Structural insights into nonheme alkylperoxoiron(III) and oxoiron(IV) intermediates by X-ray absorption spectroscopy

Article abstract of DOI:10.1021/ja047667w

Transient mononuclear low-spin alkylperoxoiron(III) and oxoiron(IV) complexes that are relevant to the activation of dioxygen by nonheme iron enzymes have been generated from synthetic iron(II) complexes of neutral tetradentate (TPA) and pentadentate (N4Py, Bn-TPEN) ligands and structurally characterized by means of Fe K-edge X-ray absorption spectroscopy (XAS). Notable features obtained from fits of the EXAFS region are Fe-O bond lengths of 1.78 A for the alkylperoxoiron(III) intermediates and 1.65-1.68 A for the oxoiron(IV) intermediates, reflecting different strengths in the Fe-O π interactions. These differences are also observed in the intensities of the 1s-to-3d transitions in the XANES region, which increase from 4 units for the nearly octahedral iron(II) precursor to 9-15 units for the alkylperoxoiron(III) intermediates to 25-29 units for the oxoiron(IV) species.

Full text of DOI:10.1021/ja047667w

Published on Web 12/02/2004
Structural Insights into Nonheme Alkylperoxoiron(III) and
Oxoiron(IV) Intermediates by X-ray Absorption Spectroscopy
Jan-Uwe Rohde,† Ste´phane Torelli,† Xiaopeng Shan,† Mi Hee Lim,†,‡
Eric J. Klinker,† Jo´zsef Kaizer,† Kui Chen,† Wonwoo Nam,‡ and
Lawrence Que, Jr.*,†
Department of Chemistry and Center for Metals in Biocatalysis, UniVersity of Minnesota,
207 Pleasant Street SE, Minneapolis, Minnesota 55455, and Department of Chemistry,
DiVision of Nano Sciences, and Center for Biomimetic Systems, Ewha Womans UniVersity,
Seoul 120-750, Korea
Received April 22, 2004; E-mail: que@chem.umn.edu
Abstract: Transient mononuclear low-spin alkylperoxoiron(III) and oxoiron(IV) complexes that are relevant
to the activation of dioxygen by nonheme iron enzymes have been generated from synthetic iron(II)
complexes of neutral tetradentate (TPA) and pentadentate (N4Py, Bn-TPEN) ligands and structurally
characterized by means of Fe K-edge X-ray absorption spectroscopy (XAS). Notable features obtained
from fits of the EXAFS region are Fe-O bond lengths of 1.78 Å for the alkylperoxoiron(III) intermediates
and 1.65-1.68 Å for the oxoiron(IV) intermediates, reflecting different strengths in the Fe-O π interactions.
These differences are also observed in the intensities of the 1s-to-3d transitions in the XANES region,
which increase from 4 units for the nearly octahedral iron(II) precursor to 9-15 units for the alkyl-
peroxoiron(III) intermediates to 25-29 units for the oxoiron(IV) species.
Introduction
formation is available for two intermediates. A structure of the
purple alkylperoxo intermediate of soybean lipoxygenase-3
Many mononuclear nonheme iron enzymes are involved in
dioxygen metabolism. They activate dioxygen to carry out
metabolically important oxidative transformations and are
involved in the detoxification of superoxide. Their active sites
have become well characterized through spectroscopic inves-
tigations and the solution of crystal structures for not only the
as isolated resting state but also for binary, ternary, and
quaternary complexes as well.1-4 Catalytic mechanisms for these
enzymes invariably invoke metal-peroxo and/or high-valent
metal-oxo species during turnover. For example, peroxo inter-
mediates have been observed in solution in the chemistry of
the antitumor drug bleomycin,5 superoxide reductase,6-9 and
lipoxygenase,10 while an oxoiron(IV) intermediate has recently
been identified in the catalytic cycle of the 2-oxoglutarate-
dependent monoiron enzyme TauD.11-13 Crystallographic in-
derived from substrate oxidation revealed an end-on bound
alkylperoxo ligand to the iron(III) center,10 while a more recent
report on naphthalene dioxygenase in complex with substrate
and O2 showed a side-on dioxygen moiety coordinated to the
iron center.14
Model compounds can play an important role in establishing
the structural descriptions of enzymic intermediates and con-
tribute to our understanding of their spectroscopic and reactivity
properties. With synthetic nonheme iron complexes, it has been
possible to generate a number of peroxoiron(III) species
stabilized at low temperature.15,16 Since crystal structures for
none of these are available, X-ray absorption spectroscopy
(XAS) has become the experimental tool of choice to determine
intramolecular distances and to gain insight into the metal
environments of these intriguing intermediates. To date, this
approach has been applied to [FeIII(SMe2N4(tren))(OOH)]+,17,18
† University of Minnesota.
‡ Ewha Womans University.
[FeIII(N4Py)(OOH)]2+ 19 [FeIII(N4Py)(η2-O2)]+,19 and [FeIII-
,
(1) Schofield, C. J.; Zhang, Z. Curr. Opin. Struct. Biol. 1999, 9, 722-731.
(2) Solomon, E. I.; Brunold, T. C.; Davis, M. I.; Kemsley, J. N.; Lee, S.-K.;
Lehnert, N.; Neese, F.; Skulan, A. J.; Yang, Y.-S.; Zhou, J. Chem. ReV.
2000, 100, 235-349.
(PaPy3)(OOH)]+ 20 and sheds some light on how the strength
of the Fe-Operoxo bond is affected by differences in the
(3) Fitzpatrick, P. F. Biochemistry 2003, 42, 14083-14091.
(4) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, L., Jr. Chem. ReV. 2004,
104, 939-986.
(11) Price, J. C.; Barr, E. W.; Tirupati, B.; Bollinger, J. M., Jr.; Krebs, C.
Biochemistry 2003, 42, 7497-7508.
(5) Burger, R. M. Chem. ReV. 1998, 98, 1153-1169.
(6) Coulter, E. D.; Emerson, J. P.; Kurtz, D. M., Jr.; Cabelli, D. E. J. Am.
Chem. Soc. 2000, 122, 11555-11556.
(12) Price, J. C.; Barr, E. W.; Glass, T. E.; Krebs, C.; Bollinger, J. M., Jr. J.
Am. Chem. Soc. 2003, 125, 13008-13009.
(13) Proshlyakov, D. A.; Henshaw, T. F.; Monterosso, G. R.; Ryle, M. J.;
Hausinger, R. P. J. Am. Chem. Soc. 2004, 126, 1022-1023.
(14) Karlsson, A.; Parales, J. V.; Parales, R. E.; Gibson, D. T.; Eklund, H.;
Ramaswamy, S. Science 2003, 299, 1039-1042.
(7) Emerson, J. P.; Coulter, E. D.; Cabelli, D. E.; Phillips, R. S.; Kurtz, D. M.,
Jr. Biochemistry 2002, 41, 4348-4357.
(8) Mathe´, C.; Mattioli, T. A.; Horner, O.; Lombard, M.; Latour, J.-M.;
Fontecave, M.; Nivie`re, V. J. Am. Chem. Soc. 2002, 124, 4966-4967.
(9) Nivie`re, V.; Asso, M.; Weill, C. O.; Lombard, M.; Guigliarelli, B.;
Favaudon, V.; Houe´e-Levin, C. Biochemistry 2004, 43, 808-818.
(10) Skrzypczak-Jankun, E.; Bross, R. A.; Carroll, R. T.; Dunham, W. R.; Funk,
M. O., Jr. J. Am. Chem. Soc. 2001, 123, 10814-10820.
(15) Girerd, J.-J.; Banse, F.; Simaan, A. J. Struct. Bonding 2000, 97, 145-177.
(16) Rohde, J.-U.; Bukowski, M. R.; Que, L., Jr. Curr. Opin. Chem. Biol. 2003,
7, 674-682.
(17) Shearer, J.; Scarrow, R. C.; Kovacs, J. A. J. Am. Chem. Soc. 2002, 124,
11709-11717.
9
16750
J. AM. CHEM. SOC. 2004, 126, 16750-16761
10.1021/ja047667w CCC: $27.50 © 2004 American Chemical Society

XAS of Alkylperoxoiron(III) and Oxoiron(IV)
A R T I C L E S
and [Fe(Bn-TPEN)(OTf)](OTf),27 1c(OTf), were synthesized according
to published procedures. Caution: perchlorate salts of metal complexes
with organic ligands are potentially explosive and should be handled
with care! The preparation of [57Fe(TPA)(OTf)2] was carried out
analogously to the natural abundance compound using 57Fe(OTf)2‚2
MeCN. [57Fe(N4Py)(NCMe)](ClO4)2 was prepared as described previ-
ously.30
supporting ligand, the peroxo binding mode, and the metal spin
state. In addition, computational methods have been useful for
assessing likely structures and predicting the Fe-Operoxo bond
length.19,21-24
In the area of high-valent oxoiron intermediates, we have
recently reported the generation and isolation of the first
examples of mononuclear nonheme oxoiron(IV) species from
synthetic iron(II) precursors. Complexes with a terminal FeIVd
O unit have now been obtained with TMC, TPA, N4Py, and
Bn-TPEN as supporting ligands via reactions with stoichiometric
peracid or iodosylbenzene as oxygen donors and extensively
characterized by various spectroscopic techniques.18,25-27 Only
for the TMC complex has a high-resolution crystal structure
been obtained to demonstrate unequivocally that a terminal
FeIVdO unit can indeed exist in a nonporphyrin ligand environ-
ment.25
To augment the very limited amount of structural information
available for metastable peroxoiron(III) and oxoiron(IV) inter-
mediates, we have conducted a detailed X-ray absorption
spectroscopic study of two low-spin [FeIII(L)(OOtBu)]2+ (L )
TPA, N4Py) and four low-spin [FeIV(O)(L)]2+ complexes (L
) TPA, N4Py, Bn-TPEN). The results reported in this paper
represent the most extensive structural examination thus far of
a related series of intermediates relevant to the catalytic cycles
of mononuclear nonheme iron enzymes involved in dioxygen
metabolism.
[Fe(N4Py)(NCMe)](OTf)2, 1b(OTf)2. 290 mg (0.38 mmol) N4Py‚
31,32
4 HClO4
were dissolved in 10 mL distilled H2O and 10 mL CH2-
Cl2. After neutralizing the aqueous phase with 5N NaOH, the organic
phase was separated, washed with brine and dried over Na2SO4. The
solvent was removed in vacuo yielding the free base ligand N4Py as a
pink oil which was immediately used for the next step. Fe(OTf)2‚2
MeCN33 (179 mg, 0.41 mmol) was then added to a solution of N4Py
(140 mg, 0.38 mmol) in 5 mL tetrahydrofuran. After stirring for 14 h,
hexane was added to afford a polycrystalline powder, which was then
recrystallized from CH2Cl2/Et2O. This compound was identified by its
crystal structure (see Supporting Information) and by comparison of
1
its H NMR spectrum with that of 1b(ClO4)2.
XAS Sample Preparation. A solid sample of 1b(ClO4)2 (2 mg in
98 mg boron nitride) was prepared by grinding the two components
into a homogeneous mixture under an inert gas atmosphere. 20 mg of
the solid mixture were then packed into a sample plate (1 mm thick)
and covered with Mylar tape.
Samples of the alkylperoxoiron(III) intermediates 2a and 2b were
generated by addition of tert-BuO2H to pre-cooled solutions of the
respective iron(II) precursors. After a suitable reaction time during
which the chromophore was observed to maximize by UV-visible
spectroscopy, ca. 0.5 mL of the solution was transferred into a pre-
cooled sample holder, covered with Mylar tape, which was then
submerged in liquid nitrogen. EPR samples of 2a and 2b were prepared
simultaneously from the same batches to assess what fraction of the
XAS samples the intermediates represent.
[FeIII(TPA)(OOtBu)(solv)]2+, 2a. 1a(ClO4)2, 10 mM in MeCN, -40
°C, + 10 equiv tert-BuO2H (70% aqueous solution), reaction time 20
min; 65% by EPR double integration.
[FeIII(N4Py)(OOtBu)]2+, 2b. 1b(OTf)2, 5 mM in MeOH, -70 °C,
+ 10 equiv tert-BuO2H (5 M in nonane), reaction time 12 h; 95% by
EPR double integration.
Samples of the mononuclear oxoiron(IV) intermediates 3a-3c were
generated by the addition of the appropriate oxidant to a pre-cooled
solution of the appropriate iron(II) precursor complex at the appropriate
temperature as described below. Formation of the intermediate was
followed by UV-visible spectroscopy. After 10 min ca. 0.5 mL of the
solution with 3a, 3b, or 3c was transferred into a pre-cooled tandem
Mo¨ssbauer/XAS cup,34 covered with Mylar tape, which was then
submerged in liquid nitrogen.
[FeIV(O)(TPA)(solv)]2+, 3a. To a solution of [Fe(TPA)(OTf)2] (0.018
mmol Fe, 33% 57Fe) in 3 mL MeCN and 10 µL H2O (doubly deionized,
Milli-Q Water System, Millipore) in a UV-visible cuvette and pre-
cooled to -40 °C was added a solution of 1.5 equiv AcO2H (0.027
mmol, 32 wt. % AcO2H) in 100 µL MeCN. (Note: Dissolution of [Fe-
(TPA)(OTf)2] in MeCN readily afforded the low-spin [Fe(TPA)-
(NCMe)2]2+ ion (1a), whose NMR spectrum was indistinguishable from
that of 1a(ClO4)2.) Mo¨ssbauer analysis of this sample showed 80(5) %
formation of 3a.
Experimental Section
Materials and General Procedures. All reagents and solvents were
purchased from commercial sources and were used as received, unless
noted otherwise. Solvents were dried according to published procedures
and distilled under Ar prior to use.28 Preparation and handling of air-
sensitive materials were carried out under an inert atmosphere by using
either standard Schlenk and vacuum line techniques or a glovebox.
Elemental analyses were performed by Atlantic Microlab, Inc., Nor-
cross, GA.
Preparation of Iron(II) Precursors. [Fe(TPA)(NCMe)2](ClO4)2,29
1a(ClO4)2, [Fe(TPA)(OTf)2],26 [Fe(N4Py)(NCMe)](ClO4)2,30 1b(ClO4)2,
(18) Abbreviations used: Bn-TPEN, N-benzyl-N, N′, N′-tris(2-pyridylmethyl)-
1,2-diaminoethane; HSMe2N4(tren), 3-[N-[2-[N, N-bis(2-aminoethyl)amino]-
ethyl]imino]-2-methyl-2-butanethiol; N4Py, N, N-bis(2-pyridylmethyl)-N-
bis(2-pyridyl)methylamine; OTf, trifluoromethylsulfonate or triflate anion;
PaPy3H, N-[2-(N, N-bis(2-pyridylmethyl)amino)ethyl]pyridine-2-carboxa-
mide; pyO, pyridine N-oxide; TMC, 1, 4, 8, 11-tetramethyl-1, 4, 8, 11-
tetraazacyclotetradecane or tetra(N-methyl)cyclam; TPA, N,N,N-tris(2-
pyridylmethyl)amine; EXAFS, extended X-ray absorption fine structure;
XANES, X-ray absorption near-edge structure; XAS, X-ray absorption
spectroscopy.
(19) Roelfes, G.; Vrajmasu, V.; Chen, K.; Ho, R. Y. N.; Rohde, J.-U.; Zondervan,
C.; la Crois, R. M.; Schudde, E. P.; Lutz, M.; Spek, A. L.; Hage, R.; Feringa,
B. L.; Mu¨nck, E.; Que, L., Jr. Inorg. Chem. 2003, 42, 2639-2653.
(20) Bukowski, M. R.; Zhu, S.; Koehntop, K. D.; Brennessel, W. W.; Que, L.,
Jr. J. Biol. Inorg. Chem. 2004, 9, 39-48.
(21) Neese, F.; Solomon, E. I. J. Am. Chem. Soc. 1998, 120, 12829-12848.
(22) Lehnert, N.; Ho, R. Y. N.; Que, L., Jr.; Solomon, E. I. J. Am. Chem. Soc.
2001, 123, 8271-8290.
(23) Lehnert, N.; Ho, R. Y. N.; Que, L., Jr.; Solomon, E. I. J. Am. Chem. Soc.
2001, 123, 12802-12816.
(24) Lehnert, N.; Neese, F.; Ho, R. Y. N.; Que, L., Jr.; Solomon, E. I. J. Am.
Chem. Soc. 2002, 124, 10810-10822.
(25) Rohde, J.-U.; In, J.-H.; Lim, M. H.; Brennessel, W. W.; Bukowski, M. R.;
Stubna, A.; Mu¨nck, E.; Nam, W.; Que, L., Jr. Science 2003, 299, 1037-
1039.
(30) Roelfes, G.; Lubben, M.; Chen, K.; Ho, R. Y. N.; Meetsma, A.;
Genseberger, S.; Hermant, R. M.; Hage, R.; Mandal, S. K.; Young, V. G.,
Jr.; Zang, Y.; Kooijman, H.; Spek, A. L.; Que, L., Jr.; Feringa, B. L. Inorg.
Chem. 1999, 38, 1929-1936.
(26) Lim, M. H.; Rohde, J.-U.; Stubna, A.; Bukowski, M. R.; Costas, M.; Ho,
R. Y. N.; Mu¨nck, E.; Nam, W.; Que, L., Jr. Proc. Natl. Acad. Sci. U.S.A.
2003, 100, 3665-3670.
(31) Lubben, M.; Meetsma, A.; Wilkinson, E. C.; Feringa, B.; Que, L., Jr. Angew.
Chem., Int. Ed. Engl. 1995, 34, 1512-1514.
(27) Kaizer, J.; Klinker, E. J.; Oh, N. Y.; Rohde, J.-U.; Song, W. J.; Stubna,
A.; Kim, J.; Mu¨nck, E.; Nam, W.; Que, L., Jr. J. Am. Chem. Soc. 2004,
126, 472-473.
(32) Roelfes, G.; Lubben, M.; Leppard, S. W.; Schudde, E. P.; Hermant, R.
M.; Hage, R.; Wilkinson, E. C.; Que, L., Jr.; Feringa, B. L. J. Mol. Catal.
A 1997, 117, 223-227.
(28) Armarego, W. L. F.; Perrin, D. D. Purification of Laboratory Chemicals;
Butterworth-Heinemann: Oxford, 1997.
(33) Hagen, K. S. Inorg. Chem. 2000, 39, 5867-5869.
(29) Zang, Y.; Kim, J.; Dong, Y.; Wilkinson, E. C.; Appelman, E. H.; Que, L.,
Jr. J. Am. Chem. Soc. 1997, 119, 4197-4205.
(34) Riggs-Gelasco, P. J.; Shu, L.; Chen, S.; Burdi, D.; Huynh, B. H.; Que, L.,
Jr.; Stubbe, J. J. Am. Chem. Soc. 1998, 120, 849-860.
9
J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004 16751

A R T I C L E S
Rohde et al.
[FeIV(O)(TPA)(pyO)]2+, 3a′. To a solution of [Fe(TPA)(OTf)2] (0.03
mmol Fe, 33% 57Fe) in 5 mL MeCN and pre-cooled to -40 °C was
added a solution of 1.0 equiv AcO2H (0.03 mmol, 32 wt. % AcO2H)
in 50 µL MeCN. After a reaction time of 5 min, 5 equiv pyO were
added (sample transfer after 2 min). Mo¨ssbauer analysis showed a
sample with 60(5) % 3a′.
[FeIV(O)(N4Py)]2+, 3b. To a solution of [Fe(N4Py)(NCMe)](ClO4)2,
1b(ClO4)2, (0.027 mmol Fe, 33% 57Fe) in 4.5 mL MeCN in a UV-
visible cuvette at 25 °C were added 6 equiv AcO2H (∼0.17 mmol, 32
wt. % AcO2H). Mo¨ssbauer analysis showed a sample with 80(5) %
3b.
[FeIV(O)(Bn-TPEN)]2+, 3c. To a solution of [Fe(Bn-TPEN)(OTf)]-
(OTf), 1c(OTf), (0.05 mmol Fe) in 2 mL MeCN at 25 °C was added
excess solid PhIO. The resulting mixture was stirred for 30 min and
filtered. Its absorbance (λmax ) 740 nm) showed 85(5) % formation of
3c.
Physical Methods. UV-visible spectra were recorded on an HP
8453A diode array spectrometer with samples maintained at low
temperature using a cryostat from Unisoku Scientific Instruments, Japan.
1H NMR spectra were recorded on a Varian Inova VXR-300 or Varian
Inova VI-300 spectrometer at ambient temperature. Chemical shifts
(ppm) were referenced to the residual solvent peaks. EPR spectra were
recorded at X-band microwave frequency on a Bruker E-500
spectrometer equipped with an Oxford liquid He cryostat. Spin
quantification of the S ) 1/2 FeIII signals was performed by double
integration of the EPR first derivative signal recorded under nonsat-
urating conditions at 20 K for comparison with a CuII(EDTA) solution
as a standard.
X-ray Absorption Spectroscopy. Data Collection. XAS data were
collected at the Stanford Synchrotron Radiation Laboratory (SSRL) of
the Stanford Linear Accelerator Center, beamline 7-3, and at the
National Synchrotron Light Source (NSLS) of the Brookhaven National
Laboratory, beamline X9B. Fe K-edge X-ray absorption spectra were
recorded on frozen solutions at 5-15 K over the energy range 6.9-
8.0 keV as described previously.35,36 Storage ring conditions: 3 GeV,
50-100 mA (SSRL), and 2.8 GeV, 100-300 mA (NSLS). The beam
vertical aperture of the pre-monochromator slits was 1 mm. Contamina-
tion of higher harmonics radiation was minimized by detuning the Si-
(220) double-crystal monochromator by 50% at ∼8 keV (beamline 7-3
at SSRL) and by a harmonic rejection mirror (beamline X9B at NSLS),
respectively. The horizontal spot size on the sample was 4-5 mm in
most cases. Spectra were measured with 10 eV steps below the edge,
0.3 eV steps in the edge region, and steps equivalent to 0.05 Å-1
increments above the edge (region borders were 6880, 7090, and 7140
eV at beamline 7-3, SSRL, and 6932, 7102, and 7137 eV at beamline
X9B, NSLS). An iron foil spectrum was recorded concomitantly for
internal energy calibration and the first inflection point of the K-edge
was assigned to 7112.0 eV. The data were obtained as fluorescence
excitation spectra (Aexp (Cf/C0)) using a solid-state germanium detector
(Canberra).
The Fe concentration in the samples and the number of scans
acquired for each sample were as follows: 1b, 70 mM Fe, 2 scans
(beamline X9B at NSLS, 13 element detector); 2a, 10 mM Fe, 2
independent samples, 13 scans, 14 scans (X9B at NSLS, 13 element
detector); 2b, 5 mM Fe, 7 scans (7-3 at SSRL, 30 element detector);
3a, 6 mM Fe, 2 independent samples, 10 scans (X9B at NSLS, 13
element detector), 10 scans (7-3 at SSRL, 30 element detector); 3a′,
6 mM Fe, 10 scans (7-3 at SSRL, 30 element detector); 3b, 6 mM
Fe, 10 scans (X9B at NSLS, 13 element detector); 3c, 25 mM Fe, 3
scans (X9B at NSLS, 13 element detector).
at SSRL were calibrated and averaged with the program EXAFSPAK.39
A background was then removed by fitting a Gaussian to the preedge
region, extrapolating and subtracting this from the entire spectrum. A
three-segment spline with fourth-order polynomial components was fit
to the EXAFS region. The resulting EXAFS data in k space were
Gaussian smoothed. Curve-fitting of XANES data to obtain preedge
peak areas and fitting of EXAFS data were carried out with the program
SSExafs according to a previously described protocol.35,40 XAS data
from beamline X9B at NSLS were treated entirely with SSExafs. Here,
baseline subtraction and determination of the edge height were
accomplished by fitting a cubic spline function (equally spaced
throughout the EXAFS region) simultaneously with edge and EXAFS
parameters (including first-shell parameters), and a correction of
fluorescence data for thickness effects. We have made comparisons
using both software packages EXAFSPAK and SSExafs for calibration
and background subtraction; the different protocols, however, did not
lead to any systematic errors and the results were identical within the
error of analysis.
The edge was modeled as an integral of a 75% Gaussian and 25%
Lorentzian peak. The heights, positions and widths (at half-height) of
preedge peaks were refined using a Gaussian function. Refinements
with multiple peaks were constrained to have a common width for all
peaks. Preedge peak areas are in % of Fe K-edge height (eV) and were
multiplied by 100.
The EXAFS refinements reported were on k3ø(k) data and the
function minimized was R ) {Σ k6(øc - ø)2/N}1/2, where the sum is
over N data points within the selected k space. Single-scattering EXAFS
theory allows the total EXAFS spectrum to be described as the sum of
shells of separately modeled atoms, e.g., øc ) Σ nA[f(k)k-1r-2 exp(-
2σ2k2) sin(2kr + R(k))], where n is the number of atoms in the shell,
k ) [8π2me (E - E0 + ∆E)/h2]1/2, and σ2 is the Debye-Waller factor.35
The amplitude reduction factor (A) and the shell-specific edge shift
(∆E) are empirical parameters that partially compensate for imperfec-
tions in the theoretical amplitude and phase functions.41 Phase and
amplitude functions were theoretically calculated using a curved-wave
formalism.42 A variation of FABM (fine adjustment based on models)
was used here in the analysis procedure with theoretical phase and
amplitude functions.43 In each shell, two parameters were refined at
one time (r and n or σ2), while A and ∆E values were determined by
using a series of crystallographically characterized model complexes.
The fitting results indicate the average metal-ligand distances, the type
and the number of scatterers, and the Debye-Waller factors which
can be used to evaluate the distribution of Fe-ligand bond lengths in
each shell. The EXAFS goodness of fit criterion applied here is
ꢀ2 ) [(Nidp/ν) ‚ Σ (øc - ø)2/σdata2]/N
as recommended by the International Committee on Standards and
Criteria in EXAFS44,45 where ν is the number of degrees of freedom
calculated as ν ) Nidp - Nvar, Nidp is the number of independent data
points, Nvar is the number of variables that are refined, and σdata is the
estimated uncertainty of the data (usually set at 1). Nidp is calculated as
(37) Scott, R. A. Methods Enzymol. 1985, 117, 414-459.
(38) Teo, B.-K.; Joy, D. C. EXAFS Spectroscopy: Techniques and Applications;
Plenum Press: New York, 1981.
(39) George, G. N.; Pickering, I. J.; Stanford Synchrotron Radiation Laboratory,
Stanford Linear Accelerator Center, 2000.
(40) Scarrow, R. C.; Trimitsis, M. G.; Buck, C. P.; Grove, G. N.; Cowling, R.
A.; Nelson, M. J. Biochemistry 1994, 33, 15023-15035.
(41) Teo, B.-K.; Lee, P. A. J. Am. Chem. Soc. 1979, 101, 2815-2832.
(42) McKale, A. G.; Veal, B. W.; Paulikas, A. P.; Chan, S.-K.; Knapp, G. S. J.
Am. Chem. Soc. 1988, 110, 3763-3768.
(43) Teo, B.-K.; Antonio, M. R.; Averill, B. A. J. Am. Chem. Soc. 1983, 105,
3751-3762.
(44) Bunker, G. A.; Hasnain, S. S.; Sayers, D. E. In X-ray Absorption Fine
Structure; Hasnain, S. S., Ed.; Ellis Horwood: New York, 1991; pp 751-
770.
(45) Riggs-Gelasco, P. J.; Stemmler, T. L.; Penner-Hahn, J. E. Coord. Chem.
ReV. 1995, 144, 245-286.
Data Analysis. The treatment of raw EXAFS data to yield ø(k) is
discussed in detail in review articles.37,38 XAS data from beamline 7-3
(35) Scarrow, R. C.; Maroney, M. J.; Palmer, S. M.; Que, L., Jr.; Roe, A. L.;
Salowe, S. P.; Stubbe, J. J. Am. Chem. Soc. 1987, 109, 7857-7864.
(36) Shu, L.; Chiou, Y.-M.; Orville, A. M.; Miller, M. A.; Lipscomb, J. D.;
Que, L., Jr. Biochemistry 1995, 34, 6649-6659.
9
16752 J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004

XAS of Alkylperoxoiron(III) and Oxoiron(IV)
A R T I C L E S
Nidp ) 2∆k∆R/π + 2.46 The use of ꢀ2 as the criterion for the goodness
of fit allows us to compare fits using different numbers of variable
parameters.
Fitting Procedure. Fitting parameters A (amplitude reduction factor)
and ∆E (phase shift) for inner-sphere N/O and outer-sphere C scatterers
had been extracted from Fe(acac)3 (Fe-O, 1.986-2.004 Å; Fe‚‚‚C,
2.935-2.971 Å).35 Since backscatterers differing in Z by 1 unit cannot
be distinguished by EXAFS, N and O scatterers were fit using the same
parameters A, ∆E, and Z ) 7 (assigned as N/O shell). However, Z )
8 was applied for the short-distance scatterer in 3a, 3a′, 3b, and 3c;
the shell was denoted as O/N. Initially, inner-sphere shells were refined
using the corresponding back-transformation range “b” (see Tables 3
and 4). Two of the parameters n, r, and ∆σ2 were allowed to float at
the same time, while the third parameter had been kept fixed.
Subsequently, these shells were refined across the broader window “c”
and outer-sphere shells were included. The resolution was calculated
from ∆r ≈ π (2 ∆k)-1 45
. The typical error of analysis for interatomic
distances (r) is approximately (0.02 Å.37,47
Results
In this study, we have obtained Fe K-edge X-ray absorption
spectra of seven low-spin iron complexes with oxidation states
+II to +IV to gain structural insight into the changes in the
iron coordination sphere as the metal center becomes oxidized
concomitant with the introduction of peroxo or oxo ligands.
Complexes of three supporting polydentate ligands have been
investigated. Our detailed analyses are presented below.
I. Iron(II) Precursor Complexes. The neutral nitrogen
ligands, tetradentate TPA and pentadentate N4Py and Bn-TPEN,
represent a series of ligands with differing ratios of pyridine
and amine functions that allow us to fine-tune the properties
and stabilities of transient iron complexes. The precursors to
such intermediates are six-coordinate iron(II) complexes of these
ligands that have, depending on the denticity of the supporting
ligand, one or two labile coordination sites: [FeII(TPA)-
(NCMe)2]2+, 1a, [FeII(N4Py)(NCMe)]2+, 1b, and [FeII(Bn-
TPEN)(OTf)]+, 1c (Figure 1). The Fe-N bond lengths in the
crystal structures of [FeII(TPA)(NCMe)2](BPh4)2,29 1a(BPh4)2,
[FeII(N4Py)(NCMe)](ClO4)2,30 1b(ClO4)2, and [FeII(N4Py)-
(NCMe)](OTf)2, 1b(OTf)2 (see Supporting Information) lie in
the range of 1.915-1.989 Å (Table S1), typical for low-spin
iron(II) complexes. On the other hand, [FeII(Bn-TPEN)(OTf)]-
(OTf), 1c(OTf), is high spin and the Fe-N bonds are longer,
2.032-2.236 Å (see Supporting Information, Table S1).
We have obtained XAS data of 1b(ClO4)2 for comparison
purposes to take advantage of the fact that its crystal structure
shows a nearly octahedral FeN6 environment with very similar
Fe-N bond lengths. The EXAFS spectrum can be best fit with
5 nitrogen scatterers at an Fe-N distance of 1.96 Å (Table S2,
Figure S4). This distance represents the average of the six Fe-N
distances. The underestimated n value and the resulting Debye-
Waller factor arise from reduction in the scattering amplitude
in the EXAFS spectrum, reflecting the static and/or dynamic
disorder inherent in the first coordination sphere; i.e., the Fe-
Figure 1. Molecular structures of complex cations in [FeII(TPA)(NCMe)2]-
(BPh4)2, 1a(BPh4)2 (top left), [FeII(N4Py)(NCMe)](ClO4)2‚MeOH, 1b(ClO4)2‚
MeOH (top right), and [FeII(Bn-TPEN)(OTf)](OTf), 1c(OTf) (bottom); color
key: pink ) Fe, blue ) N, gray ) C, red ) O, yellow ) sulfur, green )
fluorine.
Scheme 1. Generation and Interconversion of Alkylperoxoiron(III)
Complexes (2) and Oxoiron(IV) Complexes (3)a
a The 1 f 3 conversion was realized for 1a, in MeCN, -40 °C; for 1b,
in MeCN, 25 °C; for 1c, in MeCN, -20 °C; 1 f 2 for 1a, in MeCN, -40
°C; for 1b, in CH2Cl2 or MeOH, -70 °C; 2 f 3 for 2a, in MeCN, -25
°C.
react with tert-BuO2H at low temperature to form metastable
alkylperoxoiron(III) complexes [FeIII(TPA)(OOtBu)(solv)]2+ 29
,
2a, and [FeIII(N4Py)(OOtBu)]2+, 2b (Scheme 1). These inter-
mediates have different thermal stabilities. While 2a is readily
observable in MeCN at -40 °C, 2b could only be obtained at
temperatures below -60 °C. To do so, 1b(OTf)2 was synthe-
sized to enhance the solubility of this complex in solvents with
lower-temperature melting points than MeCN. As shown in
Figure 2, addition of tert-BuO2H at -70 °C to a solution of
1b(OTf)2 in CH2Cl2 results in the disappearance of absorption
features characteristic of 1b and the concomitant isosbestic
formation of an intense new chromophore 2b with λmax at 560
nm (ꢀ ) 2400 M-1cm-1); similar results are obtained in MeOH,
NNCMe distance of 1.915 Å in the crystal structure is shorter by
ca. 0.05 Å than the average Fe-NN4Py distance. On the other
hand, these distances are too close to be resolved by EXAFS
into two individual shells (Table S2).
II. Formation and Properties of Intermediates. A. Alkyl-
peroxoiron(III) Intermediates. Iron(II) precursors 1a and 1b
(46) Stern, E. A. Phys. ReV. 1993, B48, 9825-9827.
(47) Penner-Hahn, J. E. Coord. Chem. ReV. 1999, 190-192, 1101-1123.
9
J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004 16753

A R T I C L E S
Rohde et al.
Figure 2. Conversion of 1 mM 1b (s) to 2b (s) in CH2Cl2 at -70 °C by
addition of 10 equiv tert-BuO2H, as monitored by UV-visible spectroscopy.
Table 1. Optical and Mo¨ssbauer Properties of
[FeIV(O)(TMC)(NCMe)]2+ 25
3a, 3a′, 3b, and 3c
,
λ
max (nm)
δ
∆
(mm
EQ
1
1
s-
1
(ꢀ
(M-
‚
cm-1))
(mm
‚
)
‚
s-
)
[FeIV(O)(TMC)(NCMe)]2+
820 (400)
724 (300)
742 (300)
695 (400)
740 (400)
0.17
0.01
0.04
-0.04
0.01
1.24
0.92
0.90
0.93
0.87
3a
3a′
3b
3c
Figure 3. Fe K-edge X-ray absorption near-edge structures (XANES,
fluorescence excitation) of iron(III) complexes [FeIII(TPA)(OOtBu)(solv)]2+
,
2a (s), and [FeIII(N4Py)(OOtBu)]2+, 2b (- - -); and iron(IV) complexes
[FeIV(O)(TPA)(solv)]2+, 3a (red s), [FeIV(O)(N4Py)]2+, 3b (red - - -), and
[FeIV(O)(Bn-TPEN)]2+, 3c (red ‚‚‚).
which is more suitable for XAS studies. Like 2a, 2b also exhibits
an EPR spectrum with g values at 2.16, 2.10 and 1.97, indicative
of a low-spin iron(III) center. As previously established for 2a
(λmax at 600 nm; ꢀ ) 2200 M-1cm-1), low-spin alkylperoxoiron-
(III) complexes have deep blue colors that arise from an
alkylperoxo-to-iron(III) charge-transfer transition.22,29 The dif-
ferent thermal stabilities of 2a and 2b may relate to the
differences in the sixth ligand. The sixth ligand in 2b is
presumably the additional pyridine in the N4Py ligand frame-
work relative to TPA, while that for 2a is most likely MeCN.
We have previously demonstrated that coordination of pyridine
as the sixth ligand to 2a significantly diminishes its thermal
stability at -40 °C.48
B. Oxoiron(IV) Intermediates. The oxoiron(IV) intermedi-
ates 3a, 3b, and 3c can be obtained from the reaction of AcO2H
with their iron(II) precursors 1a, 1b, and 1c.26,27 These
intermediates are pale green species with absorption maxima
in the near-IR region (Table 1) whose formulations as [FeIV-
(O)(L)]2+ ions have been established by electrospray ionization
mass spectrometry. They are EPR silent and exhibit Mo¨ssbauer
properties that compare well with those reported for crystallo-
graphically characterized [FeIV(O)(TMC)(NCMe)](OTf)2 and are
characteristic of a mononuclear S ) 1 iron(IV) center in a
nitrogen-rich ligand environment.25 Because of their different
thermal stabilities, 3a-3c require different temperatures for their
generation procedure: -40 °C for 3a, 25 °C for 3b, and -20
°C for 3c. The latter two can also be formed conveniently by
the reaction of 1b or 1c with excess solid PhIO at 25 °C in
MeCN. In fact, the XAS sample of 3c for this study was
obtained by the use of the latter reagent. In addition, pyridine
N-oxide was added to a solution of 3a in MeCN to afford 3a′
with the MeCN replaced by pyridine N-oxide as the sixth ligand.
The coordination of pyridine N-oxide to the oxoiron(IV) center
is indicated by the red shift of its near-IR absorption feature to
742 nm. This collection of novel intermediates is the focus of
this XAS structural study.
III. X-ray Absorption Near-Edge Structures. A. Alkyl-
peroxoiron(III) Intermediates. X-ray absorption spectra at the
Fe K-edge were measured for two alkylperoxoiron(III) com-
plexes with different supporting ligands. The near-edge struc-
tures of [FeIII(TPA)(OOtBu)(solv)]2+, 2a, and [FeIII(N4Py)-
(OOtBu)]2+, 2b, are shown in Figure 3. The position of the rising
K-edge, where a 1s electron is promoted to the continuum,
depends on the effective charge of the metal center and is thus
related to the oxidation state, but is also influenced by the ligand
environment of the metal center. The Fe K-edges of 2a and 2b
exhibit similar shapes and lie at similar energies (based on the
first inflection point, Table 2). Relative to 1b, their edges upshift
by ca. 1.5 eV, consistent with their higher oxidation state.
The preedge regions of 2a and 2b magnified in the inset of
Figure 3 consist of 1sf3d transitions, as observed for many
iron(III) complexes,49-51 and can be fit with a minimum of two
features with energies and peak areas listed in Table 2. The
major component is a prominent peak at ca. 7113.5 eV that is
(48) Kaizer, J.; Costas, M.; Que, L., Jr. Angew. Chem., Int. Ed. 2003, 42, 3671-
3673.
9
16754 J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004

XAS of Alkylperoxoiron(III) and Oxoiron(IV)
A R T I C L E S
Table 2. XAS Preedge Peak Energies and Intensities for the Iron
Complexes in This Studya
complex
E0 (eV)
Epreedge (eV)
area
1b
2a
7122.0
7123.6
7112.9
7113.4
7115.9
7113.5
7116.0
7114.4
7114.6
7114.3
7114.1
3.9(3)
7.2(3)
1.7(2)
9.7(3)
5.3(3)
25.4(3)
24.5(4)
25.2(5)
29.3(4)
2b
7123.4
3a
3a′
3b
3c
7124.5
7124.8
7123.7
7123.7
a E0 is conventionally defined as the first inflection point of the edge
and used as the reference point for the k-space spectrum. This choice is
somewhat arbitrary, but does not significantly affect metal-ligand distances.
Preedge peak energies and intensities were determined by curve fitting to
the data as described.40,50 Intensities are reported as peak areas (normalized
to the edge jump) and were multiplied by 100.
readily observed for both complexes. The minor component
found at ca. 7116 eV is easily seen only for 2b in Figure 3
inset, while the less intense minor component of 2a can be
discerned in the background-subtracted spectrum shown in
Figure S8 (Supporting Information). Intermediates 2a and 2b
exhibit total preedge areas of 9(1) and 15(1) units, respectively,
values higher than those found for centrosymmetric six-
coordinate iron complexes.49-51 The higher intensity can be
ascribed to the expected lower symmetry of the iron sites in
the alkylperoxo complexes due to the presence of the unique
Fe-OOtBu bond, which should be much shorter than the Fe-N
bonds (2.0 Å). The distortion from centrosymmetry increases
the mixing of 3dz2 and 4pz orbitals (z axis defined as the direction
of the Fe-OOtBu bond) and thus gives rise to more intense
preedge features.
Figure 4. Fourier transforms of the Fe K-edge EXAFS data (k3ø(k)) and
Fourier-filtered EXAFS spectra (k3ø′(k), inset) of [FeIII(TPA)(OOtBu)-
(solv)]2+, 2a, Fourier transformed range k ) 2-13.9 Å-1; experimental
data (‚‚‚) and fit 1 N/O, 4 N/O, 6 C, (s); Back-transformation range: r′ )
0.60-3.40 Å (top); and of [FeIII(N4Py)(OOtBu)]2+, 2b, Fourier transformed
range k ) 2-15 Å-1; experimental data (‚‚‚) and fit 1 N/O, 5 N/O, 3 C,
(s); Back-transformation range: r′ ) 0.60-3.10 Å (bottom).
upshift of the K-edges in the FeIVdO complexes relative to that
of the FeIII-OOR intermediates.
IV. EXAFS Spectra and Fitting Results. A. Alkylperoxo-
iron(III) Intermediates. The Fourier transforms (r′ space) of
the Fe K-edge EXAFS data of alkylperoxoiron(III) complexes
2a and 2b exhibit prominent features centered at r′ ) 1.6, 2.0,
and 2.3 Å with a shoulder at 1.2 Å (Figure 4), where r′
corresponds to the actual metal-scatterer distance r after a phase
shift correction of approximately 0.4 Å (r ≈ r′ + 0.4 Å). The
first coordination spheres of the low-spin iron(III) centers of
these complexes can be fit well by two shells of N/O atoms at
absorber-scatterer distances (r) of 1.78 and 1.96-1.97 Å. The
shorter-distance shell corresponds to the oxygen scatterer of the
Fe-OOtBu moiety. This short Fe-O distance undoubtedly
arises from π interactions that stengthen the Fe-O bond,22 as
also found in the crystal structure of [FeIII(N4Py)(OMe)]2+ 30
B. Oxoiron(IV) Intermediates. The near-edge structures of
the Fe K-edge X-ray absorption spectra (XANES) of oxoiron-
(IV) complexes [FeIV(O)(TPA)(solv)]2+, 3a, [FeIV(O)(N4Py)]2+
,
3b, and [FeIV(O)(Bn-TPEN)]2+, 3c, are shown in Figure 3. The
K edges of 3a, [FeIV(O)(TPA)(pyO)]2+ (3a′), 3b and 3c are
shifted to higher energies as compared to those of corresponding
iron(III) complexes 2a and 2b, due to the increase in oxidation
state, but the effect is small. More importantly, 3a, 3a′, 3b, and
3c exhibit intense 1sf3d preedge transitions with areas of 25-
30 units (Table 2).52 Such high values have been observed for
crystallographically characterized [FeIV(O)(TMC)(NCMe)]-
(OTf)2 (30(4) units),26 and for oxoiron(IV) porphyrin complexes
(27-38 units).53,54 The occurrence of an intense 1sf3d preedge
transition indicates significant mixing of 3d and 4p orbitals. This
may be caused by strong FedO π bonding that leads to the
large deviation from centrosymmetry of the iron ligand field
imposed by a terminal oxo ligand. The strong covalency of the
FedO bond55 may also provide the rationale for the rather small
and the EXAFS analysis for [FeIII(N4Py)(OOH)]2+ 19
.
The shell at ca. 2.0 Å with 4 or 5 scatterers represents the
donor nitrogen atoms of the polydentate ligands TPA and N4Py
(Table 3). This distance is comparable to the average Fe-N
bond lengths in low-spin FeIII complexes of pyridine and/or
tertiary amine ligands.29,56-59 For 2a, good fits are achieved with
either 4 or 5 N/O atoms at 1.96 Å; the fit with 4 N/O atoms
affords a lower goodness-of-fit (GOF) value (Table 3, fits 3
and 4).60 Besides the ligating nitrogens of the tetradentate TPA
ligand, the donor atom of a sixth ligand, possibly a solvent
molecule MeCN, may contribute to this shell, as the low-spin
(49) Roe, A. L.; Schneider, D. J.; Mayer, R. J.; Pyrz, J. W.; Widom, J.; Que,
L., Jr. J. Am. Chem. Soc. 1984, 106, 1676-1681.
(50) Randall, C. R.; Shu, L.; Chiou, Y.-M.; Hagen, K. S.; Ito, M.; Kitajima, N.;
Lachicotte, R. J.; Zang, Y.; Que, L., Jr. Inorg. Chem. 1995, 34, 1036-
1039.
(51) Westre, T. E.; Kennepohl, P.; DeWitt, J. G.; Hedman, B.; Hodgson, K. O.;
Solomon, E. I. J. Am. Chem. Soc. 1997, 119, 6297-6314.
(52) We first reported 3a to have a preedge peak area of 30 units but have
revised this value downwards upon further study.
(53) Wolter, T.; Meyer-Klaucke, W.; Mu¨ther, M.; Mandon, D.; Winkler, H.;
Trautwein, A. X.; Weiss, R. J. Inorg. Biochem. 2000, 78, 117-122.
(54) Nam, W.; Choi, S. K.; Lim, M. H.; Rohde, J.-U.; Kim, I.; Kim, J.; Kim,
C.; Que, L., Jr. Angew. Chem., Int. Ed. 2003, 42, 109-111.
(55) Decker, A.; Rohde, J.-U.; Que, L., Jr.; Solomon, E. I. J. Am. Chem. Soc.
2004, 126, 5378-5379.
(56) Baker, J.; Engelhardt, L. M.; Figgis, B. N.; White, A. H. J. Chem. Soc.,
Dalton Trans. 1975, 530-534.
(57) Figgis, B. N.; Skelton, B. W.; White, A. H. Aust. J. Chem. 1978, 31, 57-
64.
(58) Marsh, R. E. Acta Crystallogr. 1987, B43, 174-178.
(59) Duelund, L.; Hazell, R.; McKenzie, C. J.; Preuss Nielsen, L.; Toftlund, H.
J. Chem. Soc., Dalton Trans. 2001, 152-156.
9
J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004 16755

A R T I C L E S
Rohde et al.
Table 3. EXAFS Fitting Results for [FeIII(TPA)(OOtBu)(solv)]2+
2a, and [FeIII(N4Py)(OOtBu)]2+, 2ba
,
significantly decreasing the GOF value and affording reasonable
Debye-Waller factors (∆σ2, Table 3, fit 7). Presumably, the
smaller spread of Fe‚‚‚C distances in 2a, as evidenced in the
crystal structure of its low-spin iron(II) precursor, 1a(BPh4)2,29
allows the backscattering contributions from this multi-carbon
shell to combine constructively to afford the more intense outer-
sphere features. Thus, like that of 2b, the spectrum of 2a can
be fit well with three shells: 1 N/O at 1.78 Å (∆σ2, 0.0029
Å2), 4 N/O at 1.96 Å (0.0023), and 6 C at 2.84 Å (0.0035)
(Table 3, fit 7, and Figure 4).
Fe−N/O
Fe
−N/O
Fe‚‚‚
C
GOF
complex fit
n
r (Å)2 ∆σ2
n
r (Å)2 ∆σ2
n
r (Å)2 ∆σ2 ꢀ2 ‚103 b
c
2a
1
2
3
4
5
6
7
(7
8
5
4
5
4
5
4
4
4
4
1.95 5.6
1.95 3.7
1.96 4.3
1.96 2.4
1.95 5.6
1.96 2.4
1.96 2.3
1.96 2.5
1.95 3.7
0.70
0.65
0.62
0.52
1
1
1.74
1.79
5
4
1.25
1.19
0.67
0.80)d
0.80
1
1
1
1.79
1.78 2.9
1.78 3.7
3
6
6
6
2.84 3.5
2.83 3.6
2.83 3.6
B. Oxoiron(IV) Intermediates. The Fourier transforms (r′
space) of the Fe K-edge EXAFS data of oxoiron(IV) complexes
3a, 3a′, 3b, and 3c exhibit a single prominent feature in the
region of r′ ) 1.5-1.6 Å and peaks of much lower intensity in
the region of 2.2-2.5 Å (Figures 5 and S7). Interestingly, unlike
in the spectra of the alkylperoxoiron(III) intermediates discussed
in the previous section, a shoulder at 1.2 Å is not observed in
the spectra of the oxoiron(IV) complexes (see also Figure S3).
Simulations of the first coordination sphere of the iron center
require two shells at 1.65-1.68 and 1.97-2.00 Å (Table 4).
The principal shell at ca. 2.0 Å consisting of 4 or 5 N/O atoms
is ascribed to the donor nitrogen atoms of the polydentate ligands
(TPA, N4Py, Bn-TPEN) as in the corresponding iron(III)
complexes. The other shell at ca. 1.65 Å consists of a single
O/N atom and its introduction results in significant decreases
in the GOF values of the spectral fits for all four complexes.
This very short absorber-scatterer distance, much shorter than
the 1.8-Å Fe-OOR distance in the alkylperoxoiron(III) com-
plexes, arises from the terminal FedO group, in excellent
agreement with the 1.646 Å FedO distance found in the
crystallographically characterized [FeIV(O)(TMC)(NCMe)]-
(OTf)2 (1.646(3) Å),25,61 and in oxoiron(IV) porphyrin com-
plexes (1.60-1.70 Å) deduced from EXAFS analysis (synthetic
porphyrins and horseradish peroxidase compounds I and
II).53,54,62-65
2b
1
2
3
4
5
6
(6
7
6
5
6
5
5
5
5
5
1.96 3.8
1.96 1.7
1.96 3.7
1.97 1.5
1.97 1.5
1.97 1.5
1.97 1.7
1.96 2.3
1.21
0.83
1
1.77 1.1
1.48
1.19
1.12
1.13
1.33)d
1.33
1
1
1
1
1.78 0.5
1.78 0.4
1.78 0.4
1.78 1.2
4
3
3
3
2.83
2.82
2.81 6.1
2.82
7
6
6
a Fourier transformed range for 2a: k ) 2-13.9 Å-1 (resolution 0.13
Å); for 2b: k ) 2-15 Å-1 (resolution 0.12 Å). r is in units Å, ∆σ2 in 103
Å2. Back-transformation range for 2a: r′ ) 0.60-2.10 Å; r′ ) 0.60-
3.40 Å; for 2b: b r′ ) 0.60-2.10 Å; c r′ ) 0.60-3.10 Å. d Fits to unfiltered
data corresponding to the best fits to filtered data.
b
c
nature of these complexes almost certainly requires the presence
of a sixth ligand. The expected coordination number of n ) 5
is consistent with this fit due to the limited accuracy for
coordination numbers determined by EXAFS analysis (20-
30%).37,47 For 2b, the low Debye-Waller factor, even with an
occupancy of n ) 5, reflects a small degree of static and
vibrational disorder in the Fe-N bonds (Table 3, fit 2). In fact,
N4Py is a quite constrained ligand. The crystallographically
determined Fe-N distances in 1b(ClO4)2 ‚ MeOH (or 1b(OTf)2),
fall in a very narrow range of 1.961-1.976 Å (1.970-1.983 Å
for OTf salt, Table S1). The strong similarities of inner-sphere
parameters between 2a and 2b are not surprising, because both
intermediates are proposed to have a [FeIIIN5(OOtBu)]2+ core.
Fits including the outer-sphere features reveal an Fe‚‚‚C shell
at approximately 2.8-2.9 Å. For 2b, the inclusion of one shell
of three carbon scatterers at 2.82 Å leads at best to only a minor
improvement in GOF (Table 3, fit 6). An examination of the
crystal structure of 1b(ClO4)2 reveals that the carbon atoms
adjacent to the ligating nitrogen atoms give rise to a wide range
of Fe‚‚‚C distances (2.650-3.023 Å, Table S1). Such a broad
distribution of backscatterers would likely lead to some destruc-
tive interference, which is expected to decrease the amplitude
of the associated components in the EXAFS spectrum. The
possibility of splitting this feature into subshells is further
complicated by the limitations of resolution (∆r >≈ 0.12 Å).
The best fit to the Fourier-filtered EXAFS spectrum (k3ø′(k))
of 2b consists of three shells: 1 N/O at 1.78 Å (∆σ2, 0.0004
Å2), 5 N/O at 1.97 Å (0.0015), and 3 C at 2.82 Å (0.006) (Table
3, fit 6, and Figure 4). These parameters are virtually the same
The EXAFS spectrum of 3a can be sufficiently described by
three shells: 1 O/N at 1.65 Å (∆σ2, 0.0003 Å2), 4 N/O at 1.98
Å (0.0024), and 6 C at 2.89 Å (0.0038) (Table 4, fit 10, and
Figure 5). A fourth absorber-scatterer shell can be refined to a
distance at ca. 2.14 Å. Inclusion of this shell models better the
beat feature at k ≈ 5 Å-1 in the observed EXAFS spectrum
(Table 4, fit 11, and Figure S6), but does not improve the global
fit (see GOF value). Analysis of an earlier independent sample
of 3a with a poorer signal-to-noise ratio, however, required a
four-shell fit with a subshell at 2.20 Å (Table S7, fit 11; see
also ref 26). This sample contains approximately 80% of an S
) 1 FeIV species and 18% of a high-spin FeIII decay product,
[FeIII2(µ-O)(µ-OAc)(TPA)2]3+, according to Mo¨ssbauer analysis.
The 4 N/O scatterers at 1.99 Å may then be attributed to nitrogen
atoms of TPA ligated to FeIV, and the single-scatterer shell at
2.20 Å to the Fe-N distance in the high-spin FeIII decay
product.66 Assignment of the 2.20-Å shell to the high-spin FeIII
as those of [FeIII(N4Py)(OOH)]2+ 19
strongly suggesting that
,
(61) We have tested our method of analysis by fitting the EXAFS spectrum of
crystallographically characterized [Fe(O)(TMC)(NCMe)](OTf)2 and were
able to obtain Fe-O and Fe-N distances in excellent agreement with the
values from the crystal structure determination. Details of this analysis will
be presented in a separate study currently in preparation.
(62) Penner-Hahn, J. E.; McMurry, T. J.; Renner, M.; Latos-Grazynsky, L.; Eble,
K. S.; Davis, I. M.; Balch, A. L.; Groves, J. T.; Dawson, J. H.; Hodgson,
K. O. J. Biol. Chem. 1983, 258, 12761-12764.
(63) Penner-Hahn, J. E.; Eble, K. S.; McMurry, T. J.; Renner, M.; Balch, A.
L.; Groves, J. T.; Dawson, J. H.; Hodgson, K. O. J. Am. Chem. Soc. 1986,
108, 7819-7825.
these two peroxo intermediates have similar structures.
The outer-sphere contributions to the r′ space spectrum of
2a can be fit well with an Fe‚‚‚C shell of 6 scatterers,
(60) The 1.8-Å shell in the two samples of 2a could also be fit with an occupancy
lower than n ) 1 (Tables S3 and S4), consistent with the lower conversion
evidenced by the intensity of the EPR signals in the g ) 2 region that
account for 65% of the Fe in a simultaneously prepared sample.
9
16756 J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004

XAS of Alkylperoxoiron(III) and Oxoiron(IV)
A R T I C L E S
Table 4. EXAFS Fitting Results for [FeIV(O)(TPA)(solv)]2+, 3a, [FeIV(O)(TPA)(pyO)]2+, 3a′, [FeIV(O)(N4Py)]2+, 3b, and [FeIV(O)(Bn-TPEN)]2+
,
3ca
Fe−O/N
Fe
−N/O
Fe
−N/O
Fe‚‚‚C
GOF
complex
fit
n
r (Å)2
∆σ2
n
r (Å)2
∆σ2
n
r (Å)2
∆σ2
n
r (Å)2
∆σ2
ꢀ2 ‚ 103 b
c
3a
1
2
3
4
5
6
7
8
9
10
(10
11
12
13
6
5
5
4
4
5
5
4
5
4
4
4
5
4
1.99
1.98
1.98
1.98
1.98
1.98
1.98
1.98
1.98
1.98
1.98
1.97
1.98
1.98
6.3
4.1
4.1
2.5
2.2
4.1
4.1
2.4
4.1
2.4
2.4
1.8
4.1
2.3
1.57
1.38
0.82
0.71
0.70
1
1
1
1.65
1.65
1.65
-0.1
0.2
0.4
1
2.18
2.14
8
1.58
1.19
1.14
0.89
0.76
0.92)d
0.79
1.46
1.26
1
1
1
1
1
1
1.66
1.65
1.65
1.65
1.65
1.65
0.1
0.4
0
0.3
0.5
0.8
6
6
6
6
6
6
2.89
2.89
2.88
2.89
2.89
2.88
3.6
3.8
3.9
4.0
4
1
3
4.1
3a′
1
2
3
4
5
6
7
8
9
10
11
(11
12
6
5
4
4
5
3
6
5
4
3
3
3
3
2.02
2.01
2.00
1.98
2.00
1.97
2.02
2.02
1.98
1.97
1.97
1.97
1.97
13
10
1.22
1.13
1.05
1.14
0.72
0.52
6.5
3.6
8.5
2.1
13
2
2
2.18
2.13
3
5
0.7
0.7
1.65
1.65
-1.0
0
1.40
1.34
1.29
0.97
0.64
0.80)d
1.04
10
3.2
1.6
1.7
1.7
1.2
2
2
2
2
2
2.17
2.13
2.12
2.13
2.14
2.1
3
3.8
3.8
2.2
0.7
0.7
0.7
1.65
1.65
1.65
0.5
0.3
0.2
6
6
6
2.90
2.90
2.90
4.5
4.6
5
3b
1
2
3
4
5
6
7
(7
8
9
6
5
4
4
6
5
5
5
5
5
1.99
1.98
1.98
1.98
1.99
1.98
1.98
1.98
1.98
1.98
2.6
1.7
0.4
0.4
2.7
1.8
1.8
1.4
1.8
1.5
1.58
0.81
0.66
0.69
1
1
1
1.68
1.67
1.67
0.3
1.3
1.5
1
2.19
15
1.67
1.04
0.98
1.32)d
0.98
1.43
1
1
1
1
1.68
1.68
1.68
1.68
0.2
0.3
1.1
0.2
3
3
4
3
2.85
2.82
2.86
2.81
4
0.1
6
3
3c
1
2
3
4
5
6
7
8
9
10
11
12
(12
13
14
6
5
5
4
4
6
5
5
4
4
5
4
4
4
4
2.02
2.01
2.00
2.00
2.00
2.02
2.01
2.00
2.00
2.00
2.00
2.00
1.99
1.99
2.00
9
1.73
1.57
0.94
0.78
0.68
5.2
5.2
3.1
2.6
9
5.3
4.9
2.9
1.9
4.8
2.9
2.6
1.9
2.7
1
1
1
1.67
1.67
1.67
-0.9
-0.6
-0.4
1
2.21
3
1.85
1.72
1.23
1.13
1.04
1.03
0.84
1.15)d
0.76
1.47
1
1
1
1
1
1
1
1.67
1.67
1.67
1.67
1.67
1.67
1.67
-0.6
-0.3
0.2
-0.6
-0.5
0
1
1
2.18
2.18
-0.7
-0.2
6
6
6
6
6
2.91
2.91
2.90
2.91
2.90
4.7
4.9
5.0
5.5
5
0.2
a Fourier transformed range for 3a and 3a′: k ) 2-15 Å-1; for 3b and 3c: k ) 2-14.9 Å-1 (resolution 0.12 Å). r is in units Å, ∆σ2 in 103 Å2.
b
c
b
c
b
Back-transformation range for 3a: r′ ) 0.70-2.05 Å; r′ ) 0.70-3.40 Å; for 3a′: r′ ) 0.75-2.05 Å; r′ ) 0.75-3.25 Å; for 3b: r′ ) 0.60-2.10
c
b
c
d
Å; r′ ) 0.60-3.05 Å; for 3c: r′ ) 0.70-2.10 Å; r′ ) 0.70-3.20 Å. Fits to unfiltered data corresponding to the best fits to filtered data.
decay product is further corroborated by the EXAFS analysis
on a sample of the pyO derivative 3a′ that contains 60% FeIV
(by Mo¨ssbauer analysis). The fitting protocol for this sample
also leads to a fit with four shells: 0.7 O/N at 1.65 Å (∆σ2,
0.0003 Å2), 3 N/O at 1.97 Å (0.0017), 2 N/O at 2.12 Å (0.0038),
and 6 C at 2.90 Å (0.0045) (Table 4, fit 11, and Figure S7).
The 2.12-Å shell may be attributed to the Fe-N distance of
high-spin FeIII byproducts that make up ca. 40% of the sample.
The EXAFS spectra of the oxoiron(IV) complexes of the
pentadentate ligands N4Py and Bn-TPEN, 3b and 3c, can be
simulated best by using three shells. They differ slightly in the
coordination numbers n of the multi-nitrogen shell near 2.0 Å
and of the multi-carbon shell near 2.9 Å, depending on the nature
of the N-donor ligand. As observed for 2b, the n value of the
2.0-Å shell in 3b is 5, reflecting a narrow range of Fe-N
distances, but the n value of the carbon shell is low, reflecting
(64) Chance, B.; Powers, L.; Ching, Y.; Poulos, T.; Schonbaum, G. R.;
Yamazaki, I.; Paul, K. G. Arch. Biochem. Biophys. 1984, 235, 596-611.
(65) Green, M. T.; Dawson, J. H.; Gray, H. B. Science 2004, 304, 1653-1656.
(66) Consistent with the Mo¨ssbauer quantification, the short-distance shell could
also be refined with an occupancy of n < 1 (Table S7).
9
J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004 16757

A R T I C L E S
Rohde et al.
Figure 6. Fe K-edge X-ray absorption near-edge structures (XANES,
fluorescence excitation) of [FeII(N4Py)(NCMe)]2+, 1b (blue s), [FeIII(N4Py)-
(OOtBu)]2+, 2b (- - -), and [FeIV(O)(N4Py)]2+, 3b (red s).
I. Information from XANES. Comparison of iron complexes
of the same oxidation state with different supporting ligands
shows that the Fe K-edge XANES features vary with the
functionality, FeIII-OOtBu or FeIVdO, as well as with the ligand
environment of the absorbing atom (Figure 3). For the alkyl-
peroxo complexes 2a and 2b, the K edges are found in a narrow
window upshifted from that of 1b, as expected for their higher
oxidation state. Their preedge regions exhibit two peaks at
7113.5(1) and 7116.0(1) eV, associated with low-spin iron(III)
centers,51 with a total peak area of about 9-15 units, indicative
of a metal center with a significant distortion from centrosym-
metry.
The FeIVdO series on the other hand exhibits a somewhat
larger variability. The K edges of the two TPA complexes 3a
and 3a′ clearly show a further 1-1.5 eV increment relative to
the alkylperoxo complexes, as may be expected for the increase
in oxidation state, but those of the two N5 complexes 3b and
3c are just upshifted by less than 0.5 eV. These smaller changes
may arise from the greater covalency of the iron(IV)-ligand
bonds, particularly that of the FedO unit.55 Unlike the alkyl-
peroxo complexes, the iron(IV) complexes exhibit only one
preedge feature, which is upshifted by about 0.6-1.1 eV with
respect to the iron(III) complexes and has an intensity of 25-
29 units.
Figure 5. Fourier transforms of the Fe K-edge EXAFS data (k3ø(k)) and
Fourier-filtered EXAFS spectra (k3ø′(k), inset) of [FeIV(O)(TPA)(solv)]2+
,
3a, Fourier transformed range k ) 2-15 Å-1; experimental data (‚‚‚) and
fit 1 O/N, 4 N/O, 6 C (s); Back-transformation range: r′ ) 0.70-3.40 Å
(top); [FeIV(O)(N4Py)]2+, 3b, Fourier transformed range k ) 2-14.9 Å-1
;
experimental data (‚‚‚) and fit 1 O/N, 5 N/O, 3 C (s); Back-transformation
range: r′ ) 0.60-3.05 Å (middle); and [FeIV(O)(Bn-TPEN)]2+, 3c, Fourier
transformed range k ) 2-14.9 Å-1; experimental data (‚‚‚) and fit 1 O/N,
4 N/O, 6 C (s); Back-transformation range: r′ ) 0.70-3.20 Å (bottom).
a wide range of Fe‚‚‚C distances: 1 O/N at 1.68 Å (∆σ2, 0.0003
Å2), 5 N/O at 1.98 Å (0.0018), and 3 C at 2.85 Å (0.004) (Table
4, fit 7, and Figure 5). On the other hand, 3c has parameters
similar to 3a: 1 O/N at 1.67 Å (∆σ2, -0.0005 Å2), 4 N/O at
2.00 Å (0.0029), and 6 C at 2.91 Å (0.0049) (Table 4, fit 12,
and Figure 5).
Our systematic study provides an unprecedented opportunity
to compare the XANES features of complexes of the same
supporting ligand in three different oxidation states, all with
low-spin configurations. As illustrated by the series of N4Py
complexes 1b, 2b, and 3b in Figure 6, the Fe K edges have
very similar shapes and the edge shifts by less than 2 eV with
an increase of the iron oxidation state from +II to +IV: FeII-
NCMe, FeIII-OOtBu, FeIVdO. More striking is the progressive
increase in the preedge peak area, from 4 units for 1b to 15
units for 2b and finally to 25 units for 3b. This trend reflects
the increased distortion from centrosymmetry as the MeCN
ligand in the highly symmetric FeN6 core in 1b is progressively
replaced with an alkylperoxo ligand in 2b with a 1.78 Å Fe-O
bond and then with the oxo ligand in 3b with a 1.68 Å Fe-O
bond.
Discussion
We have analyzed the X-ray absorption spectra of two
alkylperoxoiron(III) and four oxoiron(IV) complexes in order
to obtain structural information on these thermally labile species.
These complexes have been generated as intermediates in the
reactions of iron(II) precursors with peroxides, representing the
first examples of such species to be observed. With the exception
of 3a, for which a preliminary analysis has been reported,26 this
study presents the first structural information available on this
series of novel intermediates. This systematic analysis enables
us to establish the signatures of FeIII-OOtBu and FeIVdO
groups.
II. Information from EXAFSsMetrical Trends. The first
coordination spheres of the alkylperoxoiron(III) and oxoiron-
(IV) complexes discussed here can generally be described by
two shells of metal-ligand scattering interactions: a longer
9
16758 J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004

XAS of Alkylperoxoiron(III) and Oxoiron(IV)
A R T I C L E S
Table 5. Summary of EXAFS Fitting Results for the Iron
Complexes in This Study
as in the case of porphyrins. The consistently short Fe-O
distance found for oxoiron(IV) complexes support DFT calcula-
tions that invariably describe the FeIVdO bond as a highly
covalent interaction between the oxo atom and the 3dz2, 3dxz,
and 3dyz orbitals of the iron.55,67,68
Fe−N/O
Fe
−N/O
Fe
−N/O
Fe‚‚‚
C
complex
n
r (Å)
∆σ2
n
r (Å) ∆σ2
n
r (Å) ∆σ2
n
r (Å) ∆σ2
iron(II)
1b
alkylperoxoiron(III)
5
1.96 3.6
3
2.82 4
Additional shells are needed to fit the EXAFS features
observed under 3 Å for 3a, 3a′, 3b, and 3c. Common to all
four complexes is a shell of low-Z atoms at 2.9 Å that are
assigned to the carbon atoms adjacent to the ligating nitrogen
atoms. Scrutiny of the fit protocols in Table 4 reveals that the
addition of this shell makes a more marked improvement in
the GOF value for TPA and Bn-TPEN complexes than for the
N4Py complex. This difference may be due to the larger spread
of Fe‚‚‚C distances near 2.9 Å associated with N4Py, as found
2a
2b
1
1
1.78
1.78
2.9
0.4
4
5
1.96 2.3
1.97 1.5
6
3
2.84 3.5
2.82
6
oxoiron(IV)
3a
3a′
3b
3c
1
1.65
0.3
0.3
0.3
4
3
5
4
1.98 2.4
1.97 1.7
1.98 1.8
2.00 2.9
6
6
3
6
2.89 3.8
2.90 4.5
2.85
0.7 1.65
1
1
2
2.12 3.8
1.68
1.67 -0.5
4
2.91 4.9
Table 6. Structural Parameters of Synthetic High-valent FedO
in the crystal structures of 1b and [FeIII(N4Py)(OMe)]2+ 30
,
Complexesa
resulting in some destructive interference among the contribu-
tions to this shell.
complex
Fe
−
N
Fe
−O
method ref.
non-porphyrin
The alkylperoxo intermediates 2a and 2b can, like the
oxoiron(IV) complexes, be fit with three shells of scatterers.
Besides the 2.0 Å principal shell that arises from the Fe-N
bonds of the polydentate ligand, there is a single-scatterer shell
at 1.78 Å assigned to the peroxo ligand and a multiscatterer
shell at 2.8 Å due to the carbon atoms adjacent to the ligating
N atoms. The single low-Z scatterer at 1.78 Å can be assigned
to the FeIII-OOtBu bond, in agreement with DFT calculations
K2[FeVIO4]
1.660-1.671 XRD 69
1.790-1.825 XRD 70
Na4[FeIVO4]b
c
[FeIV(O)(TMC)(NCMe)](OTf)2
2.058-2.117 1.646
XRD 25
porphyrin
2.02
[FeIV(O)(TMP)+•(MeOH)]+ d
1.63-1.66
1.66, 1.65
1.67
XAS 63
XAS 53
XAS 54
XAS 63
[FeIV(O)(TMP)+•(X)], X ) Cl, Brd 1.99, 2.01
[FeIV(O)(F8TPP)+•
]
2.00
2.03
+ e
[FeIV(O)(TTP)(Me-im)]f
1.65-1.66
a Distances are in Å. powder diffraction. TMC, 1,4,8,11-tetramethyl-
1,4,8,11-tetraazacyclotetradecane. dTMP, meso-tetramesitylporphinato di-
anion. eF8TPP, meso-tetrakis(2,6-difluorophenyl)porphinato dianion. fTTP,
meso-tetra-m-tolylporphinato dianion; Me-im, N-methylimidazole. There is
a crystal structure of an FeIV(O)(porphyrin) complex with an Fe-O bond
length of 1.604 Å that was briefly mentioned in a footnote of Schappacher
et al.,71 but this result has not been subsequently presented in greater detail.
b
c
on [FeIII(TPA)(OOtBu)(OH2)]2+ 22
. This distance is comparable
to the 1.76 Å FeIII-OOH distance found for low-spin [FeIII-
(N4Py)(OOH)]2+ by EXAFS analysis19 but is shorter than those
found for low-spin FeIII-OOH complexes with pentadentate
ligands having one anionic group, [FeIII(SMe2N4(tren))(OOH)]+
(1.86 Å)17 and [FeIII(PaPy3)(OOH)]+ (∼1.97 Å).20 The ad-
ditional anionic functionality in the latter two complexes
probably competes with the peroxide ligand for the lone t2g
orbital on the low-spin iron(III) center available for π bonding.72
Thus, unlike the oxo group in complexes 3, the FeIII-O
distances of the peroxo ligands appear to be strongly modulated
by the nature of the other ligands in the metal coordination
sphere.
The binding mode of the alkylperoxo ligand deserves some
comment. Most crystal structures of synthetic alkylperoxo
complexes of 3d transition metals (e.g., TiIV (d0), MnII (d5), CoII
(d7), CoIII (d6), CuII (d9)) have revealed end-on binding of the
alkylperoxo group with M-OOR distances in the range 1.82-
1.96 Å and M-O-O angles of 105-122°, with the distal
oxygen atom of the M-OOR fragment found at distances of
2.68-2.95 Å from the metal center (Table 7). On the other hand,
side-on binding can be observed with two complexes of 3d0
metal ions, TiIV and VV, giving rise to M-OOR distances of
1.913 and 1.872 Å and M‚‚‚OR distances of 2.269 and 1.999
Å, respectively (Table 7). We have considered whether our data
allow us to distinguish between the two binding modes for 2a
and 2b. The observed 1.78-Å distances and the fact that 2a and
2b are both low-spin complexes would certainly favor an end-
distance that accounts for 4-5 nitrogen atoms of the supporting
ligand, and a shorter distance that arises from only 1 nitrogen/
oxygen atom. The longer Fe-N distances fall within a very
narrow range of 1.96-2.00 Å for the S ) 1/2 FeIII and S ) 1
FeIV complexes, consistent with their low-spin nature (Table
5). In the series of N4Py complexes, 1b, 2b, and 3b, the 2.0-Å
shell remains nearly unchanged with the increase in oxidation
state from +II to +IV; furthermore the 5-atom shell is
consistently fit with a low Debye-Waller factor, reflecting the
small range of Fe-N bond lengths. In contrast, there appears
to be a greater variability of Fe-N bond lengths for the TPA
and Bn-TPEN complexes that leads to some destructive interfer-
ence, as the coordination numbers for the main shell in the best
fits consistently have a value of 4. Nevertheless, the average
Fe-N distance remains near 2 Å for these complexes.
Complexes 3a, 3a′, 3b, and 3c all have single scatterers with
short Fe-O distances in the range of 1.65-1.68 Å, which are
assigned to the FeIVdO group (Table 5). These distances closely
match that found in the crystallographically characterized [FeIV-
(O)(TMC)(NCMe)](OTf)2 (1.646(3) Å).25 Furthermore, it agrees
well with the corresponding distances in synthetic oxoiron(IV)
porphyrins53,54,62,63 (Table 6) and horseradish peroxidase Com-
pounds I and II62-65 determined by EXAFS analysis. The
nonheme complexes reported here significantly augment the
structural database for oxoiron(IV) complexes and demonstrate
that the FedO distance of the S ) 1 oxoiron(IV) unit is rather
insensitive to the nature of the other ligands in the iron
coordination sphere, be they neutral and pentadentate as in the
case of N4Py or Bn-TPEN or dianionic and planar tetradentate
(67) Zhang, Y.; Oldfield, E. J. Am. Chem. Soc. 2004, 126, 4470-4471.
(68) Ghosh, A.; Almlo¨f, J.; Que, L., Jr. J. Phys. Chem. 1994, 98, 5576-5579.
(69) Hopper, M. L.; Schlemper, E. O.; Murmann, R. K. Acta Crystallogr. 1982,
B38, 2237-2239.
(70) Weller, M. T.; Hector, A. L. Angew. Chem., Int. Ed. 2000, 39, 4162-
4163.
(71) Schappacher, M.; Weiss, R.; Montiel-Montoya, R.; Trautwein, A.; Tabard,
A. J. Am. Chem. Soc. 1985, 107, 3736-3738.
(72) Neese, F.; Zaleski, J. M.; Zaleski, K. L.; Solomon, E. I. J. Am. Chem. Soc.
2000, 122, 11703-11724.
9
J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004 16759

A R T I C L E S
Rohde et al.
Table 7. Structural Parameters of Fe(OOH), Fe(η2-O2), and Fe(OOR) Complexes and M(OOR) Complexes of Other 3d Metals (nonbridging
peroxo ligands)a
complex
spin state
M
−
Nb
M
−O
M
‚‚‚O
method
ref.
end-on peroxo complexes of Fe
[FeIII(SMe2N4(tren))(OOH)]+ c
[FeIII(N4Py)(OOH)]2+ d
[FeIII(qn)2(OOC(O)O)]-e
purple lipoxygenase
low spin
low spin
high spin
high spin
2.01
1.96
2.182
2.2-2.3
1.86
1.76
1.936
2.0
2.78
XAS
XAS
XRD
XRD
17
19
73
10
2.871
2.8
(1IK3.pdb; 2.0 Å resolution)f
side-on peroxo complexes of Fe
[FeIII(N4Py)(η2-O2)]+ d
NDO:indole:dioxygen
high spin
2.20
2.1-2.2
1.93
1.8
XAS
XRD
19
14
2.0
(1O7N.pdb; 1.4 Å resolution)g
end-on alkylperoxo complexes of other 3d metals
[TiIVClCp2(OOtBu)]
1.909
2.952
2.700
2.71
2.737
2.683
-2.826
XRD
XRD
XRD
XRD
XRD
74
75
76
77
i
[MnII(TptBu,iPr)(OOCMe2Ph)]h
[CoII(TptBu,iPr)(OOCMe2Ph)]h
[CuII(TpiPr,iPr)(OOCMe2Ph)]h
[CoIII(L)(OOR)]i
high spin
high spin
2.132-2.161
2.01-2.06
1.949-2.161
1.917-2.024
1.964
1.86
1.816
1.838
-1.905
low spin
side-on alkylperoxo complexes of other 3d metals
{[TiIV(tea)(η2-O2 Bu)]}2
2.299
1.913
1.872
2.269
1.999
XRD
XRD
78
79
t
j
t
[VV(L)(O)(η2-O2 Bu)(OH2)] k
a Distances are in Å; average values are reported for crystallographic disorder and independent molecules. b M-N distance of amine, imidazole, and
pendant pyridine groups. c HSMe2N4(tren), 3-[N-[2-[N,N-bis(2-aminoethyl)amino]ethyl]imino]-2-methyl-2-butanethiol. d N4Py, N,N-bis(2-pyridylmethyl)-N-
bis(2-pyridyl)methylamine. e Hqn, 2-quinolinecarboxylic acid. f Purple lipoxygenase, from soybean lipoxygenase-3 and (9Z,11E)-13(S)-hydroperoxy-9,11-
octadecadienoic acid. g NDO, naphthalene dioxygenase from Pseudomonas sp. h TptBu,iPr, hydrotris(3-tert-butyl-5-isopropylpyrazol-1-yl)borate; TpiPr,iPr
,
hydrotris(3,5-diisopropylpyrazol-1-yl)borate. i Selected examples: [CoIII(acac)2(OOtBu)(L)], L ) py, Me-im;80 [CoIII(dbm)2(OOtBu)(py)];80 [CoIII(Py3P)(OOR)],
t
R ) Bu, CMe2Ph, CMe2Bn;81,82 [CoIII(BPI)(OOtBu)(OBz)];83 [CoIII(Hdmg)2(OOCMe2Ph)(py)];84 [CoIII(tren)(OOCHMeCO2)]+;85 abbreviations: Hacac,
2,5-pentanedione; Me-im, N-methylimidazole; Hdbm, 1,3-diphenyl-1,3-propanedione; Py3PH2, N,N-bis[2-(2-pyridyl)ethyl]pyridine-2,6-dicarboxamide; BzOH,
benzoic acid; BPI ) 1,3-bis(2′-pyridylimino)isoindoline; H2dmg, 2,3-butanedione dioxime; tren, tris(2-aminoethyl)amine. j H3tea, 2,2′,2′′-nitrilotriethanol.
k H2L ) 2,6-pyridinedicarboxylic acid.
on binding mode. In such a binding mode, the distal oxygen
atom would be expected to be 2.7-3.0 Å from the metal center.
It would thus be included with the scatterers in the 2.8-Å multi-
carbon shell from the TPA and N4Py ligand backbones, unlike
in the case of [FeIII(SMe2N4(tren))(OOH)]+, where the lack of
such a carbon shell allows an oxygen scatterer at 2.78 Å to be
discerned in the EXAFS analysis.17 On the other hand, a side-
on bound alkylperoxide would be expected to introduce an
additional oxygen scatterer in the range of 2.0-2.3 Å. Such an
additional shell is not required in the analyses. It is also possible
that the distal oxygen could have a short enough Fe-O distance
to be part of the principal sub-shell at 2 Å. If one compares
different intermediates with the same ligand, e. g. 2a and 3a
(TPA) or 2b and 3b (N4Py), the coordination number for either
the 2.0 or 2.8 Å shell remains constant with little change in the
Debye-Waller factor (Table 5), suggesting the same number
of scatterers in the two pairs of intermediates in the same series.
The strongest argument against a side-on peroxo binding mode
for 2a and 2b is the fact that such complexes are to date all
high-spin, even in the case of porphyrins. Thus, we favor end-
on coordination of the alkylperoxo group in 2a and 2b.
To date, only one oxoiron(IV) intermediate has been identified
for a mononuclear nonheme iron oxygenase.11-13 Very recent
EXAFS data on such an intermediate derived from TauD
(taurine:2-oxoglutarate dioxygenase) revealed a 1.62 Å Fe-O
bond,86 comparable in length to those found for the synthetic
oxoiron(IV) complexes reported here. Mo¨ssbauer analysis how-
ever determined the iron(IV) center in this species to be high
spin,11,86 as opposed to low spin for the synthetic complexes.
Despite the difference in spin states, the comparable bond
lengths are consistent with DFT calculations on the oxoiron-
(IV) unit and can be rationalized by the fact that the dxy and
the dx2-y2 orbitals, which are the orbitals affected by the spin
state change, are not involved in FedO bonding interactions.55
(73) Hashimoto, K.; Nagatomo, S.; Fujinami, S.; Furutachi, H.; Ogo, S.; Suzuki,
M.; Uehara, A.; Maeda, Y.; Watanabe, Y.; Kitagawa, T. Angew. Chem.,
Int. Ed. 2002, 41, 1202-1205.
(74) DiPasquale, A. G.; Kaminsky, W.; Mayer, J. M. J. Am. Chem. Soc. 2002,
124, 14534-14535.
(75) Komatsuzaki, H.; Sakamoto, N.; Satoh, M.; Hikichi, S.; Akita, M.; Moro-
oka, Y. Inorg. Chem. 1998, 37, 6554-6555.
(76) Hikichi, S.; Komatsuzaki, H.; Akita, M.; Moro-oka, Y. J. Am. Chem. Soc.
1998, 120, 4699-4710.
(77) Kitajima, N.; Katayama, T.; Fujisawa, K.; Iwata, Y.; Moro-oka, Y. J. Am.
Chem. Soc. 1993, 115, 7872-7873.
(78) Boche, G.; Mo¨bus, K.; Harms, K.; Marsch, M. J. Am. Chem. Soc. 1996,
118, 2770-2771.
(79) Mimoun, H.; Chaumette, P.; Mignard, M.; Saussine, L.; Fischer, J.; Weiss,
R. NouV. J. Chim. 1983, 7, 467-475.
Perspective
(80) Chavez, F. A.; Briones, J. A.; Olmstead, M. M.; Mascharak, P. K. Inorg.
Chem. 1999, 38, 1603-1608.
We have used X-ray absorption spectroscopy to gain struc-
tural insight into thermally labile low-spin alkylperoxoiron(III)
and oxoiron(IV) complexes that model possible intermediates
in the dioxygen activation cycles of nonheme iron enzymes.
Except for the crystallographically characterized [FeIV(O)-
(TMC)(NCMe)]2+ 25
, the data reported here constitute the only
collection of metrical data available for these novel intermedi-
ates. These results serve as benchmarks with which to compare
data on new enzyme intermediates.
(81) Chavez, F. A.; Nguyen, C. V.; Olmstead, M. M.; Mascharak, P. K. Inorg.
Chem. 1996, 35, 6282-6291.
(82) Chavez, F. A.; Rowland, J. M.; Olmstead, M. M.; Mascharak, P. K. J. Am.
Chem. Soc. 1998, 120, 9015-9027.
(83) Saussine, L.; Brazi, E.; Robine, A.; Mimoun, H.; Fischer, J.; Weiss, R. J.
Am. Chem. Soc. 1985, 107, 3534-3540.
(84) Giannotti, C.; Fontaine, C.; Chiaroni, A.; Riche, C. J. Organomet. Chem.
1976, 113, 57-65.
(85) Kojima, M.; Akhter, F.; Nakajima, K.; Yoshikawa, Y. Bull. Chem. Soc.
Jpn. 1996, 69, 2889-2895.
(86) Riggs-Gelasco, P. J.; Price, J. C.; Guyer, R. B.; Brehm, J. H.; Barr, E. W.;
Bollinger, J. M., Jr.; Krebs, C. J. Am. Chem. Soc. 2004, 126, 8108-8109.
9
16760 J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004

XAS of Alkylperoxoiron(III) and Oxoiron(IV)
A R T I C L E S
Thus an obvious target for future synthetic efforts is the
generation and characterization of a high-spin oxoiron(IV)
complex.
Acknowledgment. This work was supported by grants from
the National Institutes of Health (GM-33162 to L.Q.) and the
Ministry of Science and Technology of Korea through the
Creative Research Initiative Program (to W.N.) and a fellowship
from the Deutsche Forschungsgemeinschaft (to J.-U.R.). We
are grateful to Audria Stubna and Prof. Eckard Mu¨nck of the
Carnegie Mellon University for Mo¨ssbauer analyses of the XAS
samples. We thank Dr. Neil R. Brooks of the University of
Minnesota X-ray Crystallographic Facility for the crystal
structure determination of 1b(OTf)2. XAS data were collected
on beamline 7-3 at the Stanford Synchrotron Radiation Labora-
tory and beamline X9B at the National Synchrotron Light
Source, both supported by the US Department of Energy and
the National Institutes of Health. We thank Dr. Matthew J.
Latimer at SSRL and Dr. Nebojsa S. Marinkovic at NSLS for
their excellent technical support of our synchrotron experiments.
An EXAFS analysis of chloroperoxidase Compound II has
also been recently reported.65 In this case, the Fe-O distance
was determined to be 1.82 Å, significantly longer than those
found for horseradish peroxidase Compound II and the synthetic
complexes described in this paper. This lengthening has been
ascribed to the protonation of the oxoiron(IV) unit, a phenom-
enon also observed for other,87-89 but not all,63,65 Compounds
II. The oxoiron(IV) centers in peroxidase Compounds II may
be expected to be more prone to protonation because the
dianionic porphinate should diminish the Lewis acidity of the
metal center; for chloroperoxidase, the additional thiolate ligand
further enhances this effect. On the other hand, the oxoiron-
(IV) centers of complexes 3 have neutral ligands, making the
metal center significantly more Lewis acidic and the oxo group
more difficult to protonate. The availability of well characterized
oxoiron(IV) complexes with a range of supporting ligands thus
provides us with a better basis for understanding the fundamental
properties of these novel high-valent species.
Supporting Information Available: Fe K-edge EXAFS
spectra of all compounds (Figures S1 and S2) and detailed
EXAFS fitting protocols (Tables S2-S10). Best fits (Fourier
transforms of Fe K-edge EXAFS data and Fourier-filtered
EXAFS spectra) for 1b, duplicate 2a, and 3a′ (Figures S4-S7,
PDF). Details of the crystal structure determinations of 1b(OTf)2
and 1c(OTf) (Tables S1, S11, S12, and Figure S9, PDF) and
X-ray crystallographic files (CIF). This material is available free
of charge via the Internet at http://pubs.acs.org.
(87) Berglund, G. I.; Carlsson, G. H.; Smith, A. T.; Szo¨ke, H.; Henriksen, A.;
Hajdu, J. Nature 2002, 417, 463-468.
(88) Bonagura, C. A.; Bhaskar, B.; Shimizu, H.; Li, H.; Sundaramoorthy, M.;
McRee, D. E.; Goodin, D. B.; Poulos, T. L. Biochemistry 2003, 42, 5600-
5608.
(89) Hersleth, H.-P.; Dalhus, B.; Go¨rbitz, C. H.; Andersson, K. K. J. Biol. Inorg.
Chem. 2002, 7, 299-304.
JA047667W
9
J. AM. CHEM. SOC. VOL. 126, NO. 51, 2004 16761