Synthesis, characterization, and physicochemical properties of manganese(III) and manganese(V)-oxo corrolazines

Article abstract of DOI:10.1021/ic0503636

The structural and physicochemical properties of the manganese-corrolazine (Cz) complexes (TBP₈Cz)Mn^V≡O (1) and (TBP ₈Cz)Mn^III (2) (TBP = p-tert-butylphenyl) have been determined. Recrystallization of 2 from toluene/ MeOH resulted in the crystal structure of (TBP₈Cz)Mn^III(CH₃OH) (2·MeOH). The packing diagram of 2·MeOH reveals hydrogen bonds between MeOH axial ligands and meso N atoms of adjacent molecules. Solution binding studies of 2 with different axial ligands (Cl^-, Et ₃PO, and Ph₃PO) reveal strong binding, corroborating the preference of the Mn^III ion for a five-coordinate environment. High-frequency and field electron paramagnetic resonance (HFEPR) spectroscopy of solid 2·MeOH shows that 2·MeOH is best described as a high-spin (S = 2) Mn^III complex with zero-field splitting parameters typical of corroles. Structural information on 1 was obtained through an X-ray absorption near-edge structure (XANES)/extended X-ray absorption fine structure (EXAFS) study and compared to XANES/ EXAFS data for 2·MeOH. The XANES data for 1 shows an intense pre-edge transition characteristic of a high-valent metal-oxo species, and a best fit of the EXAFS data gives a short Mn-O bond distance of 1.56 A, confirming the structure of the metal-oxo unit in 1. Detailed spectroelectrochemical studies of 1 and 2 were performed revealing multiple reversible redox processes for both complexes, including a relatively low potential for the Mn^V → Mn^IV process in 1 (near 0.0 V vs saturated calomel reference electrode). Chemical reduction of 1 results in the formation of a Mn^IIIMn^IV(μ-O) dimer as characterized by electron paramagnetic resonance spectroscopy.

Full text of DOI:10.1021/ic0503636

Inorg. Chem. 2005, 44, 4485−4498
Synthesis, Characterization, and Physicochemical Properties of
Manganese(III) and Manganese(V) Oxo Corrolazines
−
David E. Lansky,† Beaven Mandimutsira,† Bobby Ramdhanie,† Maria Clause´n,§ James Penner-Hahn,§
S. A. Zvyagin,‡ Joshua Telser,¶ J. Krzystek,‡ Riqiang Zhan,| Zhongping Ou,| Karl M. Kadish,*,|
Lev Zakharov, Arnold L. Rheingold, and David P. Goldberg*,†
Department of Chemistry, Johns Hopkins UniVersity, 3400 North Charles Street,
Baltimore, Maryland 21218, Department of Chemistry, UniVersity of Michigan,
Ann Arbor, Michigan 48109, Department of Chemistry and Biochemistry, UniVersity of California,
San Diego, 9500 Gilman DriVe, La Jolla, California 92093, National High Magnetic Field
Laboratory, Florida State UniVersity, Tallahassee, Florida 32310, Chemistry Program, RooseVelt
UniVersity, Chicago, Illinois 60605, and Department of Chemistry, UniVersity of Houston,
Houston, Texas 77204-5003
Received March 9, 2005
The structural and physicochemical properties of the manganese
(1) and (TBP8Cz)MnIII (2) (TBP
p-tert-butylphenyl) have been determined. Recrystallization of 2 from toluene/
MeOH resulted in the crystal structure of (TBP8Cz)MnIII(CH3OH) (2
MeOH). The packing diagram of 2 MeOH reveals
− O
corrolazine (Cz) complexes (TBP8Cz)MnVt
)
‚
‚
hydrogen bonds between MeOH axial ligands and meso N atoms of adjacent molecules. Solution binding studies
of 2 with different axial ligands (Cl-, Et3PO, and Ph3PO) reveal strong binding, corroborating the preference of the
MnIII ion for a five-coordinate environment. High-frequency and field electron paramagnetic resonance (HFEPR)
spectroscopy of solid 2‚MeOH shows that 2‚MeOH is best described as a high-spin (S )
2) MnIII complex with
zero-field splitting parameters typical of corroles. Structural information on 1 was obtained through an X-ray absorption
near-edge structure (XANES)/extended X-ray absorption fine structure (EXAFS) study and compared to XANES/
EXAFS data for 2
valent metal oxo species, and a best fit of the EXAFS data gives a short Mn
the structure of the metal oxo unit in 1. Detailed spectroelectrochemical studies of 1 and 2 were performed revealing
multiple reversible redox processes for both complexes, including a relatively low potential for the MnV MnIV
process in 1 (near 0.0 V vs saturated calomel reference electrode). Chemical reduction of 1 results in the formation
of a MnIIIMnIV(
-O) dimer as characterized by electron paramagnetic resonance spectroscopy.
‚
MeOH. The XANES data for 1 shows an intense pre-edge transition characteristic of a high-
−
−O bond distance of 1.56 Å, confirming
−
f
µ
Introduction
one of the meso atoms has been deleted and replaced with
a direct pyrrole-pyrrole bond. The remaining three meso
Interest in synthetic porphyrins and analogous porphy-
rinoid species stems from their many applications in catalysis
and materials and medicinal chemistry and from their use
as models for biological systems. Corrolazines,1-7 and their
related corroles,7-11 are porphyrinoid compounds in which
(2) Ramdhanie, B.; Zakharov, L. N.; Rheingold, A. L.; Goldberg, D. P.
Inorg. Chem. 2002, 41, 4105-4107.
(3) Mandimutsira, B. S.; Ramdhanie, B.; Todd, R. C.; Wang, H. L.;
Zareba, A. A.; Czernuszewicz, R. S.; Goldberg, D. P. J. Am. Chem.
Soc. 2002, 124, 15170-15171.
(4) Wang, S. H. L.; Mandimutsira, B. S.; Todd, R.; Ramdhanie, B.; Fox,
J. P.; Goldberg, D. P. J. Am. Chem. Soc. 2004, 126, 18-19.
(5) Ramdhanie, B.; Telser, J.; Caneschi, A.; Zakharov, L. N.; Rheingold,
A. L.; Goldberg, D. P. J. Am. Chem. Soc. 2004, 126, 2515-2525.
(6) Fox, J. P.; Ramdhanie, B.; Zareba, A. A.; Czernuszewicz, R. S.;
Goldberg, D. P. Inorg. Chem. 2004, 43, 6600-6608.
* To whom correspondence should be addressed. E-mail: dpg@jhu.edu.
(D.P.G.); kkadish@uh.edu (K.M.K.).
† Johns Hopkins University.
§ University of Michigan.
University of California, San Diego.
‡ Florida State University.
¶ Roosevelt University.
(7) Gryko, D. T.; Fox, J. P.; Goldberg, D. P. J. Porphyrins Phthalocyanines
2004, 8, 1091-1105.
(8) Gryko, D. T. Eur. J. Org. Chem. 2002, 1735-1743.
(9) Gross, Z. JBIC, J. Biol. Inorg. Chem. 2001, 6, 733-738.
| University of Houston.
(1) Ramdhanie, B.; Stern, C. L.; Goldberg, D. P. J. Am. Chem. Soc. 2001,
123, 9447-9448.
10.1021/ic0503636 CCC: $30.25
Published on Web 05/24/2005
© 2005 American Chemical Society
Inorganic Chemistry, Vol. 44, No. 13, 2005 4485

Lansky et al.
positions are occupied by carbon atoms in the case of corroles
and nitrogen atoms in the case of corrolazines. Corroles have
been the focus of significant synthetic efforts in recent years,
which has led to the discovery of facile synthetic methods
that have made corroles and their metal complexes more
accessible.8,12-15 Corrolazines, which are new entries in the
general corrole family, are also easy to prepare, and the first
example of a corrolazine was synthesized by some of us a
few years ago.1 Corroles and corrolazines exhibit physico-
chemical properties and reactivity patterns that resemble, but
are distinctly different from, porphyrins and other porphyrin
analogues. Determining these properties for corroles and
corrolazines should provide insight into the physicochemical
and reactivity properties of porphyrins in general, including
their biological counterparts in heme-containing enzymes and
proteins.
In a previous paper we reported the synthesis of a
manganese(V)-oxo corrolazine, (TBP8Cz)MnVtO (TBP )
p-tert-butylphenyl; Cz ) corrolazine), (1).3 This compound,
which was found to be stable at room temperature, was
characterized by a variety of spectroscopic techniques, but
no direct structural information from X-ray studies were
reported. High-valent metal-oxo porphyrinoid compounds
are of particular importance because of their putative roles
in catalysis16,17 and heme enzyme mechanisms.18,19 In
particular, high-valent iron-oxo species are implicated as
intermediates in synthetic and biological porphyrin-mediated
oxidations, such as in cytochrome P450.20,21 In related
studies, manganese porphyrins have been examined as
catalysts for P450-type transformations (epoxidation, hy-
droxylation, N-dealkylation), and a high-valent manganese-
oxo porphyrin intermediate is usually invoked.22-25 Mecha-
nistic proposals for the water oxidation process in photo-
synthesis have also included putative MnVtO species as
intermediates.26-28 In many of these earlier studies, a high-
valent metal-oxo porphyrin species is not observed directly
but is invoked from indirect evidence. In this regard, the
isolation, direct characterization, and spectroscopic properties
of 1 are of interest. In addition, our earlier report described
the synthesis of a manganese(III)-corrolazine complex,
which was formulated as the four-coordinate [(TBP8Cz)MnIII]
(2). This compound was spectroscopically characterized by
UV-vis and fast atom bombardment mass spectrometry
(FAB-MS), but its structure and other physical properties
were not described. Although a four-coordinate environment
for MnIII is rare, it was observed in manganese(III) corroles,29
and thus we wanted to better define the structure and axial
ligand preferences of 2.
In this work, further spectroscopic measurements on 1 and
2 are described with the goal of unambiguously assigning
their structures as well as determining in detail several of
their physicochemical properties, which have implications
for their stability and reactivity. A complete structural
characterization of 2 by single-crystal X-ray diffraction has
been carried out, and the axial ligand binding properties of
2 have also been investigated. Although X-ray quality single
crystals of 1 have thus far eluded preparation, direct structural
analysis of 1 has been obtained through an X-ray absorption
near-edge structure (XANES)/extended X-ray absorption fine
structure (EXAFS) study. High-field electron paramagnetic
resonance (HFEPR) spectroscopy has been employed to
conclusively assign both the spin and the oxidation states of
the Mn ion in 2 and to better understand the ligand field
induced by the corrolazine macrocycle. The electrochemical
and spectroelectrochemical behaviors of both 1 and 2 have
been determined to define the general redox behavior of these
Mn corrolazines and compare their behavior with that of Mn
corroles. In addition, these measurements provide insight into
the ability of corrolazines to stabilize high-valent species such
as 1, which is an important feature of corroles in general.
(10) Erben, C.; Will, S.; Kadish, K. M. In The Porphyrin Handbook; Kadish,
K. M., Smith, K. M., Guilard, R., Eds.; Academic Press: New York,
2000; Vol. 2, pp 233-300.
(11) Sessler, J. L.; Weghorn, S. J. Expanded, Contracted, & Isomeric
Porphyrins; Elsevier Science Inc.: New York, 1997; Vol. 15.
(12) Collman, J. P.; Decre´au, R. A. Tetrahedron Lett. 2003, 44, 1207-
1210.
(13) Paolesse, R.; Jaquinod, L.; Nurco, D. J.; Mini, S.; Sagone, F.; Boschi,
T.; Smith, K. M. Chem. Commun. 1999, 1307-1308.
(14) Gross, Z.; Galili, N.; Saltsman, I. Angew. Chem., Int. Ed. 1999, 38,
1427-1429.
(15) Guilard, R.; Barbe, J. M.; Stern, C.; Kadish, K. M. In The Porphyrin
Handbook; Kadish, K. M., Smith, K. M., Guilard, R., Eds.; Elsevier:
San Diego, CA, 2003; Vol. 18, pp 303-349.
(16) Sheldon, R. A. Metalloporphyrins in Catalytic Oxidations; Marcel
Dekker: New York, 1994.
(17) Sheldon, R. A.; Kochi, J. K. Metal-Catalyzed Oxidations of Organic
Compounds; Academic Press: New York, 1981.
Experimental Section
General Methods and Instrumentation. The starting material
octa-tert-butylphenyl corrolazine [(TBP8Cz)H3] was synthesized
according to published procedures.1 The manganese(V)-oxo
complex 1 appeared to be stable to air, light, and moisture in the
solid state, but in solution it was handled under air-free conditions
as a precaution. The manganese(III) complex 2 was, in general,
stable to air, light, and moisture, and no special precautions were
necessary in its handling. Tetrahydrofuran was distilled from
sodium/benzophenone, and pyridine was distilled from CaH2. Other
solvents and reagents were of reagent-grade quality and used as
received from commercial sources. NMR spectra were recorded
on a Varian Unity FT-NMR instrument at 400 MHz (1H). All
spectra were recorded in 5-mm o.d. NMR tubes, and chemical shifts
were reported as δ values from standard solvent peaks. Fourier
(18) Collman, J. P.; Boulatov, R.; Sunderland, C. J.; Fu, L. Chem. ReV.
2004, 104, 561-588.
(19) English, A. M.; Tsaprailis, G. AdV. Inorg. Chem. 1995, 43, 79-125.
(20) Meunier, B.; de Visser, S. P.; Shaik, S. Chem. ReV. 2004, 104, 3947-
3980.
(21) Ortiz de Montellano, P. R.; De Voss, J. J. Nat. Prod. Rep. 2002, 19,
477-493.
(22) McLain, J. L.; Lee, J.; Groves, J. T. In Biomimetic Oxidations
Catalyzed by Transition Metal Complexes; Meunier, B., Ed.; Imperial
College Press: London, 2000; pp 91-169.
(26) Carrell, T. G.; Tyryshkin, A. M.; Dismukes, G. C. JBIC, J. Biol. Inorg.
Chem. 2002, 7, 2-22.
(27) Pecoraro, V. L.; Hsieh, W.-Y. In Manganese and Its Role in Biological
Processes; Sigel, A., Sigel, H., Eds.; Marcel Dekker: New York, 2000;
Vol. 37, pp 429-504.
(23) Jin, N.; Bourassa, J. L.; Tizio, S. C.; Groves, J. T. Angew. Chem., Int.
Ed. 2000, 39, 3849-3851.
(24) Jin, N.; Groves, J. T. J. Am. Chem. Soc. 1999, 121, 2923-2924.
(25) Groves, J. T.; Lee, J.; Marla, S. S. J. Am. Chem. Soc. 1997, 119, 6269-
6273.
(28) Hoganson, C. W.; Babcock, G. T. Science 1997, 277, 1953-1956.
(29) Licoccia, S.; Morgante, E.; Paolesse, R.; Polizio, F.; Senge, M. O.;
Tondello, E.; Boschi, T. Inorg. Chem. 1997, 36, 1564-1570.
4486 Inorganic Chemistry, Vol. 44, No. 13, 2005

Manganese(III) and Manganese(V)-Oxo Corrolazines
transform infrared (FT-IR) spectra were recorded on a Perkin-
Elmer RX I FT-IR spectrometer. Electron paramagetic resonance
(EPR) spectra were obtained on a Bruker EMX EPR spectrometer
controlled with a Bruker ER 041 X G microwave bridge. The EPR
spectrometer was equipped with an Oxford Instruments continuous-
flow liquid helium cryostat and an ITC503 temperature controller.
We simulated X-band EPR spectra using the program DDPOWH3
by Telser. UV-vis spectral studies were performed with an Agilent
8453 diode array spectrometer equipped with HPChemstation
software and with a Varian Cary Bio 50 UV-vis spectrophotometer.
We obtained mass spectrometry data using a Kratos SEQ MALDI-
TOF mass spectrometer. Elemental analyses were performed by
Desert Analytics, Tucson, AZ.
dimensions: 0.50 × 0.20 × 0.10 mm-3, 2θmax ) 56.6°, total number
of reflections ) 64340, independent reflections ) 23476 (Rint
)
0.0568)]. The structure was solved by direct methods, completed
by subsequent difference Fourier syntheses, and refined by full
matrix least-squares procedures on F2. All non-hydrogen atoms were
refined with anisotropic displacement coefficients except the
disordered carbon atoms of one phenyl ring in the main molecule
2 and the solvent toluene molecules that were refined with isotropic
thermal parameters. In the crystal structure, there were three toluene
solvent molecules that were refined as riding groups. The H atoms
were placed at calculated positions and refined in a riding group
model. The number of refinement parameters is 1004. The relatively
large thermal parameters of solvent toluene molecules and the
carbon atoms of the methyl groups in 2 indicate significant
flexibility of the methyl groups and solvent molecules in the crystal
structure. The final R factors [I > 2σ(I)]: R1 ) 0.1125, wR2 )
0.3224, GOF ) 1.169. All software and sources of the scattering
factors are contained in the SHELXTL (5.10) program package
(G. Sheldrick, Bruker XRD, Madison, WI).
Binding Studies. The addition of axial ligands to 2 was
performed in a 250-mL Schlenk flask that was specially modified
to contain a 1-cm path-length quartz cell (total volume ) 3 mL)
fused to a sidearm assembly, which could then be filled by tilting
the flask and allowing the solution to drain from the quartz cell
into the sidearm attachment. This setup allowed for the convenient
addition of aliquots of stock solutions of the axial ligands (Cl-,
Et3PO, Ph3PO) to 2. A typical experiment involving the addition
of Cl- is as follows: complex 2 was dissolved in CH2Cl2 to give
a 3.3 µM solution. A stock solution of tetraethylammonium chloride
(Et4NCl) in CH2Cl2 (0.33 µM) was added in successive aliquots
(0-1.3 equiv) by syringe to the solution of 2 until no further change
was observed in the UV-vis spectrum. The titrations were
performed at different fixed concentrations of 2 to confirm that
the binding constants were not dependent on the concentration of
2.
[(TBP8Cz)MnIII] (2). The initial synthesis of 2 involved the
addition of MnIII(acac)3 (acac ) acetylacetonate) to the metal-free
corrolazine [(TBP8Cz)H3] in pyridine under high-temperature
conditions.3 Since that first report, we have found an alternative,
low-temperature procedure that gives comparable yields of 2, which
is as follows: an amount of MnIII(acac)3 (0.257 g, 0.738 mmol)
was added to [(TBP8Cz)H3] (0.100 g, 0.0738 mmol) in CH2Cl2/
CH3OH (90:10 v:v). The reaction mixture was allowed to stir at
room temperature for 18 h, after which time the solvents were
removed on a rotary evaporator to give a dark blue solid. The crude
product was purified by chromatography (CH2Cl2/CH3OH) (99:1
v:v; silica gel) to give 2 as a dark green-brown solid (0.098 g, 95%).
Crystals suitable for X-ray crystallography were grown from the
slow vapor diffusion of MeOH into a solution of 2 in toluene. These
crystals were also used for elemental analysis. Rf ) 0.88, CH2Cl2/
MeOH (99:1) on silica gel. UV-vis (CH2Cl2): λmax, nm (ꢀ × 10-4
M-1 cm-1): 435 (5.85), 562 (1.23), 634 (1.46), 685 (3.49). FAB-
MS (m/z) (NBA matrix): isotopic cluster centered at 1411.4 (M+),
1
1434.4 (M + Na, w/ NaI added to the NBA matrix). H NMR
(400 MHz, CD2Cl2): δ (ppm) 12.89 (br), 9.25 (br), 7.13 (br), 1.62
(br); FT-IR (thin film) (cm-1) 2961 vs, 2868 m, 1735 w, 1655 w,
1460 m, 1394 w, 1363 m, 1268 m, 1200 w, 1162 m, 1106 s, 1016
m, 999 vs, 864 w, 835 m. Anal. Calcd for C97H108N7MnO
(2‚CH3OH): C, 80.74; H, 7.54; N, 6.79. Found: C, 80.56; H, 7.32;
N, 6.79.
A general equilibrium expression is given in eq 1, along with a
definition of the binding constant in eq 2.
(TBP8Cz)Mn + L h (TBP8Cz)Mn(L)
(1)
(2)
[(TBP8Cz)MnV(O)] (1). In a previous report, 1 was synthesized
by the addition of meta-chloroperbenzoic acid (mCPBA) to 2 in
CH2Cl2.3 An alternative method in which the oxidant has been
changed to iodosylbenzene (PhIO) has been developed to avoid
difficulties encountered from overoxidation by mCPBA. Excess
PhIO (0.763 g, 3.471 mmol) was added to a solution of 2 (0.50 g,
0.347 mmol) in CH2Cl2 (25 mL), and the reaction mixture was
stirred at room temperature for 30 min. PhIO is only sparingly
soluble in CH2Cl2, and thus an excess was added to ensure the
complete oxidation of 2. The solution changed from a dark green-
brown to a deep green color. The reaction was monitored by UV-
vis spectroscopy to confirm the complete conversion of 2 to 1. The
solvent was then removed on a rotary evaporator, and the residue
was purified by column chromatography (CH2Cl2, silica gel) to give
1 as a dark green solid. Any unreacted PhIO, which is mostly
insoluble, was easily removed by the column purification of 1. The
Ka ) [(TBP8Cz)Mn(L)]/{[(TBP8Cz)Mn][L]}
An analytical expression for relating the concentration of the
bound complex to the amount of ligand added for a 1:1 binding
event can be derived from eq 2 by using mass balance relationships
and is given in eq 3
[(TBP8Cz)Mn(L)] ) [{[L]0 + [Mn]0 + (1/Ka)} +
({[L]0 + [Mn]0 + (1/Ka)}2 - 4[Mn]0[L]0)1/2]/2 (3)
in which [L]0 and [M]0 are the total ligand and Mn corrolazine
concentrations, respectively. The measured absorbance is then
related to [(TBP8Cz)Mn(L)] through eq 4
1
complex 1 is diamagnetic, and its H NMR spectrum and other
(A - A0)/(ꢀML - ꢀM) ) [(TBP8Cz)Mn(L)]
(4)
analytical data have been previously reported.3
X-ray Crystallography. Crystal data for 2‚(CH3OH)(C7H8)3 at
173(2) K: C118H132MnN7O, Mr ) 1719.25, monoclinic, C2/c, a )
39.253(3) Å, b ) 21.6263(18) Å, c ) 25.784(2) Å, â )
113.556(2)°, V ) 20064(3) Å3, Z ) 8, F ) 1.138 g cm-3, F(000)
) 7376, and µ(Mo KR) ) 0.185 mm-1. X-ray diffraction intensities
were collected on a Bruker SMART APEX CCD diffractometer
[T ) 173(2) K, Mo KR radiation (λ ) 0.71073 Å), crystal
in which A ) absorbance throughout the titration, A0 ) initial
absorbance before the addition of L, and (ꢀML - ꢀM) is the difference
in the molar absorptivities for the bound and free complex,
respectively. We calculated the Ka from a nonlinear least-squares
fit of the binding isotherm using eqs 3-4. Because ꢀML is not known
with certainty, the quantity (ꢀML - ꢀM) ) ∆ꢀ was also calculated
as part of the fitting procedure.
Inorganic Chemistry, Vol. 44, No. 13, 2005 4487

Lansky et al.
Scheme 1
X-ray Absorption Spectroscopy (XAS) Data Collection.
150-700 GHz range (∼5-23 cm-1 energy) were provided by a
set of four backward-wave oscillators. We precalibrated the
frequency using a Fabry-Perot resonator. The magnet used was
the resistive “Keck” magnet (0-25 T) of improved homogeneity
(12 ppm in a 1-cm diameter spherical volume) and temporal
stability. We precalibrated the field using an NMR probe and
checked the field during the experiment using a DPPH marker. The
oversized-pipes wave propagation system was home-built along the
principles outlined before.34 Detection was provided with an InSb
hot-electron bolometer. Modulation for detection purposes was
provided alternatively either by modulating the magnetic field (1
kHz frequency, 1 mT max amplitude) or by chopping the sub-THz
wave beam at ∼300 Hz. A Stanford SR830 lock-in amplifier
converted the modulated signal to DC voltage.
Manganese K-edge XAS data were collected at the National
Synchrotron Light Source, beam line X9-B (2.8 GeV, ∼300 mA).
We measured data at 14 K in transmission mode using a Si(111)
double crystal monochromator. Samples were diluted in boron
nitride (approximately 1:1 volume ratio), ground to homogeneity,
and packed into 1-mm-thick Al cells with Kapton windows. We
calibrated the energy using KMnO4 as an internal standard, with
the KMnO4 pre-edge transition defined as 6543.3 eV, and averaged
three or four scans. We normalized the XANES spectra by fitting
the data to the McMaster absorption coefficients30 below and above
the edge using a single background polynomial and scale factor.31
The EXAFS was extracted by first removing a polynomial pre-
edge and then subtracting a three region cubic spline with the
EXAFS normalized to the Victoreen absorption falloff.
Electrochemistry and Spectroelectrochemistry. Benzonitrile
(PhCN) and pyridine were obtained from Sigma-Aldrich Co., and
PhCN was distilled over phosphorus pentoxide (P2O5) under
vacuum prior to use. CH2Cl2 was obtained from EM Science, and
was stabilized with an alkene. High purity N2 was used to
deoxygenate the solution before the electrochemical and spectro-
electrochemical experiments. Tetra-n-butylammonium perchlorate
(TBAP) was purchased from Sigma-Aldrich Co., recrystallized
from ethyl alcohol, and dried under vacuum at 40 °C for at least
one week prior to use. Cyclic voltammetry was conducted with an
EG&G model 173 potentiostat. A three-electrode system was used
consisting of a glassy carbon working electrode, a platinum counter
electrode, and a saturated calomel reference electrode (SCE). Half-
wave potentials were calculated from E1/2 ) (Epa + Epc)/2 and were
referenced to the SCE. For spectroelectrochemical measurements,
an optically transparent thin-layer platinum electrode was used as
described previously.35 Time-resolved UV-vis spectra were re-
corded with a Hewlett-Packard model 8453 diode-array rapid-
scanning spectrophotometer.
Absorber-scatterer information was obtained from fitting the
EXAFS data to eq 5
NsAas(k)
ø(k) ) S20
exp(-2k2σ2as) sin[2kRas + φas(k)] (5)
∑
kR2as
s
in which ø(k) is the fractional modulation in the absorption
coefficient above the edge, Ns is the number of scatterers at a
distance Ras, Aas(k) is the effective backscattering amplitude, S2 is
0
a scale factor, σ2as is the mean-square deviation in Ras, φas is the
phase-shift that the photoelectron wave undergoes in passing
through the potentials of the absorbing and scattering atoms, and
the sum is taken over all shells of the scatterers. The k3-weighted
EXAFS was truncated at 11.7 Å-1 because of the presence of a
small Fe edge; spectra were analyzed over k ) 3-11.7 Å-1
.
The program FEFF 7.0232 was used to calculate theoretical
amplitude and phase functions from the crystallographic data of
2‚(CH3OH). For the evaluation of the data for 1, the Mn-O bond
length was shortened to 1.55 Å. The scale factor, S20, was held
fixed at 0.9. The shift in the threshold energy, ∆E0, was allowed
to vary but was limited to a single ∆E0 value for all the different
shells.
Results and Discussion
Synthesis. The manganese(III) and manganese(V)-oxo
complexes were synthesized according to Scheme 1. The
syntheses of both 1 and 2 have been reported in an earlier
communication, but in both cases modifications to the
procedures have improved the syntheses in terms of con-
venience and/or yield. In the case of 2, the metalation
procedure initially involved high-temperature conditions in
pyridine as solvent. We have since discovered that it is
unnecessary to use pyridine, which is toxic and difficult to
remove at the end of the reaction, because one can use a
CH2Cl2/MeOH mixture at room temperature instead; how-
ever, the reaction time needs to be increased from 1 h to 18
HFEPR Spectroscopy. A sample of 2‚(CH3OH) was obtained
by recrystallization and used for HFEPR studies. The typical amount
of 2‚(CH3OH) was 10 to 15 mg, and the material was trapped in
n-eicosane mull to prevent the mechanical torquing of crystallites
in the high magnetic field. A recently constructed tunable-frequency
HFEPR spectrometer was employed.33 Tunable frequencies in the
(30) McMaster, W. H.; Del Grande, N. K.; Mallett, J. H.; Hubbell, J. H.
U.S. Department of Commerce Report Number UCRL-50174-SEC
2-R1, U.S. Goverment Printing Office: Washington, DC, 1969.
(31) Weng, T.-C.; Waldo, G. S.; Penner-Hahn, J. E. J. Synchrotron Radiat.,
in press.
(32) Zabinsky, S. I.; Rehr, J. J.; Ankudinov, A.; Albers, R. C.; Eller, M. J.
Phys. ReV. B 1995, 52, 2995-3009.
(33) Zvyagin, S. A.; Krzystek, J.; van Loosdrecht, P. H. M.; Dhalenne,
G.; Revcolevschi, A. Phys. B (Amsterdam, Neth.) 2004, 346-347,
1-5.
(34) Hassan, A. K.; Pardi, L. A.; Krzystek, J.; Sienkiewicz, A.; Goy, P.;
Rohrer, M.; Brunel, L. C. J. Magn. Reson. 2000, 142, 300-312.
(35) Lin, X. Q.; Kadish, K. M. Anal. Chem. 1985, 57, 1498-1501.
4488 Inorganic Chemistry, Vol. 44, No. 13, 2005

Manganese(III) and Manganese(V)-Oxo Corrolazines
Table 1. Selected Bond Lengths (Å) and Bond Angles (deg) for
2‚MeOH
Mn-N(1)
1.892(4)
1.876(3)
1.885(3)
1.884(3)
0.373(2)
0.582(1)
1.392(1)
1.440(1)
1.445(6)
2.107(3)
3.694(5)
3.683(5)
122.8(3)
92.76(15)
88.58(14)
80.59(14)
89.10(14)
96.92(14)
100.92(14)
106.45(14)
101.83(14)
Mn-N(2)
Mn-N(3)
Mn-N(4)
Mn-[N(1)-N(4)]plane
Mn-(23-atom)core
Câ-Câ (av)
CR-Câ (av)
CR-CR
Mn-O
N(1)‚‚‚N(3)
N(2)‚‚‚N(4)
Mn-O-C(97)
N(1)-Mn-N(2)
N(2)-Mn-N(3)
N(3)-Mn-N(4)
N(4)-Mn-N(1)
N(1)-Mn-O
N(2)-Mn-O
N(3)-Mn-O
N(4)-Mn-O
Figure 1. ORTEP diagram of 2‚MeOH. Thermal ellipsoids are drawn at
a 30% probability level. Solvent molecules and hydrogen atoms have been
omitted for clarity.
interactions between the axial methanol molecule and the
peripheral TBP groups, which may be enhanced by the
hydrogen bonded dimer pairs seen in the packing diagram
(Figure 2). The manganese ion is located 0.582(1) Å above
the plane of the macrocycle toward the axial MeOH ligand.
The metal ion in this case is much further out of the plane
than in the analogous five-coordinate cobalt corrolazines,
with Co-(plane) ) 0.442 Å for [(TBP8Cz)Co(PPh3)] and
Co-(plane) ) -0.076(1) Å for [(TBP8Cz)Co(CCSiPh3)].
The Mn-N(ave) distance is 1.884(7) Å, which is also longer
than the average CoIII-N distances of 1.837(8) and 1.838(3)
Å for [(TBP8Cz)Co(PPh3) ] and [(TBP8Cz)Co(py)2], respec-
tively. The effects of the smaller cavity of a corrolazine as
compared to that of a corrole is usually manifested in short
M-Npyrrole bond lengths, but for the MnIII complex, the
average Mn-N distance is nearly the same as that for the
corroles [(E-7,13-MC)MnIII (1.894 Å)29 and (OEC)MnIII(py)
(1.917 Å)].10 The longer Mn-Npyrrole distances are clearly a
consequence of the large out-of-plane displacement of the
metal ion.
h. Chromatographic purification and isolation of 2 were
performed in the usual manner. We have also further
characterized this compound using elemental analysis, IR
data, and a 1H NMR spectrum, which reveals broad,
paramagnetically shifted peaks typical of a high-spin man-
ganese(III) complex. After recrystallization from toluene/
MeOH, the elemental analysis is consistent with one MeOH
present in the crystalline material, as shown in the X-ray
structure. However, in routine preparations of 2, the ma-
terial obtained after chromatography is typically dried under
high vacuum, and UV-vis spectroscopy indicates that this
material does not contain coordinated MeOH, as discussed
below.
The MnVtO complex 1 was initially prepared from
oxidation of 2 with mCPBA, but care had to be taken to not
add excess mCPBA as to avoid the degradation of 2 to
unidentified products. The replacement of mCPBA with
PhIO gives 2 in comparable yields, and no degradation is
seen even with a large excess of PhIO. The limited solubility
of PhIO in CH2Cl2 also makes it easy to separate from the
product. Both of these complexes have now been character-
ized by X-ray techniques, as described below.
The Mn-O bond distance to the axial methanol ligand is
2.107(3) Å and is similar to that found for [(OEP)MnIII-
X-ray Crystal Structure of [(TBP8Cz)MnIII(MeOH)]‚
(C7H8)3. An ORTEP diagram of 2‚(MeOH) is shown in
Figure 1, and selected bond distances and angles are given
in Table 1. The manganese(III) ion is five-coordinate with a
methanol molecule bound in the axial position. In other
metallocorrolazines such as [(TBP8Cz)Co(PPh3)],2 [(TBP8Cz)-
Co(py)2] (py ) pyridine),2 and [(TBP8Cz)Co(CCSiPh3)],5 the
macrocycle is relatively planar, with the largest deviation
from the least-squares-plane of the 23-atom core being
0.22(1) Å as found in [(TBP8Cz)Co(PPh3)]. For the man-
ganese(III) complex, there is significantly more distortion
in the corrolazine core, as shown by the large deviation from
the 23-atom plane of ∼0.29 Å for N(3) and N(4) and ∼0.32
Å for C(14) and C(15). This distortion may come from steric
Figure 2. A fragment of the crystal structure of 2‚MeOH showing the
hydrogen bonding interaction between two complexes. Solvent and hydrogen
atoms (except for the axial MeOH) have been omitted for clarity.
Inorganic Chemistry, Vol. 44, No. 13, 2005 4489

Lansky et al.
(HOEt)]ClO4 [Mn-(HOEt) ) 2.145 Å].36 As seen in the
packing diagram in Figure 2, the axial MeOH ligands on
neighboring corrolazines form hydrogen bonds with adjacent
meso nitrogen atoms to give hydrogen bonded dimer pairs.
The hydrogen atom of the MeOH ligand was located in the
electron density map and refined in position. The distance
between the O atom and the meso N atom of 2.764(5) Å
and the O-H‚‚‚N angle of 176.2° are clearly consistent with
a hydrogen bond interaction. Corroles10 and porphyrins37
often form π-stacked dimer pairs in the crystal lattice, and
the first example of a π-stacked corrolazine was recently
reported.5 The hydrogen bonded dimer motif seen for
2‚(CH3OH) is much less common and demonstrates that the
meso nitrogen atoms in the corrolazine are potentially good
Lewis basic sites for protonation or possibly even peripheral
metalation.
Binding of Axial Ligands to 2. The addition of MeOH
to the MnIII complex causes significant changes in the UV-
vis spectrum, as seen in Supporting Information Figure S1.
These data show that the MnIII starting material is most likely
a four-coordinate complex, which converts to the five-
coordinate MeOH adduct upon the titration of MeOH in
CH2Cl2. Although we cannot rule out an alternative inter-
pretation corresponding to the conversion of a five-coordinate
(Cz)Mn(MeOH) species to a six-coordinate (Cz)Mn(MeOH)2
species, we believe this mechanism is highly unlikely because
of the five-coordinate nature of the X-ray structure for
2‚(CH3OH) and the similarity of the UV-vis spectra with
other donor ligands, such as Ph3PdO (vide infra).38 The
binding of MeOH is complicated by the possibility of
intermolecular hydrogen bonding as indicated by the X-ray
structure, and thus a binding constant was not calculated for
MeOH. Titration of heterocyclic bases such as pyridine and
1-methylimidazole (1-MeIm) to 2 leads to very small changes
(<5 nm shift in the Soret band) in the UV-vis spectrum
(data not shown). These small changes appear to increase
as the concentration of pyridine or 1-MeIm is increased,
suggesting weak coordination to the MnIII ion. However, the
changes in the spectrum are too small to give a reliable
binding isotherm, and thus quantification of the binding
constants was not possible. In contrast, the binding of Cl-,
Et3PO, or Ph3PO gives significant changes in the UV-vis
spectrum, leading to binding isotherms that can be fit to give
binding constants.
Figure 3. UV-vis titration of tetraethylammonium chloride (0-1.3 equiv)
into a solution of 2 (3.33 × 10-6 M) in CH2Cl2.
Figure 4. Binding isotherm for the addition of Et4NCl to 2 in CH2Cl2
plotted from the data in Figure 3 at 480 nm. The red line is the best fit of
the data to eq 4 in the text, which gives Ka ) 1.34 × 107 M-1 and ∆ꢀ )
1.4 × 104 M-1cm-1. Inset: Job’s plot showing a maximum at [Cl-]/([Cl-]
+ [Mn]) ≈ 0.5, indicating 1:1 binding stoichiometry.
data, the binding isotherm shown in Figure 4 was obtained,
which was fit directly to obtain Ka. We determined the
stoichiometry of binding by using Job’s method, which
yielded a maximum in the Job’s plot that indicates a
corrolazine to Cl- ratio of 1:1, as seen in the inset of Figure
4. The equilibrium expression is given in eq 1, along with a
definition of the binding constant in eq 2. The Ka was
obtained as described in the experimental section. Simpler
methods involving a linearization of the equilibrium expres-
sion to obtain Ka, such as a Hill plot, are commonly employed
for ligand binding to porphyrins and corroles. However, such
methods are only useful in cases of relatively weak binding
in which [L0] - [L] ≈ [L0] and as such would not be useful
for the strong binding behavior observed for 2.39
The addition of Et4NCl to 2 in CH2Cl2 resulted in changes
in the UV-vis spectra as shown in Figure 3, and good
isosbestic behavior was observed. The Soret band splits into
two peaks at 435 and 480 nm while the Q-band increases in
intensity, and a weak shoulder appears at 720 nm. From these
(36) Cheng, B. S.; Scheidt, W. R. Acta Crystallogr., Sect. C 1996, 52, 585-
588.
(37) Scheidt, W. R. In The Porphyrin Handbook; Kadish, K. M., Smith,
K. M., Guilard, R., Eds.; Academic Press: New York, 2000; Vol. 3,
pp 49-112.
(38) We have characterized a five-coordinate 2‚DMF (DMF ) dimethyl-
formamide) complex by X-ray crystallography. The UV-vis spectrum
observed for the addition of DMF to 2 is the same as that seen for the
addition of MeOH. All of these data are consistent with a five-
coordinate complex for both DMF and MeOH. (Lansky, D. E.,
Goldberg, D. P. unpublished results).
(39) Webb, S. J.; Sanders, J. K. M. Inorg. Chem. 2000, 39, 5912-5919.
4490 Inorganic Chemistry, Vol. 44, No. 13, 2005

Manganese(III) and Manganese(V)-Oxo Corrolazines
The best fit to the data for Cl- gave Ka ) 1.3(4) × 107
M-1. The binding constant for Cl- was significantly higher
than that for the phosphine oxides Ph3PO and Et3PO, which
gave binding constants of 7.8(5) × 105 and 8.6(4) × 105
M-1, respectively. In comparison, binding constants for four-
coordinate manganese corroles have been reported for a
series of pyridine ligands and are much smaller (in the range
6-80 M-1), although it was suggested that for these
compounds interactions between macrocycles satisfied the
preference for five-coordination.29 For 2, intermolecular
interactions are discouraged by the sterically bulky peripheral
phenyl substituents. The high affinity for all three ligands
studied is consistent with a “bare” four-coordinate MnIII
center.
HFEPR Spectroscopy of 2‚(CH3OH). HFEPR spectros-
copy was first used by some of us to obtain EPR spectra
from integer-spin (non-Kramers) MnIII (S ) 2) porphyrin
and porphyrazine compounds.40 This study showed that EPR
spectra could be obtained from “EPR-silent” species such
as high-spin MnIII compounds, and some detailed properties
of the electronic structure of these species were determined.
Since that work, multifrequency HFEPR has proven to be a
valuable tool for studying other integer-spin, EPR-silent
species, including the manganese(III) corroles (tpfc)Mn-
(OPPh3), [tpfc ) 5,10,15-tris-(pentafluorophenyl)corrolato]41
and (DEHMC)Mn (DEHMC ) 8,12-diethyl-2,3,7,13,17,18-
hexamethyl-corrolato).42 Recently, the multifrequency ex-
periment was extended to a tunable-frequency version, which
allows greater precision in determining spin Hamiltonian
parameters. By using swept fields (0-25 T) in combination
with high (sub-THz) frequencies, we performed a two-
dimensional (resonant field vs transition energy) EPR study
of the manganese(III)-corrolazine complex, as shown in
Figure 5. Each data point corresponds to a single peak in an
individual spectrum, and the inset to Figure 5 shows a typical
spectrum recorded at 503 GHz (transition energy 16.77
cm-1). At higher resolution, a single-frequency spectrum such
as that in the inset to Figure 5 displays a small splitting of
the perpendicular turning points in the powder spectrum (B0
| x or y) which is indicative of a small but nonzero rhombic
zero-field splitting parameter E. We simulated the data in
Figure 5 according to established procedures40,43 using the
spin Hamiltonian for an S ) 2 system:
Figure 5. Magnetic field vs transition energy dependence of HFEPR
resonances (squares) and simulations using the spin Hamiltonian parameters,
as presented in the text. Black lines: parallel turning points; blue lines:
perpendicular turning points; and red lines: off-axis turning points in the
powder pattern. The small splittings in the perpendicular transitions due to
a nonzero E term are not shown for clarity. Inset: single-frequency spectrum
at 503 GHz and 10 K. The peak at 19.35 T is due to a g ) 2.00 impurity
that is not reflected in the main plot.
simulations using the best fit values: S ) 2, isotropic g )
2.00(1), D ) -2.60(2) cm-1, and |E| ) 0.015(5) cm-1.
These spin Hamiltonian parameters are within the range
of those observed for related MnIII systems, such as
Mn(TPP)Cl,43,44 confirming that 2‚(CH3OH) is a high-spin
S ) 2, MnIII complex with no evidence for the contribution
of a MnII(corrolazine+•) configuration. In particular, the zero-
field splitting parameters (D ) -2.60(2) cm-1; |E| )
0.015(5) cm-1) are similar to those found for the related MnIII
corroles {(tpfc)Mn(OPPh3): D ) -2.69(2) cm-1, E )
0.030(3) cm-1;41 and (DEHMC)Mn: D ) -2.64(1) cm-1,
E ) 0.015(5) cm-1}.42 The isotropic g value of ∼2.00 is
common for most complexes of manganese(III). In contrast
to porphyrins and porphyrazines,40 corrole and corrolazine
complexes of MnIII are not rigorously axial spin systems.
More importantly, the magnitude of D is larger for the latter
class of complexes. It has been pointed out that the smaller
cavity size and higher negative charge of a corrole compared
to that of a porphyrin implies a stronger equatorial ligand
field, which in turn should lead to a smaller magnitude for
D.42 The larger D values seen for corroles as compared to
those of MnIII porphyrins (1.5 < |D| < 2.5 cm-1) have been
explained as one of the triplet (S ) 1) excited states
H ) â B‚g‚S + D[S2z - S(S + 1)/3] + E(Sx2 - S2y) (6)
The observed resonances were unequivocally attributed
to particular transition branches in the S ) 2 spin manifold.
The complex in the solid state is characterized by the
following spin Hamiltonian parameters obtained from the
1
(dxy dxz,yz3) lying closer in energy to the quintet (S ) 2) ground
1
2
21
state (dxy dxz,yz dz ) in MnIII corrole complexes, which is not
the case in porphyrin complexes.42,43 The ordering of the d
orbital energies for these systems is that which was used
previously for MnIII corroles42 and is based on the ordering
proposed by Dugad et al. for MnIII porphyrins.44 The smaller
cavity size of the MnIII corrolazine as compared to that of
(40) Goldberg, D. P.; Telser, J.; Krzystek, J.; Montalban, A. G.; Brunel,
L. C.; Barrett, A. G. M.; Hoffman, B. M. J. Am. Chem. Soc. 1997,
119, 8722-8723.
(41) Bendix, J.; Gray, H. B.; Golubkov, G.; Gross, Z. Chem. Commun.
2000, 1957-1958.
(42) Krzystek, J.; Telser, J.; Hoffman, B. M.; Brunel, L. C.; Licoccia, S.
J. Am. Chem. Soc. 2001, 123, 7890-7897.
(43) Krzystek, J.; Telser, J.; Pardi, L. A.; Goldberg, D. P.; Hoffman, B.
M.; Brunel, L. C. Inorg. Chem. 1999, 38, 6121-6129.
(44) Dugad, L. B.; Behere, D. V.; Marathe, V. R.; Mitra, S. Chem. Phys.
Lett. 1984, 104, 353-356.
Inorganic Chemistry, Vol. 44, No. 13, 2005 4491

Lansky et al.
Figure 6. Normalized Mn XANES spectra for 2 (blue line) and 1 (red
line). Both spectra are plotted on the same scale.
Figure 7. Fourier transform of 2 (blue line) and 1 (red line). Inset: EXAFS
data (solid line) and fit (dashed line) for 2 and 1. Spectra are offset vertically
for clarity.
corroles is alleviated at the manganese center by increasing
the distance of the manganese ion to the plane formed by
the pyrrole N atoms; the Mn displacement in 2‚(CH3OH) is
0.373(2) Å, compared to 0.29 Å in (tpfc)Mn(OPPh3), and
the average Mn-N distances in these two complexes are
approximately the same. Thus, the equatorial and axial
(MeOH vs Ph3PO) ligand fields are similar in these two
complexes, yet the magnitude of D is slightly smaller for
2‚(CH3OH). This difference may come from a larger energy
separation of the triplet-quintet states in the corrolazine
complex. The HFEPR data confirm the spin and oxidation
state of the Mn ion in 2‚(CH3OH) as MnIII (S ) 2)
and demonstrate that the corrolazine ligand field is dif-
ferent from that of porphyrins, being much closer to that of
corroles.
XANES/EXAFS Results. The XANES spectra of 1 and
2‚(CH3OH) are shown in Figure 6. The sample of the MnIII
complex was crystallized from CH3OH/toluene as described
in the experimental section and thus contained the axial
MeOH ligand seen in the crystal structure. There is a shift
of ∼4 eV going from MnIII to MnV, which is consistent with
the oxidation of Mn.45 One of the most striking features of
the XANES spectra is the intense pre-edge transition that is
observed for 1. Such an intense pre-edge feature has been
seen for oxo- and nitrido-porphyrin complexes, as well as
for the crystallographically characterized complex Na[MnVt
O(HMPAB)] [HMPAB ) 1,2-bis(2-hydroxy-2-methyl-
propan-amido)benzene], which was synthesized by Collins
and co-workers.45 This feature can be attributed to a 1s f
(M3d + O2p-π) transition.46
Table 2. Curve Fitting Results for XANES/EXAFS Dataa
coord Rab
coord Rab
coord Rab
sample
Mn(III) 4N 1.89 1.5
no. (Å)b σ2c no. (Å)b σ2c no. (Å)b σ2c ∆E0 Fd
8C 2.89 3.5
3.0 211
3.1 190
4.0 205
4N 1.90 1.6 1O 2.19 6.4 8C 2.89 3.6
4N 1.89 1.5 1O 1.47 4.5 8C 2.90 3.5
Mn(V)
4N 1.86 1.3
4N 1.88 2.8 1O 1.56 1.8 6C 2.90 4.1
4N 1.85 1.5 1O 2.20 4.8 6C 2.83 4.1 -6.9 348
6C 2.84 3.7 -3.5 387
7.3 117
a Fits to unfiltered data. Best fits are shown in bold. b Absorber-scatterer
distance. c Debye-Waller factor (Å2) × 103. d F ) ∑(øexp - øcalc)2k6. The
accuracy in the determination of the bond lengths is approximately 0.02 Å,
and the precision is 0.004 Å.
These are due to scattering from four first shell nitrogens.
Although both Fourier transforms show only a single first
shell peak, it was not possible to fit the data for either
complex without including an additional nearest neighbor
scatterer together with at least one shell of outer shell
scatterers representing the Mn-Ccorrolazine interaction. The best
three shell fits are shown in Figure 7, and the fit results are
summarized in Table 2. The refined Mn-N distances were
virtually identical for the two samples, reflecting the rigidity
of the corrolazine ring. In the case of the MnIII complex, the
first shell peak is asymmetric at higher R, suggesting the
presence of a long Mn-O distance as is seen in the crystal
structure. Consistent with this, the data could be modeled
with a Mn-O distance of 2.19 Å. The agreement with the
Mn-O distance found crystallographically (2.107 Å) is
relatively poor, perhaps reflecting the difficulty of refining
the position of one weak scatterer in the presence of four
stronger scatterers. Attempts to include a short Mn-O
interaction in place of the 2.2-Å Mn-O distance gave worse
fits and unrealistically large Debye-Waller factors.
The most prominent feature of the Fourier transforms of
both samples is the peaks at R + R ≈ 1.5 Å (Figure 7).
To properly fit the (Cz)MnVtO data, it was necessary to
include a short Mn-O distance that refined to 1.56 Å.
Inclusion of a long Mn-O distance in place of the short
Mn-O gave fits that were three times worse and resulted in
an unrealistic shift in ∆E0. The 1.56-Å distance obtained from
(45) Weng, T.-C.; Hsieh, W.-Y.; Uffelman, E. S.; Gordon-Wylie, S. W.;
Collins, T. J.; Pecoraro, V. L.; Penner-Hahn, J. E. J. Am. Chem. Soc.
2004, 126, 8070-8071.
(46) Penner-Hahn, J. E.; Benfatto, M.; Hedman, B.; Takahashi, T.; Doniach,
S.; Groves, J. T.; Hodgson, K. O. Inorg. Chem. 1986, 25, 2255-
2259.
4492 Inorganic Chemistry, Vol. 44, No. 13, 2005

Manganese(III) and Manganese(V)-Oxo Corrolazines
the best fit for 1 is in excellent agreement with the Mn-O
distance determined by X-ray crystallography for Na[MnVt
O(HMPAB)] (Mn-O ) 1.548 Å)47 and the related com-
plexes [MnVtO(PHAB)]- [PHAB ) 1,2-bis(2,2-diphenyl-
2-hydroxyethanamido)benzene; Mn-O ) 1.558(4) Å]48 and
[MnVtO(tetraamido)]- [Mn-O ) 1.549(3) Å]49 and is thus
consistent with the presence of a triply bonded MnVtO unit.
The features at higher R for both data sets are due to single
and multiple scattering involving the corrolazine. Six and
eight carbons were also included in the models, but no
attempt was made to model the multiple scattering.
Although inclusion of a short Mn-O distance is necessary
to fit the EXAFS data for 1, there is no well-defined peak
for the MnVtO unit in the low-R region in Figure 7. Similar
results were obtained for Na[MnVtO(HMPAB)], which was
fit with a short metal-oxo bond but did not exhibit a MnVt
O peak in the Fourier transform, and it was concluded that
it is difficult to determine the presence of a short Mn-O
bond on the basis of EXAFS fitting alone.45 However, as
mentioned above, MtE species with d0-d2 configurations
(e.g., M ) Ti, V, Cr; E ) O, N) and short MtE distances
exhibit the common feature of an intense pre-edge peak in
the XANES spectrum, which provides a definitive signature
for the presence of a MnVtO moiety.45 In contrast, XANES
spectra for a MnIV-oxo (d3) porphyrin showed a significantly
attenuated pre-edge peak.50 The combined results from the
EXAFS fitting and the XANES data provide unambiguous
identification of the MnVtO unit in 1, with a metal-oxo
bond distance of 1.56(2) Å. The area under the pre-edge
peak is identical to that found previously for Na[MnVt
O(HMPAB)], indicating that the MnVtO species accounts
for essentially all of the manganese present in the bulk
material.
Figure 8. Cyclic voltammograms of 2 in PhCN and pyridine containing
0.1 M TBAP. Scan rate ) 0.10 V/s.
Table 3. Half-Wave Potentials (V vs SCE) in Different Solvents
Containing 0.1 M TBAP
oxidation
solvent 2nd 1st
reduction
2nd 3rd
compound
1st
4th
(TBP8Cz)MnIII (2)
PhCN
pyridine
1.24 0.78 -0.91 -1.50a -1.68a -1.92a
0.90b -0.91 -1.50
(TBP8Cz)MnVtO (1) PhCN
1.52b 1.09
0.02 -1.17 -1.66a -1.89a
0.03 -0.91 -1.15 -1.70a
pyridine
CH2Cl2 1.57b 1.02 -0.05 -1.20 -1.71
b
a Epc at scan rate ) 0.1 V/s. Epa at scan rate ) 0.1 V/s.
V in PhCN and E1/2 ) -1.50 V in pyridine. The peak
currents for the first two reductions (at -0.91 and -1.50 V)
are approximately equal in PhCN, but in pyridine the first
reduction peak current is approximately double that of the
second, which may suggest two electrons are added in the
first step under these solution conditions.
Time-resolved thin-layer UV-vis spectra were obtained
during the first reduction of 2 at a controlled-potential of
-1.30 V vs SCE. Two sets of spectral changes with different
isosbestic points are seen in PhCN (Figure 9a), suggesting
two different sets of equilibria and a possible ECE type
electrochemical mechanism [an electron transfer (E) followed
by a chemical step (C) and then another electron transfer
(E)]. The first set of spectral changes occurs from 0 to 150
s of controlled-potential electrolysis, and the second occurs
from 150 to 340 s. In the initial process, the Soret band at
448 nm decreases and splits into two bands at 433 and 474
nm. The Q-band at 693 nm also decreases slightly in intensity
and blue-shifts to 685 nm. In the second step (t ) 150-340
s), the Soret band becomes broader and decreases in intensity
as a shoulder grows in at 532 nm. More significant changes
are seen in the visible region of the spectrum where the band
at 685 nm disappears and is replaced by a broad band
centered at 766 nm. The final spectrum of the reduction
process in PhCN after 340 s is similar to the final UV-vis
spectrum obtained in pyridine via the single one-step
reduction process at an applied potential of -1.20 V. This
is shown in Figure 9b. Further reduction of 2 at an applied
potential of -1.60 V in either solvent (PhCN or pyridine)
gives identical spectral changes, with a loss of the broad
peaks in both the UV (∼460 nm) and the visible (∼766 nm)
Electroreduction of 2. The electrochemistry of 2 was
examined in both PhCN and pyridine. Representative cyclic
voltammograms are shown in Figure 8, and the half-wave
or peak potentials (referenced to SCE) are summarized in
Table 3. A proposed mechanism for the different electro-
chemical transformations of 2 is shown in Scheme 2 and is
based in part on the spectroelectrochemistry data in which
Eapp ) the applied potential in the thin-layer cell (vide infra).
Two reversible oxidations and one reversible reduction are
seen in PhCN, whereas in pyridine there are two reversible
reductions and one irreversible oxidation. There are also
several additional irreversible reduction processes in PhCN
as seen in Figure 8a. The first reversible reduction of 2 occurs
at E1/2 ) -0.91 V in both PhCN and pyridine. A second
reduction is also seen at similar potentials of Epc ) -1.50
(47) Collins, T. J.; Gordon-Wylie, S. W. J. Am. Chem. Soc. 1989, 111,
4511-4513.
(48) MacDonnell, F. M.; Fackler, N. L. P.; Stern, C.; O’Halloran, T. V. J.
Am. Chem. Soc. 1994, 116, 7431-7432.
(49) Miller, C. G.; Gordon-Wylie, S. W.; Horwitz, C. P.; Strazisar, S. A.;
Peraino, D. K.; Clark, G. R.; Weintraub, S. T.; Collins, T. J. J. Am.
Chem. Soc. 1998, 120, 11540-11541.
(50) In this work, an intense pre-edge transition was observed for a Mnv(N)
porphyrin and compared to the weak pre-edge feature of a MnIVdO
porphyrin. Ayougou, K.; Bill, E.; Charnock, J. M.; Garner, C. D.;
Mandon, D.; Trautwein, A. X.; Weiss, R.; Winkler, H. Angew Chem.,
Int. Ed. Engl. 1995, 34, 343-346.
Inorganic Chemistry, Vol. 44, No. 13, 2005 4493

Lansky et al.
Scheme 2
regions and a slight increase near 420 nm (Supporting
Information Figure S2).
µ-oxo dimer in solution (Scheme 2). Related dimer forma-
tions have been shown in the case of (TPP)Mn (TPP )
tetraphenylporphyrin).53-55 Although the mechanism of dimer
formation in Scheme 2 is speculative, and indeed the
intermediate in the first step in Figure 9a may be an unidenti-
fied species, it is reasonable to assume that this first step is
associated with a metal-centered redox process given the
small changes in the UV-vis spectrum, which contrast with
the large changes that are associated with a ligand-centered
process (π-anion radical formation at 340 s in Figure 9a).
Electrooxidation of 2. The first and second oxidations
of 2 in PhCN appear at E1/2 ) 0.78 and 1.24 V, respectively,
and are reversible one-electron-transfer processes. This
contrasts with what is seen in pyridine in which the first
The spectroelectrochemical results indicate that the final
product of the first reduction of 2 (-0.91 V) is the same
species in both PhCN and pyridine. This product is charac-
terized by a broad band at 766 nm that points to a compound
with π-anion-radical character, [(Cz)MnIII]•-.51,52 In PhCN,
an intermediate species appears from 0 to 150 s (Figure 9a),
which has a spectrum similar to that seen for the product
formed by the chemical reduction of 1 with cobaltocene in
a toluene/CH2Cl2 mixture, and is tentatively assigned as a
mixed-valence dimeric species, [(Cz)MnIV(µ-O)MnIII(Cz)]-,
on the basis of EPR spectroscopy (vide infra). A mixed-
valence dimer that is more oxidized than the starting material
could form if the initial reduction of 2 in PhCN produces a
transient, highly unstable [(Cz)MnII]- species that is then
immediately reoxidized to [(Cz)MnIV(O)]- by adventitious
oxygen, which in turn combines with (Cz)MnIII to give a
oxidation is an irreversible multielectron process at Epa
)
0.90 V (Figure 8b). Quite different UV spectral changes are
also observed during the first controlled-potential oxidation
of the MnIII complex in the two solvents, and these spectral
changes are illustrated in Figure 10a.
The reversible one-electron oxidation of 2 at Eapp ) 1.00
V in PhCN leads to a UV-vis spectrum having a broad band
at 784 nm, which suggests formation of a corrolazine
π-cation radical. In contrast, controlled-potential oxidation
of the same compound in pyridine at 0.90 V involves more
than one electron and leads to a species with blue-shifted
Soret and Q-bands located at 420 and 633 nm, respectively.
No π-cation-radical-type bands are observed between 700
and 900 nm, and the final spectrum closely resembles the
spectrum of unreacted 1 in the same solvent (Figure 10b).
These data clearly suggest that the irreversible electrooxi-
dation of 2 in pyridine leads to the electrochemically initiated
conversion of 2 to 1.
Further evidence for a MnIII to MnV-oxo conversion is
provided by the cyclic voltammogram of 2 in pyridine
(Figure 8 and inset of Figure 10a). Here an irreversible
cathodic wave at Epc ) 0.00 V is seen after one scans through
the first irreversible oxidation. The peak potential of this new
reduction process matches closely the peak potential for the
(51) Mack, J.; Stillman, M. J. J. Am. Chem. Soc. 1994, 116, 1292-1304.
(52) Kadish, K. M.; Liu, H. Y.; Anderson, J. E. Inorg. Chim. Acta 1989,
163, 201-205. Similar spectral changes are observed during the one-
electron reduction of a phosphorus(V) corrolazine, and the formation
of a π-anion radical product was verified by ESR spectroscopy
(unpublished results).
(53) Razenberg, J. A. S. J.; Van der Made, A. W.; Smeets, J. W. H.; Nolte,
R. J. M. J. Mol. Catal. 1985, 31, 271-287.
(54) Camenzind, M. J.; Schardt, B. C.; Hill, C. L. Inorg. Chem. 1984, 23,
1984-1986.
(55) Schardt, B. C.; Hollander, F. J.; Hill, C. L. J. Am. Chem. Soc. 1982,
104, 3964-3972.
Figure 9. Time-resolved thin-layer UV-vis spectra during (a) the first
reduction of 2 at -1.30 V in PhCN from 0 to 150 s (i), and from 150 to
340 s (ii) and (b) the first reduction in pyridine at -1.20 V.
4494 Inorganic Chemistry, Vol. 44, No. 13, 2005

Manganese(III) and Manganese(V)-Oxo Corrolazines
Figure 11. Cyclic voltammograms of 1 in PhCN (a), pyridine (b), and
CH2Cl2 (c), 0.1 M TBAP at a scan rate of 0.1 V/s.
Figure 10b for the first reduction of 1 in pyridine (Eapp
)
-0.30 V) show quite clearly that 2 is the final product. This
species has bands at 447, 480, and 691 nm, and these bands
are identical to those observed in Figure 10a for the MnIII
compound in pyridine. Formation of the MnIII complex is
also supported by thin-layer cyclic voltammetry in pyridine,
in which the bulk material is undergoing redox conversions
(Supporting Information Figure S3). On the first scan, four
reductions at -0.02, -0.93, -1.18, and -1.52 V are
observed. However, on the second scan, the two reductions
at -0.02 and -1.18 V disappear, and only the two located
at E1/2 ) -0.93 and -1.52 V remain. Both E1/2 values are
close to the half-wave potentials for the reduction of 2 in
the same solvent, that is, -0.90 and -1.50 V vs SCE (see
Figure 8).
Different results are seen during the reduction of 1 in
PhCN. As illustrated in Figure 12a, the Soret band at 425
nm decreases and red-shifts to 436 nm as a new band at 475
nm grows in; at the same time the visible band at 633 nm
disappears and a new, more intense visible band at 685 nm
is formed. The final UV-vis spectrum has bands at 434,
472, and 685 nm and looks quite similar to the spectrum of
the in-situ-generated [(Cz)MnIV(µ-O)MnIII(Cz)]- obtained
after the chemical reduction of 1 with cobaltocene (see Figure
13a). However, attempts to generate the same µ-oxo dimer
by the bulk electrolysis of 1 were unsuccessful.
The second controlled-potential reduction of 1 at -1.30
V shows two sets of spectral changes, each of which is
accompanied by well-defined isosbestic points (Figure 12b).
The first occurs from 0 to 60 s, and the second occurs from
60 to 160 s. The initial set of spectral changes is characterized
by a decrease in the split Soret band at 436 nm, an increase
in the band at 473 nm, and a slight increase in the intensity
of the visible band at 683 nm. The shape and position of the
bands in this spectrum closely resemble those in the spectrum
obtained during either the first reduction of 1 (Figure 12a)
or the first reduction of 2 (Figure 9). In contrast, the spectral
changes between 60 and 160 s are more pronounced (Figure
12b), and the final spectrum has bands at 478, 518, and 792
nm.
Figure 10. Time-resolved thin-layer spectral changes of (a) 2 during the
first oxidation in PhCN at 1.00 V and pyridine at 0.90 V and (b) 1 during
reduction in pyridine at -0.30 V, 0.1 M TBAP.
first reduction of 1 in either solvent (vide infra). Taken
together, the electrochemical and spectroelectrochemical data
are self-consistent and point to the formation of 1 in pyridine
upon the oxidation of the MnIII corrolazine. The fact that
this conversion takes place in pyridine and not in PhCN is
also consistent with the properties of the two solvents. Trace
water, which is the most likely source of the terminal oxo
group in 1, can be expected to be in higher concentration in
pyridine than in PhCN. In addition, pyridine, as opposed to
PhCN, can serve as a base to help deprotonate the coordi-
nated water and form the oxo complex.
Electroreduction of 1. The electrochemistry of 1 was
examined in PhCN, pyridine, and CH2Cl2. Cyclic voltam-
mograms of these three solvents are illustrated in Figure 11,
and the half-wave or peak potentials are summarized in Table
3. A proposed mechanism for the different electrochemical
transformations of 1 is shown in Scheme 3 in which Eapp
represents the applied potential in the thin-layer spectroelec-
trochemistry experiments described below. In all three
solvents the compound exhibits a reversible to quasireversible
reduction at potentials close to 0.0 V vs SCE, and the position
of this reduction strongly suggests a metal-centered process,
that is, MnV f MnIV.
Time-resolved thin-layer spectra obtained during the
reduction of the MnV-oxo complex in pyridine and PhCN
are shown in Figures 10b and 12, respectively. The data in
Inorganic Chemistry, Vol. 44, No. 13, 2005 4495

Lansky et al.
Scheme 3
All of the above data are consistent with a mechanism in
-1.30 V), the reduction of [(Cz)MnIV(µ-O)MnIII(Cz)]- to
[(Cz)MnIII(µ-O)MnIII(Cz)]2- is possibly the first step (0 to
60 s) in Figure 12b, followed by further reduction to a
π-anion-radical species (60 to 160 s) that is either monomeric
or dimeric in nature.
It is interesting to note that the electrochemically generated
µ-oxo dimer only forms during the electroreduction of 1 or
2 in PhCN and not in pyridine. This result is consistent with
which the first reduction product of 1 is a transient, highly
unstable MnIV-oxo complex. It is reasonable to expect that
the MnIV-oxo species would be highly reactive because it
is a high-spin d3 metal-oxo, as compared to the low-spin
d2 configuration of the metal-oxo unit of 1. In the case of
pyridine, this species then abstracts an electron from the
solvent to give a MnIII complex, resulting in a smooth
transition from (Cz)MnVtO to (Cz)MnIII. However, in the
case of PhCN, evidence of the formation of the dimeric
[(Cz)MnIV(µ-O)MnIII(Cz)]- is observed at Eapp ) -0.30 V
(Figure 12a). As shown in Scheme 3, this mixed-valence
dimer may form via the reduction of (Cz)MnVtO to
[(Cz)MnIV(O)]-, which can either abstract an electron from
the solvent to give (Cz)MnIII as it does in pyridine or react
with the (Cz)MnIII as it is formed in PhCN to give the mixed-
valence dimer. At a more negative applied potential (Eapp
)
Figure 13. Spectral data for the reaction mixture 1 + excess Cp2Co in
toluene/CH2Cl2 (90:10; v:v). (a) UV-vis spectrum and (b) X-band EPR
spectrum at 23 K (red line) and simulation (blue line). Instrument settings:
microwave frequency ) 9.476 GHz; microwave power ) 2.0 mW; 100
kHz modulation amplitude ) 0.10 G. Simulation parameters: S ) 1/2, g
) 2.015, g| ) 2.035, A(55MnIII) ) 390 MHz, and A(55MnIV) ) 185 MHz;
single-crystal Gaussian line widths: 25 MHz.
Figure 12. Time-resolved, thin-layer UV-vis spectra of 1 during
reductions in PhCN. (a) First reduction at -0.30 V and (b) second reduction
at -1.30 V.
4496 Inorganic Chemistry, Vol. 44, No. 13, 2005

Manganese(III) and Manganese(V)-Oxo Corrolazines
the fact that PhCN is a poor axial ligand compared to
pyridine. The latter solvent most likely binds to one or both
axial ligand positions of the intermediates formed during the
electroreduction of 1 or 2, preventing dimer formation.
Electrooxidation of 1. Complex 1 exhibits two oxidations
in both PhCN and CH2Cl2 (Figure 11). The first is reversible,
and the second is irreversible. The reversible oxidation
potentials are located at E1/2 ) 1.09 V in PhCN and 1.02 V
in CH2Cl2, whereas the irreversible peak potentials are
located at Epa ) 1.52 and 1.57 V in either PhCN or CH2Cl2.
No oxidations can be seen in pyridine because of the limited
anodic potential window in this solvent. The time-resolved
UV spectral changes obtained during the first oxidation of
1 at 1.20 V in PhCN are illustrated in Supporting Information
Figure S4. There is a decrease in the intensity of the Soret
and visible bands, and a new, weak band appears at 718 nm.
These spectral changes are consistent with a ring-centered
oxidation.
Chemical Reduction of 1 to Generate [(Cz)MnIV(µ-O)-
MnIII(Cz)]-. The chemical reduction of 1 was attempted with
the goal of generating and trapping a MnIV species. Man-
ganese(IV) (S ) 3/2) porphyrins have been characterized
previously by EPR spectroscopy.56,57 The addition of one
equivalent of Cp2Co (E1/2 ) -1.33 V)58 to 1 in a solution of
toluene/CH2Cl2 (90:10) led to an immediate color change
from green to brown, and the UV-vis spectrum of this
solution is shown in Figure 13a. This solution was allowed
to mix for 1-2 min before being transferred to an EPR tube
and rapidly frozen in liquid nitrogen. The X-band EPR spec-
trum of this solution at 23 K is shown in Figure 13b. To our
surprise, the pseudo-16-line pattern shown in Figure 13b was
obtained. This pattern is not consistent with a monomeric
MnIV porphyrin but is quite similar to the spectrum obtained
for mixed-valent MnIIIMnIV(µ-oxo) dimers that contain an S
) 3/2 MnIV ion antiferromagnetically coupled with an S )
2 MnIII ion to give Stotal ) 1/2 as the ground state.54 On the
basis of this precedent, the UV-vis spectrum in Figure 13a
is assigned to [(Cz)MnIV(µ-O)MnIII(Cz)]-, which is the same
species proposed to form during the electroreduction of 1
and 2 (vide supra).
values are nearly identical with those reported by Hureau et
al. for similar species.59 The hyperfine coupling constants
found here for both MnIII and MnIV are roughly 10-20%
smaller than those found for MnIIIMnIV bis(µ-oxo) dimers
that lack ligands with extensive (macrocyclic) π-
conjugation [e.g., 2,2′-bipyridine (∼470, 220 MHz)60 or N,N′-
dimethyl-N,N′-bis(2-pyridylmethyl)ethane-1,2-diamine (436
MHz, 234 MHz)61]. The reduction in manganese hyperfine
coupling can be qualitatively assigned to a greater covalency
in ligand-metal bonding for the corrolazine ligand and hence
larger spin delocalization.
The UV-vis spectrum in Figure 13a exhibits a split Soret
band at 435 and 482 nm and a Q-band at 685 nm. This
spectrum is similar to what is seen upon the electrochemical
reduction of 1 in PhCN (Figure 12). Thus, the initial product
of the one-electron reduction by Cp2Co is likely a MnIV-
oxo species, which is highly unstable and undergoes further
reactions, ultimately producing the MnIIIMnIV (µ-oxo) dimer
observed by X-band EPR. It is also possible that there are
other products from this reaction that are EPR-silent at
X-band.
Conclusions
The improved synthesis of manganese(III) and manga-
nese(V)-oxo corrolazine complexes has been accomplished.
Characterization of the manganese(III) complex 2 by X-ray
crystallography revealed the preference of the MnIII ion in a
corrolazine for five-coordination. This structure is different
from that of a MnIII corrole, in which four-coordination was
found.29 The preference for five-coordination in MnIII cor-
rolazines is also clearly revealed by the strong binding of
axial ligands (Cl-, Et3PO, and Ph3PO) to 2 in solution. The
electronic structure of 2 has been investigated by HFEPR,
which has demonstrated that the complex has an S ) 2
ground state, but a relatively low lying S ) 1 excited state,
indicative of the strong ligand field of corrolazines.
The manganese(V)-oxo complex, 1, has not yet been
characterized by X-ray crystallography; however, a XANES/
EXAFS study has shown that it contains a short metal-oxo
bond consistent with a triply bonded terminal oxo group,
which gives rise to a pre-edge feature in the XANES
spectrum. This feature has become a fingerprint for the
presence of high-valent metal-oxo groups.
A detailed spectroelectrochemical study of both 1 and 2
has revealed multiple reversible redox processes for both
compounds. Of particular interest is the relatively low
potential for the MnV f MnIV process in 1 (near 0.0 V vs
SCE), which shows that the MnV-oxo complex is not as
strong an oxidant as might be expected from the high Mn
oxidation state. In addition, the putative MnIV-oxo product
species should be much less stable than a MnV-oxo species
on the basis of electronic grounds (a d3 vs d2 metal-oxo),
and this prediction is borne out by the data, which indicate
that the product of the reduction near 0.0 V is extremely
A satisfactory simulation of the EPR spectrum (blue line
in Figure 13b) was calculated for a S ) 1/2 system with
hyperfine coupling to the two different 55Mn nuclei treated
via exact diagonalization of the spin Hamiltonian matrix.
Slight axial anisotropy in the g value was included (g )
2.015, g| ) 2.035). A good simulation was obtained without
including anisotropy in the hyperfine coupling to either
55MnIII or 55MnIV [A(55MnIII) ) 390 MHz; A(55MnIV) ) 185
MHz], although in reality there is likely some anisotropy
for 55MnIII.59 The ratio of the hyperfine coupling of the two
55Mn nuclei is 2.1:1, which is exactly what was found for
other MnIIIMnIV oxo-bridged dimers.54 The isotropic A(55Mn)
(56) Groves, J. T.; Stern, M. K. J. Am. Chem. Soc. 1988, 110, 8628-
8638.
(57) Camenzind, M. J.; Hollander, F. J.; Hill, C. L. Inorg. Chem. 1983,
22, 3776-3784.
(58) Connelly, N. G.; Geiger, W. E. Chem. ReV. 1996, 96, 877-910.
(59) Hureau, C.; Anxolabehere-Mallart, E.; Nierlich, M.; Gonnet, F.;
Riviere, E.; Blondin, G. Eur. J. Inorg. Chem. 2002, 2710-2719.
(60) Cooper, S. R.; Dismukes, G. C.; Klein, M. P.; Calvin, M. J. Am. Chem.
Soc. 1978, 100, 7248-7252.
(61) Hureau, C.; Blondin, G.; Cesario, M.; Un, S. J. Am. Chem. Soc. 2003,
125, 11637-11645.
Inorganic Chemistry, Vol. 44, No. 13, 2005 4497

Lansky et al.
unstable, undergoing a conversion to another species on the
electrochemical time scale. Thus, isolation or characterization
of the elusive MnIV-oxo species needs a special approach,
perhaps through low-temperature methods. An unexpected
finding was the easy conversion of the MnIII complex to the
MnV-oxo species in pyridine by electrooxidation. This result
points to possible new synthetic/electrochemical methods for
the bulk preparation of MnV-oxo and other high-valent
metal-oxo corrolazine compounds.
obtained at the National Synchrotron Light Source beam line
X9B, which is supported by the NIH-RR program; NSLS is
operated by the US Department of Energy. HFEPR experi-
ments were performed at the National High Magnetic Field
Laboratory, which is funded by the NSF through the
Cooperative Agreement No. DMR-0084173 and the State
of Florida. The 25-T resistive magnet was funded by the W.
M. Keck Foundation.
Supporting Information Available: X-ray crystallographic file
for compound 2‚(CH3OH) (CIF); UV-vis spectra of the addition
of methanol to 2 (Figure S1), time-resolved thin-layer UV-vis
spectra for 2 (Figure S2) and 1 (Figure S4), and thin-layer cyclic
voltammograms for 1 (Figure S3). This material is available free
of charge via the Internet at http://pubs.acs.org.
Acknowledgment. The support of the NSF (CHE0094095
to D.P.G.), NIH (GM-45205 to J.E.P.H.), and the Robert A.
Welch Foundation (Grant E-680, K.M.K.) is gratefully
acknowledged. M.C. is grateful for financial support from
The Foundation Blanceflor Boncompangi-Ludovisi, ne´e Bildt
and the Stiftelsen La¨ngmanska Kulturfonden. XAS data were
IC0503636
4498 Inorganic Chemistry, Vol. 44, No. 13, 2005