Generation of a high-valent iron imido corrolazine complex and NR group transfer reactivity

Article abstract of DOI:10.1021/ic400280x

The generation of a new high-valent iron terminal imido complex prepared with a corrolazine macrocycle is reported. The reaction of [Fe ^III(TBP₈Cz)] (TBP₈Cz = octakis(4-tert- butylphenyl)corrolazinato) with the commercially available chloramine-T (Na ⁺TsNCl^-) leads to oxidative N-tosyl transfer to afford [Fe^IV(TBP₈Cz^+?)(NTs)] in dichloromethane/acetonitrile at room temperature. This complex was characterized by UV-vis, Moessbauer (δ = -0.05 mm s^-1, ΔE _Q = 2.94 mm s^-1), and EPR (X-band (15 K), g = 2.10, 2.00) spectroscopies, and together with reactivity patterns and DFT calculations has been established as an iron(IV) species antiferromagnetically coupled with a Cz-π-cation-radical (S_total = ¹/₂ ground state). Reactivity studies with triphenylphosphine as substrate show that [Fe^IV(TBP₈Cz^+?)(NTs)] is an efficient NTs transfer agent, affording the phospharane product Ph₃P=NTs under both stoichiometric and catalytic conditions. Kinetic analysis of this reaction supports a bimolecular NTs transfer mechanism with rate constant of 70(15) M^-1 s^-1. These data indicate that [Fe ^IV(TBP₈Cz^+?)(NTs)] reacts about 100 times faster than analogous Mn terminal arylimido corrole analogues. It was found that two products crystallize from the same reaction mixture of Fe ^III(TBP₈Cz) + chloramine-T + PPh₃, [Fe ^IV(TBP₈Cz)(NPPh₃)] and [Fe ^III(TBP₈Cz)(OPPh₃)], which were definitively characterized by X-ray crystallography. The sequential production of Ph ₃P=NTs, Ph₃P=NH, and Ph₃P=O was observed by ³¹P NMR spectroscopy and led to a proposed mechanism that accounts for all of the observed products. The latter Fe^III complex was then rationally synthesized and structurally characterized from Fe ^III(TBP₈Cz) and OPPh₃, providing an important benchmark compound for spectroscopic studies. A combination of Moessbauer and EPR spectroscopies led to the characterization of both intermediate spin (S = ³/₂) and low spin (S = ¹/₂) Fe^III corrolazines, as well as a formally Fe^IV corrolazine which may also be described by its valence tautomer Fe^III(Cz ^+?).

Full text of DOI:10.1021/ic400280x

Article
pubs.acs.org/IC
Generation of a High-Valent Iron Imido Corrolazine Complex and NR
Group Transfer Reactivity
Pannee Leeladee,† Guy N. L. Jameson,*,‡ Maxime A. Siegler,† Devesh Kumar,⊥ Sam P. de Visser,*,∥
and David P. Goldberg*,†
†Department of Chemistry, The Johns Hopkins University, 3400 North Charles Street, Baltimore, Maryland 21218, United States
‡Department of Chemistry and MacDiarmid Institute for Advanced Materials and Nanotechnology, University of Otago, P.O. Box 56,
Dunedin 9054, New Zealand
⊥Department of Applied Physics, Babasaheb Bhimrao Ambedkar University, Vidya Vihar, Rai Bareilly Road, Lucknow 226 025, India
∥Manchester Institute of Biotechnology and School of Chemical Engineering and Analytical Science, The University of Manchester,
131 Princess Street, Manchester M1 7DN, United Kingdom
S
* Supporting Information
ABSTRACT: The generation of a new high-valent iron
terminal imido complex prepared with a corrolazine macro-
cycle is reported. The reaction of [FeIII(TBP8Cz)] (TBP8Cz =
octakis(4-tert-butylphenyl)corrolazinato) with the commer-
cially available chloramine-T (Na+TsNCl−) leads to oxidative
N-tosyl transfer to afford [FeIV(TBP8Cz+•)(NTs)] in dichloro-
methane/acetonitrile at room temperature. This complex was
characterized by UV−vis, Mossbauer (δ = −0.05 mm s−1, ΔE
̈
Q
= 2.94 mm s−1), and EPR (X-band (15 K), g = 2.10, 2.00)
spectroscopies, and together with reactivity patterns and DFT
calculations has been established as an iron(IV) species antiferromagnetically coupled with a Cz-π-cation-radical (Stotal = /2
1
ground state). Reactivity studies with triphenylphosphine as substrate show that [FeIV(TBP8Cz+•)(NTs)] is an efficient NTs
transfer agent, affording the phospharane product Ph3PNTs under both stoichiometric and catalytic conditions. Kinetic
analysis of this reaction supports a bimolecular NTs transfer mechanism with rate constant of 70(15) M−1 s−1. These data
indicate that [FeIV(TBP8Cz+•)(NTs)] reacts about 100 times faster than analogous Mn terminal arylimido corrole analogues. It
was found that two products crystallize from the same reaction mixture of FeIII(TBP8Cz) + chloramine-T + PPh3,
[FeIV(TBP8Cz)(NPPh3)] and [FeIII(TBP8Cz)(OPPh3)], which were definitively characterized by X-ray crystallography. The
sequential production of Ph3PNTs, Ph3PNH, and Ph3PO was observed by 31P NMR spectroscopy and led to a proposed
mechanism that accounts for all of the observed products. The latter FeIII complex was then rationally synthesized and
structurally characterized from FeIII(TBP8Cz) and OPPh3, providing an important benchmark compound for spectroscopic
studies. A combination of Mossbauer and EPR spectroscopies led to the characterization of both intermediate spin (S = 3/ ) and
̈
2
low spin (S = 1/2) FeIII corrolazines, as well as a formally FeIV corrolazine which may also be described by its valence tautomer
FeIII(Cz+•).
metal-nitrido and metal-imido complexes, including iron-based
systems, for similar potential applications in atom or group
transfer processes.12−17
INTRODUCTION
■
Iron−nitrogen multiply bonded complexes have been impli-
cated in biological systems and industrial processes. For
instance, in the Haber−Bosch process, iron-bound surface
nitrides have been postulated as the reactive intermediates
responsible for ammonia production.1,2 The iron-containing
metalloenzyme nitrogenase likely relies upon one or more
multiply bonded Fe−N intermediates during the dinitrogen
reduction process.3−5 An FeNR porphyrin species was
suggested as the reactive intermediate for reactions involving
NR insertion into C−H bonds by the heme enzyme
Cytochrome P450-LM 3,4.6 In analogy to this chemistry,
much effort has gone into accessing transient Fe−N(R)
multiply bonded species as potent catalysts for C−H activation
and functionalization.7−11 There is great interest in preparing
The synthesis of mid- to late-transition metal-imido
complexes has been quite challenging. The stability of these
compounds depends on their geometry, the oxidation state of
the metal, and the nature of the ancillary ligands. In particular,
5- or 6-coordinate Fe-imido complexes are usually considered
highly unstable, making it difficult to generate and/or
characterize such species. It can be seen from qualitative d-
orbital splitting diagrams that these high coordination number
iron complexes would normally have filled, or partially filled,
Received: February 2, 2013
Published: March 25, 2013
© 2013 American Chemical Society
4668
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
Chart 1. Representative Examples of Multiply-Bonded, Terminal-Imido/Nitrene Fe and Mn Complexes (Ad = adamantyl; Ts =
tosyl; Mes = mesityl)
dπ-pπ antibonding orbitals with respect to the Fe−N
interaction. The Fe ion needs to be in an unusually high
oxidation state, e.g., Fe(6+), to avoid this destabilizing
influence,18 and iron compounds in relatively low oxidation
states, e.g. Fe(2+), Fe(3+), are destabilized due to the
population of the antibonding dπ(Fe)−pπ(N) orbitals.9,10,17,19
To our knowledge, there is only one example of an Fe(NR)
complex with a coordination number ≥5 that has been
crystallographically characterized, and this complex is an iron
porphyrin derivative.20 Interestingly, this complex was for-
mulated as a high-spin FeII-nitrene compound, as indicated by
and X-ray crystallographic characterization of a high-valent
MnV-imido complex, [MnV(TBP8Cz)(NMes)] (Mes = 2,4,6-
trimethylphenyl; TBP8Cz = octakis(4-tert-butylphenyl)-
corrolazinato) (Chart 1).32 The related corrole (Cor) complex,
[MnV(tpfc)(NMes)] (tpfc = meso-tris(pentafluorophenyl)-
corrole), was also isolated and fully characterized, including
an X-ray structure determination.33 Isoelectronic MnV terminal
oxo complexes have also been generated with Cz and Cor
supporting ligands, and the first example of this class,
MnV(O)(TBP8Cz), was isolable and stable as a solid at room
temperature.34−38 In these cases the paradigm does not involve
enforcing low-coordinate environments at the metal center;
rather the Cz and Cor ligands are designed to provide strong,
in-plane σ donation that helps stabilize metal ions in high-
valent states, depopulating the key, dπ−pπ antibonding orbitals
involved in terminal imido/oxo ligation.
Mossbauer spectroscopy. In addition, a 6-coordinate, nonheme
̈
FeIV-imido complex supported by a tetrapyridylamine ligand
was described, but this complex could not be characterized by
X-ray crystallography (Chart 1).21
An alternative, elegant strategy to prepare stable Fe terminal
imido complexes is to construct them from compounds with
low coordination number (Chart 1). Peters and co-workers
have shown that triply bonded FeII-, FeIII-, and FeIV-imido
complexes can be stabilized in pseudotetrahedral coordination
environments by sterically encumbered tris(phosphino)borane,
tris(phosphino)borate, and bis(phosphine)pyrazolyl li-
gands.22−26 Similarly, Smith and co-workers introduced strong
field, bulky tris(carbene)borate ligands to stabilize FeIII- and
FeIV-imido complexes with 3-fold symmetry.27 Iron-imide
complexes with redox-active bis(imino)pyridine (BIP) ligands
have been constructed from distorted square planar geo-
metries,28,29 although the assignment of the iron oxidation
states in these compounds is not straightforward.28 Highly
unsaturated, three-coordinate iron(III) terminal imido com-
plexes have also been prepared with either bulky bidentate, β-
diketiminate ligands,30,31 or bidentate, dipyrromethene-based
systems.8 In all of these cases it seems that the steric
encumbrance of the surrounding ligands is essential to
engender the low-coordinate environments that stabilize iron-
imido species.
The synthesis of these rare, manganese-imido/oxo complexes
suggested to us that the Cz scaffold might support the
preparation of analogous high-valent, iron-imido complexes.
Indeed, we had already described the generation, low-
temperature stabilization, and C−H activation chemistry of a
high-valent, iron-oxo corrolazine π-cation-radical (FeIV(O)-
(TBP8Cz+•)), equivalent to the compound-I intermediate in
heme enzymes.39,40 It has been shown that FeIII corroles can
catalyze aziridinations using PhINTs as the nitrogen source,
and a high-valent Fe-imido species was postulated as the
reactive intermediate.41 In the same study, an FeIV corrole was
also found to be a potent catalyst for aziridinations when
PhINTs was replaced with chloramine-T (Na+TsNCl−) as the
nitrogen source. However, no direct spectroscopic evidence for
the existence of high-valent Fe-imido corroles in this study, or
any other, has thus far been obtained.
Herein, we report the generation of a high-valent iron-imido
corrolazine, (TBP8Cz+•)FeIV(NTs), and its characterization
through a combination of UV−vis, EPR, and Mossbauer
̈
spectroscopies. Density functional theory (DFT) calculations
were performed to support and analyze the electronic structure
Our group has focused on the stabilization and character-
ization of high-valent manganese and iron complexes with
multiply bonded terminal ligands encapsulated in the heme-like
corrolazine (Cz) platform. The strongly donating and
oxidatively resistant Cz macrocycle has allowed for the isolation
assignment, and simulate the Mossbauer parameters for the
̈
high-valent iron-imido complex. This complex is a rare example
of a metastable Fe-imido porphyrinoid compound, and the first
example of such a species housed in a corrole-type macrocycle.
4669
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
described here were the result of a full optimization without geometric
constraints, and the local minima were ascertained with an analytical
frequency calculation that established real frequencies only. These
optimizations and frequencies were done with an LACVP basis set on
Fe that includes a core potential and a 6-31G basis set on the rest of
the atoms: basis set B1.52 This basis set was shown to give adequate
structures for the investigation of a potential energy profile of a
chemical reaction by an iron(IV)-oxo species.53 To improve the
energetics we did single point calculations using a triple-ζ quality
LACV3P+ basis set on Fe (with core potential) and 6-311+G* on the
rest of the atoms: basis set B2. Energies reported in here are taken
from the UB3LYP/B2 results and are corrected for zero-point energies
at UB3LYP/B1. Free energies use UB3LYP/B2 energies and
UB3LYP/B1 zero-point energies, thermal and entropic corrections.
We also did single point calculations in Jaguar using the Polarized
Continuum Model using a dielectric constant of ε = 5.708 and probe
radius of 2.722 Å.
It is shown that this compound is capable of NR group transfer
to PPh3 to give the TsNPPh3 product with concomitant two-
electron reduction of the iron-imido complex. Kinetic studies
show that the reactivity of the FeNTs complex is
dramatically enhanced over MnNAr corroles. However,
(TBP8Cz+•)FeIV(NTs) is, in general, significantly more stable
(and less reactive) than expected for a multiply bonded, iron-
imido species. Crystal structures were determined for the PPh3-
derived products [FeI V (TBP8 Cz)(NPPh3 )] and
[FeIII(TBP8Cz)(OPPh3)], and these structures have aided in
our understanding of the mechanism of NR transfer and
subsequent product transformations. Combined EPR and
Mossbauer studies on the series of iron complexes described
̈
here provide insights into the oxidation state assignments and
ligand noninnocence for such compounds.
Spectroscopic parameters (Mossbauer and EPR parameters) were
̈
obtained from single point calculations on the optimized geometries
using the ORCA program package.54 These calculations used the
B3LYP method in combination with the CP(PPP) basis set on Fe
coupled to a TZVP basis set on the rest of the atoms.55 The electric
field gradients Vi (i = x, y, or z) were used to calculate the quadrupole
splitting (ΔEQ) using eqs 2 and 3 with e the elementary charge of a
proton/electron, Q the nuclear quadrupole moment, Q(57Fe) = 0.16
barn, and η the asymmetry parameter of the nuclear quadrupole
tensor.56
EXPERIMENTAL DETAILS
■
All reactions were performed using dry solvents and standard Schlenk
or drybox techniques at room temperature unless otherwise noted.
The complex [FeIII(TBP8Cz)] (1)42 and metal-free corrolazine
43
TBP8CzH3 were synthesized as reported previously. The 57FeCl2
was prepared according to a published procedure.44 Dichloromethane
and acetonitrile were purified via a Pure-Solv solvent purification
system from Innovative Technologies, Inc. Deuterated solvents for
NMR samples were purchased from Cambridge Isotope Laboratories,
Inc. Chloramine-T trihydrate (Na+TsNCl−·3H2O) was purchased
from Sigma-Aldrich and used as received. Triphenylphosphine (PPh3)
was received from Sigma-Aldrich and recrystallized from diethyl ether.
Triphenylphosphine oxide (Ph3PO) was purchased from Sigma-
Aldrich and recrystallized from ethanol. Authentic samples of (N-
tosylimino)triphenylphosphorane (Ph3PNTs) and Ph3PNH were
synthesized according to published methods.45,46
1
2
1
3
ΔEQ
=
eQ·V · 1 + η2
z
(2)
(3)
η = (V − Vy)/V
x
z
The isomer shift δ was calculated from the spin density at the iron
nucleus ρ0(Fe) using previously determined fit-parameters in ORCA.57
The magnetic hyperfine parameters, Ai (i = x, y, or z), were obtained
using the scalar relativistic zero-order regular approximation (ZORA)
at the B3LYP level of theory. These methods were shown to reproduce
Instrumentation. UV−vis spectroscopy was performed on a
Varian Cary 50 Bio spectrophotometer. Kinetic experiments were run
on a Hewlett-Packard 8542 diode-array spectrophotometer equipped
with HPChemstation software. Electron paramagnetic resonance
(EPR) spectra were obtained on a Bruker EMX EPR spectrometer
controlled with a Bruker ER 041 X G microwave bridge at 15 K, and
equipped with a continuous-flow liquid helium cryostat (ESR900)
coupled to a TC503 temperature controller made by Oxford
well experimentally determined Mossbauer and EPR parameters of
̈
transition metal complexes.
Generation of [FeIV(TBP8Cz+•)(NTs)] (2) in Situ. To a solution of
[FeIII(TBP8Cz)] (16 μM) in CH2Cl2 was added Na+TsNCl−·3H2O (1
equiv dissolved in CH3CN). A color change of the solution from green
to brown was noted. The reaction was monitored by UV−vis
spectroscopy, revealing isosbestic conversion from the spectrum for 1
1
Instruments, Inc. H NMR and 31P NMR spectra were recorded on
a Bruker Avance 400 MHz FT-NMR spectrometer. Chemical shifts
were referenced to the residual solvent peaks for 1H NMR and to the
H3PO4 peak for 31P NMR. Electrospray ionization mass spectra (ESI-
MS) were collected on a Thermo Finnigan LCQ Duo ion-trap mass
spectrometer fitted with an electrospray ionization source in positive
ion mode. Samples were infused into the instrument at a rate of 25
μL/min using a syringe pump via a silica capillary line. The spray
voltage was set at 5 kV, and the capillary temperature was held at 250
°C. 57Fe Mossbauer spectra were recorded on a Mossbauer
to that of the new brown species [FeIV(TBP8Cz+•)(NTs)] (2) (λmax
=
395, ε = 4.3 × 104 M−1 cm−1). The spectrum for 2 is stable for at least
2 h at room temperature. An EPR sample was prepared as follows: to a
solution of [FeIII(TBP8Cz)] (2.0 mM) in toluene (300 μL) was added
chloramine-T (1 equiv dissolved in CH3CN) to give a final
concentration of 1.5 mM of the iron complex. A color change from
green to brown was observed. The reaction mixture was then
transferred to a Wilmad quartz EPR tube (3 mm i.d.), and the sample
was frozen and stored at 77 K.
̈
̈
spectrometer from SEE Co. (Science Engineering & Education Co.,
MN) equipped with a closed cycle refrigerator system from Janis
Research Co. and SHI (Sumitomo Heavy Industries Ltd.). Data were
collected in constant acceleration mode in transmission geometry. The
Precipitation of Solid [FeIV(TBP8Cz+•)(NTs)] (2). To an amount
of [FeIII(TBP8Cz)] (20.6 mg, 14.6 μmol) in CH2Cl2 (1.00 mL) was
added Na+TsNCl−·3H2O (6.2 mg, 22 μmol, 1.5 equiv) dissolved in
CH3CN (0.40 mL). The reaction mixture was stirred for 30 min, and
the color change from green to brown was observed. An amount of
CH3CN was added until a solid precipitate had formed in the reaction
mixture. The resulting solid was filtered, isolated on filter paper, and
washed with CH3CN to afford a dark brown-green solid (8.0 mg).
Kinetics of NR Group Transfer to Triphenyl Phosphine. In a
custom-made 100-mL round-bottom flask adapted with a 1-cm path-
length side arm UV−vis cell was prepared a solution of
[FeIV(TBP8Cz+•)(NTs)] (2) by addition of Na+TsNCl−·3H2O (100
μL of an 0.8 mM stock solution in CH3CN, 1 equiv) to a solution of 1
(16 μM) dissolved in CH2Cl2 (5 mL). The cell was placed in the
spectrophotometer, and a 400 nm long-pass filter was placed between
the spectrophotometer light source and the sample solution in order to
zero velocity of the Mossbauer spectra refers to the centroid of the
̈
room temperature spectrum of a 25 μm metallic iron foil. Analysis of
the spectra was conducted using the WMOSS program (SEE Co,
formerly WEB Research Co. Edina, MN). Magnetic spectra were fitted
using the spin Hamiltonian:
⎡
⎤
2
z
1
3
2
x
2
̂
̂
̂
y
̂
̂
H = D S − S(S + 1) + E(S − S ) + βS·g·B
⎢
⎣
⎥
⎦
(1)
Computational Calculations and Spectroscopic Parameter
Determination. All calculations were done using previously
described methods47 with density functional theory at the unrestricted
hybrid density functional level UB3LYP48,49 as implemented in the
Jaguar 7.9 and Gaussian-03 program packages.50,51 All geometries
4670
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
avoid any potential photodegradation of iron corrolazine. After
complete conversion of 1 to 2 was noted by UV−vis, excess PPh3
(30−75 equiv dissolved in CH2Cl2) was added to start the reaction,
and an immediate color change from brown to green was noted,
indicating the recovery of the starting material 1. The time-course of
the reaction was monitored by UV−vis and showed a smooth
conversion of 2 to 1 over a 100−200 s period depending on the
concentration of PPh3. The pseudo-first-order rate constants, kobs’s, for
this reaction were obtained by nonlinear least-squares fitting of the
plots of absorbance (Abs) at 440 nm versus time (t) to the following
equation:
aryl group is disordered over two orientations, and the occupancy
factor for the major component refines to 0.876(5) for 3, 0.870(5) for
4a, and 0.895(4) for 4b. In the structure of 3, disorder was modeled
for the four crystallographically independent dichloromethane
molecules, and the occupancy factors for the major components are
0.831(4), 0.880(2), 0.695(5), and 0.529(8). In the structures of 4a and
4b, free variables for the major occupancy factors of one disordered
CH2Cl2 and one ordered CH3CN solvent molecules were used in the
final refinement. Their refined values are 0.753(4), 0.131(3), and
0.902(8) for 4a, and 0.792(4), 0.113(3), and 0.923(8) for 4b. Other
solvent molecules were found in the crystal lattice but were
significantly more disordered, and therefore, their contributions were
taken out by using the program SQUEEZE (Spek, 2003) for the final
refinement.
Abst = Absf − (Absf − Abs0)e−k
obst
(4)
Data for 3 follow: fw = 2068.75, black block, 0.34 × 0.31 × 0.26
mm3, monoclinic, P21/c (No. 14), a = 17.2124(2) Å, b = 33.4786(5)
Å, c = 19.9855(2) Å, β = 102.6307(12)°, V = 11237.9(2) Å3, Z = 4, Dx
= 1.223 g cm−3, μ = 0.390 mm−1, abs corr range 0.894−0.923. There
Here, Absf = final absorbance and Abs0 = initial absorbance. The
reaction was repeated over a range of substrate concentrations, and a
plot of kobs versus substrate concentration was found to be linear. The
second-order rate constant (k) was obtained from the slope of the
best-fit line to this plot.
were 65 453 reflections measured up to a resolution of (sin θ/λ)max
=
0.59 Å−1. There were 19 786 unique reflections (Rint = 0.0377), of
which 13 559 were observed [I > 2σ(I)]. There were 1354 parameters
refined using 180 restraints. R1/wR2 [I > 2σ(I)]: 0.0666/0.1901. R1/
wR2 [all reflns]: 0.0961/0.2026. S = 1.069. Residual electron density
found between −0.93 and 0.78 e Å−3.
Monitoring of NR Group Transfer by 31P NMR Spectroscopy.
To a solution of 1 (12−15 mM) in CD2Cl2 was added
Na+TsNCl−·3H2O (1−3 equiv dissolved in CD3CN). A color change
from green to brown was noted, consistent with formation of 2. An
aliquot of the reaction mixture was checked by UV−vis spectroscopy
to ensure the complete formation of 2. Excess PPh3 (5−10 equiv
dissolved in CD2Cl2) was added, causing an immediate color change
from brown to green. An aliquot of the reaction mixture was removed
and analyzed by UV−vis spectroscopy, revealing the return of 1. The
solution was then transferred to an NMR tube, and the reaction was
monitored by 31P NMR over 14 d.
Data for 4a follow (asterisk indicates exclusion of the contribution
of the unresolved residual electron density): *fw = 1802.09, *black
block, 0.62 × 0.51 × 0.30 mm3, monoclinic, P21/c (No. 14), a =
17.2614(8) Å, b = 33.4143(11) Å, c = 19.9714(8) Å, β = 102.610(4)°,
V = 11241.2(8) Å3, Z = 4, Dx = 1.065 g cm−3, *μ = 0.239 mm−1, *abs
corr range 0.899−0.941. *There were 59 539 reflections measured up
to a resolution of (sin θ/λ)max = 0.59 Å−1. There were 19 795 unique
reflections (Rint = 0.0449), of which 15 006 were observed [I > 2σ(I)].
There were 1221 parameters refined using 63 restraints. R1/wR2 [I >
2σ(I)]: 0.0605/0.1733. R1/wR2 [all reflns]: 0.0759/0.1822. S = 1.069.
Residual electron density found between −0.47 and 0.66 e Å−3.
Data for 4b follow: *fw = 1804.71, *black block, 0.35 × 0.24 × 0.20
mm3, monoclinic, P21/c (No. 14), a = 17.2502(2) Å, b = 33.4217(5)
Å, c = 19.9821(2) Å, β = 102.7119(12)°, V = 11237.9(2) Å3, Z = 4, Dx
= 1.067 g cm−3, *μ = 0.240 mm−1, *abs corr range 0.900−0.937.
There were *64 199 reflections measured up to a resolution of (sin θ/
λ)max = 0.59 Å−1. There were 19 786 unique reflections (Rint = 0.0375),
of which 14 933 were observed [I > 2σ(I)]. There were 1221
parameters refined using 63 restraints. R1/wR2 [I > 2σ(I)]: 0.0611/
0.1662. R1/wR2 [all reflns]: 0.0791/0.1750. S = 1.073. Residual
electron density found between −0.40 and 0.66 e Å−3.
Synthesis of [Fe(TBP8Cz)(NPPh3)] (3). A reaction of 2 with PPh3
substrate was performed as described above for 31P NMR experiments,
and then the reaction mixture was allowed to stand in an NMR tube
under ambient conditions. Good quality single crystals appeared after
2 weeks. A crystal was selected for X-ray structure determination and
revealed the structure for [Fe(TBP8Cz)(NPPh3)] (3). ESI-MS, EPR,
and Mossbauer data indicated that the bulk material of these crystals
̈
was contaminated with a small amount of [Fe(TBP8Cz)(OPPh3)], and
a separate X-ray structure determination of a second crystal confirmed
the presence of this complex, whose structure was solved as
[Fe(TBP8Cz)(OPPh3)] (4a).
Synthesis of [Fe(TBP8Cz)(OPPh3)] (4b). To a solution of 1 (9.8
mg, 6.9 μmol) in CH2Cl2 (400 μL) was added Ph3PO (5 equiv in 100
μL of CH2Cl2). An amount of CH3CN (100 μL) was added, and good
quality single crystals (black blocks) (8 mg, 68%) were obtained by
slow evaporation of this mixture for one week. UV−vis (CH2Cl2): λmax
= 439, 607, 732 nm. ESI-MS (m/z): isotopic cluster centered at 1689.7
(M+). 1H NMR (CD2Cl2): δ (ppm) 13.05 (br), 11.26 (br), 9.95 (br),
7.47 (br), 2.00, 1.82, 1.66, 1.57, 1.26, 0.88. 31P NMR (CD2Cl2): δ
(ppm) 27.4. Anal. Calcd for C114H119FeN7OP: C, 81.02: H, 7.10; N,
5.80. Found: C, 81.18; H, 7.03; N, 5.78.
X-ray Crystallography. All reflection intensities were measured at
110(2) K using a KM4/Xcalibur (detector: Sapphire3) with enhanced
graphite-monochromated Mo Kα radiation (λ = 0.710 73 Å) under the
program CrysAlisPro (Version 1.171.33.55 or 1.171.35.11, Agilent
Technologies, 2010/2011). The temperature of the data collection was
controlled using the system Cryojet (manufactured by Oxford
Instruments). The program CrysAlisPro (Version 1.171.33.55 or
1.171.35.11, Agilent Technologies, 2010/2011) was used to refine the
cell dimensions and data reduction. The structure was solved with the
program SHELXS-97 and was refined on F2 with SHELXL-97.
Analytical numeric absorption corrections based on a multifaceted
crystal model were applied using CrysAlisPro (Version 1.171.33.55 or
1.171.35.11, Agilent Technologies, 2010/2011). The H-atoms were
placed at calculated positions using the instructions AFIX 23, AFIX 43,
or AFIX 137 with isotropic displacement parameters having values 1.2
or 1.5 times Ueq of the attached C atoms.
Synthesis of [57FeIII(TBP8Cz)]. The metal-free corrolazine
TBP8CzH3 (62 mg, 0.046 mmol) was combined with 57FeCl2 (29
mg, 0.23 mmol) in dimethylformamide (10 mL) and heated at reflux
for 2 h. After removing the solvent in vacuo, the crude product was
purified by flash column chromatography (CH2Cl2/MeOH 99:1 v/v;
silica gel) to obtain [57FeIII(TBP8Cz)] as a dark-green solid (51 mg,
79%). UV−vis (CH2Cl2): λmax = 441, 610, 746 nm.
Samples for Mossbauer Studies. The 57Fe-labeled complexes
̈
[57FeIII(TBP8Cz)] and [57FeIII(TBP8Cz)(NPPh3)] were used for
Mossbauer measurements. An isotopically labeled sample of 3
̈
([57FeIII(TBP8Cz)(NPPh3)]) was synthesized from [57FeIII(TBP8Cz)]
starting material. In the case of [FeIII(TBP8Cz)(OPPh3)] (4b),
Mossbauer data were collected on samples that were not enriched in
̈
57Fe. Samples were either measured as solid-state samples (∼10 mg)
or as frozen solutions (∼400 μL) in toluene or pyridine in custom-
made Teflon sample holders.
RESULTS AND DISCUSSION
■
Synthesis and Characterization of a High-Valent Iron
Imido Corrolazine Complex. A variety of discrete metal-
imido complexes have been synthesized by the reaction of the
appropriate metal precursors and alkyl- or aryl-
azides.9,32,33,58−61 This strategy was used to synthesize the
MnV-imido complex [MnV(TBP8Cz)NMes], in which the MnIII
The asymmetric unit of 3, 4a, and 4b contains one Fe complex, and
some lattice dichloromethane and acetonitrile solvent molecules. All
structures are partly disordered. The t-butyl group C73→C76 of one
4671
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
precursor [MnIII(TBP8Cz)] was reacted with mesityl azide
(2,4,6-(CH3)3C6H2N3) at elevated temperature.32 Corrole
analogues [Mn(tpfc)NAr] (Ar = 2,4,6-(CH3)3C6H2, 2,4,6-
Cl3C6H2) have been similarly prepared.33 Our initial efforts to
gain access to a high-valent Fe-imido corrolazine followed the
same methodology. Attempts were made to react the previously
synthesized FeIII complex [FeIII(TBP8Cz)] with both alkyl- and
aryl-azides. For example, [FeIII(TBP8Cz)] was stirred with 50
equiv of mesityl azide in toluene for 24 h, but even at elevated
temperature (90 °C) no evidence of reaction in the form of a
color change, UV−vis spectroscopic conversion, or TLC
change was noted. The azides AdN3 (Ad = adamantyl),
PhN3, and Me3SiN3 were examined under similar conditions,
but none led to any noticeable reaction with the FeIII
corrolazine. We speculated that the coordination of the azide
reagents, or the triplet nitrene that potentially forms upon
thermal activation of the azides,61 was not occurring because
even small changes in the UV−vis spectrum of [FeIII(TBP8Cz)]
were not observed. The failure of the azide reagents to react
with [FeIII(TBP8Cz)] prompted us to examine alternative NR
group transfer reagents.
Figure 2. UV−vis spectral change upon addition of pentafluoroiodo-
sylbenzene (PFIB) to [FeIII(TBP8Cz)] at −78 °C in CH2Cl2/CH3OH.
two electrons, as shown in Scheme 1. The stoichiometry of the
reaction was determined by titration of chloramine-T followed
by monitoring by UV−vis spectroscopy, as shown in Figures 1
and S1. Addition of chloramine-T in 0.1 equiv increments
results in isosbestic conversion of the FeIII complex up to a total
of 1 equiv, at which point a loss of isosbestic behavior is
observed together with a decrease at 395 nm corresponding to
the oxidized product. This spectral titration confirms that
chloramine-T and [FeIII(TBP8Cz)] (1) react in a ratio of 1:1 to
form [FeIV(TBP8Cz+•)(NTs)] (2). The imido complex 2 is
thus remarkably stable in solution at room temperature, and
this stability is likely a consequence of the previously
established ability of the corrolazine platform to stabilize
high-valent, multiply bonded terminal oxo and terminal imido
complexes.
The inexpensive, commercially available reagent chloramine-
T (Na+TsNCl−·3H2O) was shown to be a good NTs-transfer
agent in catalytic aziridinations involving Fe corrole
[(FeIV(tpfc)],41 and both chloramine-T and bromamine-T
have been used as a nitrene source in metal-catalyzed C−H
insertions.41,62−65 Addition of a stock solution of chloramine-T
in CH3CN to [FeIII(TBP8Cz)] dissolved in CH2Cl2 at room
temperature led to the conversion of green [FeIII(TBP8Cz)]
(λmax = 440, 611, 747 nm) to a new, brown species with λmax
=
395 nm (Figure 1). Isosbestic points are observed at 392, 490,
Several attempts were made to grow single crystals of the
product from chloramine-T for X-ray structure determination,
but thus far these efforts have remained unsuccessful. However,
the complex can be isolated as a solid by solvent precipitation.
Relatively concentrated solutions of [FeIV(TBP8Cz+•)(NTs)]
(10 mM) can be generated in CH2Cl2 from [FeIII(TBP8Cz)]
and chloramine-T, and the resulting FeNTs product is then
precipitated from the reaction mixture by the addition of excess
CH3CN. Formation of a brown-green solid and a brown
solution results, and the solid is then isolated by filtration. This
solid exhibited similar UV−vis and other spectroscopic
properties as seen for a sample of [FeIV(TBP8Cz+•)(NTs)]
freshly generated in CH2Cl2, indicating that the precipitated
material contains the [FeIV(TBP8Cz+•)(NTs)] complex pre-
pared in situ. However, Mossbauer spectroscopic measure-
̈
Figure 1. UV−vis spectral change upon addition of chloramine-T
trihydrate (0−1.0 equiv) to [FeIII(TBP8Cz)] at 23 °C in CH2Cl2/
CH3CN.
ments show that this solid cannot be isolated in pure form, and
contains up to ∼25% of the starting FeIII complex (vide infra).
Autoreduction of the iron-imido complex likely occurs upon
precipitation, which is similar to what we have observed
previously in attempts to isolate high-valent Mn-oxo
corrolazines in the solid state.
NR Group Transfer. The brown complex 2, generated from
the reaction of chloramine-T and [FeIII(TBP8Cz)], was next
examined for its ability to transfer the NTs group to phosphine
substrates. The analogous reaction with corrole derivatives (e.g.
[MnV(tpfc)(NAr)]) and phosphine substrates occurs smoothly,
and has yielded kinetic and mechanistic information regarding
NR group transfer.66 Reaction of [FeIV(TBP8Cz+•)(NTs)]
(freshly prepared in situ) with excess triphenylphosphine
(Scheme 2) led to the isosbestic return of the spectrum for
[FeIII(TBP8Cz)] as shown in Figure 3. Production of the two-
538, and 792 nm. The new species with λmax = 395 nm appears
surprisingly stable for more than 2 h at room temperature as
monitored by UV−vis. The changes in the UV−vis following
addition of chloramine-T are similar to those observed for the
reaction of [FeIII(TBP8Cz)] and pentafluoroiodosylbenzene to
give the high-valent iron-oxo complex [FeIV(TBP8Cz+•)(O)],
although in this case the iron-oxo product is thermally unstable
and had to be observed at low temperature (−78 °C) (Figure
2). The UV−vis data are consistent with the formation of the
expected product, [FeIV(TBP8Cz+•)(NTs)] (2), in which the
tosylimido group replaces the terminal oxo ligand of
[FeIV(TBP8Cz+•)(O)] and the FeIII complex is oxidized by
4672
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
Scheme 1. Reaction of FeIII(TBP8Cz) with Chloramine-T
Scheme 2. Reaction of [FeIV(TBP8Cz+•)(NTs)] with PPh3
Preliminary investigation of the reaction of the imido
complex 2 with styrene was also performed to determine its
potential reactivity toward aziridination of alkene substrates. As
seen in Figure S14, a titration experiment with styrene leads to
the reduction of 2 back to the starting FeIII complex 1, but a
very large excess of substrate (48 000 equiv) is needed for this
reaction to go to completion over a 25 min period. These data
suggest that 2 is a very sluggish NR transfer agent for alkene
substrates, and this finding is consistent with the general,
inherent stability of 2 in solution at room temperature
compared to other high-valent Fe-imido complexes. Given
the low reactivity of 2 toward styrene, further studies regarding
aziridination of styrene or other substrates were not pursued at
this time.
Turnover. The UV−vis and NMR data suggested that the
reactions involving the generation of the FeNTs complex
and the subsequent NR transfer to PPh3 were relatively fast,
and therefore we attempted to determine if catalytic turnover
could be achieved. The NR transfer reaction was run under
catalytic conditions with a small excess of oxidant (3 equiv) and
PPh3 substrate (10 equiv). The initial formation of 2 from 1 +
chloramine-T was first confirmed by UV−vis spectroscopy, and
then the substrate was added immediately to initiate the NR
transfer reaction. As seen in Figure 4, monitoring of this
Figure 3. Time-resolved UV−vis spectra of [FeIV(TBP8Cz+•)(NTs)]
+ PPh3 (30 equiv) at 23 °C in CH2Cl2/CH3CN (100:1 v:v).
electron-reduced FeIII product is consistent with the
formulation of [FeIV(TBP8Cz+•)(NTs)], which is two oxidizing
equivalents above that of the FeIII starting material. Analysis of
the reaction mixture by 31P NMR following addition of PPh3
reveals a single peak at 14.8 ppm (Figure S11). This peak
matches that seen for the expected phospharane product,
Ph3PNTs, providing confirmation of NTs transfer to the
phosphine substrate. No signal for unreacted PPh3 was
observed by 31P NMR at the end of the NTs transfer reaction.
This observation is consistent with the lack of a 31P signal for
independent solutions of [FeIII(TBP8Cz)] + excess PPh3, which
is likely due to paramagnetic effects from the FeIII complex on
PPh3. The PPh3 also may coordinate to one or both axial
positions at the FeIII center, enhancing the paramagnetic
influence.
Figure 4. 31P NMR spectrum showing the production of Ph3PNTs
from the reaction of [FeIII(TBP8Cz)] + chloramine-T trihydrate (3
equiv) + PPh3 (10 equiv) after 15 min in CD2Cl2/CD3CN at 23 °C.
The generation of [FeIV(TBP8Cz+•)(NTs)] and subsequent
1
reaction with PPh3 was monitored by H NMR spectroscopy
(Figure S13). The initial spectrum for the FeIII complex 1
reveals paramagnetically shifted peaks between −10 and +15
ppm. Addition of a slight excess of chloramine-T causes the
clear disappearance of the paramagnetic signals for 1, revealing
a featureless paramagnetic spectrum for the iron-imido complex
2. Subsequent addition of excess PPh3 to the reaction mixture
causes an immediate return of the peaks from −10 to 15 ppm
associated with the starting complex 1. These data are nicely
consistent with the proposed mechanism of formation of the
iron-imido complex and subsequent NTs transfer to PPh3,
which also causes the concomitant return of the FeIII starting
material.
reaction by 31P NMR showed the rapid production of Ph3P
NTs with no other phospharane products. The catalytic cycle is
summarized in Scheme 3. A background reaction does occur in
the absence of 1 under these conditions; however, a significant
amount of triphenylphosphine oxide is also produced in this
case (Figure S12). The OPPh3 is presumably obtained from
oxidation of PPh3 by hypochlorite anion (OCl−), which is
known to readily form from chloramine-T and exogenous H2O.
There is no OPPh3 observed by 31P NMR in Figure 4,
indicating that the catalytic NR transfer must be highly efficient,
preventing any background reaction from occurring. We can
4673
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
underwent NR group transfer to PPh3 with a rate constant of
0.71(5) M−1 s−1.66 Thus, the FeNTs corrolazine exhibits a
100-fold rate enhancement over an MnV imido corrole
analogue. These findings are in line with the enhanced
reactivity of high-valent iron-oxo versus manganese-oxo
corrolazines.40 In addition, the electron-withdrawing nature of
a tosyl group, as compared to a 2,4,6-trichlorophenyl
substituent, can be expected to enhance the electrophilicity of
2 and make it more reactive for NR group transfer to
nucleophiles such as PPh3.
Scheme 3. Catalytic Turnover
Structural Characterization of an NR Group Transfer
Product. Efforts were made to obtain single crystals for
structural characterization of the phosphine product following
reaction of [FeIV(TBP8Cz+•)(NTs)] with PPh3. An NMR
sample of this reaction mixture in CD2Cl2/CD3CN was allowed
to stand for two weeks, affording black crystalline blocks
suitable for X-ray diffraction. The structure revealed an iron
corrolazine complex with a phospharaniminato ligand coordi-
nated in the axial position, corresponding to [FeIV(TBP8Cz)-
(NPPh3)] (3). A displacement ellipsoid plot for complex 3 is
shown in Figure 6, and selected bond distances and angles are
conclude that 1 is capable of catalytic turnover in the NR
transfer to phosphines, and that the high-valent iron imido
complex 2 is the likely active oxidant in the catalytic cycle.
Kinetics of NR Group Transfer. The kinetics of the
reaction between [FeIV(TBP8Cz+•)(NTs)] and PPh3 were
examined at 23 °C to gain further insight into the reaction
mechanism for NR group transfer. The FeNTs complex was
generated in situ, and excess PPh3 (30−75 equiv) was added to
yield pseudo-first-order conditions. An example of time-
resolved UV−vis spectra used for kinetic measurements is
shown in Figure 3. Monitoring this transformation at 440 nm
over time gives the plot in Figure 5, which was satisfactorily fit
Figure 6. Displacement ellipsoid plot (50% probability level) of
[Fe(TBP8Cz)(NPPh3)] (3). H atoms and disorder are omitted for
clarity.
given in Table 1. The tosyl substituent is absent, presumably
having been hydrolytically cleaved from the initial Ph3PNTs
product during crystallization. The remaining anionic Ph3P
N− axial ligand indicates that the iron is formally in the +4
oxidation state. We attempted to model the donor atom
attached to Fe as an oxygen atom given the possibility of
forming OPPh3 in the reaction mixture. However, substitution
of O for N in the axial position invariably leads to an unusually
large thermal ellipsoid (Figure S3). In addition, substitution of
O for N does not alter bond distances significantly, and the
resulting PO bond length would be unusually short (Table
2). These results rule out the presence of an OPPh3 axial donor
in 3. The structural parameters for 3 are also not reasonable for
a protonated Ph3PNH ligand, and no evidence for a
hydrogen atom could be found in the final difference Fourier
maps.
Figure 5. Representative change in absorbance at 440 nm vs time for 2
+ PPh3 corresponding to the return of FeIII(TBP8Cz) (open circles),
and best fit line (red) (top). Plot of the pseudo-first-order rate
constants (kobs) vs [PPh3] (bottom). A second-order rate constant of
70(15) M−1 s−1 was obtained from the best fit line (black).
with a first-order kinetics model and yielded a kobs = 0.022 s−1.
The concentration of PPh3 was varied and led to a linear
correlation of kobs versus [PPh3] as shown in Figure 5,
consistent with an overall second-order rate law of rate =
k2[FeIV(TBP8Cz+•)(NTs)][PPh3]. The slope of this plot gave a
derived second-order rate constant of 70(15) M−1 s−1. In
comparison, the MnV-imido corrole [MnV(tpfc)(NMes)]
To our knowledge there are only two other crystallo-
graphically characterized mononuclear FeNPPh3 com-
plexes reported to date, and relevant structural data for these
complexes are given in Table 2. These iron complexes are four-
coordinate iron(II) species derived from sterically encumbered
tris(carbene)borate ancillary ligands. The Fe−N distance for
4674
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
prepared crystal from the PPh3 reaction mixture revealed a
similar cell to that obtained for 3. However, we speculated that
an FeIII−OPPh3 complex might have nearly identical unit cell
dimensions to those of 3, and thus, another full set of data was
collected. The structure solution for this second crystal revealed
an OPPh3 rather than NPPh3 axial ligand. The crystal structure
of this complex, [FeIII(TBP8Cz)(OPPh3)] (4a), is provided in
the Supporting Information, and relevant bond distances and
angles are given in Table 1. In this case, the presence of an O
atom was confirmed, but the refinement of an N atom in the
axial position was attempted. As expected, an N atom in this
position led to a significantly smaller thermal ellipsoid, and
three residual electron density peaks as large as 0.69 e− Å−3
were found at 0.43−0.48 Å from the N atom, which strongly
suggests that not enough electron density has been assigned to
this atomic site. The metrical parameters for the second crystal
structure from the reaction of [Fe(TBP8Cz+•)(NTs)] + PPh3
are fully consistent with an FeIII−OPPh3 complex. These
findings indicate that a mixture of Fe-NPPh3 and Fe-OPPh3
products ultimately form in the PPh3 reaction and both
crystallize under the same conditions.
Table 1. Selected Bond Distances (Å) and Angles (deg) for
3, 4a, and 4b
3
4a
4b
Fe(1)−N(1)
Fe(1)−N(2)
Fe(1)−N(4)
Fe(1)−N(6)
Fe(1)−(Npyrrole
1.840(3)
1.844(3)
1.862(3)
1.866(3)
0.292
1.843(2)
1.849(2)
1.867(2)
1.866(2)
0.291
1.840(2)
1.848(2)
1.864(2)
1.865(2)
0.291
)
plane
Fe(1)−(23-atom)core
Cβ−Cβ (av)
0.271
0.262
0.259
1.393
1.388
1.390
Cα−Cβ (av)
1.446
1.448
1.447
Cα−Cα (C4−C5)
Cα−Npyrrole(av)
Cα−Nmeso (av)
1.425(5)
1.373
1.436(4)
1.367
1.430(4)
1.369
1.338
1.338
1.339
Fe1−X (X = N8 or O1)
P(1)−X
1.963(3)
1.427(3)
80.9(1)
90.2(1)
92.9(1)
90.4(1)
97.9(1)
95.9(1)
100.1(1)
101.9(1)
154.5(2)
2.002(2)
1.494(2)
81.12(9)
90.45(9)
92.62(9)
90.17(9)
98.60(9)
96.13(8)
99.30(8)
101.72(8)
154.5(1)
2.001(2)
1.500(2)
81.1(1)
90.3(1)
92.6(1)
90.3(1)
98.50(9)
95.95(9)
99.47(9)
101.85(9)
154.4(1)
N(1)−Fe(1)−N(2)
N(2)−Fe(1)−N(4)
N(4)−Fe(1)−N(6)
N(6)−Fe(1)−N(1)
N(1)−Fe(1)−X
N(2)−Fe(1)−X
N(4)−Fe(1)−X
N(6)−Fe(1)−X
Fe(1)−X−P(1)
Synthesis of [FeIII(TBP8Cz)(OPPh3)]. The FeIII(OPPh3)
complex was independently prepared for comparison purposes.
Addition of excess OPPh3 to a solution of [FeIII(TBP8Cz)] led
to the crystalline complex [FeIII(TBP8Cz)(OPPh3)] (4b) in
good yield (Scheme 4). A displacement ellipsoid plot is shown
[FeIV(TBP8Cz)(NPPh3)] of 1.963(3) Å is considerably longer
than the Fe−N distances seen for the tris(carbene)borate
complexes (Fe−N = 1.807(2)−1.894(2) Å). The shorter Fe−N
bonds in the latter complexes may be assigned to the NPPh3
ligand functioning as both a σ- and π-donor to the metal. In
these approximate C3-symmetric complexes, the dxz/dyz orbitals
can serve as π-acceptors, whereas in the tetragonal Cz complex
the orbitals of similar parentage are π* in character with respect
to the Fe−N bond and are partly filled, leading to the elongated
Fe−N distance. In contrast, the N−P distance for 3 is
significantly shorter than the N−P distances found in the
other FeNPPh3 complexes in Table 2. The origin of this
discrepancy is not known at this time, although the N−P
distance for 3 is clearly shorter than that expected for either
M−OPPh3 or M−(H)NPPh3 complexes, as seen in Table 2.
Efforts to characterize a bulk crystalline sample of
[FeIV(TBP8Cz)(NPPh3)] by ESIMS(+) revealed an isotopic
cluster centered at m/z 1689.7, whose main peak was +2 above
that expected for [Fe(TBP8Cz)(NPPh3)]+•, but a close match
for [Fe(TBP8Cz)(OPPh3)]+• (theoretical m/z 1689.8, see
Supporting Information). However, the isotope pattern did
not fit well for the OPPh3 complex, and suggested that a
possible mixture of Fe−NPPh3 and Fe−OPPh3 products
crystallized in the same bulk material. This result prompted
the analysis of a new batch of crystals from the PPh3 reaction by
X-ray diffraction. A unit cell check on an independently
Scheme 4. Synthesis of FeIII(TBP8Cz)(OPPh3)
in Figure 7, and selected bond distances and angles are given in
Table 1. The metrical parameters match well with those found
for 4a. The UV−vis and ESIMS spectra for 4b are shown in
Figure 8. The mass spectral data show a dominant cluster
centered at m/z 1689.7, and its position and isotope pattern
match nicely for [Fe(TBP8Cz)(OPPh3)]+. This result can be
contrasted with Figure S4, in which a different isotope pattern
was obtained for the mixture of 3 + 4a from the chloramine-T
reaction. The UV−vis spectrum is similar to that of the FeIII
starting material, but a blue-shift of ∼15 nm is seen for the 732
nm band, resulting from coordination of OPPh3 to FeIII.
Elemental analysis confirmed the high purity of 4b as a bulk
crystalline solid, and this material provided a key benchmark for
further studies, including 31P NMR and Mossbauer spectros-
̈
copies, while also confirming our assignment of the 3 + 4a
mixture from the PPh3 oxidation reaction.
Monitoring Production of Hydrolyzed Products by
31P NMR Spectroscopy. The reaction of [FeIV(TBP8Cz+•)-
Table 2. Comparison of Selected Bond Distances (Å) for Metal-XPPh3 Compounds (X = N, O)
M−X
1.963(3)
P−X
1.427(3)
ref
[FeIV(TBP8Cz)(NPPh3)] (3)
[FeIII(TBP8Cz)(OPPh3)] (4b)
M-(H)NPPh3 (M = Os, Cu, Zr, Ta)
FeIII−OPPh3
this study
this study
67, 81−84
85−89
59
2.001(2)
1.500(2)
1.931−2.367
1.912−2.056
1.894(2)
1.511−1.843
1.483−1.507
1.527(2)
PhB(tBuIm)3FeII−NPPh3
PhB(MesIm)3FeII−NPPh3
1.807(2)−1.855(2)
1.524(2)−1.549(2)
90
4675
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
Figure 7. Displacement ellipsoid plot (50% probability level) of
[FeIII(TBP8Cz)(OPPh3)] (4b). H atoms and disorder are omitted for
clarity.
Figure 9. 31P NMR spectra of the reaction of [FeIII(TBP8Cz)] +
chloramine-T trihydrate (3 equiv) + PPh3 (10 equiv) in CD2Cl2/
CD3CN over 2 weeks.
(Figure S7). Thus, the 31P NMR time-course shows the
production of all three phosphorus products, Ph3PNTs,
Ph3PN(H), and Ph3PO, over a two-week period.
Mechanism. A proposed mechanism for NR group transfer
and generation of the different phosphine products is shown in
Scheme 5. Initial formation of the FeNTs complex is
followed by nucleophilic attack of PPh3 to give Ph3PNTs and
an iron(III) corrolazine. The Ph3PNTs may be activated by
FeIII toward hydrolytic cleavage of the tosyl group, resulting in
production of Ph3PNH. Coordination of this phosphinimine
product can then lead to the formation of FeIV(NPPh3),
following oxidation of an FeIII precursor. Although the identity
of the oxidant is not known, oxidation of FeIII to FeIV is likely
facilitated by a strong axial donor such as Ph3PN−. The
generation of Ph3PO occurs upon hydrolysis of Ph3PNH, which
is known to readily hydrolyze to Ph3PO and NH4OH in the
presence of residual water.67 The presence of Ph3PO can then
give rise to the final FeIII(OPPh3) complex. This mechanism
accounts for both the NMR time-course experiments and the
observation of the different products from X-ray crystallo-
graphy.
̈
EPR and Mossbauer Spectroscopies. To obtain
information about the oxidation states and spin states of the
Fe corrolazine complexes described here, EPR and Mossbauer
̈
spectra were collected. Previous work on iron corrole
complexes68−71 has shown that it is often difficult to determine
their electron configurations due to the noninnocent nature of
the corrolate ligand. Similar complications arise with the related
iron corrolazines presented in this study. Further complexity is
Figure 8. (a) UV−vis and (b) ESIMS spectra for [FeIII(TBP8Cz)-
(OPPh3)] (4b) as a crystalline solid dissolved in CH2Cl2.
to be expected when Mossbauer spectra of frozen solutions are
̈
(NTs)] (2) + PPh3 was monitored by 31P NMR spectroscopy
over an extended time-course to gain insight into the
mechanism of NR transfer and formation of the final
phosphine-derived products [FeIV(TBP8Cz)(NPPh3)] (3) and
[FeIII(TBP8Cz)(OPPh3)] (4). Examination of the reaction
mixture after day 1 shows only the Ph3PNTs product, as
expected (Figure 9). Following this initial spectrum, there is a
slow build-up of the detosylated Ph3PNH compound at 21.8
ppm (verified by comparison with an authentic sample). By day
7, a broad peak appears at 27.9 ppm, which then grows in
intensity after a further week of reaction (day 14). An
independent sample of pure [FeIII(TBP8Cz)(OPPh3)] (4b)
exhibited the same broad peak in the 31P NMR spectrum
collected because corrolazines, like other porphyrinoid
compounds, are prone to aggregation, which can lead to
intermolecular spin−spin relaxation.72−74 For these reasons the
Mossbauer and EPR data have to be treated as a whole. To
̈
support the Mossbauer and EPR assignments we ran a set of
̈
DFT calculations in tandem with the experiments.
A solid state, crystalline sample of 4b formed from the
reaction of [FeIII(TBP8Cz)] with Ph3PO provides a well-
defined reference compound for spectroscopic measurements
because the structure has been determined with high precision
by single-crystal X-ray diffraction, and the oxidation state of the
iron center can be definitively assigned as +3. A sample of 4
exhibits a Mossbauer spectrum at 5.3 K with a single, sharp
̈
4676
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
Scheme 5. Proposed Mechanism
quadrupole doublet (δ = 0.13 mm s−1; ΔEQ = 4.40 mm s−1,
and the large quadrupole splitting is in part due to the square
pyramidal geometry (see bond angles above). Indeed, the
square pyramidal geometry of the iron center revealed by the
crystal structure of [FeIII(TBP8Cz)(OPPh3)] (Figure 7) and
the presence of the strong-field equatorial corrolazine ligand
Figure 10). The isomer shift is consistent with an intermediate
3
spin (S = /2) FeIII ion as seen by comparison with other
intermediate spin iron(III) complexes (see Table 5 in ref 56),
2
2
should greatly destabilize the dx ‑y orbital relative to the other d
3
orbitals and therefore stabilize the S = /2 state relative to the
5
high spin S = /2 state. The intermediate spin state of this
compound is further confirmed by EPR spectroscopy. The X-
band EPR spectrum of a crystalline sample of 4 dissolved in
toluene and measured at 15 K (Figure 11) shows an axial signal
3
with g⊥ = 4.43, g∥ = 1.99, which is typical for an S = /2 ion.
In comparison, the Mossbauer spectrum of solid
̈
[FeIII(TBP8Cz)], which lacks an OPPh3 axial ligand, was
measured at 293 K and consists of two quadrupole doublets (δ
= 0.06 mm s−1, ΔEQ = 4.00 mm s−1 and δ = 0.30 mm s−1, ΔEQ
= 1.00 mm s−1, Figure 10) in a ratio of approximately 3:1.
Figure 10. Mossbauer spectra of a crystalline sample of
̈
[FeIII(TBP8Cz)(OPPh3)] (4b) measured at 5.3 K and a solid
sample of [FeIII(TBP8Cz)] (1) measured at 293 K (solid black line =
high spin FeIII; dashed black line = intermediate spin FeIII) and 5.3 K
with no applied magnetic field (dashed line = slow relaxing
intermediate spin FeIII). The red lines show the best fit to the data.
Figure 11. X-band EPR spectrum at 15 K of crystalline
[FeIII(TBP8Cz)(OPPh3)] (4b) dissolved in toluene. Experimental
conditions: frequency = 9.49 GHz, microwave power = 0.201 mW,
modulation amplitude 10.0 G, modulation frequency = 100 kHz.
4677
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
These parameters and intensities are similar to those recently
reported for [FeIII(TBP8Cz)] by Nakamura and co-workers,75
which revealed an intermediate spin iron(III) spectrum at 15
K.42 Careful comparison of the Mossbauer spectra in the
̈
who measured temperature-dependent Mossbauer spectra
presence and absence of a small magnetic field (47 mT) allows
this species to be fitted with the values δ = 0.15 mm s−1; ΔEQ =
+3.98 mm s−1; D = −4 cm−1; E/D = 0; Ax = Ay = 19 T, Az = 6
T, which are comparable with intermediate spin FeIII
tetraamido complexes.77 We can conclude that dissolution in
toluene provides a uniform environment for complex 1 and
leads to the observation of a single iron(III) intermediate spin
ground state, in contrast to the solid state, in which stacking or
other aggregation effects appear to give rise to a mixture of the
̈
down to 77 K. In this study the quadrupole splitting was
seen to be temperature-independent, but the isomer shift
increased with a decrease in temperature, as one would expect,
as a result of the second-order Doppler shift.76 As seen in
Figure 10, the spectrum at 5.3 K for 1 consists of two
quadrupole doublets with a slight increase in the isomer shifts
(δ = 0.15 mm s−1 and δ = 0.47 mm s−1) as compared to the
data at 293 K. These isomer shifts can be assigned to a mixture
5
of intermediate spin FeIII and high spin FeIII (S = /2) ions.
3
5
two different spin states S = /2 and S = /2 for 1.
The third possible spin state for an FeIII species is a low-spin
(S = /2) configuration, and this state can be accessed for
Additionally, there is a broad component (∼20%, dashed line
1
Figure 10) caused by slow relaxation of the intermediate spin
iron(III) centers relative to the Mossbauer time scale of ∼107
̈
iron(III) corrolazine by addition of excess pyridine. Previous
work has shown that addition of pyridine to [FeIII(TBP8Cz)]
gives the axially ligated complex [FeIII(TBP8Cz)(py)2], which
reveals an intense rhombic EPR spectrum (g 2.39, 2.20, 1.90)
that is characteristic of a low spin FeIII porphyrinoid
s−1, resulting in splitting of the nuclear magnetic spin states and
giving rise to the broad spectrum observed. The parameters
used to simulate this magnetic spectrum were those used to fit
the frozen solution spectrum of [FeIII(TBP8Cz)] described
below.
compound.42 The Mossbauer spectrum of this species at 100
̈
Dissolution of the [FeIII(TBP8Cz)] complex in toluene
K consists of a symmetrical quadrupole doublet (δ = 0.01 mm
s−1, ΔEQ = 3.99 mm s−1, Figure 13). The isomer shift is
consistent with low spin FeIII, and indeed the value is lower
than that of the intermediate spin species discussed above. At
5.3 K, the spectrum consists of both slow (dashed line, Figure
13) and fast relaxing low spin FeIII signals (δ = 0.03 mm s−1,
ΔEQ = 4.03 mm s−1), as seen for the intermediate spin complex
(Figure 10). These parameters are similar to, but slightly larger
than, those recently described75 for the analogous bis(cyano)
complex [FeIII(TBP8Cz)(CN)2]2− (δ = −0.09 mm s−1, ΔEQ =
2.74 mm s−1 at 77 K).
provides Mossbauer spectra that help to clarify the spectral
̈
assignments. A frozen toluene solution of the FeIII complex
produces a single quadrupole doublet at 100 K (δ = 0.11 mm
s−1, ΔEQ = 3.96 mm s−1, Figure 12), consistent with the
The Mossbauer spectrum of bulk, crystalline 3 consists of a
̈
quadrupole doublet at 293 K (δ = 0.13 mm s−1, ΔEQ = 4.25
mm s−1, Figure 13). However, a bulk sample of this material
consists of both [FeIV(TBP8Cz)(NPPh3)] and [FeIII(TBP8Cz)-
(OPPh3)] as shown by independent X-ray analysis. An EPR
spectrum of a bulk sample of 3 at 15 K revealed only a weak
signal with broad features at g 5.39, 2.02 (Figure S10),
consistent with the presence of [FeIII(TBP8Cz)(OPPh3)] as a
minor impurity. The Mossbauer spectrum of 3 at 5.3 K can be
̈
fitted to an FeIV (S = 1) species (δ = 0.13 mm s−1; ΔEQ = 4.25
mm s−1; D = +5 cm−1; E/D = 0; Ax = Ay = 8 T, Az = 10 T;
Figure 13) together with [FeIII(TBP8Cz)(OPPh3)] as a minor
component (∼10%, fitted with the parameters found for the
pure FeIII(OPPh ) complex). Thus, Mossbauer and EPR
̈
3
spectroscopy are in agreement with the assignment of the
bulk crystalline material for 3. Interestingly, the isomer shift for
the FeIV complex (δ = 0.13 mm s−1) is comparable to the
isomer shift found for the intermediate spin FeIII corrolazines,
which suggests that the FeIV−NPPh3 complex may be better
described as an intermediate spin FeIII center coupled to a
corrolazine π-cation radical. This dichotomy for iron(IV)
corroles has been discussed at length by Walker and co-
workers,70 but these two options cannot be distinguished for
the current complex from the available data. We note that the
quadrupole splitting parameters are distinct for the FeIV
complex 3 (4.25 mm s−1) and the intermediate spin-FeIII
complex 4 (4.40 mm s−1), although not by as much as might
be expected. Future experimental/computational work may
shed light on this issue.
Figure 12. Mossbauer spectra of [FeIII(TBP Cz)] dissolved in toluene
̈
8
and measured at 100 K with an applied magnetic field of 47 mT
parallel to the γ-rays, 5.3 K with no magnetic field and 5.3 K with an
applied magnetic field of 47 mT parallel to the γ-rays. The red line
through the data shows the best fit using the parameters given in the
text.
presence of a pure intermediate spin species. If the sample is
then suitably diluted and carefully frozen,72 the spectra
collected at the lower temperature of 5.3 K are clearly in the
slow relaxation regime (Figure 12), and can be fitted to a single
Finally, Mossbauer spectroscopy was also employed to
̈
3
S = /2 species. This result is in good agreement with EPR
analyze a 57Fe-labeled sample of [FeIV(TBP8Cz+•)(NTs)],
which was isolated as a precipitated solid from the reaction
spectra of the same complex in a frozen solution of toluene,
4678
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
unambiguously defined, but if positive, a rotation of the electric
field gradient (EFG) relative to the principle axes is required. In
fact there is no reason why the principal axes should be
collinear and the EFG tensor can be rotated by a set of Euler
angles α, β, and γ (0, 105, 0). The low isomer shift and smaller
quadrupole splitting of this species as compared to those found
for the FeIII complexes strongly suggest that this complex
contains an iron(IV) center as proposed.68 In addition, EPR
spectroscopy of this complex at 15 K gives a sharp axial signal
1
with g⊥ = 2.10, g∥ = 2.00, consistent with an overall S = /2
ground state as expected for an FeIV−Cz-π-cation-radical and in
good agreement with the Mossbauer analysis.
̈
Altogether, the Mossbauer data show consistent findings
̈
across all species as summarized in Table 3. The intermediate
spin S = /2 FeIII species all have isomer shifts of ∼+0.13 mm
3
s−1. The isomer shift for [FeIV(TBP8Cz)(NPPh3)] also falls in
this range, and this complex may be best described as an FeIII
coupled to a π-radical-cation. Conversion to a low spin FeIII
center by addition of pyridine to form [FeIII(TBP8Cz)(py)2]
lowers the isomer shift to +0.03 mm s−1. Further oxidation to
form [FeIV(TBP8Cz+•)(NTs)] leads to an isomer shift of −0.05
mm s−1, consistent with a high-valent iron(IV) ion.
Computational Studies. To further confirm the spectro-
scopic data and gain structural insight into the iron-imido
complex we have calculated the Mossbauer and EPR
̈
parameters of [FeIV(H8Cz+•)(NTs)] in both the doublet and
quartet spin states using DFT. To reduce calculation time a
t
model where the eight peripheral Bu-phenyl groups from the
corrolazine ring were replaced by hydrogen atoms was
implemented, H8Cz. Initial geometric optimization led to a
structure with reasonable bond lengths (Figure 15). The Fe−N
Figure 13. Mossbauer spectra of [FeIII(TBP Cz)] dissolved in
̈
8
pyridine at 100 and 5.3 K with no applied field (dashed line = slow
relaxing low spin FeIII). Mossbauer spectra of a solid sample of
̈
[FeIII(TBP8Cz)(N/OPPh3)] measured at 293 and 5.3 K (dotted line =
[FeIII(TBP8Cz)(OPPh3)]; dashed line = [FeIV(TBP8Cz)(NPPh3)]).
Frozen solution spectrum of [FeIV(TBP8Cz+•)(NTs)] measured at 5.3
K (dashed line = ∼25% [FeIII(TBP8Cz)]). The red line through the
data shows the best fit using the parameters given in the text.
mixture of [FeIII(TBP8Cz)] and chloramine-T. The spectrum
measured at 5.3 K in toluene (Figure 13) can be fitted to two
species: ∼25% of unreacted [FeIII(TBP8Cz)] (dashed line,
parameters given above), and an FeIV (S = 1) ion
antiferromagnetically coupled to a Cz-π-cation-radical (δ =
−0.05 mm s−1; ΔEQ = 2.94 mm s−1; Ax = Ay = 26 T, Az = 3.5 T;
J ∼ 30 cm−1). The sign of the quadrupole splitting cannot be
Figure 14. Mossbauer spectra of [FeIV(TBP Cz+•)(NTs)] measured at
5.3 K. Overlaid is a simulation (red line) using the parameters
determined by DFT with the EFG and A rotated by the following
Euler angles (19, 100, 51) and (70, 92, 3).
̈
8
Table 3. Mossbauer Parameters for Iron Corrolazine Complexes
̈
a
a
a
spin state δ (mm s−1
)
ΔEQ (mm s−1
)
Γ (mm s−1
)
D (cm−1
−4
)
E/D Ax/gnμn (T) Ay/gnμn (T) Az/gnμn (T)
[FeIII(TBP8Cz)]
/
/
/
0.15
+3.98
4.03
0.6
0.7
0.3
0.7
0.3
0
19
19
6
3
2
2
2
[FeIII(TBP8Cz)(py)2]
[FeIII(TBP8Cz)(OPPh3)]
[FeIV(TBP8Cz)(NPPh3)]
[FeIV(TBP8Cz•+)(NTs)]
0.03
1
3
0.13
4.40
b
1
0.13
4.25
+5
0
8
8
10
c
1
expt
/
−0.05
−0.12
2.95
26
26
24.0
3.5
30.9
2
d
1
calcd
/
2
+2.92
−4.8
a
b
The sign of the hyperfine coupling constants cannot be determined. The spin state is not fully determined and can be considered an intermediate
c
spin FeIII coupled to a radical cation, as the isomer shift suggests. The S = 1 FeIV is antiferromagnetically coupled to a S = 1/2 radical cation with an
d
exchange coupling constant of ∼30 cm−1. The electric field gradient (EFG) is rotated using the following Euler angles (0, 105, 0). η = 0.38. To fit
experiment, both the EFG and A tensors have to be rotated by the following Euler angles (19, 100, 51) and (70, 92, 3), respectively, because the
principal axes used are different.
4679
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
Figure 15. DFT calculated structures of 2,4[FeIV(H8Cz+•)(NTs)] with bond lengths in angstroms and group spin densities (ρ) in atomic units. Also
given are the spin state energy differences in kcal mol−1 as calculated at UB3LYP/B2//UB3LYP/B1 including solvent and zero-point corrections.
Orbital drawings of the singly occupied molecular orbitals are given on the right.
bond length of 1.722 Å is comparable to other iron-imido
complexes both experimental and calculated.8,78 The doublet
spin state is the ground state and is found to lie 5.7 kcal mol−1
below the quartet spin state, consistent with the spectroscopic
results given above. The spin densities were then calculated and
molecular orbitals analyzed, which revealed two unpaired
electrons in the π*xz/yz orbitals antiferromagnetically coupled to
a single electron in the a″-type Cz orbital. The orbital shapes
and the electronic configuration show resemblance to that
calculated for [MnV(H8Cz+•)(O)], which also gave single
occupation of the a″ orbital.38 Due to mixing of the atomic
3dxz/3dyz orbitals on Fe with 2px/2py orbitals on nitrogen, the
π*xz and π*yz orbitals are mixed iron/nitrogen antibonding
orbitals, and consequently, considerable spin density is
observed on the nitrogen and tosylate group. This mixing is
similar to that observed on the oxo group in heme and
nonheme iron(IV)-oxo species.79,80
CONCLUSIONS
■
The reaction of [FeIII(TBP8Cz)] with the NR transfer agent
chloramine-T results in the formation of the high-valent
iron(IV) terminal imido complex [FeIV(TBP8Cz+•)(NTs)] (2).
Previously iron terminal imido complexes have been stabilized
through a combination of steric encumbrance and the
enforcement of relatively low-coordinate geometries about the
iron center. The stability of 2 is likely a consequence of the
oxidatively resistant, trianionic corrolazine supporting ligand
favoring high-valent iron(IV) together with the strong σ and π
donation of the terminal tosylimido donor. The stability of 2 is
significant because high-valent, multiply bonded terminal imido
complexes of the mid- to late-transition metals are normally
considered to be highly reactive, but the data presented here
indicates that 2 is much more stable (and consequently less
reactive) than is expected for an iron-imido complex. Thus,
catalytic mechanisms that rely on proposed intermediates of
this type, especially in porphyrinoid chemistry, may need to be
carefully scrutinized when identifying the actual oxidizing
species.
Finally, the Mossbauer and EPR parameters were then
̈
calculated and are given in Table 3. The parameters compare
favorably with the experimentally determined ones and provide
further strong evidence that our understanding of the electronic
state is correct. The calculated g values (2.00, 2.03, 2.07) are
Mossbauer and EPR spectroscopies show that
̈
[FeIV(TBP8Cz+•)(NTs)] contains a high-valent FeIV ion with
an extra oxidizing equivalent stored on the Cz ring as a π-
cation-radical, analogous to the previously described com-
pound-I-like terminal oxo complex [FeIV(TBP8Cz+•)(O)].39
This view has been further corroborated by DFT calculations.
The spin state of both the imido and oxo complexes is
proposed to be a doublet with iron(IV) antiferromagnetically
coupled to the Cz π-cation-radical. It was demonstrated that 2
is capable of NR group transfer to triphenylphosphine to give
Ph3PNTs, leading to the two-electron reduction of 2 to give the
FeIII complex 1. Catalytic NR transfer to PPh3 by 1 was also
successful, suggesting that further design of suitable metal-
locorrolazines (or corroles) may yield potent NR transfer
rhombic rather than axial, but the Mossbauer isomer shift and
̈
quadrupole splitting reproduce the experimental data within the
error of the calculations. A simulation of the Mossbauer
̈
spectrum with these parameters is given in Figure 14 and shows
graphically how close the match is. To obtain a reasonable fit
both the electric field gradient (EFG) and hyperfine (A)
tensors had to be rotated relative to each other. The following
Euler angles were used: (19, 100, 51) for the EFG and (70, 92,
3) for A.
4680
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
(15) Eikey, R. A.; Abu-Omar, M. M. Coord. Chem. Rev. 2003, 243,
83−124.
catalysts. The kinetic data For NR transfer from 2 to PPh3 was
in agreement with the proposed bimolecular mechanism, and a
comparison of rate constants shows that 2 exhibits a dramatic
rate enhancement (∼100-fold) for NR transfer as compared to
MnV(NAr) corrole analogues. Structural analyses of complexes
formed following initial NR transfer reveal that both
[FeIV(TBP8Cz)(NPPh3)] and [FeIII(TBP8Cz)(OPPh3)] crys-
tallize from the same reaction mixture. A mechanism was
described that can account for the NR transfer reactivity of 2 as
well as the subsequent formation of the former complexes.
(16) Abu-Omar, M. M. Dalton Trans. 2011, 40, 3435−3444.
(17) Smith, J. M.; Subedi, D. Dalton Trans. 2012, 41, 1423−1429.
(18) To our knowledge, there are only two reported Fe compounds
with +6 oxidation state: Berry, J. F.; Bill, E.; Bothe, E.; DeBeer George,
S.; Mienert, B.; Neese, F.; Wieghardt, K. Science 2006, 312, 1937−
1941.
(19) Hohenberger, J.; Ray, K.; Meyer, K. Nat. Commun. 2012, 3, 720.
(20) Mahy, J. P.; Battioni, P.; Mansuy, D.; Fisher, J.; Weiss, R.;
Mispelter, J.; Morgensternbadarau, I.; Gans, P. J. Am. Chem. Soc. 1984,
106, 1699−1706.
(21) Klinker, E. J.; Jackson, T. A.; Jensen, M. P.; Stubna, A.; Juhasz,
ASSOCIATED CONTENT
■
G.; Bominaar, E. L.; Munck, E.; Que, L., Jr. Angew. Chem., Int. Ed.
̈
S
* Supporting Information
2006, 45, 7394−7397.
Spectroscopic and crystallographic data (CIF), and details of
the DFT calculations, including relative and absolute energies,
group spin densities, charges, and Cartesian coordinates of all
structures. This material is available free of charge via the
Internet at http://pubs.acs.org.
(22) Brown, S. D.; Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2003,
125, 322−323.
(23) Moret, M. E.; Peters, J. C. Angew. Chem., Int. Ed. 2011, 50,
2063−2067.
(24) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2003, 125, 10782−
10783.
(25) Brown, S. D.; Peters, J. C. J. Am. Chem. Soc. 2005, 127, 1913−
AUTHOR INFORMATION
■
1923.
Corresponding Author
*E-mail: dpg@jhu.edu (D.P.G.); gjameson@chemistry.otago.
ac.nz (G.N.L.J.); sam.devisser@manchester.ac.uk (S.P.d.V.).
(26) Thomas, C. M.; Mankad, N. P.; Peters, J. C. J. Am. Chem. Soc.
2006, 128, 4956−4957.
(27) Nieto, I.; Ding, F.; Bontchev, R. P.; Wang, H. B.; Smith, J. M. J.
Am. Chem. Soc. 2008, 130, 2716−2717.
Notes
(28) Bowman, A. C.; Milsmann, C.; Bill, E.; Turner, Z. R.;
Lobkovsky, E.; DeBeer, S.; Wieghardt, K.; Chirik, P. J. J. Am. Chem.
Soc. 2011, 133, 17353−17369.
The authors declare no competing financial interest.
(29) Bart, S. C.; Lobkovsky, E.; Bill, E.; Chirik, P. J. J. Am. Chem. Soc.
2006, 128, 5302−5303.
ACKNOWLEDGMENTS
■
This work was supported by the NSF (CHE0909587 and
CHE1213836 to D.P.G.). P.L. is grateful for the Queen Sirikit
Scholarship (Thailand). We thank Dr. Yosra M. Badiei for the
synthesis of 57FeCl2 and Dr. Victoria J. A. Jameson for help with
(30) Cowley, R. E.; Eckert, N. A.; Vaddadi, S.; Figg, T. M.; Cundari,
T. R.; Holland, P. L. J. Am. Chem. Soc. 2011, 133, 9796−9811.
(31) Cowley, R. E.; Holland, P. L. Inorg. Chem. 2012, 51, 8352−8361.
(32) Lansky, D. E.; Kosack, J. R.; Sarjeant, A. A. N.; Goldberg, D. P.
Inorg. Chem. 2006, 45, 8477−8479.
Mossbauer sample preparation. The computational work was
̈
(33) Eikey, R. A.; Khan, S. I.; Abu-Omar, M. M. Angew. Chem., Int.
Ed. 2002, 41, 3592−3595.
supported by CPU time provided by the National Service of
Computational Chemistry Software (NSCCS). D.K. holds a
Ramanujan Fellowship from the Department of Science and
Technology (DST), New Delhi (India), and acknowledges its
financial support (Research Grants SR/S2/RJN-11/2008 and
SR/S1/PC-58/2009).
(34) 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.
́
(35) Lansky, D. E.; Mandimutsira, B.; Ramdhanie, B.; Clausen, M.;
Penner-Hahn, J.; Zvyagin, S. A.; Telser, J.; Krzystek, J.; Zhan, R. Q.;
Ou, Z. P.; Kadish, K. M.; Zakharov, L.; Rheingold, A. L.; Goldberg, D.
P. Inorg. Chem. 2005, 44, 4485−4498.
REFERENCES
■
(36) Prokop, K. A.; Goldberg, D. P. J. Am. Chem. Soc. 2012, 134,
8014−8017.
(1) Ertl, G. Angew. Chem., Int. Ed. 1990, 29, 1219−1227.
(2) Smil, V. Enriching the Earth: Fritz Haber, Carl Bosch, and the
Transformation of World Food Production; MIT Press: Cambridge,
2004.
(37) Prokop, K. A.; de Visser, S. P.; Goldberg, D. P. Angew. Chem.,
Int. Ed. 2010, 49, 5091−5095.
(38) Prokop, K. A.; Neu, H. M.; de Visser, S. P.; Goldberg, D. P. J.
Am. Chem. Soc. 2011, 133, 15874−15877.
(3) Hoffman, B. M.; Dean, D. R.; Seefeldt, L. C. Acc. Chem. Res.
2009, 42, 609−619.
(39) McGown, A. J.; Kerber, W. D.; Fujii, H.; Goldberg, D. P. J. Am.
Chem. Soc. 2009, 131, 8040−8048.
(4) Seefeldt, L. C.; Hoffman, B. M.; Dean, D. R. Annu. Rev. Biochem.
2009, 78, 701−722.
(40) Cho, K.; Leeladee, P.; McGown, A. J.; DeBeer, S.; Goldberg, D.
P. J. Am. Chem. Soc. 2012, 134, 7392−7399.
(5) Pickett, C. J. J. Biol. Inorg. Chem. 1996, 1, 601−606.
(6) Svastits, E. W.; Dawson, J. H.; Breslow, R.; Gellman, S. H. J. Am.
Chem. Soc. 1985, 107, 6427−6428.
(41) Simkhovich, L.; Gross, Z. Tetrahedron Lett. 2001, 42, 8089−
8092.
(7) King, E. R.; Betley, T. A. Inorg. Chem. 2009, 48, 2361−2363.
(8) King, E. R.; Hennessy, E. T.; Betley, T. A. J. Am. Chem. Soc. 2011,
133, 4917−4923.
(42) Kerber, W. D.; Ramdhanie, B.; Goldberg, D. P. Angew. Chem.,
Int. Ed. 2007, 46, 3718−3721.
(43) Ramdhanie, B.; Stern, C. L.; Goldberg, D. P. J. Am. Chem. Soc.
2001, 123, 9447−9448.
(9) Berry, J. F. Comments Inorg. Chem. 2009, 30, 28−66.
(10) Mehn, M. P.; Peters, J. C. J. Inorg. Biochem. 2006, 100, 634−643.
(11) Saouma, C. T.; Peters, J. C. Coord. Chem. Rev. 2011, 255, 920−
937.
(44) Aresta, M.; Nobile, C. F.; Petruzzelli, D. Inorg. Chem. 1977, 16,
1817−1818.
(45) Bittner, S.; Assaf, Y.; Krief, P.; Pomerantz, M.; Ziemnicka, B. T.;
Smith, C. G. J. Org. Chem. 1985, 50, 1712−1718.
(46) Davidson, M. G.; Goeta, A. E.; Howard, J. A. K.; Lehmann, C.
W.; McIntyre, G. M.; Price, R. D. J. Organomet. Chem. 1998, 550,
449−452.
(12) Zalatan, D. N.; Du Bois, J. Top. Curr. Chem. 2010, 292, 347−
378.
(13) Muller, P.; Fruit, C. Chem. Rev. 2003, 103, 2905−2919.
̈
(14) Chow, T. W. S.; Chen, G. Q.; Liu, Y. G.; Zhou, C. Y.; Che, C.
M. Pure Appl. Chem. 2012, 84, 1685−1704.
4681
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682

Inorganic Chemistry
Article
(47) Kumar, D.; Karamzadeh, B.; Sastry, G. N.; de Visser, S. P. J. Am.
Chem. Soc. 2010, 132, 7656−7667.
(81) Stephens, J. C.; Khan, M. A.; Houser, R. P. Inorg. Chem. 2001,
40, 5064−5065.
(82) Chen, G.; Man, W. L.; Yiu, S. M.; Wong, T. W.; Szeto, L.;
Wong, W. T.; Lau, T. C. Dalton Trans. 2011, 40, 1938−1944.
(48) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(49) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37, 785−
789.
(83) Grob, T.; Geiseler, G.; Harms, K.; Greiner, A.; Dehnicke, K. Z.
̈
Anorg. Allg. Chem. 2002, 628, 217−221.
(50) Jaguar, version 7.9; Schrodinger, LLC: New York, 2011.
(51) Frisch, M. J.; et al. Gaussian 03, revision C.02; Gaussian, Inc.:
Wallingford, CT, 2004. See Supporting Information.
(52) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270−283.
(53) de Visser, S. P. J. Am. Chem. Soc. 2010, 132, 1087−1097.
(54) Neese, F. ORCAAn Ab Initio, DFT and Semiempirical SCF-
MO Package, Version 2.9; Bonn, Germany, 2009.
(55) Neese, F. Inorg. Chim. Acta 2002, 337, 181−192.
(56) Sinnecker, S.; Svensen, N.; Barr, E. W.; Ye, S.; Bollinger, J. M.,
Jr.; Neese, F.; Krebs, C. J. Am. Chem. Soc. 2007, 129, 6168−6179.
(57) Karamzadeh, B.; Kumar, D.; Sastry, G. N.; de Visser, S. P. J.
Phys. Chem. A 2010, 114, 13234−13243.
(84) Schrumpf, F.; Roesky, H. W.; Noltemeyer, M. Z. Naturforsch., B
1990, 45, 1600−1602.
(85) Tomi, F.; Wah, H. L. K.; Postel, M. New J. Chem. 1988, 12,
289−292.
(86) Ding, Z. R.; Bhattacharya, S.; Mccusker, J. K.; Hagen, P. M.;
Hendrickson, D. N.; Pierpont, C. G. Inorg. Chem. 1992, 31, 870−877.
(87) Chekhlov, A. N. Russ. J. Inorg. Chem. 2005, 50, 1207−1211.
(88) Atkinson, F. L.; Blackwell, H. E.; Brown, N. C.; Connelly, N. G.;
Crossley, J. G.; Orpen, A. G.; Rieger, A. L.; Rieger, P. H. J. Chem. Soc.,
Dalton. Trans. 1996, 3491−3502.
(89) Olejnik, Z.; Lis, T.; Ondrejkovicova, I. Acta Crystallogr., Sect. C
1995, 51, 2246−2249.
(90) Scepaniak, J. J.; Harris, T. D.; Vogel, C. S.; Sutter, J.; Meyer, K.;
Smith, J. M. J. Am. Chem. Soc. 2011, 133, 3824−3827.
(58) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 6252−
6254.
(59) Scepaniak, J. J.; Fulton, M. D.; Bontchev, R. P.; Duesler, E. N.;
Kirk, M. L.; Smith, J. M. J. Am. Chem. Soc. 2008, 130, 10515−15017.
(60) Mankad, N. P.; Muller, P.; Peters, J. C. J. Am. Chem. Soc. 2010,
132, 4083−4085.
(61) Abu-Omar, M. M.; Shields, C. E.; Edwards, N. Y.; Eikey, R. A.
Angew. Chem., Int. Ed. 2005, 44, 6203−6207.
(62) Fructos, M. R.; Trofimenko, S.; Diaz-Requejo, M. M.; Perez, P.
J. J. Am. Chem. Soc. 2006, 128, 11784−11791.
(63) Fantauzzi, S.; Caselli, A.; Gallo, E. Dalton Trans. 2009, 5434−
5443.
(64) Vyas, R.; Gao, G. Y.; Harden, J. D.; Zhang, X. P. Org. Lett. 2004,
6, 1907−1910.
(65) Harden, J. D.; Ruppel, J. V.; Gao, G. Y.; Zhang, X. P. Chem.
Commun. 2007, 4644−4646.
(66) Edwards, N. Y.; Eikey, R. A.; Loring, M. I.; Abu-Omar, M. M.
Inorg. Chem. 2005, 44, 3700−3708.
(67) Bennett, B. K.; Saganic, E.; Lovell, S.; Kaminsky, W.; Samuel, A.;
Mayer, J. M. Inorg. Chem. 2003, 42, 4127−4131.
(68) Cai, S.; Walker, F. A.; Licoccia, S. Inorg. Chem. 2000, 39, 3466−
3474.
(69) Walker, F. A.; Licoccia, S.; Paolesse, R. J. Inorg. Biochem. 2006,
100, 810−837.
(70) Zakharieva, O.; Schunemann, V.; Gerdan, M.; Licoccia, S.; Cai,
̈
S.; Walker, F. A.; Trautwein, A. X. J. Am. Chem. Soc. 2002, 124, 6636−
6648.
(71) Schwalbe, M.; Dogutan, D. K.; Stoian, S. A.; Teets, T. S.;
Nocera, D. G. Inorg. Chem. 2011, 50, 1368−1377.
(72) Amusa, A.; Debrunner, P.; Frauenfe, H.; Munck, E.; DePasqua,
̈
G. J. Phys. C 1974, 7, 1881−1891.
(73) Teschner, T.; Yatsunyk, L.; Schunemann, V.; Paulsen, H.;
Winkler, H.; Hu, C. J.; Scheidt, W. R.; Walker, F. A.; Trautwein, A. X.
J. Am. Chem. Soc. 2006, 128, 1379−1389.
(74) Simonneaux, G.; Schunemann, V.; Morice, C.; Carel, L.;
Toupet, L.; Winkler, H.; Trautwein, A. X.; Walker, F. A. J. Am. Chem.
Soc. 2000, 122, 4366−4377.
(75) Kurahashi, S.; Ikeue, T.; Sugimori, T.; Takahashi, M.; Mikuriya,
M.; Handa, M.; Ikezaki, A.; Nakamura, M. J. Porphyrins Phthalocyanines
2012, 16, 518−529.
(76) Gutlich, P.; Bill, E.; Trautwein, A. X. Mossbauer Spectroscopy and
̈
̈
Transition Metal Chemistry: Fundamentals and Applications; Springer-
Verlag: Berlin, 2011.
(77) Kostka, K. L.; Fox, B. G.; Hendrich, M. P.; Collins, T. J.;
Rickard, C. E. F.; Wright, L. J.; Munck, E. J. Am. Chem. Soc. 1993, 115,
̈
6746−6757.
(78) Jaccob, M.; Rajaraman, G. Dalton Trans. 2012, 41, 10430−
10439.
(79) de Visser, S. P. J. Am. Chem. Soc. 2006, 128, 15809−15818.
(80) Iron-Containing Enzymes: Versatile Catalysts of Hydroxylation
Reactions in Nature; de Visser, S. P., Kumar, D., Eds.; Royal Society of
Chemistry Publishing: Cambridge, U.K., 2011.
4682
dx.doi.org/10.1021/ic400280x | Inorg. Chem. 2013, 52, 4668−4682