaction conditions, the total initial concentration of the
manganese(IV) complexes was 4 mM, and the substrate was
present in excess (0.04 M), imposing pseudo-first-order condi-
tions. The overall pseudo-first-order rate law for the disappear-
ance of manganese(IV) species is expressed as
-d[MnIV] ⁄ dt ) {k1[A] + k2[B] + k3[C]}[S]
(13)
Incorporation of the equilibria, material balance, and [H+]
(overall, eqs 7-13) leads to
k1[H+] + k2Ka1[H+] + k3Ka1Ka2
[H+]2 + Ka1[H+] + Ka1Ka2
d[MnIV]
-
)
[S][MnIV]
dt
(14)
(15)
Thus:
k1[H+]2 + k2Ka1[H+] + k3Ka1K
kobs
)
a2[S]
[H+]2 + Ka1[H+] + Ka1Ka2
Rate Law and Kinetics of Hydrogen Abstraction by Three
Active Manganese(IV) Species. Since the composition of the
mixture of three active manganese forms, including MnIV(Me2EBC)-
(OH)22+, MnIV(Me2EBC)(O)(OH)+ and MnIV(Me2EBC)(O)2, in the
solution is highly pH dependent (eq 1 and eq 2), it is essential to
investigate the influence of proton concentration on the reaction
rate. The kinetic correlation between the [H+] and the pseudo-
first-order rate constant (kobs) was studied with 9,10-dihydroan-
thracene as the substrate (eqs 4 and 5). For this substrate, the first
hydrogen abstraction step, which generates the radical intermediate,
is the rate determining step and the overall reaction is second order,
with first-order dependences on both the substrate and the man-
ganese(IV). The second, follow-up step is fast compared to the
initial hydrogen abstraction so that eqs 4 and 5 can be abbreviated
as eq 6.
Under the limiting conditions, at pH 4.0, the percent of
MnIV(Me2EBC)(OH)22+ exceeds 99%; k1[H+]2 . k2Ka1[H+] +
k3Ka1Ka2, and [H+]2 . Ka1[H+] + Ka1Ka2, thus, kobs ≈ k1[S].
Similarly, at pH 13.4, the presence of MnIV(Me2EBC)(O)2 is
greater than 99%; k3Ka1Ka2 . k1[H+]2 + k2Ka1[H+], and Ka1Ka2
. [H+]2 + Ka1[H+], thus kobs ≈ k3[S]. Figure 6 displays the
influence of the initial proton concentration (pH) in the reaction
mixture on the pseudo-first-order rate constant for hydrogen
abstraction from 9,10-dihydroanthracene (DHA) with the man-
ganese(IV) complex. The data represented by the black squares
is made up of experimental observed rate constants, and the
red dots represent calculated rate constants obtained by applying
eq 15, beginning with the limiting second order rate constants,
k1 ) 3.52 × 10-4, k2 ) 0.00384, and k3 ) 0.01496 M-1 s-1
(Table 3), and pKa1 ) 6.86 and pKa2 ) 10 (eqs 10 and 11). In
the range from weakly acidic to neutral pH, the two curves
overlap very well, supporting the expectation that eq 15
describes the actual reaction system; that is, abstraction of the
first hydrogen atom from the substrate by the manganese(IV)
Table 3. The Related Activation Parameters for Hydrogen
Abstraction from 9,10-Dihydroanthracene and Various Monomeric
Metal Complexes
activation enthalpy
∆Hq (kcal mol-1
activation entropy
∆Sq (cal K-1 mol-1
)
complexa
ref
)
[FeIII(Hbim)(H2bim)2]2+
[MnO4]-
11.6 ( 0.5
13.8 ( 1
13.2 ( 0.5
9.3 ( 0.5
13.1 ( 0.7
12.1 ( 1.8
-36 ( 5
-15 ( 7
26b
17b
26a
25
this work
this work
[FeIII(PY5)OH]2+
-26 ( 5
Mn(IV) + S f P + Mn(III)
(6)
(7)
[MnIII(PY5)OH]2+
[MnIV(Me2EBC)(OH)2]2+
[MnIV(Me2EBC)(O)2]
-36 ( 5
-35.0 ( 2.2
-30.3 ( 5.9
A + S f P + Mn(III)
B + S f P + Mn(III)
C + S f P + Mn(III)
k1
k2
k3
(8)
(9)
a H2bim: 2,2′-bi-1,4,5,6-tetrahydropyrimidine. PY5: 2,6-bis(bis(2-pyridyl)-
methoxymethane)pyridine.
9
J. AM. CHEM. SOC. VOL. 130, NO. 48, 2008 16251

A R T I C L E S
Yin et al.
rate constants listed in eqs 16-18, increase from 1.26 × 10-6
for k1obs, to 1.19 × 10-4 for k2obs to 3.36 × 10-4 s-1 for k3obs
.
For the initial concentration of 0.04 M 9,10-dihydroanthracene
(DHA), these intercepts represent 8%, 43%, and 36% of the
total kobs, values, respectively. Such big intercepts for k2obs and
k3obs cannot be regarded as experimental error, clearly indicating
a growing misfit of the chemistry and the assumed rate law.
The extreme deviation at high pH values is consistent with the
known slow decomposition of the deprotonated species,
MnIV(Me2EBC)(O)(OH)+ and MnIV(Me2EBC)(O)2, in basic
media, a supposition supported by the fact that the deviation is
independent of substrate composition. Because the degradation
of the manganese(IV) oxo complex at pH 13.4 is relatively slow
in the absence of substrate (kobs ) ∼2 × 10-5 s-1; corrected on
the basis of the blank experiment performed in the absence of
substrate), as shown in blank experiments, the large intercept
(3.36 × 10-4 s-1) for the k3obs plot represents a relatively rapid
reduction of manganese(IV) that has to be explained. One
possibility is that the degradation of manganese(IV) is acceler-
ated, or “catalyzed”, by radical intermediates generated in the
course of the catalytic oxidation of the substrate by manga-
nese(IV) (for example, eq 4). The small steady state amounts
of radical intermediates may initiate a rapid degradation of
manganese(IV), a complicated process that has been described
elsewhere.16 The detailed investigation of the degradation of
the manganese(IV) complex in base is beyond the scope of this
study.
Figure 6. The influence of pH dependence of the pseudo-first-order rate,
kobs, for hydrogen abstraction from 9,10-dihydroanthracene (40 mM) by
the manganese(IV) complex (4 mM). Solvent: 4:1 acetone/water; 288 K.
Perspective. The difference between the MnIV-OH and
MnIVdO functional groups in this manganese(IV) complex
obviously derives from the dissociation of a proton from the
same manganese(IV) complex. Since proton transfer is very
rapid in polar media, especially in aqueous media, the equilib-
rium between the MdO and M-OH functional groups with
the same metal ion in the same oxidation state, and the same
ligand, must be very facile. Further, this property of proton
lability must carry over to the active centers of some redox
enzymes. On the other hand, the chemical literature chooses to
distinguish between and select each of these groups Mn+dO,2a,c,d
and Mn+-OH,4a-c,8a,10,25,26 as the key active intermediate in
specific series of oxidation events in chemical or in biological
systems. For example, in horseradish peroxidases, it has been
argued that the relatively unreactive compound II involves the
FeIV-OH moiety in performing hydrogen abstractions.27 The
discoveries presented here on the active species in such systems
provide an enhanced focus on the immediate issue of distin-
guishing the reactivity difference between certain metal oxo and
metal hydroxo moieties, Mn+d)O and Mn+-OH. The study
of this manganese(IV) complex has revealed that, given the same
metal ion, the same ligand, and the same oxidation state, the
oxidizing powers of metal oxo and metal hydroxo functions
are not very different. As one might expect, the rates of their
reactions (at least for hydrogen abstraction) are, in a relative
sense, distinctly different, and the Mn+dO group can perform
oxidations substantially more rapidly than the corresponding
Mnn+-OH group. Chemists and biologists should bear this in
mind when considering the mechanisms mediated by the redox
Figure 7. Correlation of substrate concentration with the pseudo-order rate
constant for hydrogen abstraction from 9,10-dihydroanthracene by man-
ganese(IV) complex at pH 4.0 (for k1), 8.4 (for k2) and 13.4 (for k3). Sovent:
acetone/water (ratio 4:1); [MnIV] ) 4 mM, 288 K.
oxidant determines the reaction dynamics for those values of
the variables. However, at pH values more positive than the
neutral point, the experimental and calculated curves deviate
dramatically. This deviation is apparent also in the correlation
of the second order rate constants, k2 and k3, with the
experimental pseudo-first-order rate constant (kobs) data (Figure
7, eqs 16-18, see Supporting Information for details).
k1obs ) 1.26 × 10-6 + 3.52 × 10-4[DHA],
k1 ) 3.52 × 10-4 M-1 s-1 (16)
k2obs ) 1.19 × 10-4 + 0.00384[DHA],
k2 ) 0.00384 M-1 s-1 (17)
(25) Goldsmith, C. R.; Cole, A. P.; Stack, T. D. P. J. Am. Chem. Soc.
2005, 127, 9904.
k3obs ) 3.36 × 10-4 + 0.01496[DHA],
k3 ) 0.01496 M-1 s-1 (18)
(26) (a) Goldsmith, C. R.; Stack, T. D. P. Inorg. Chem. 2006, 45, 6048.
(b) Roth, J. P.; Mayer, J. M. Inorg. Chem. 1999, 38, 2760.
(27) (a) Nilsson, K.; Hersleth, H.; Rod, T. H.; Andersson, K. K.; Ryde, U
Biophys. J. 2004, 87, 3437. (b) Frey, P. A.; Hegeman, A. D. Enzymatic
Reaction Mechanisms; Oxford University Press: New York, 2007; p
208.
The intercepts from the graph of k1obs, k2obs to k3obs plotted
against the concentration of substrate, using the second order
9
16252 J. AM. CHEM. SOC. VOL. 130, NO. 48, 2008

Oxidative Reactivity among Transition Metal Moieties
A R T I C L E S
active transition metal ions in various enzymes, including the
oxidases, dehydrogenases, and electron transfer enzymes.
the MnIVdO group proceeds ∼40 times faster than that of
MnIV-OH with the same substrate. The difference in reactivity
in this case is attributed to differences in both activation
entropies and enthalpies, for the case of hydrogen abstraction
from 9,10-dihydroanthracene. The distinctive reactivity differ-
ence revealed by this exemplary manganese(IV) activated
catalyst may provide new insights into oxidation events occur-
ring in nature and provide guidance in the design of new
oxidation catalysts for laboratory and industrial use.
Conclusions
A recently synthesized and characterized manganese(IV)
complex having two hydroxide ligands, [MnIV(Me2EBC)-
(OH)2](PF6)2, has been studied to investigate the difference in
oxidative reactivity of the metal oxo and metal hydroxo
functional groups for the same metal ion with the same ligands
and the same oxidation states, that is, LMn+dO and LMn+-OH.
For the systems under investigation, the calculated BDEOH
values for the MnIII-(HO-H) and MnIII-OH moieties in the
manganese(III) complexes, that are produced by hydrogen
abstraction from a substrate, show that, thermodynamically,
MnIV-OH and MnIVdO have very similar oxidizing powers
(83.0 vs 84.3 kcal/mol). Close agreement was found in experi-
ments using batch reactions, conducted for well defined reaction
times and identical conditions, performing oxidations that
involve the abstraction of hydrogen atoms from selected
substrates having known BDECH values. Neither of these
functional groups will oxidize substrates having BDECH values
of 82 kcal/mol or higher. However, hydrogen abstraction by
Acknowledgment. Support by the Procter and Gamble Com-
pany is deeply appreciated, and we also acknowledge the National
Science Foundation Engineering Research Center Grant (EEC-
0310689) for partial support. GC-MS analysis of reaction products
was performed by R. C. Drake.
Supporting Information Available: Kinetic results of hydro-
gen abstraction from different substrates under various reaction
conditions. This material is available free of charge via the
Internet at http://pubs.acs.org.
JA804305X
9
J. AM. CHEM. SOC. VOL. 130, NO. 48, 2008 16253