Correlation between structural, spectroscopic, and reactivity properties within a series of structurally analogous metastable manganese(III)-alkylperoxo complexes

Article abstract of DOI:10.1021/ja308915x

Manganese-peroxos are proposed as key intermediates in a number of important biochemical and synthetic transformations. Our understanding of the structural, spectroscopic, and reactivity properties of these metastable species is limited, however, and correlations between these properties have yet to be established experimentally. Herein we report the crystallographic structures of a series of structurally related metastable Mn(III)-OOR compounds, and examine their spectroscopic and reactivity properties. The four reported Mn(III)-OOR compounds extend the number of known end-on Mn(III)-(η¹-peroxos) to six. The ligand backbone is shown to alter the metal-ligand distances and modulate the electronic properties key to bonding and activation of the peroxo. The mechanism of thermal decay of these metastable species is examined via variable-temperature kinetics. Strong correlations between structural (O-O and Mn···N^py,quin distances), spectroscopic (E(π_v(O-O) → Mn CT band), ν_O-O), and kinetic (ΔH^a and ΔS^a) parameters for these complexes provide compelling evidence for rate-limiting O-O bond cleavage. Products identified in the final reaction mixtures of Mn(III)-OOR decay are consistent with homolytic O-O bond scission. The N-heterocyclic amines and ligand backbone (Et vs Pr) are found to modulate structural and reactivity properties, and O-O bond activation is shown, both experimentally and theoretically, to track with metal ion Lewis acidity. The peroxo O-O bond is shown to gradually become more activated as the N-heterocyclic amines move closer to the metal ion causing a decrease in π-donation from the peroxo π_v(O-O) orbital. The reported work represents one of very few examples of experimentally verified relationships between structure and function.

Full text of DOI:10.1021/ja308915x

Article
pubs.acs.org/JACS
Correlation Between Structural, Spectroscopic, and Reactivity
Properties Within a Series of Structurally Analogous Metastable
Manganese(III)−Alkylperoxo Complexes
Michael K. Coggins,† Vlad Martin-Diaconescu,⧧ Serena DeBeer,⧧,§ and Julie A. Kovacs*,†
†The Department of Chemistry, University of Washington, Box 351700, Seattle, Washington 98195-1700, United States
⧧Max-Planck-Institute for Chemical Energy Conversion, Stiftstrasse 34-36, 45470 Mulheim an der Ruhr, Germany
̈
§Department of Chemistry and Chemical Biology, Cornell University, Ithaca, New York 14853, United States
S
* Supporting Information
ABSTRACT: Manganese−peroxos are proposed as key
intermediates in a number of important biochemical and
synthetic transformations. Our understanding of the structural,
spectroscopic, and reactivity properties of these metastable
species is limited, however, and correlations between these
properties have yet to be established experimentally. Herein
we report the crystallographic structures of a series of
structurally related metastable Mn(III)−OOR compounds,
and examine their spectroscopic and reactivity properties. The
four reported Mn(III)−OOR compounds extend the number
of known end-on Mn(III)−(η1-peroxos) to six. The ligand backbone is shown to alter the metal−ligand distances and modulate
the electronic properties key to bonding and activation of the peroxo. The mechanism of thermal decay of these metastable
species is examined via variable-temperature kinetics. Strong correlations between structural (O−O and Mn···Npy,quin distances),
spectroscopic (E(πv*(O−O) → Mn CT band), νO−O), and kinetic (ΔH⧧ and ΔS⧧) parameters for these complexes provide
compelling evidence for rate-limiting O−O bond cleavage. Products identified in the final reaction mixtures of Mn(III)−OOR
decay are consistent with homolytic O−O bond scission. The N-heterocyclic amines and ligand backbone (Et vs Pr) are found to
modulate structural and reactivity properties, and O−O bond activation is shown, both experimentally and theoretically, to track
with metal ion Lewis acidity. The peroxo O−O bond is shown to gradually become more activated as the N-heterocyclic amines
move closer to the metal ion causing a decrease in π-donation from the peroxo πv*(O−O) orbital. The reported work represents
one of very few examples of experimentally verified relationships between structure and function.
mononuclear Mn(III)−peroxos,32,34,38,41−43 five contain the
INTRODUCTION
■
peroxo in a side-on (η2) binding mode. The O−O bond
Manganese (Mn)-containing enzymes promote a number of
distances associated with the Mn(III)−(η2-peroxos) all fall
within a fairly narrow range (1.403(4)−1.428(7) Å). Not
surprisingly, the spectroscopic properties of these metastable
species are also relatively invariant.15,17,35,36,41−43 The only
structural evidence for an end-on (η1) peroxo binding mode
with mononuclear manganese(III) is for an alkylperoxo
compound.32 The O−O bond in this end-on Mn(III)(η1-
alkylperoxo)32 is significantly more activated (1.457(7) Å) than
the side-on Mn(III)−(η2-peroxos).34,38,41−43 More examples of
end-on Mn(III)(η1-peroxo) compounds are needed, however,
in order to establish a relationship between binding mode and
levels of O−O bond activation. Ultimately, we wish to
understand how structure influences reactivity and the factors
that favor O−O vs Mn−O bond cleavage and formation.
Previously, we reported the first example of a Mn(III)−
alkylperoxo complex, [MnIII(SMe2N4(QuinoEN))(OOtBu)]-
important biochemical1−14 and synthetic15−19 transformations
involving dioxygen and reduced derivatives thereof. Many of
these reactions, including two of the most fundamental
processes of life, photosynthetic H2O splitting7,20 and DNA
biosynthesis,1 are proposed to involve Mn−peroxo intermedi-
ates. These peroxo intermediates form either en route to more
reactive high-valent metal-oxo intermediates,1,21,22 or as a result
of O−O coupling.7,23 Alkylperoxo Mn−OOR intermediates are
proposed to play a key role in the reactions catalyzed by
manganese lipoxygenase3,18,24,25 and homoprotocatechuate 2,3-
dioxygenase enzymes,8,26,27 as well as in industrial pro-
cesses.16,28−31 With the exception of one system,14 biological
Mn−peroxo intermediates have yet to be observed. Our
understanding of the structural and electronic properties that
favor O−O bond formation vs O−O bond cleavage is
extremely limited for Mn. This is because very few Mn−
peroxos, biological or synthetic, have been character-
ized,14,15,17,32−43 due in part to their thermal and photo-
chemical instability. Of the six structurally characterized
Received: September 14, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(BPh4) (1a).32 This complex is high-spin (S = 2) and intensely
blue in color. A lower-resolution structure of an end-on
cumenyl peroxo derivative, [MnIII(SMe2N4(QuinoEN))-
(OOCm)](BPh4), was also included. Preliminary evidence
suggested that the thermal decomposition of these species
proceeded via a mechanism involving homolytic O−O bond
scission.32 This reaction pathway was unprecedented with
synthetic Mn−peroxos which are typically shown to undergo
heterolytic O−O bond cleavage.44 Our observations32 further
contrast with those made for high-spin Fe(III)−OOR
complexes, which are shown to preferentially undergo Fe−O
bond cleavage.45−48 With synthetic Fe(III)−OOR complexes, a
low-spin state is required in order for homolytic O−O bond
cleavage to occur.46,49 With end-on iron hydroperoxo
intermediates proton-assisted heterolytic O−O bond cleavage
has been shown to occur when the Fe(III)−OOH is low
spin,50,51 whereas heterolytic Fe−O bond cleavage occurs when
the Fe(III)−OOH is high spin.47
Herein we report the synthesis as well as spectroscopic and
structural characterization of a series of four new structurally
related high-spin (S = 2) manganese(III)−alkylperoxo
complexes. The ligand environment is shown to subtly vary
the extent of O−O bond activation in these complexes, and
theoretical calculations provide an explanation for these
observations. Variable-temperature kinetics studies are also
described, which allow us to investigate the mechanism of
thermal decay. Strong correlations between experimentally
measured and theoretically calculated structural, spectroscopic,
and kinetic parameters for these complexes provide compelling
evidence for rate-limiting O−O bond cleavage.
of the requisite Mn(II) precursor was prepared in CH2Cl2 (3−4 mL)
inside a glovebox. The resulting solution was transferred via gastight
syringe to a custom-made two-neck vial equipped with a stir bar and
septum cap and threaded dip-probe feed-through adaptor that had
previously been purged with argon. The solution was then cooled to
258 K using an ethylene glycol/dry ice bath. Tert-butyl hydroperoxide
(tBuOOH; 1.5 equiv from a 0.1 M anaerobic stock solution (CH2Cl2))
was added via syringe to a stirring Mn(II) solution, along with 2.0
equivalents of triethylamine. Scans were recorded at low temperatures
from 300 to 1000 nm in one minute intervals until changes to the
absorption spectra ceased for approximately 10 min.
Preparation of FT-IR Samples of [MnIII(SMe2N4(QuinoPN))-
(OOtBu)](PF6)·pentane (2a), [MnIII(SMe2N4(6-Me-DPEN))-
(OOtBu)](BPh4) (3a), [MnIII(SMe2N4(6-Me-DPPN))(OOtBu]-
(BPh4)·Et2O (4a). Solid material of 2a−4a was obtained from in situ
reaction mixtures via low-temperature precipitation. In a typical
experiment, a 20−35 mM supersaturated solution of the appropriate
Mn(II) precursor was prepared in MeCN (2−3 mL) in a glovebox.
The resulting solution was then transferred to an anaerobic vial, cooled
to −40 °C using a ethylene glycol/dry ice bath, followed by the
t
addition of 2.0 equiv of BuOOH from an anaerobic 1 M stock
solution (CH2Cl2). The resulting reaction mixture was allowed to stir
for 5−7 min at −40 °C. Pre-cooled Et2O (10−15 mL) was then
quickly added to the reaction mixture to promote the rapid
precipitation of solid material. The resulting solid was isolated via
filtration through a cold frit, mixed with cold Nujol, and immediately
analyzed by FT-IR. The identity of the sample was verified by
redissolving the solid material in cold CH2Cl2, and recording the UV/
vis spectra, which revealed spectral features identical to those observed
in an in situ low-temperature generated sample.
t
Determination of the Solution Magnetic Moment of Butyl-
Peroxos 2a−4a. Crystalline samples of 2a−4a were obtained by
layering cold pentane (or Et2O) onto a concentrated in situ CH2Cl2
reaction mixture generated at −40 °C and cooled to −80 °C.
Following diffusion of the pentane (or Et2O) into the CH2Cl2 the
resulting crystals were filtered through a cold frit, washed with pre-
cooled Et2O, weighed on an analytical balance in an NMR tube, and
then transported down to the NMR room. Once in the NMR room, a
known volume of pre-cooled (in a bed of dry ice) CD2Cl2 containing
TMS, was then added to the crystalline samples immediately prior data
collection. Magnetic moments obtained in this manner, (μeff = 4.78
BM (2a), 4.79 BM (3a), 4.89 BM (3b)) were close to the spin-only
value (4.90 BM) for high-spin (S=2) Mn(III), indicating that very
little, if any, decomposition had occurred over the course of the data
collection.
EXPERIMENTAL SECTION
■
General Methods. All reactions were performed under an inert
atmosphere in a glovebox, using standard Schlenk techniques, or using
a custom-made solution cell equipped with a threaded glass connector
sized to fit an ATR dip probe. Reagents purchased from commercial
vendors were of the highest purity and used without further
p u r i fi c a t i o n . 3 - M e t h y l - 3 - m e r c a p t o - 2 - b u t a n o n e ,
III
[MnII(SMe N4(QuinoEN))](BPh4) (1), [Mn (SMe2N4(QuinoEN))-
2
(OOtBu)](BPh4) (1a), [MnIII(SMe2N4(QuinoEN))(OOCm)](BPh4)
(1b), [MnII(SMe2N4(QuinoPN))](PF6) (2), [MnII(SMe2N4(6-Me-
DPPN))](BPh4) (4), Bu18O18OH, and Et2POtBu were synthesized
t
Ab Initio Density Functional Theory (DFT) Calculations. All
calculations were performed using the ORCA program package.57
Geometries were optimized based on the crystal structures, and a high-
spin state, using the TPSS functional and the def2-TZVP(-f) basis set
in combination with the def2-TZV/J auxiliary basis sets for the
Coulombic density fitting approximation.58−65 Scalar relativistic effects
were accounted for using the ZORA approximation and a conductor
like screening model was applied using dichloromethane as the solvent
of choice (COSMO, CH2Cl2 ε = 9.08). Minimum deviations from the
crystallographically determined geometries, bond angles and bond
distances were obtained when intramolecular van der Waals
interactions were accounted for using the vdw10 keyword. All
calculations were performed using an integration grid of 5 (Grid5)
for the complex and a grid of 7 for the Mn(III) metal center. Relaxed
surface scans were performed on the O−O bond length from 1.1 to
1.6 Å to explore the impact of the alkylperoxo bond length on the
metal ion’s coordination sphere.
as previously described.52−54 [MnII(SMe2N4(6-Me-DPEN))](BPh4)
(3) was synthesized as previously described, using NaBPh4 instead
of NaBF4.54 Acetonitrile (MeCN) and diethyl ether (Et2O) were
rigorously degassed and purified using solvent purification columns
housed in a custom stainless steel cabinet, and dispensed via a stainless
steel Schlenk line (GlassContour). Methanol (MeOH) and methylene
chloride (CH2Cl2) were distilled from magnesium methoxide and
calcium hydride, respectively, prior to use. IR spectra were recorded on
1
a Perkin-Elmer 1700 FT-IR spectrometer as nujol mulls. H NMR
spectra were recorded on a Bruker AV 301 FT-NMR spectrometer and
referenced to residual solvent. Magnetic moments (solution state)
were obtained using the Evans method as modified for super-
conducting solenoids.55 Temperatures were obtained using Van Geet’s
method.56 Electronic absorption spectra were recorded using a Varian
Cary 50 spectrophotometer equipped with a fiber optic cable
connected to a “dip” ATR probe (C-technologies) with a custom-
built two-neck solution sample holder equipped with a threaded glass
connector. Mass spectra data were recorded on either a Bruker Esquire
liquid chromatograph − ion trap mass spectrometer or Hewlett-
Packard 5971A gas chromatograph − mass spectrometer. X-ray
diffraction data were collected on a Bruker APEX II single crystal X-ray
diffractometer with Mo-radiation.
X-ray Crystal Structures of tButyl Peroxos 1a-4a, and
Cumenyl Peroxo 3b. A blue prism of 2a immersed in oil with
dimensions 0.15 × 0.15 × 0.05 mm3 was mounted on a glass capillary.
Data were collected at −173 °C. The crystal-to-detector distance was
40 mm, and the exposure time was 10 s per degree for all sets. The
scan width was 0.5°. Data collection was 100% complete to 25° in ϑ. A
total of 49,617 merged reflections were collected covering the indices h
= −12 to 12, k = −17 to 17, and l = −28 to 28; 6557 reflections were
Monitoring the Reaction of 2−4 with tBuOOH via Electronic
Absorption Spectroscopy. In a typical reaction, a 1−2 mM solution
B
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
t
t
Table 1. Crystal Data for [MnIII(SMe N4(QuinoPN))(OO Bu)](PF6)·C5H12 (2a), [MnIII(SMe N4(6-Me-DPEN))(OO Bu)](BPh4)
2
2
(3a), [MnIII(SMe N4(6-Me-DPEN))(OOCm](BPh4) (3b), [Mn (S N4(6-Me-DPPN))(OO Bu](BPh4)·Et2O (4a)
III Me2
t
2
2a
3a
3b
4a
formula
MW
C69H92F12Mn2N8P2S2
1561.45
100(2)
C53H68BMnN4O3S
906.92
C56H64.65BCl1.34Mn N4O2.33
976.36
S
C50H60BMnN4O2S
846.83
T, K
100(2)
orthorhombic
17.265(2)
25.499(3)
11.2048(14)
90
100(2)
100(2)
a
unit cell
a, Å
monoclinic
10.5559(8)
14.0937(10)
23.9583(16)
90
triclinic
triclinic
12.8928(18)
14.433(2)
15.069(3)
71.085(10)
77.463(10)
82.463(9)
2583.7(7)
2
9.5420(7)
14.5846(11)
16.4547(13)
77.689(4)
83.682(4)
78.144(4)
2184.3(3)
2
b, Å
c, Å
α, deg
β, deg
γ, deg
V, Å3
Z
94.560(4)
90
90
90
3553.0(4)
2
4932.7(11)
4
d(calc), g/cm3
1.460
1.221
1.255
1.288
space group
P21/n
Pna21
P1
P1
̅
̅
b
R
0.0688
0.0696
0.1004
0.2391
0.952
0.0667
0.1786
1.008
c
Rw
0.2088
0.1246
GOF
1.013
1.000
a
b
c
2
In all cases: Mo Kα(λ = 0.71070 Å) radiation. R = ∑||Fo| − |Fc||/∑|Fo|. Rw = [∑w(|Fo| = |Fc|)2/∑wFo ]1/2, where w−1 = [σcount2 + (0.05F2)2]/4F2.
symmetry independent, and the Rint = 0.0875, indicating the data was
slightly less than average. Indexing and unit cell refinement indicated a
monoclinic P lattice, and the space group was found to be P21/n (No.
l4).
(SHELXS, SIR97) produced a complete heavy atom phasing model
consistent with the proposed structure. The structure was completed
by difference Fourier synthesis with SHELXL97. Scattering factors
were obtained from Waasmair and Kirfel.66 Hydrogen atoms were
placed in geometrically idealized positions and constrained to ride on
their parent atoms with C−H distances in the range 0.95−1.00 Å.
Isotropic thermal parameters (Ueq) for the hydrogens were fixed such
that they were 1.2Ueq of their parent atom Ueq for CH’s and 1.5Ueq of
their parent atom Ueq in case of methyl groups. All non-hydrogen
atoms were refined anisotropically by full-matrix least-squares. Crystal
data for 2a, 3a, 3b, and 4a are provided in Table 1, and selected
metrical parameters are provided in Table 3.
Variable-Temperature Kinetics and Reactivity of Alkyl
Peroxos 1a, 2a, 3a, 3b, and 4a. In a typical experiment, 4 mL
(CH2Cl2) of a 1−2 mM solution of a Mn(II) complex was prepared
under an inert atmosphere in a glovebox. The resulting solution was
then transferred via gastight syringe to a custom-made two-neck vial
equipped with a stir bar and septum cap and a threaded dip-probe
feed-through adaptor that had previously been purged with argon. The
solution was brought to the desired temperature using a low-
temperature bath (258 K (ethylene glycol/dry ice), 268 K (NaCl/ice
water), 273 K (ice water)). Gentle stirring was maintained throughout
the course of each experiment to maintain solution homogeneity. An
immediate production of the corresponding Mn(III)−alkylperoxo
intermediate as indicated by its UV/vis absorption spectrum was
caused by the addition of 1.5 equiv of tBuOOH in a single addition to
the Mn(II) solution. Decay of the respective intermediate was
monitored at 592 nm (1a), 580 nm (2a), 594 nm (3a), 590 nm (3b),
or 450 nm (4a) once formation of the corresponding Mn(III)−OOR
complex was complete. UV/vis scans were recorded at 1 min intervals
until no further change in absorbance intensity occurred for at least 10
min. First-order rate constants were calculated by plotting ln[(Ax −
A∞)/(A0 − A∞)] vs time. Experiments were conducted in at least
triplicate under any given set of conditions.
A faint blue prism of 3a immersed in oil with dimensions 0.15 ×
0.10 × 0.05 mm3 was mounted on a glass capillary. Data were collected
at −173 °C. The crystal-to-detector distance was 40 mm and the
exposure time was 120 s per degree for all sets. The scan width was
0.5°. Data collection was 98.9% complete to 25° in ϑ. A total of 76,962
merged reflections were collected covering the indices h = −20 to 20, k
= −30 to 30, and l = −13 to 13; 4716 reflections were symmetry
independent, and the Rint = 0.4200, indicating the data were
problematic. Indexing and unit cell refinement indicated a primitive
orthorhombic lattice, and the space group was found to be Pna21 (No.
33).
A blue plate of 3b immersed in oil with dimensions 0.10 × 0.05 ×
0.05 mm3 was mounted on a glass capillary. Data were collected at
−173 °C. The crystal-to-detector distance was 40 mm, and the
exposure time was 20 s per degree for all sets. The scan width was 0.5°.
Data collection was 98.8% complete to 25° in ϑ. CELL_NOW54 was
used to perform a 4.3° rotation twin operation about (0 0 1) to
prevent considerable overlap of diffraction peak intensities from
different domains of the sample. Multidomain integration with SAINT
within the APEX2 software package and an absorption correction with
TWINABS55 removed the overlap. A total of 66,667 merged
reflections were collected covering the indices h = −15 to 15, k =
−16 to 11, and l = 0 to 18; 9356 reflections were symmetry
independent, and the Rint = 0.1265, indicating the twin-refined data
were acceptable. Indexing and unit cell refinement indicated a triclinic
lattice, and the space group was found to be P1 (No. 2). Molecules of
̅
CH2Cl2 and Et2O were found to occupy the same space with a 2:1
ratio.
A purple prism of 4a immersed in oil with dimensions 0.25 × 0.20
× 0.10 mm3 was mounted on a glass capillary. Data were collected at
−173 °C. The crystal-to-detector distance was 40 mm, and the
exposure time was 10 s per degree for all sets. The scan width was 0.5°.
Data collection was 98% complete to 25° in ϑ. A total of 32,237
merged reflections were collected covering the indices h = −11 to 11, k
= −14 to 17, and l = −19 to 19; 7856 reflections were symmetry
independent, and the Rint = 0.0766, indicating the data were of average
quality. Indexing and unit cell refinement indicated a monoclinic P
Reactions between the Mn(III)−OOR complexes and various
substrates were performed in an analogous fashion by adding the
substrate to the fully formed intermediate in a single aliquot at either
273 or 298 K. Concentrated stock solutions of each substrate were
prepared under an inert atmosphere in a glovebox and transferred to
the reaction solution with a gastight syringe. After changes in
absorbance values ceased for at least 10 min, insolubles were removed
from the crude reaction mixtures by filtration through a fine-fritted
filter. Soluble organics were diluted with an appropriate amount of
CH2Cl2 and analyzed by GC/MS. Products were identified by
comparison with authentic standards and product yields were
lattice, and the space group was found to be P1 (No. 2).
̅
All data were integrated and scaled using SAINT, SADABS within
the APEX2 software package by Bruker. Solutions by direct methods
C
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 1
determined using calibration curves referenced to 1,4-dioxane as an
internal standard.
Scheme 2
General Workup Procedure for Analysis of the Thermal
Decomposition Products From [MnIII(SMe N4(QuinoEN))-
2
t
(OOtBu)](BPh4) (1a), [MnIII(SMe N4(QuinoPN))(OO Bu)]-
2
(PF6)·pentane (2a), [MnIII(SMe N4(6-Me-DPEN))(OO Bu)](BPh4)
t
2
(3a), [MnIII(SMe N4(6-Me-DPEN))(OOCm](BPh4) (3b), and
2
[MnIII(SMe N4(6-Me-DPPN))(OO Bu](BPh4)·Et2O (4a). In a typical
t
2
reaction, a 1−2 mM solution of the desired intermediate was prepared
in a fashion identical to that described above at 273 K in CH2Cl2.
Reactions were allowed to progress until the absorption features of the
respective Mn(III)−OOR intermediate bleached and no further
changes to the absorption spectrum were observed for at least 1 h.
The final reaction mixtures were observed to contain white insoluble
solids, black insoluble solids, in addition to CH2Cl2 soluble
components. Insolubles were removed by filtering each reaction
through a fine-fritted filter. In order to indentify the CH2Cl2 soluble
reaction mixture components, all volatiles were distilled in vacuo to a
separate container. Nonvolatile components remaining in the initial
+
(3), and [MnII(SMe N4(6-Me-DPPN))] (4). Dichloro methane
2
solutions of five-coordinate 2−4 (Scheme 2) change from
colorless to blue upon the anaerobic addition of BuOOH. At
t
ambient temperature, these blue solutions persist for only
∼10−30 min before they turn pale yellow (vide infra),
indicating that the species formed are metastable. As was
noted previously for 1a and 1b (Scheme 1), base (Et3N) was
found to increase the stability of these metastable species; in
the absence of base the blue color fades more rapidly. When
these reactions are monitored via electronic absorption
spectroscopy at low temperatures (258 K), two distinct
absorption bands grow in between 350 and 420 nm and
560−605 nm (Table 2). The relative energies and intensities of
these absorption bands (Figures S-1−S-4 in Supporting
Information [SI]) are comparable to those previously reported
for alkylperoxo-ligated 1a,32 suggesting that the species formed,
2a−4a, are also alkylperoxos. Solid samples of 2a−4a for the
measurement of extinction coefficients and IR spectra were
obtained via the addition of cold Et2O (10−15 mL) to a low-
temperature (−40 °C), in situ generated reaction mixture,
followed by filtration through a cold frit immediately (2−3
min) prior to data collection. Compounds 2a−4a are slightly
more stable in the solid state, but still require the low-
temperature generation of freshly prepared samples immedi-
ately prior to each experiment. Titration experiments show that
1
sample container were then analyzed by H NMR and ESI-MS, while
1
volatile components were analyzed by H NMR and GC/MS. The
clear and black insoluble reaction mixture components were separated
by suspending these products in CHCl3, which caused the white
product to dissolve, leaving behind the insoluble black precipitate.
After filtration to separate the black insoluble precipitate, the resulting
CHCl3 solution was dried in vacuo, redissolved in fresh CDCl3, and
1
analyzed by H NMR, ESI-MS, and/or GC/MS.
RESULTS AND DISCUSSION
■
C o o r d i n a t i v e l y u n s a t u r a t e d , r e d u c e d
+
[MnII(SMe N4(QuinoEN))] (1) was shown previously to
2
react with tert-butyl hydroperoxide (tBuOOH) and cumene
hydroperoxide (CmOOH) to form the corresponding oxidized
high-spin (S = 2) Mn(III)−alkylperoxo complexes
[MnI I I (SM e N4 (QuinoEN))(OO Bu)] (1a) and
t
+
2
[MnIII(SMe N4(QuinoEN))(OOCm)] (1b), respectively
+
2
(Scheme 1).32 Although metastable, both 1a and 1b proved
stable enough to permit crystallographic characterization,
thereby providing the first examples of end-on Mn(III)(η1-
peroxo) complexes. Five-coordinate 1 was also recently shown
to react with dioxygen to afford a rare example of a structurally
t
1.5−1.8 equiv of BuOOH are required to fully maximize
(Figures S-5, S-6 in SI) the electronic absorption features of
2a−4a as is illustrated in Figure 1 for compound 4.
Evidence to support the formulation of blue metastable 2a−
4a as Mn(III)−OOtBu alkylperoxo species was obtained via
mass spectrometry, FT-IR, and solution magnetic moment
measurements. Electrospray ionization mass spec (ESI-MS)
experiments were performed using aliquots of low temperature
in situ generated reaction mixtures (1−4 + 1.8 equiv of
tBu16O16OH). Prominent peaks are observed in the ESI-MS at
m/z = 598.8 (2a, m/z = 598.6 (calc)), 513.0 (3a, m/z = 513.6
(calc)), and 526.9 (4a, m/z = 526.6 (calc)), respectively
(Figures S-7−S-12 in SI). Upon preparation with tBu18O18OH,
each of these peaks increases by approximately 4 mass units to
m/z = 602.7 (2a), 516.9 (3a), and 530.8 (4a), confirming the
presence of two 18O atoms in each complex (Figures S-7−S-12
c h a r a c t e r i z e d m o n o - o x o b r i d g e d d i m e r ,
54
[MnIII(SMe N4(QuinoEN)]2-(μ-O)(PF6)2·Et2O. Three addi-
2
tional coordinatively unsaturated Mn(II) complexes (Scheme
II Me2
2), [MnII(SMe N4(2-QuinoPN))](PF6) (2), [Mn (S N4(6-
2
Me-DPEN))](BF4) (3), and [MnII(SMe N4(6-Me-DPPN))]-
2
(BPh4) (4), were included in this report,54 each of which was
shown to react with O2 to afford an unsupported oxo-bridged
dimer. Given the similar structural and reactivity properties of
1−4, we decided to explore the reactivity of 2−4 with alkyl
hydroperoxides in order to provide a more structurally diverse
library of end-on Mn(III)(η1-peroxo) compounds.
Reaction Between tert-Butyl Hydroperoxide and
+
II Me2
+
[MnII(SMe N4(QuinoPN))] (2), [Mn (S N4(6-Me-DPEN))]
2
D
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 2. Electronic Absorption and IR Parameters for Alkylperoxo Compounds 1a−4a in CH2Cl2 at 258 K
λmax (ε, cm−1 M−1) nm
λmax (ε, cm−1 M−1) nm
νO−O (Δ18O) (cm−1) nm
a
t
[MnIII(SMe N4(6-Me-DPPN))(OO Bu](BPh4) (4a)
420(240)
415(510)
385(640)
355(1060)
585(320)
590(465)
590(465)
600(575)
893(58)
895(64)
888(57)
875(59)
2
[MnIII(SMe N4(QuinoPN))(OO Bu)](PF6) (2a)
t
2
[MnIII(SMe N4(QuinoEN))(OO Bu)](BPh4) (1a)
t
2
[MnIII(SMe N4(6-Me-DPEN))(OO Bu)](BPh4) (3a)
t
2
a
Arranged in order of decreasing λmax in the near UV region.
indicating that the observed vibrational modes involve more
than a simple diatomic O−O stretch.46,49 Samples for solution
magnetic moment determination (i.e., the Evan’s method55)
were freshly prepared immediately (∼2 min) prior to data
collection by filtering low-temperature (−80 °C) grown crystals
through a cold frit, washing with precooled Et2O, quickly
weighing them on an analytical balance in an NMR tube, and
then adding precooled (−78 °C) CH2Cl2 immediately prior to
the insertion of the sample into the spectrometer. Solution
magnetic moments were estimated on the basis of a molecular
weight that included the manganese cation, anion, and no
solvents of crystallization. The resulting moments (μeff = 4.78
μB (2a); 4.79 μB (3a); 4.89 μB (4a)) were close to the spin-only
value (4.90 μB) for high-spin (S = 2) Mn(III), indicating that
the samples were reasonably pure and that the manganese ions
were in the +3 oxidation state.
Figure 1. Titration of 0−2.1 equiv of tBuOOH (in 0.3-equiv aliquots)
Structural Characterization of Metastable
to a CH2Cl2 solution [MnII(SMe N4(6-Me-DPPN))](BPh4) (4; 2.2
2
t
[MnIII(SMe N4(QuinoPN))(OO Bu)](PF6)·pentane (2a),
2
mM), in the presence of 2 equiv of Et3N, at −15 °C as monitored by
visible absorption spectroscopy.
[MnIII(SMe N4(6-Me-DPEN))(OO Bu)](BPh4) (3a),
t
2
t
[MnIII(SMe N4(6-Me-DPPN))(OO Bu](BPh4)·Et2O (4a). Evi-
2
t
in SI). The presence of a BuOO− ligand in 2a−4a is also
dence to definitively support the description of metastable
2a−4a as Mn(III)−OOR compounds was obtained via X-ray
crystallography. X-ray-quality single crystals were obtained by
layering either pentane or Et2O onto concentrated CH2Cl2
solutions of 2a−4a at −80 °C. ORTEP diagrams for 2a−4a are
shown in Figure 3, and selected metrical parameters are
provided in Table 3. Although the metrical parameters for 1a
have been reported elsewhere,32 they are also included in Table
3 for comparison. Each akylperoxo-bound complex, 2a−4a, is
monocationic as is indicated by the presence of a single anionic
supported by vibrational data (FT-IR) for both the 16O- and
18O- isotopic derivatives. In order to ensure that the counterion
did not dominate the IR spectra in the region of interest (900−
−
700 cm−1), the BPh4 salt of each complex was prepared. As
illustrated in Figure 2 for compound 3a, isotopically sensitive
−
−
counterion (PF6 for 2a, and BPh4 for 3a and 4a) per Mn
complex. The Mn(III) oxidation state in each complex is clearly
evident from the Mn(III)−S(1) bond distances (Table 3),
which are ∼0.1 Å shorter than those of the corresponding
reduced Mn(II) complexes, 1−4,54 and close to that of
alkylperoxo-bound 1a.32 Each Mn(III) ion contains an
akylperoxo ligand coordinated through its terminal oxygen
O(1) trans to an imine nitrogen N(1), and cis to both a thiolate
sulfur S(1) and tertiary amine N(2). Two N-heterocyclic amine
nitrogens, N(3) and N(4) (quinolines for 2a, 6-Me-pyridines
for 3a and 4a), lie trans to one another at a distance from the
Mn(III) ion (2.349(7)−2.522(8) Å; Table 2), which is
significantly greater than the sum of their ionic and covalent
radii (2.105 Å).67 The geometry of complexes 1a−4a is
therefore close to being four-coordinate square planar.
Elongation of the two trans Mn···N(3) and Mn···N(4)
interactions is most likely driven by the stabilization of an
odd electron in the “eg*” set of d-orbitals (akin to Jahn−Teller
distortion) and provides additional evidence that the Mn(III)
ions are high-spin d4. Computational studies (vide infra) show,
however, that the elongated Mn···N(3) and Mn···N(4)
distances also provide relief from steric interactions between
hydrogens located on the N-heterocyclic amines and the gem-
dimethyls adjacent to the sulfur. Sterics do not directly
Figure 2. FT-IR (nujol mull) of metastable [MnIII(SMe N4(6-Me-
2
DPEN))(16O16OtBu)]+ (3a; blue) vs metastable [MnIII(SMe N4(6-Me-
2
DPEN))(18O18OtBu)]+ (3a; red).
stretches (Table 2) are observed in the range expected for a
peroxo.47,68−75 Each of these stretches (Figures S-13, S-14 in SI,
Figure 2) falls in the reported range (885−896 cm−1) for the
handful of vibrationally characterized Mn(III)−peroxos,15,42,43
including alkylperoxo 1a (888 cm−1).32 The isotopic shifts
(Δ18O18OR) vary by as much as 14−20 cm−1 from that
predicted (44 cm−1) based upon a harmonic oscillator model,
E
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
methylene linker introduces flexibility that allows the metal ion
to attain a more idealized pseudo-octahedral geometry with
angles closer to 180° and 90°, respectively (Table 3). This
results in better orbital overlap with the alkylperoxo ligand (vide
infra).
t
Metrical parameters associated with the BuOO− ligands in
2a−4a (Table 3) are all within the anticipated ranges for
metal−alkylperoxo species,68−72 roughly correlate with vibra-
tional parameters (Table 4) and are consistent with an end-on
coordination mode. The peroxo Mn(III)−O(1) bond lengths
in 2a−4a (range: 1.840(4)−1.853(6) Å) are slightly shorter
than that of 1a (1.861(5) Å), and fall at the short end of the
limited number of reported Mn(III)−peroxos (range: 1.85−
1.90 Å),34,38,41−43 all of which happen to contain a side-on η2-
i
bound peroxo. Compared to reduced [TptBu, PrMn(II)-
(OOtBu)] (1.964 Å),72 the peroxo Mn−O(1) bonds in 2a−
4a are especially short. The former would have an additional
antibonding eg* electron which would contribute to a
lengthening of the Mn(II)−O bond. The lower coordination
number, and resulting increase in metal ion Lewis acidity,
would also contribute to the shorter Mn−O(1) bonds in 1a−
4a. The relatively obtuse Mn(III)−O(1)−O(2) angles, and
short Mn−O(1) bond distances would be consistent with at
least some MnOOtBu double bond character.46,49 The
O(1)−O(2) peroxo bond lengths in 1a−4a span the range
1.431(5)−1.468(7) Å, which in the case of Et-linked 1a and 3a
(1.457(7) and 1.468(7) Å, respectively) is significantly longer
than all structurally characterized examples of mononuclear
Mn(III)−peroxos (range O−O = 1.403(4)−1.428(7) Å),
[(TPP)Mn(O2)](K(K222)),41 [Mn(14-TMC)(O2)]+,38 [Mn-
Figure 3. ORTEP diagrams (50% probability ellipsoids) of
t
+
[MnIII(SMe N4(QuinoPN))(OO Bu)] (2a), [MnIII(SMe N4(6-Me-
2
2
DPEN))(OOtBu)]+ (3a), and [MnIII(SMe N4(6-Me-DPPN))-
2
(OOtBu]+ (4a) showing exceedingly long Mn··N(3,4) interactions
(Table 3) as dashed lines. Hydrogen atoms, counterions, and solvents
of crystallization have been omitted.
t
influence the BuOO− ligand, however (vide infra), which is
spatially quite separated from both the gem-dimethyls and 6-
Me-pyridine substituents. The Mn(III)···N(3,4) separation
represents the largest structural variable ( 0.17 Å) within
this particular series; the Mn(III)−O(1), M(III)−N(1),
Mn(III)−N(2), and Mn(III)−S(1) distances are well within
acceptable bonding ranges, and display little variation ( 0.02 Å,
(13-TMC)(O2)]+,34 Mn(O2)(3,5-iPr2pzH)(HB(3,5-iPrpz)3),42
i
and TpPr Mn(η -O2)(im H),43 all of which contain a side-on
2
Me
2
η2-peroxo. An end-on binding mode of the BuOO− ligand in
t
2a−4a is confirmed by the long distance separating the Mn(III)
ion from the distal oxygen (Mn···O(2) = 2.861 Å, 2.769 Å, and
2.901 Å), which is nearly a full 1 Å longer than that of the side-
on Mn(III)(η2-O2) complexes mentioned above.38,41−43 The
less acute Mn−O(1)−O(2) angles of 2a−4a (121.1(3)°,
112.44(4)°, and 124.25(3)°, respectively) are also consistent
with an end-on binding mode. These angles are similar to,
0.03 Å,
0.04 Å, and 0.03 Å, respectively). As was
observed with the five-coordinate Mn(II) precursors 1−4,54 the
insertion of an extra methylene unit into the ligand backbone is
found to much more noticeably influence the structures than
does the identity of the N-heterocyclic amine (6-Me-pyridine vs
quinoline). In propyl-linked 2a and 4a (Figure 3), the extra
t
Table 3. Selected Bond Distances (Å) and Bond Angles (deg) for [MnIII(SMe N4(QuinoEN))(OO Bu)](BPh4) (1a),
2
[MnIII(SMe N4(QuinoPN))(OO Bu)](PF6)·pentane (2a), [Mn (S N4(6-Me-DPEN))(OO Bu)](BPh4) (3a), [Mn (S N4(6-
t
III Me2
t
III Me2
2
t
Me-DPEN))(OOCm](BPh4) (3b), and [MnIII(SMe N4(6-Me-DPPN))(OO Bu](BPh4)·Et2O (4a)
2
1a
2a
3a
3b
4a
Mn−S(1)
Mn−N(1)
Mn−N(2)
Mn−N(3)
Mn−N(4)
Mn−O(1)
O(1)−O(2)
Mn··O(2)
2.270(3)
2.034(7)
2.173(7)
2.349(7)
2.522(8)
1.861(5)
1.457(7)
2.714
2.2693(15)
2.046(4)
2.182(4)
2.450(5)
2.518(5)
1.840(4)
1.438(5)
2.861
2.241(3)
2.015(8)
2.163(7)
2.354(8)
2.471(7)
1.853(6)
1.468(7)
2.769
2.268(2)
2.018(5)
2.145(5)
2.499(6)
2.389(5)
1.848(4)
1.457(5)
2.756
2.2603(13)
2.061(4)
2.178(4)
2.517(4)
2.504(4)
1.843(3)
1.431(5)
2.901
Mn−O(1)−O(2)
O(1)−Mn−N(1)
S(1)−Mn−N(2)
O(1)−Mn−S(1)
O(1)−Mn−N(2)
S(1)−Mn−N(1)
S(1)−Mn−N(4)
Mn−O(1)−O(2)−C
109.2(4)
175.4(3)
163.1(2)
97.5(2)
97.9(3)
82.3(2)
105.6(3)
139.51°
121.1(3)
176.12(16)
177.62(13)
93.69(12)
84.16(17)
82.44(12)
108.98(14)
131.85°
112.4(4)
175.7(3)
162.6(2)
94.9(2)
112.4(3)
176.9(2)
164.95(14)
99.92(14)
94.23(19)
83.11(16)
107.23(16)
142.22°
124.2(3)
174.27(15)
176.78(11)
91.79(11)
85.02(14)
83.32(11)
108.32(10)
126.68°
102.2(3)
83.6(2)
109.13(18)
163.69
F
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
t
Table 4. Selected metrical, spectroscopic, and kinetic data for [MnIII(SMe N4(QuinoEN))(OO Bu)](BPh4) (1a),
2
[MnIII(SMe N4(QuinoPN))(OO Bu)](PF6)·pentane (2a), [Mn (S N4(6-Me-DPEN))(OO Bu)](BPh4) (3a), and
t
III Me2
t
2
t
[MnIII(SMe N4(6-Me-DPPN))(OO Bu](BPh4)·Et2O (4a)
2
ligand backbone
Mn···N(3,4)avg (Å)
O−O (Å)
λmax (ε, cm−1 M−1) nm
νO−O(Δ18O) (cm−1
)
ΔH⧧ (kcal/mol)
4a
2a
1a
3a
Pr
Pr
Et
Et
2.510
2.484
2.435
2.411
1.431(5)
1.438(5)
1.457(7)
1.468(7)
420(240)
415(510)
385(640)
355(1060)
893(58)
895(64)
888(57)
875(59)
15.9(5)
13.8(4)
11.6(4)
10.5(3)
albeit larger than, that of 1a (109.14°), but nearly twice as large
as those (66°−70°) found in the side-on Mn(III)(η2−O2)
complexes mentioned above.34,38,41−43
the alkylperoxo O(1)−O(2) bond length (Figure 5, Table 4) of
1a−4a. The strong correlation between these parameters
Correlation Between Structural and Spectroscopic
Parameters. Changes to the ligand backbone of the series of
N-heterocyclic amine Mn−alkylperoxo compounds described
herein causes subtle variations in the metal−ligand bond
distances and angles, and systematic changes to the electronic
structure. The long Mn···N(3) and Mn···N(4) distances
increase the metal ion Lewis acidity (vide infra) resulting in
shorter metal ligand bonds (Table 3). As shown in Table 4, the
average Mn···N(3,4) distance in complexes 1a−4a decreases as
a methylene unit is removed from the ligand backbone.
Although exceedingly long, these nitrogens (N(3) and N(4))
are found to perturb the Mn ion enough to influence its
properties. For example, the highest energy electronic
absorption band (Figures S-1−S-4 in SI) systematically blue-
shifts as the average Mn····N(3,4) distance decreases (Table 4,
Figure S-16 in SI). The N-heterocyclic amines also influence
the alkyl peroxo’s metrical parameters. As N(3) and N(4) move
in closer to the metal ion, the Mn−O(1) and O(1)− O(2)
distances increase and the peroxo becomes more activated
(Figure 4 and Figure S-17 in SI). This observed trend78 can be
Figure 5. Plot of O(1)−O(2) bond length vs λmax (nm) for the high-
energy absorption bands displayed by 1a−4a.
provides evidence to support the crystallographically deter-
mined differences in O−O bond lengths, and provides clues
regarding the assignment of this electronic absorption feature.
Although the O−O bond distances for Et-linked 1a and 3a are
indistinguishable from one another (Figure 5) based on the
crystallographic error (esds in Table 3), these distances are
distinctly different from those of Pr-linked 2a and 4a.
The observed systematic blue shift in λmax upon elongation of
the peroxo O−O bond in 1a−4a agrees with TD-DFT-
predicted trends for the πv*(O−O) → Mn charge transfer band
of peroxo-ligated [MnIII(O2)(L7py2Me)]+,35 and the theoret-
ically calculated energy (λmax = 442 nm) approximately matches
our experimentally measured transition energy (Table 4). On
this basis, we have tentatively assigned the highest energy band
associated with 1a−4a as a πv*(O−O) → Mn charge transfer
band (Figures S-1−S-4 in SI). The extinction coefficients for
these transitions display a roughly linear relationship with the
Mn(III)−O(1)−O(2)−R dihedral angles (139.5° (1a), 131.8°
(2a), 163.6° (3a), 126.6° (4a)), with larger dihedral angles
resulting in higher extinction coefficients (Figure S-19 in SI).
This relationship can be explained by the fact that extinction
coefficients are expected to increase as orbital overlap (between
the πv*(O−O) orbital and the symmetry-appropriate Mn d-
orbital, vide infra) increases, with maximum overlap at a
dihedral angle of 180°.79 There are no direct correlations
between the lower-energy visible absorption features of 1a−4a
(λmax = 560−605 nm; Figures S-1−S-4 in SI) and any metrical
parameter that would permit a tentative spectral assignment;
however, the relatively low extinction coefficients for these
bands would be at least consistent with ligand field transitions.
Figure 4. Plot of O(1)−O(2) bond length (Å) vs Mn(III)−N(3,4)avg
distance (Å) for Et-linked (1a, 3b, and 3a), and Pr-linked (2a and 4a)
structures. Data for [MnIII(SMe N4(6-Me-DPEN))(OOCm](BPh4)
2
(3b) is represented by the pink triangle and is not included in the
linear fit correlation coefficient (R2).
explained by empirical and computational studies which show
that electron-donating ligands promote longer, more activated
peroxo O−O bonds.33−35,71,73−77 As the Mn····N(3,4)
separation decreases in 1a−4a electron density increases at
the Mn ion. As shown in Figure 4, the inverse correlation
between the Mn····N(3,4) and O(1)−O(2) distances in 1a−4a
is strong (R2 = 0.99). There is very little correlation between
the Mn−S(1) and O(1)−O(2) distances (Figure S-22 in SI).
We also see a strong correlation between the energy of λmax and
G
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
In addition to the observed shifts to the πv*(O−O) → Mn
charge transfer bands, the ν(O−O) vibrational energies also
shift to higher energies as the O(1)−O(2) bond lengths in 1a−
4a decrease (Table 4). However the correlation coefficient
(0.81) for ν(O−O) vs O(1)−O(2) distance (Figure S-16 in SI)
is modest, suggesting that the force constants differ from
complex to complex, possibly due to a change in hybridization
of O(1). A change in O(1) hybridization (from sp3 to sp2) is
indicated by the systematic change in Mn(III)−O(1)−O(2)
bond angles (from 109.2(4)° to 124.2(3)°; Figures S-20 and S-
21 in SI). There is also a wide range of Mn(III)−O(1)−
O(2)−tBu dihedral angles (139.5° (1a), 131.8° (2a), 163.3°
(3a), 126.6° (4a)) suggesting that varying degrees of coupling
to the tBu-group might perturb the vibrational modes.46,49 The
strong correlation between crystallographically determined
alkylperoxo O−O bond distances and spectroscopic parameters
for 1a−4a (Table 4, Figure 5) provides experimental evidence
supporting the accuracy of the observed crystallographic
differences in O−O bond length. Computational studies
support this
changes in charge density at the Mn ion were found to be
relatively small. This may be attributed to the fact that the
entire ligand framework is involved in modulation of the charge
distribution. As the calculated apical Mn−N bond length
increases (from ∼2.03 Å (1a/3a) to ∼2.11 Å (2a/4a)), more
charge density is predicted to be localized on the apical
nitrogen, and less on the thiolate sulfur. This would suggest
that sulfur acts as a charge donor that compensates80 for
electron density fluctuations at the metal center.
The bonding molecular orbitals of [MnIII(SMe N4(6-Me-
2
DPPN))(OOtBu)]+ (4a) illustrate why an increased local-
ization of charge on the peroxo would lengthen the O−O bond.
The Mn ion contribution to the Mn−O(1) π-bonding MO of
Figure 6 decreases (from ∼33% to ∼27%) as the O−O bond
Ab Initio Density Functional Theory (DFT) Calcula-
tions. Geometry optimizations for 1a−4a were carried out
using ab initio density functional theory (DFT), and potential
correlations between peroxo O−O bond length and N-
heterocyclic amine Mn···N(3,4) distances were examined (for
3a and 4a) using relaxed surface scans. Initial geometry
optimizations showed deviations in Mn···N(3,4) separations
( 0.05 Å), and Mn−O−O angles (up to 10°), relative to the
crystallographic structures (Table S-1 in SI). However, the
inclusion of van der Waals corrections greatly improved the
agreement between calculated and experimental parameters
( 0.03 Å difference in Mn···N(3,4) separation, and less than
∼3° difference in Mn−O−O angle), indicating that steric
interactions are likely responsible for the long Mn···N(3,4)
distances and coordinative unsaturation in 1a−4a. The
calculated peroxo O−O bond lengths (Table S-1 in SI) were
found to be indistinguishable for the two N-heterocyclic amines
(6-Me-pyridine vs quinoine), but fall into two distinct groups,
depending on the ligand backbone (Pr- vs Et-linker). As is
observed experimentally, Et-linked 1a and 3a are predicted to
have shorter Mn···N(3,4) distances and longer O−O bonds (by
∼0.01 Å) relative to those of Pr-linked 2a and 4a (Table S-1 in
SI). However, the theoretically calculated difference is not as
large as that seen in the crystal structures. To further
understand the differences between structures containing an
Et- vs Pr-backbone, Mulliken charge densities were considered.
The Mn center of Et-linked 3a, which has the longest O−O
bond in the series (1.468(7) Å; Table 2), was theoretically
determined to have the lowest Mulliken charge density (+0.42),
and hence the lowest Lewis acidity (Table S-2 in SI). Whereas,
the Mn center of Pr-linked 4a, which has the shortest
crystallographically determined O−O bond in the series
(1.431(5) Å), was theoretically determined to have the highest
Mulliken charge density (+0.50) and, hence, the highest Lewis
acidity (Table S-2 in SI).
Figure 6. Bonding π(Mn−O) orbitals (isovalue = 0.035) for
alkylperoxo compound 4a at the O−O = 1.2 Å (A) and 1.5 Å (B)
bond limits. Both involve an antibonding peroxo πv*(O−O) orbital
interacting with a Mn dxz orbital. The Mn contribution drops from
24.2% (A) to 11.4% (B) with increasing O−O bond length. The
change in Mn contribution is largely compensated by donation from
the sulfur πp orbital (0.2% at 1.2 Å vs 11.9% at 1.5 Å).
gets longer (from 1.2 Å to 1.5 Å), whereas the peroxo
contribution increases from ∼9.4% to 10.3%.81 The length-
ening of the O−O bond is thus a consequence of an increased
localization of charge density in a π(Mn−O) orbital (shown in
Figure 6) that is antibonding with respect to the peroxo O−O
bond. These theoretical results are supported by literature
precedent33−35,71,73−76 and agree with the bonding picture,
established for Fe(III)−OOR compounds, which also includes
a π(Fe−O) bond formed via overlap between a πv*(O−O)
orbital and a symmetry-appropriate metal d orbital.46,49 In both
cases, the end-on η1-peroxo behaves as a π-donor ligand. With
Lewis acidic metal ions, π-donation from the filled peroxo
πv*(O−O) orbital results in a stronger O−O bond. Longer
peroxo O−O bonds are observed with more electron-rich metal
centers not only with the members of the series, 1a and 3a,
containing shorter Mn···Npy,quin interactions, but also with
Cu(II), Co(III), Fe(III), and other Mn(III) com-
plexes.33−35,71,73−76,78
Variable-Temperature Decay Kinetics for Alkylperoxo
Compounds 1a−4a. Temperature-dependent kinetic studies
were performed in order to provide additional experimental
evidence supporting the structural and spectroscopic trends
described above. As described below, we observe strong
correlations between structural and kinetic parameters
consistent with a decay mechanism involving rate-limiting
O−O bond cleavage. Each metastable Mn(III)−OOR complex
1a−4a was found to decay in solution over a relatively short
period of time, even at low temperatures. The details regarding
the mechanism of this decay reaction were obtained using
In order to understand the relationship between the O−O
bond length and the metal ion coordination sphere, relaxed
surface scans were performed. For this study, the O−O bond
length was varied from 1.1 to 1.6 Å, while optimizing all other
coordinates for Et-linked 3a vs Pr-linked 4a, (Table S-3 in SI).
As the O−O bond was increased in length, charge density
became more localized on the proximal peroxo oxygen (varying
from −0.22 (O−O = 1.1 Å) to −0.40 (O−O = 1.6 Å). The
H
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
t
Table 5. Kinetics data for the Decay of Alkylperoxos [MnIII(SMe N4(QuinoEN))(OO Bu)] (BPh4) (1a),
2
[MnIII(SMe N4(QuinoPN))(OO Bu)](PF6) (2a), [Mn (S N4(6-Me-DPEN)) (OO Bu)](BPh4) (3a), [Mn (S N4(6-Me-
t
III Me2
t
III Me2
2
t
DPEN))(OOCm](BPh4) (3b), [MnIII(SMe N4(6-Me-DPPN))(OO Bu](BPh4) (4a)
2
3a
1a
3b
2a
4a
k1 (s−1), 293 K
k1 (s−1), 273 K
k1 (s−1), 268 K
k1 (s−1), 258 K
ΔH⧧ (kcal/mol)
ΔS⧧ (cal/mol·K)
0.00278
0.00101
0.000623
0.000205
10.4
0.00189
0.00043
0.00033
0.00013
10.8
0.00257
0.00079
0.000698
0.00015
11.0
0.0051
0.001
0.00473
0.000528
0.000308
0.000105
15.9
0.000559
0.000172
13.9
−34
−34
−32
−21
−15
variable-temperature kinetics probed by electronic absorption
spectroscopy in CH2Cl2. Previously, we showed that 1a
undergoes a first-order thermal decomposition process at 258
K in CH2Cl2 with a rate constant that is insensitive to the
presence of electrophilic and nucleophilic organic substrates, as
well as dioxygen.32 Herein we examine the reaction kinetics for
the other members of the series 2a−4a, in addition to the
temperature-dependent kinetics for all of the compounds
described herein including 1a. Time-dependent changes in
absorption (at 592 nm (1a), 580 nm (2a), 594 nm (3a), or 450
nm (4a)) were measured at several temperatures between 258
and 293 K, over at least three half-lives. Linear fits to ln[(At −
A∞)/(Ai − A∞)] vs time plots indicated that the reaction was
first-order in Mn−alkylperoxo complex at all of the temper-
atures examined. Correlation coefficients were significantly
lower (R2 = 0.4−0.7) for second-order ([(Ai − A∞)/(At −
A∞)] vs time) plots. First-order rate constants were obtained as
described in the Experimental Section and are assembled in
Table 5. A representative time-dependent absorption vs
wavelength stack plot (Figure S-23 in SI), and the
corresponding first order kinetics plot, are shown in Figure 7.
Half-lives were found to be independent of the initial peroxo
intermediate concentration (1−2 mM), providing additional
support for a first-order reaction. Activation parameters (Table
5) for the thermal decay of each Mn(III)−OOR complex
(Figure S-23 in SI) were obtained from linear fits to Eyring
plots (ln(k/T) vs T−1; Figure 8, and Figures S-24−S-26 in SI).
Initial inspection of the data (Table 5) shows that both ΔH⧧
and ΔS⧧ each fall within a relatively narrow range of one
another, consistent with the thermal decay of 1a−4a
proceeding via a similar mechanism. If we include a cumenyl
Figure 7. Electronic absorption spectra displaying the thermal decay of
2a in CH2Cl2 at 258 K with individual scans representing 5-min
intervals (top). Kinetics plot of ln[(At − A∞)/(Ai − A∞)] vs time for
the thermal decay of 2a in CH2Cl2 at 258 K (bottom).
peroxo derivative of 3a, [MnIII(SMe N4(6-Me-DPEN))-
2
(OOCm](BPh4) (3b), in this study, then we see very similar
activation parameters (Table 5) suggesting that the R-group (of
Mn(III)−OOR) does not dramatically influence the decay
mechanism. In order to confirm that 3b was indeed a peroxo
and to determine whether the R group influences the structure,
single crystals were grown for crystallographic characterization
of metastable 3b. As shown by the ORTEP diagram of
O(2) bond lengths for 1a−4a and 3b provide the most
significant clues regarding a likely mechanism for the observed
thermal decay process. As shown in Figure 10, the activation
parameters ΔH⧧ and ΔS⧧ progressively decrease as the O(1)−
[MnIII(SMe N4(6-Me-DPEN))(OOCm](BPh4) (3b) in Figure
2
O(2) bond length increases. Additionally, the half-life (τ1/2
=
9, the metrical parameters of Table 3, electronic absorption
spectrum of Figure S-27 in SI, and Eyring plot of Figure S-28 in
SI, it is clear that the R group has very little influence on the
structural and spectroscopic properties. An enthalpy−entropy
compensation effect82 is also apparent from a plot of ΔH⧧ vs
ΔS⧧ for all five compounds 1a−4a and 3b (Figure S-29 in SI),
providing additional evidence for a common decay mechanism
(although we recognize that the validity of this particular
relationship is debated and has been scrutinized).83 Strong
correlations between the activation parameters (ΔH⧧ and ΔS⧧)
and the crystallographically determined alkylperoxo O(1)−
56.4 min) of the peroxo complex (Et-linked 3a) with the
longest O−O bond (1.468(7) Å) is significantly shorter than
that (τ1/2 = 110 min) of the peroxo complex (Pr-linked 4a)
with the shortest O−O bond (1.431(5) Å). This provides
additional evidence that the crystallographically determined
bond lengths are accurate and the error is overestimated. The
decay occurs much more slowly for compounds (2a, 4a)
containing an extra methylene unit and longer Mn···N(3,4)avg
separation (Table 5), showing that the ligand backbone and N-
heterocyclic amines indeed influence reactivity. There is very
little correlation between the activation parameters ΔH⧧ (or
I
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 8. Eyring plot from the variable-temperature decay kinetics
t
data for [MnIII(SMe N4(6-Me-DPPN))(OO Bu](BPh4) (4a) in
2
CH2Cl2.
Figure 10. Plot of activation enthalpy (kcal/mol) (top) and activation
entropy (cal/mol·K) (bottom) vs O(1)−O(2) bond length (Å).
1
1a−4 and 3b were analyzed via H NMR, GC/MS, and ESI-
MS. For each complex, final reaction mixtures were found to
contain white and black solids, in addition to CH2Cl2-soluble
products. The white solids were identified as the X− (X =
Figure 9. ORTEP diagram (50% probability ellipsoids) of
+
[MnIII(SMe N4(6-Me-DPEN))(OOCm)] (3b) showing exceptionally
2
long Mn···N(3,4) intgeractions as dashed lines. Hydrogen atoms,
counterions, and solvent of crystallization have been omitted for
clarity.
−
−
BPh4 for 1a, 3a, 3b, and 3a, PF6 for 2a) salts of either
protonated N,N-bis(2-quinoline)ethane-1,2-diamine (1a), N,N-
bis(2-quinoline)propane-1,3-diamine (2a), N,N-bis(6-methyl-
2-pyridilmethyl)ethane-1,2-diamine (3a and 3b), or N,N-bis(6-
methyl-2-pyridilmethyl)propane-1,3-diamine (4a), each of
which was obtained in 70−85% yields. The CH2Cl2-soluble
component was shown to consist of 3-methyl-2-butanone-3-
disulfide (80−90%; Figures S-33, S-34 in SI), as well as either
tert-butanol (70−80%, from 1a to 4a) or acetophenone (68−
72%, 3b). The disulfide product may form via a reaction
between a highly reactive tert-butoxy radical and the thiolate
ligand moiety, suggesting that the pentadentate ligands of 1a−
4a may serve as an internal trapping reagent in these reactions.
Identification of the insoluble black precipitate was less
conclusive; however, we suspect that this product is likely a
manganese oxide of some sort.
Analysis of the anaerobic thermal decay products from 3b
was of particular interest, given that cumene hydroperoxide is
well-recognized as a mechanistic probe for homolytic vs
heterolytic O−O bond-cleaving pathways. Thermal decay of
1b was previously reported to exclusively produce acetophe-
none (68%), as opposed to cumenol (0%), which strongly
suggested a homolytic, as opposed to heterolytic, O−O bond-
cleaving pathway was involved. With cumeneperoxo-ligated 3b,
we again only observed exclusive formation of acetophenone
(72%), as opposed to cumenol. This supports our conclusions
ΔS⧧) and the Mn(III)−O(1) bond lengths, on the other hand
(Figures S-31, S-32 in SI). All of this data would be consistent
with a mechanism involving rate-determining cleavage of the O−O,
as opposed to Mn−O bond.82,84,85 The activation enthalpy ΔH⧧
decreases as νO−O decreases (Figure S-30 in SI), which is also
consistent with this mechanism. One would expect that a rate-
determining reaction step involving O−O bond cleavage would
display a positive activation entropy, which is contrary to our
experimental results. Negative activation entropies are also seen
in O−O bond-cleaving reactions involving high-spin (S = 5/2)
Fe(III)−OOtBu complexes,86 and peroxo-bridged (TMP)Fe-
(III)−O−O−Fe(III)(TMP).87 One possible explanation for
the negative activation entropies seen with the peroxo
compounds described in this study is that the transition state
leading to successful cleavage of the O−O bond has a shorter,
more ordered interaction between the Mn ion and N(3,4). This
would make sense given that longer O−O bonds were found to
result from shorter Mn(III)−N(3,4)avg distances. It is also
possible that solvent effects are responsible.
Thermal Decomposition Products from 1a−4a and
3b, Both in the Presence and Absence of Organic
Substrates. The products formed following thermal decay of
J
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
based on kinetics, and suggests that O−O bond homolysis is the
predominant reaction pathway for 3b. In light of the kinetic
trends established for 3b vs 1a−4a, this observation further
supports a homolytic O−O bond-cleavage pathway in the
thermal decay of 1a−4a. It is worth noting that
tBuOO−/tBuOOH and CmOO−/CmOOH were not observed
even in trace amounts, providing evidence to suggest that
homolytic or heterolytic Mn−O bond cleavage is not an
operative process in these systems. To our knowledge, these are
the only known Mn(III)−peroxo compounds that decay via
homolytic, as opposed to heterolytic, O−O bond cleavage.
Given that Et3P has been shown to serve as a probe for tert-
butoxide radicals,53 we examined reactions between 1a to 4a
and Et3P (10 equiv) under anaerobic conditions. Analysis of the
final reaction mixture by GC/MS showed a 24% yield of
Et2POtBu and 31% yield of Et3PO. These products are
consistent with a decay mechanism involving homolytic O−O
bond cleavage, followed by the formation of a tert-butoxy
radical, as well as a transient Mn(IV)O. We have been unable
to probe or independently synthesize the Mn(IV)O, most
likely because it is highly reactive.
O−O bond lengths that the ligand-centered orbital involved in
the π(Mn−O) bond is an antibonding πv*(O−O) orbital, and
this orbital is involved in the high-energy charge transfer
transition seen in the electronic spectrum. The peroxo O−O
bond was shown to gradually become more activated as the N-
heterocyclic amines moved closer to the metal ion, due to
decreased shifting of electron density out of the πv*(O−O)
orbital onto the metal (i.e., π-donation). Theoretical calcu-
lations support this. An increased interaction with the N-
heterocyclic amines would decrease the metal ion Lewis acidity,
making it a poor π-acceptor. Theoretical studies showed that
the Lewis acidity of the metal ion directly influences the extent
of O−O bond activation, with longer O−O bonds occurring in
the less Lewis acidic metal complexes. The identity of the
ROO− alkyl substituent was shown to have no effect upon
these structural and reactivity trends. Temperature-dependent
kinetic parameters for the decay of the compounds described
herein were shown to correlate with the peroxo O−O bond
length, consistent with a mechanism involving rate-determining
O−O bond cleavage. The products of decay were shown to be
consistent with homolytic O−O bond cleavage.
Finally, we sought to determine whether 1a−4a and 3b
display any reactivity toward exogenous organic molecules.
Reactions were examined using a wide variety of substrates
including cyclohexane, cyclohexane carboxaldehyde, tetracyano-
ethylene, styrene, triphenylphosphine, triethylphosphine, thio-
anisole, 1-hydroxy-2,2,6,6-tetramethyl-piperidine (TEMPO-H),
dihydroanthracene (DHA), and 2,4-di-tert-butylphenol. These
reactions were monitored by electronic absorption spectrosco-
py at 273 and 293 K in a fashion analogous to the kinetics
experiments described above. The decay rates of 1a−4a and 3b
were unperturbed by the presence of these substrates, even at
500-fold excess amounts. Each reaction displayed a zero-order
dependence on substrate concentration, suggesting that
alkylperoxo 1a−4a and 3b are themselves relatively unreactive
toward external substrates (Figures S-35−S-49 in SI). Analysis
of the final reaction mixtures did, however, show that substrate
oxidation occurs. The invariance in thermal decay rates both in
the presence and absence of substrates indicates that the
reactive oxidant forms following rate-limiting O−O bond
cleavage. Since this was not the focus of our study and many of
the substrates analyzed could react with either a metal-based or
organic radical oxidant, most of these reactions were not
pursued further.
ASSOCIATED CONTENT
■
S
* Supporting Information
Quantitative UV/vis of alkylperoxo 1a−4a and titration
t
between BuOOH + five-coordinate 2, 3, ESI-MS and FT-IR
of Mn−16O16OtBu and Mn−18O18OtBu compounds, plots
showing correlation between structural, spectroscopic, and
kinetic parameters, electronic spectra and Eyring plots
associated with the thermal decay kinetics of alkylperoxo
compounds 1a−4a, and plots showing zero-order dependence
on substrate concentration. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
kovacs@chem.washington.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
J.K. acknowledges the NIH (RO1GM45881-20) for funding.
S.D. acknowledges the Max Planck Society for funding, and the
Sloan Foundation for a fellowship.
SUMMARY AND CONCLUSIONS
■
Four new Mn(III)−alkylperoxo complexes were synthesized
and structurally and spectroscopically characterized. This
extends the number of crystallographically characterized
monomeric Mn(III)−peroxo compounds to 10. The ligand
backbone of the series described herein was shown to alter
metal−ligand distances and modulate electronic properties and
the degree of peroxo O−O bond activation. Although not
technically bonded, the N-heterocyclic amines were shown to
systematically move toward the metal ion upon removal of a
methylene unit and alteration of the base (py→quinoline), and
this was shown to gradually shift both the electronic absorption
energies and associated extinction coefficients. The increased
metal ion Lewis acidity caused by elongation of the N-
heterocyclic amine Mn···N(3,4) distances was found to shorten
the Mn(III)−O(peroxo) bonds and facilitate π(Mn−O)
bonding. It was concluded on the basis of a correlation
between shifts to the electronic absorption band and peroxo
REFERENCES
■
(1) Cotruvo, J. A., Jr.; Stubbe, J. Biochemistry 2010, 49, 1297−1309.
(2) Umena, Y.; Kawakami, K.; Shen, J.-R.; Kamiya, N. Nature 2011,
473, 55−60.
(3) Oliw, E. H.; Jerneren, F.; Hoffmann, I.; Sahlin, M.; Garscha, U.
Biochim. Biophys. Acta 2011, 1811, 138−147.
(4) Perry, J. J. P.; Hearn, A. S.; Cabelli, D. E.; Nick, H. S.; Tainer, J.
A.; Silverman, D. N. Biochemistry 2009, 48, 3417−3424.
(5) Boal, A. K.; Cotruvo, J. A., Jr.; Stubbe, J.; Rosenzweig, A. C.
Science 2010, 329, 1526−1530.
(6) Barnese, K.; Sheng, Y.; Stich, T.; Gralla, E. B.; Britt, R. D.;
Cabelli, D.; Valentine, J. S. J. Am. Chem. Soc. 2010, 132, 12525−12527.
(7) Armstrong, F. A. Philos. Trans. R. Soc. London, Ser. B 2008, 363,
1263−1270.
(8) Emerson, J. P.; Kovaleva, E. G.; Farquhar, E. R.; Lipscomb, J. D.;
Que, L., Jr. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 7347−7352.
(9) McEvoy, J. P.; Brudvig, G. W. Chem. Rev. 2006, 106, 4455−4483.
K
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(10) Hearn, A. S.; Stroupe, M. E.; Cabelli, D. E.; Lepock, J. R.;
Tainer, J. A.; Nick, H. S.; Silverman, D. N. Biochemistry 2001, 40,
12051−12058.
(11) Jackson, T. A.; Karapetian, A.; Miller, A.-F.; Brunold, T. C.
Biochemistry 2005, 44, 1504−1520.
(45) Namuswe, F.; Hayashi, T.; Jiang, Y.; Kasper, G. D.; Narducci
Sarjeant, A. A.; Moenne-Loccoz, P.; Goldberg, D. P. J. Am. Chem. Soc.
̈
2010, 132, 157−167.
(46) Lehnert, N.; Ho, R. Y. N.; Que, L. J.; Solomon, E. I. J. Am.
Chem. Soc. 2001, 123, 12802−12816.
(12) Miller, A.-F. Curr. Opin. Chem. Biol. 2004, 8, 162−168.
(13) Yachandra, V. K.; Sauer, K.; Klein, M. P. Chem. Rev. 1996, 96,
2927−2950.
(47) Kitagawa, T.; Dey, A.; Lugo-Mas, P.; Benedict, J.; Kaminsky, W.;
Solomon, E.; Kovacs, J. A. J. Am. Chem. Soc. 2006, 128, 14448−14449.
(48) Shan, X.; Rohde, J.-U.; Koehntop, K. D.; Zhou, Y.; Bukowski, M.
R.; Costas, M.; Fujisawa, K.; Que, L., Jr. Inorg. Chem. 2007, 46, 8410−
8417.
(14) Bull, C.; Niederhoffer, E. C.; Yoshida, T.; Fee, J. A. J. Am. Chem.
Soc. 1991, 113, 4069−4076.
(15) Shook, R. L.; Gunderson, W. A.; Greaves, J.; Ziller, J. W.;
Hendrich, M. P.; Borovik, A. S. J. Am. Chem. Soc. 2008, 130, 8888−
8889.
(16) Murphy, A.; Stack, D. T. P. J. Mol. Catal. A: Chem. 2006, 251,
78−88.
(17) Shook, R. L.; Peterson, S. M.; Greaves, J.; Moore, C.; Rheingold,
A. L.; Borovik, A. S. J. Am. Chem. Soc. 2011, 133, 5810−5817.
(18) Goldsmith, C. R.; Cole, A. P.; Stack, T. D. P. J. Am. Chem. Soc.
2005, 127, 9904−9912.
(19) Betley, T. A.; Surendranath, Y.; Childress, M. V.; Alliger, G. E.;
Fu, R.; Cummins, C. C.; Nocera, D. G. Philos. Trans. R. Soc. London,
Ser. B 2008, 363, 1293−1303.
(20) Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U.S.A. 2006,
103, 15729−15735.
(21) Zhang, Y.; Stubbe, J. Biochemistry 2011, 50, 5615−5623.
(22) Skulan, A. J.; Brunold, T. C.; Baldwin, J.; Saleh, L.; Bollinger, J.
M., Jr; Solomon, E. I. J. Am. Chem. Soc. 2004, 126, 8842−8855.
(23) Clausen, J.; Junge, W. Nature 2004, 430, 480−483.
(24) Hamberg, M.; Su, C.; Oliw, E. J. Biol. Chem. 1998, 273, 13080−
13088.
(49) Lehnert, N.; Ho, R. Y. N.; Que, L. J.; Solomon, E. I. J. Am.
Chem. Soc. 2001, 123, 8271−8290.
(50) Denisov, I. G.; Makris, T. M.; Sligar, S. G.; Schlichting, I. Chem.
Rev. 2005, 105, 2253−2277.
(51) Que, L.; Tolman, W. B. Nature 2008, 455, 333−340.
(52) Walling, C.; Buckler, S. A. J. Am. Chem. Soc. 1955, 77, 6032−
6038.
(53) DiPasquale, A. G.; Hrovat, D. A.; Mayer, J. M. Organometallics
2006, 25, 915−924.
(54) Coggins, M. K.; Toledo, S.; Shaffer, E.; Kaminsky, W.; Shearer,
J.; Kovacs, J. A. Inorg. Chem. 2012, 51, 6633−6644.
(55) Live, D. H.; Chan, S. I. Anal. Chem. 1970, 42, 791.
(56) Van Geet, A. L. Anal. Chem. 1968, 40, 2227−2229.
(57) Neese, F. ORCA-an ab initio,density functional and semi-
empirical program package, Version 2.7; University of Bonn: Bonn,
Germany, 2010.
(58) Roemelt, M.; Beckwith, M. A.; Duboc, C.; Collomb, M.-N.;
Neese, F.; DeBeer, S. Inorg. Chem. 2012, 51, 680−687.
(59) Buhl, M.; Kabrede, H. J. Chem. Theory Comput. 2006, 2, 1282−
1290.
(25) Skrzypczak-Jankun, E.; Bross, R. A.; Carroll, R. T.; Dunham, W.
R.; Funk, M. O. J. Am. Chem. Soc. 2001, 123, 10814−10820.
(26) Kovaleva, E. G.; Lipscomb, J. D. Science 2007, 316, 453−456.
(27) Emerson, J. P.; Kovaleva, E. G.; Farquhar, e. R.; Lipscomb, J. D.;
Que, L., Jr. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 7347−7352.
(28) McGarrigle, E. M.; Gilheany, D. G. Chem. Rev. 2005, 105,
1536−1602.
(60) Lenthe, E. v.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys. 1993,
99, 4597.
(61) van Wullen, C. J. Chem. Phys. 1998, 109, 392.
(62) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297.
(63) Klamt, A.; Schuurmann, G. J. Chem. Soc., Perkin Trans. 1993,
̈
̈
799.
(64) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Phys.
Rev. Lett. 2003, 91, 146401.
(65) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Phys.
Rev. Lett. 2004, 120, 6898−6911.
(66) Waasmaier, D.; Kirfel, A. Acta Crystallogr. A. 1995, 51, 416.
(67) Shannon, R. D. Acta Crystallogr. 1976, A32, 751−767.
(68) Kitajima, N.; Katayama, T.; Fujisawa, K.; Iwata, Y.; Moro-oka, Y.
J. Am. Chem. Soc. 1993, 115, 7872−7873.
(69) Hikichi, S.; Komatsuzaki, H.; Akita, M.; Moro-oka, Y. J. Am.
Chem. Soc. 1998, 120, 4699−4710.
(70) Chavez, F. A.; Mascharak, P. K. Acc. Chem. Res. 2000, 33, 539−
545.
(29) Boisvert, L.; Goldberg, K. I. Acc. Chem. Res. 2012, 45, 899−910.
(30) Zhang, C.; Xu, Z.; Shen, T.; Wu, G.; Zhang, L.; Jiao, N. Org.
Lett. 2012, 14, 2362−2365.
(31) Brinksma, J.; Hage, R.; Kerschner, J. L.; Feringa, B. L. Chem.
Commun. 2000, 537−538.
(32) Coggins, M. K.; Kovacs, J. A. J. Am. Chem. Soc. 2011, 133,
12470−12473.
(33) Geiger, R. A.; Wijeratne, G.; Day, V. W.; Jackson, T. A. Eur. J.
Inorg. Chem. 2012, 10, 1598−1608.
(34) Annaraj, J.; Cho, J.; Lee, Y.-M.; Kim, S. Y.; Latifi, R.; de Visser, S.
P.; Nam, W. Angew. Chem., lnt. Ed. 2009, 48, 4150−4153.
(35) Geiger, R. A.; Chattopadhyay, S.; Day, V. W.; Jackson, T. A. J.
Am. Chem. Soc. 2010, 132, 2821−2831.
(71) Chavez, F. A.; Rowland, J. M.; Olmstead, M. M.; Mascharak, P.
K. J. Am. Chem. Soc. 1998, 120, 9015.
(72) Komatsuzaki, H.; Sakamoto, N.; Satoh, M.; Hikichi, S.; Akita,
M.; Moro-oka, Y. Inorg. Chem. 1998, 37, 6554−6555.
(73) Henson, M. J.; Vance, M. A.; Zhang, C. X.; Liang, H.-C.; Karlin,
K. D.; Solomon, E. I. J. Am. Chem. Soc. 2003, 125, 5186−5192.
(74) Hatcher, L. Q.; Vance, M. A.; Narducci Sarjeant, A. A.;
Solomon, E. I.; Karlin, K. D. Inorg. Chem. 2006, 3004−3013.
(75) Hatcher, L. Q.; Lee, D.-H.; Vance, M. A.; Milligan, A. E.;
Sarangi, R.; Hodgson, K. O.; Hedman, B.; Solomon, E. I.; Karlin, K. D.
Inorg. Chem. 2006, 45, 10055−10057.
(36) Geiger, R. A.; Leto, D. F.; Chattopadhyay, S.; Dorlet, P.;
́ ̀
Anxolabehere-Mallart, E.; Jackson, T. A. Inorg. Chem. 2011, 50,
10190−10203.
(37) Hearn, A. S.; Tu, C. K.; Nick, H. S.; Silverman, D. N. J. Biol.
Chem. 1999, 274, 24457−24460.
(38) Seo, M. S.; Kim, J. Y.; Annaraj, J.; Kim, Y.; Lee, Y.-M.; Kim, S.-J.;
Nam, W. Angew. Chem., lnt. Ed. 2007, 46, 377−380.
(39) Bossek, U.; Weyhermuller, T.; Wieghardt, K.; Nuber, B.; Weiss,
J. J. Am. Chem. Soc. 1990, 112, 6387−6388.
(76) Lee, Y.; Park, G. Y.; Lucas, H. R.; Vajda, P. L.; Kamaraj, K.;
Vance, M. A.; Milligan, A. E.; Woertink, J. S.; Siegler, M. A.; Narducci
Sarjeant, A. A.; Zakharov, L. N.; Rheingold, A. L.; Solomon, E. I.;
Karlin, K. D. Inorg. Chem. 2009, 48, 11297−11309.
(77) Roelfes, G.; Vrajmasu, V.; Chen, K.; Ho, R. Y. N.; Rohde, J.-U.;
Zondervan, C.; Crois, R. M.; Schudde, E. P.; Lutz, M.; Spek, A. L.;
Hage, R.; Feringa, B. L.; Munck, E.; Que, L., Jr Inorg. Chem. 2003, 42,
2639−2653.
(40) Groni, S.; Dorlet, P.; Blain, G.; Bourcier, S.; Guillot, R.;
́ ̀
Anxolabehere-Mallart, E. Inorg. Chem. 2008, 47, 3166−3172.
(41) VanAtta, R. B.; Strouse, C. E.; Hanson, L. K.; Valentine, J. S. J.
Am. Chem. Soc. 1987, 109, 1425−1434.
(42) Kitajima, N.; Komatsuzaki, H.; Hikichi, S.; Osawa, M.; Moro-
oka, Y. J. Am. Chem. Soc. 1994, 116, 11596−11597.
(43) Singh, U. P.; Sharma, A. K.; Hikichi, S.; Komatsuzaki, H.; Moro-
oka, Y.; Akita, M. Inorg. Chim. Acta 2006, 359, 4407−4411.
(44) Liu, S.-Y.; Soper, J. D.; Yang, J. Y.; Rybak-Akimova, E. V.;
Nocera, D. G. Inorg. Chem. 2006, 45, 7572−7574.
(78) Cho, J.; Sarangi, R.; Nam, W. Acc. Chem. Res. 2012, 45, 1321−
1330.
L
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(79) Solomon, E. I.; Brunold, T. C.; Davis, M. I.; Kemsley, J. N.; Lee,
S. K.; Lehnert, N.; Neese, F.; Skulan, A. J.; Yang, Y. S.; Zhou, J. Chem.
Rev. 2000, 100, 233−349.
(80) Lugo-Mas, P.; Dey, A.; Xu, L.; Davin, S. D.; Benedict, J.;
Kaminsky, W.; Hodgson, K. O.; Hedman, B.; Solomon, E. I.; Kovacs, J.
A. J. Am. Chem. Soc. 2006, 128, 11211−11221.
(81) This shift in peroxo contribution is diminished by a shift in
contributions from the thiolate sulfur, which is predicted to rise from
∼5.7% to ∼11%.
(82) Halpern, J. Bull. Chem. Soc. Jpn. 1988, 61, 13−15.
(83) Liu, L.; Guo, Q.-X. Chem. Rev. 2001, 101, 673−696.
(84) Jones, P. G.; Kirby, A. J. J. Am. Chem. Soc. 1984, 106, 6207−
6212.
(85) Mayer, J. M. Acc. Chem. Res. 1998, 31, 441−450.
(86) Jensen, M. P.; Mairata i Payeras, A.; Fielder, A. T.; Costas, M.;
Kaizer, J.; Stubna, A.; Munck, E.; Que, L., Jr. Inorg. Chem. 2007, 46,
̈
2398−2408.
(87) Chin, D.-H.; La Mar, G. N.; Balch, A. L. J. Am. Chem. Soc. 1980,
102, 4344−4350.
M
dx.doi.org/10.1021/ja308915x | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX