Further insights into the spectroscopic properties, electronic structure, and kinetics of formation of the heme-peroxo-copper complex [(F 8TPP)FeIII-(O22-)-Cu II(TMPA)]+

Article abstract of DOI:10.1021/ic061726k

In the further development and understanding of heme-copper O ₂-reduction chemistry inspired by the active-site chemistry in cytochrome c oxidase, we describe a dioxygen adduct, [(F₈TPP)Fe ^III-(O₂^2-)-Cu^II(TMPA)](ClO ₄) (3), formed by addition of O₂ to a 1:1 mixture of the porphyrinate-iron(II) complex (F₈TPP)Fe^II (1a) {F ₈TPP = tetrakis(2,6-difluorophenyl)porphyrinate dianion} and the copper(I) complex [(TMPA)Cu^I(MeCN)](ClO₄) (1b) {TMPA = tris(2-pyridylmethyl)amine}. Complex 3 forms in preference to heme-only or copper-only binuclear products, is remarkably stable {t_1/2 (RT; MeCN) ≈ 20 min; λ_max = 412 (Soret), 558 nm; EPR silent}, and is formulated as a peroxo complex on the basis of manometry {1a/1b/O₂ = 1:1:1}, MALDI-TOF mass spectrometry {¹⁶O₂, m/z 1239 [(3 + MeCN)⁺]; ¹⁸O₂, m/z 1243}, and resonance Raman spectroscopy {ν_(O-O) = 808 cm^-1; Δ ¹⁶O₂/¹⁸O₂ = 46 cm^-1; Δ¹⁶O₂/^16/18O₂ = 23 cm ^-1}. Consistent with a μ-η²:η¹ bridging peroxide ligand, two metal-O stretching frequencies are observed {ν_(Fe-O) = 533 cm^-1, ν_(Fe-O-Cu) = 511 cm^-1}, and supporting normal coordinate analysis is presented. ²H and ¹⁹F NMR spectroscopies reveal that 3 is high-spin {also μ_B = 5.1 ± 0.2, Evans method} with downfield-shifted pyrrole and upfield-shifted TMPA resonances, similar to the pattern observed for the structurally characterized μ-oxo complex [(F₈TPP)Fe ^III-O-Cu^II(TMPA)]⁺ (4) (known S = 2 system, antiferromagnetically coupled high-spin Fe^III and Cu^II). Moessbauer spectroscopy exhibits a sharp quadrupole doublet (zero field; δ = 0.57 mm/s, |ΔE_Q| = 1.14 mm/s) for 3, with isomer shift and magnetic field dependence data indicative of a peroxide ligand and S = 2 formulation. Both UV-visible-monitored stopped-flow kinetics and Moessbauer spectroscopic studies reveal the formation of heme-only superoxide complex (S)(F₈TPP)Fe^III-(O₂ ^-) (2a) (S = solvent molecule) prior to 3. Thermal decomposition of μ-peroxo complex 3 yields μ-oxo complex 4 with concomitant release of ～0.5 mol O₂ per mol 3. Characterization of the reaction 1a/1b + O₂ → 2 → 3 → 4, presented here, advances our understanding and provides new insights to heme/Cu dioxygen-binding and reduction.

Full text of DOI:10.1021/ic061726k

Inorg. Chem. 2007, 46, 3889−3902
Further Insights into the Spectroscopic Properties, Electronic Structure,
and Kinetics of Formation of the Heme Peroxo Copper Complex
−
−
2-
+
[(F8TPP)FeIII
−(O2
)−
CuII(TMPA)]
Reza A. Ghiladi,†,£ Eduardo E. Chufa´n,† Diego del R´ıo,‡ Edward I. Solomon,‡ Carsten Krebs,§,#
Boi Hanh Huynh,§ Hong-wei Huang,| Pierre Moenne-Loccoz,| Susan Kaderli, Marcus Honecker,
Andreas D. Zuberbu
hler, Lisa Marzilli,∇ Robert J. Cotter,∇ and Kenneth D. Karlin*,†
1
1
Department of Chemistry, The Johns Hopkins UniVersity, Charles and 34th Streets, Baltimore,
Maryland 21218, Department of Chemistry, Stanford UniVersity, Stanford, California 94305,
Department of Physics, Emory UniVersity, Atlanta, Georgia 30322, Department of EnVironmental
& Biomolecular Systems, OGI School of Science and Engineering, Oregon Health and
Science UniVersity, BeaVerton, Oregon 97006, Department of Chemistry, UniVersity of Basel,
CH-4056 Basel, Switzerland, and Department of Pharmacology and Molecular Sciences,
The Johns Hopkins School of Medicine, Baltimore, Maryland 21205
Received September 12, 2006
In the further development and understanding of heme
chemistry in cytochrome c oxidase, we describe a dioxygen adduct, [(F8TPP)FeIII
formed by addition of O2 to a 1:1 mixture of the porphyrinate
iron(II) complex (F8TPP)FeII (1a)
(2,6-difluorophenyl)porphyrinate dianion
and the copper(I) complex [(TMPA)CuI(MeCN)](ClO4) (1b)
(2-pyridylmethyl)amine . Complex 3 forms in preference to heme-only or copper-only binuclear products, is remarkably
stable t1 (RT; MeCN) 412 (Soret), 558 nm; EPR silent , and is formulated as a peroxo
complex on the basis of manometry
MeCN)+]; 18O2, m/z 1243
−copper O2-reduction chemistry inspired by the active-site
2-
−(O2
)−
CuII(TMPA)](ClO4) (3),
F8TPP tetrakis-
TMPA tris-
−
{
)
}
{
)
}
{
≈
20 min; λmax
)
}
/
2
{
1a/1b/O2
)
1:1:1
}
, MALDI-TOF mass spectrometry
{
16O2, m/z 1239 [(3
16O2/18O2 46 cm-
+
;
1
1
}
, and resonance Raman spectroscopy {ν(O
)
808 cm-
;
−
∆
)
-O)
/
1
∆
16O2/16 18O2
)
23 cm-
}
. Consistent with a
µ
-
)
η2
:
η1 bridging peroxide ligand, two metal
O stretching frequencies
1
1
are observed {ν(Fe
)
533 cm-
,
ν(Fe
511 cm-
}
, and supporting normal coordinate analysis is presented.
also µB ) 5.1 0.2, Evans method with downfield-
shifted pyrrole and upfield-shifted TMPA resonances, similar to the pattern observed for the structurally characterized
-O)
-O-Cu)
2H and 19F NMR spectroscopies reveal that 3 is high-spin
{
±
}
µ
-oxo complex [(F8TPP)FeIII CuII(TMPA)]+ (4) (known S
−O− ) 2 system, antiferromagnetically coupled high-spin
FeIII and CuII). Mo¨ssbauer spectroscopy exhibits a sharp quadrupole doublet (zero field; δ ) 0.57 mm/s, |∆EQ| )
1.14 mm/s) for 3, with isomer shift and magnetic field dependence data indicative of a peroxide ligand and S
)
2 formulation. Both UV
formation of heme-only superoxide complex (S)(F8TPP)FeIII
decomposition of -peroxo complex 3 yields -oxo complex 4 with concomitant release of
3. Characterization of the reaction 1a/1b 4, presented here, advances our understanding and
provides new insights to heme/Cu dioxygen-binding and reduction.
−
visible-monitored stopped-flow kinetics and Mo¨ssbauer spectroscopic studies reveal the
(O2-) (2a) (S
solvent molecule) prior to 3. Thermal
0.5 mol O2 per mol
−
)
µ
µ
∼
+ O2 f 2 f 3 f
Introduction
sequentially by cytochrome c to catalyze the 4e-/4H+
reduction of dioxygen to water.1-9 This membrane-bound
multimetallic enzyme couples this exergonic process to the
movement of four additional protons across the mitochondrial
membrane, resulting in a net translocation of four charges.
The proton pumping mechanism of CcO generates the
As a member of the heme-copper oxidase superfamily,
cytochrome c oxidase (CcO) utilizes electrons provided
* To whom correspondence should be addressed. E-mail: karlin@
jhu.edu.
† Department of Chemistry, The Johns Hopkins University.
‡ Stanford University.
§ Emory University.
(1) Ferguson-Miller, S.; Babcock, G. T. Chem. ReV. 1996, 96, 2889-
2907.
|
Oregon Health and Science University.
University of Bazel.
(2) Michel, H.; Behr, J.; Harrenga, A.; Kannt, A. Annu. ReV. Biophys.
∇ Department of Pharmacology and Molecular Sciences, The Johns
Hopkins University.
Biomol. Struct. 1998, 27, 329-356.
(3) Kitagawa, T. J. Inorg. Biochem. 2000, 82, 9-18.
(4) Collman, J. P.; Boulatov, R.; Sunderland, C. J. Porphyrin Handbook
2003, 11, 1-49.
(5) Kim, E.; Chufa´n, E. E.; Kamaraj, K.; Karlin, K. D. Chem. ReV. 2004,
104, 1077-1133.
# Current address: Department of Chemistry, Pennsylvania State Uni-
versity, University Park, PA 16802.
£ Current address: Department of Chemistry, North Carolina State
University, Raleigh, NC 27695.
10.1021/ic061726k CCC: $37.00
Published on Web 04/20/2007
© 2007 American Chemical Society
Inorganic Chemistry, Vol. 46, No. 10, 2007 3889

Ghiladi et al.
Chart 1
Figure 1. Rasmol depiction of the heme a3-CuB binuclear active site of
dioxygen reduction from bovine cytochrome c oxidase. Coordinates (1OCR)
were obtained from the Protein Data Bank (Brookhaven, NY).13
membrane potential used by ATP synthase to drive the
formation of ATP from ADP, and in light of its biologically
important relevance, has been the subject of numerous and
intense biophysical and spectroscopic investigations.
Confirming earlier biochemical and molecular biological
suppositions are the recent X-ray crystallographic studies
which show that dioxygen reduction occurs at an active site
consisting of a heme a3-CuB heterobinuclear center with an
intermetal separation of 4.5-5.2 Å, depending on the protein
derivative (Figure 1).5-9 As it is generally considered,
dioxygen binding and reduction occurs as follows:5,10-12 An
initial CuB-O2 interaction precedes transfer of O2 to heme
a3, forming a ferrous-oxy (FeIII-superoxo) observable
intermediate (A). The next intermediate detected has already
undergone O-O cleavage, giving the ferryl FeIVdO species
II
(P) and CuB -OH, with either tyrosinate or tyrosyl radical
(Tyr•), depending on the enzyme form used for the O2-
chemistry. Rapid O-O cleavage seems an attractive mecha-
nistic feature as such a reaction course bypasses the formation
electron) donor, and/or as an important element of the proton-
pumping mechanism. Despite the wealth of knowledge,
complete insights are lacking, including a full understanding
of the nature of O2-intermediates (and their protonation
states) forming prior to the O-O cleavage process.
-
of potentially cytotoxic (if leaked) superoxide (as O2 or
-
HO2) or peroxide (as HO2 or H2O2) products. However,
(hydro)peroxide metal-bridged or associated FeIII-O-
O(H)‚‚‚CuB transients or intermediates have also been
discussed, based on structural, spectroscopic, and theoretical
considerations.5,7,13,14 The post-translationally formed amino-
acid His-Tyr cross-link (Figure 1)5,15 may be key in the
reduction mechanism as a hydrogen atom (i.e., proton plus
There has also been intense interest from inorganic
chemists to provide coordination chemistry insights into CcO
heme a3-CuB physical properties and reactivity. Dioxygen
reactivity studies have recently undergone considerable
advancement, as approached either via electrocatalytic O2-
reduction with heme-Cu assemblies4,16 or by the generation
of O2-adducts of discrete heme-Cu complexes or heme and
Cu-ligand components.5,17 Some recent examples are given
in Chart 1. Collman18 and more recently Naruta19 have
(6) Kannt, A.; Michel, H. Bacterial cytochrome c oxidase. In Handbook
of Metalloproteins; Messerschmidt, A, Huber, R., Poulos T., Wieghardt,
K., Eds.; Wiley: New York, 2001; Vol. 1; pp 331-347.
(7) Yoshikawa, S. AdV. Protein Chem. 2002, 60, 341-395.
(8) Yoshikawa, S.; Shinzawa-Itoh, K.; Yamashita, E.; Tsukihara, T.
Mitochodrial cytochrome c oxidase. In Handbook of Metalloproteins;
Messerschmidt, A, Huber, R., Poulos T., Wieghardt, K., Eds.; Wiley:
New York, 2001; Vol. 1; pp 348-362.
(9) Than, M. E.; Tewfik, S. ba3-Cytochrome c oxidase from Thermus
thermophilus. In Handbook of Metalloproteins; Messerschmidt, A,
Huber, R., Poulos T., Wieghardt, K., Eds.; Wiley: New York, 2001;
Vol. 1; pp 363-378.
(10) Babcock, G. T. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 12971-12973.
(11) Brunori, M.; Giuffre, A.; Sarti, P. J. Inorg. Biochem. 2005, 99, 324-
336.
-
generated heme-superoxide (FeIII-O2 ) moieties, I and II,
respectively, which are stable in the presence of the
neighboring Cu(I) complex; these species mimic the forma-
tion of the initial O2-adduct A intermediate in the CcO O2-
reduction cycle. Likewise, we have spectroscopically and/
or kinetically observed solutions containing both Fe-O2 and
CuI species which exist prior to forming µ-peroxo FeIII-
(O22-)-CuII products.20-25 An exciting development was
Naruta’s crystal structure determination of binuclear complex
(12) Barry, B. A.; Einarsdottir, O. J. Phys. Chem. B 2005, 109, 6972-
6981.
(13) Yoshikawa, S.; Shinzawa-Itoh, K.; Nakashima, R.; Yaono, R.;
Yamashita, E.; Inoue, N.; Yao, M.; Jei-Fei, M.; Libeu, C. P.;
Mizushima, T.; Yamaguchi, H.; Tomizaki, T.; Tsukihara, T. Science
1998, 280, 1723-1729.
(16) Collman, J. P.; Boulatov, R.; Sunderland, C. J.; Fu, L. Chem. ReV.
2004, 104, 561-588.
(17) Karlin, K. D.; Kim, E. Chem. Lett. 2004, 33, 1226-1231.
(18) Collman, J. P.; Sunderland, C. J.; Berg, K. E.; Vance, M. A.; Solomon,
E. I. J. Am Chem. Soc. 2003, 125, 6648-6649.
(19) Liu, J. G.; Naruta, Y.; Tani, F. Angew. Chem., Int. Ed. 2005, 44, 1836-
1840.
(14) Blomberg, M. R. A.; Siegbahn, P. E. M.; Wikstro¨m, M. Inorg, Chem.
2003, 42, 5231-5243.
(15) Buse, G.; Soulimane, T.; Dewor, M.; Meyer, H. E.; Blu¨ggel, M. Protein
Sci. 1999, 8, 985-990.
3890 Inorganic Chemistry, Vol. 46, No. 10, 2007

Further Insights into [(F8TPP)FeIII-(O22-)-CuII(TMPA)]+
(III) demonstrating that an η2:η1 FeIII-(O22-)-CuII coordi-
nation is a viable structure, at least in a high-spin complex.26
These systems employ binucleating superstructured ligands,
i.e., hemes with covalently linked chelates for copper. We
have also studied closely related systems, including O2-
adducts which lead to high-spin µ-peroxo FeIII-(O22-)-CuII
complexes [(6L)FeIII-(O22-)-CuII]+ (IV)20,22 and [(2L)FeIII-
(O22-)-CuII]+ (V).23,24 In combination, studies on these
complexes reveal that the O-O bond strength (as indicated
by νO-O) can be dramatically lowered (by ∼40-50 cm-1)
by changing the peroxo-binding from a tetradentate to
tridentate Cu-chelate and this also changes the peroxo
structure to µ-η2:η2 as observed for complex V,23,24 and also
for an ‘untethered’ analogue VI (Chart 1).25
Scheme 1
Thus, such investigations involving systematic changes in
heme/Cu ligand environment provide insight into the O2-
binding process and properties of the resulting new types of
dioxygen adducts. To these ends, we have also studied heme/
Cu/O2 chemistry of the 1:1 mixture of (F8TPP)FeII (1a)27
and [(TMPA)CuI(MeCN)]+ (1b)28-30 (Scheme 1). As a close
analogue of [(6L)FeIII-(O22-)-CuII]+ (IV) (Chart 1), we
reported the preliminary characterization of the high-spin
heme/Cu µ-peroxo complex, [(F8TPP)FeIII-(O22-)-
CuII(TMPA)]+ (3),21 which forms in lieu of heme-only or
Cu-only products prior to its thermal transformation to the
structurally characterized µ-oxo species [(F8TPP)FeIII-O-
CuII(TMPA)]+ (4).31 Deeper insights in the structure of
peroxo complex 3 have recently come from both X-ray
absorption spectroscopy (i.e., EXAFS) and DFT calcula-
tions,32 showing that the bound peroxide assumes a
µ-η2:η1 geometry like that of Naruta’s complex (III, Chart
1). The computational work also reveals insights into the
bonding in 3 and the magnetic orbitals which mediate
the antiferromagnetic coupling between the high-spin
iron(III) heme and the Cu(II)-TMPA moiety via the
peroxide bridging ligand.32
(20) Ghiladi, R. A.; Huang, H. W.; Moe¨nne-Loccoz, P.; Stasser, J.;
Blackburn, N. J.; Woods, A. S.; Cotter, R. J.; Incarvito, C. D.;
Rheingold, A. L.; Karlin, K. D. J. Biol. Inorg. Chem. 2005, 10, 63-
77.
(21) Ghiladi, R. A.; Hatwell, K. R.; Karlin, K. D.; Huang, H.-w.; Moe¨nne-
Loccoz, P.; Krebs, C.; Huynh, B. H.; Marzilli, L. A.; Cotter, R. J.;
Kaderli, S.; Zuberbu¨hler, A. D. J. Am. Chem. Soc. 2001, 123, 6183-
6184.
(22) Ghiladi, R. A.; Ju, T. D.; Lee, D.-H.; Moe¨nne-Loccoz, P.; Kaderli,
S.; Neuhold, Y.-M.; Zuberbu¨hler, A. D.; Woods, A. S.; Cotter, R. J.;
Karlin, K. D. J. Am. Chem. Soc. 1999, 121, 9885-9886.
(23) Kim, E.; Helton, M. E.; Lu, S.; Moe¨nne-Loccoz, P.; Incarvito, C. D.;
Rheingold, A. L.; Kaderli, S.; Zuberbu¨hler, A. D.; Karlin, K. D. Inorg,
Chem. 2005, 44, 7014-7029.
(24) Kim, E.; Shearer, J.; Lu, S.; Moe¨nne-Loccoz, P.; Helton, M. E.;
Kaderli, S.; Zuberbu¨hler, A. D.; Karlin, K. D. J. Am Chem. Soc. 2004,
126, 12716-12717.
(25) Kim, E.; Helton, M. E.; Wasser, I. M.; Karlin, K. D.; Lu, S.; Huang,
H.-w.; Moe¨nne-Loccoz, P.; Incarvito, C. D.; Rheingold, A. L.;
Honecker, M.; Kaderli, S.; Zuberbu¨hler, A. D. Proc. Natl. Acad. Sci.
U.S.A. 2003, 100, 3623-3628.
(26) Chishiro, T.; Shimazaki, Y.; Tani, F.; Tachi, Y.; Naruta, Y.; Karasawa,
S.; Hayami, S.; Maeda, Y. Angew. Chem., Int. Ed. 2003, 42, 2788-
2791.
(27) Ghiladi, R. A.; Kretzer, R. M.; Guzei, I.; Rheingold, A. L.; Neuhold,
Y.-M.; Hatwell, K. R.; Zuberbu¨hler, A. D.; Karlin, K. D. Inorg. Chem.
2001, 40, 5754-5767.
The focus of this report is to provide a completed picture
for the chemistry of 3 and of its formation (Scheme 1) with
results and experimental details not previously described: (i)
new resonance Raman spectroscopic measurements reveal
the presence of two inequivalent metal-oxo stretching
vibrations consistent with the η2:η1 peroxo bridging structure
deduced from EXAFS spectroscopic and computational
results,32 (ii) 2H NMR spectroscopy employing a deuterated
F8TPP porphyrinate illustrates the clean direct reduced-
to-peroxo-to-oxo complex conversion and 19F NMR spec-
troscopy provides additional new characterization of the
system, (iii) new Mo¨ssbauer spectroscopic measurements
provide parameters which define the system and provide
insights into the electronics/bonding of [(F8TPP)FeIII-
(O22-)-CuII(TMPA)]+ (3), and a detailed analysis is pro-
vided, and (iv) dynamics information is provided via low-
temperature stopped-flow kinetics for the reduced-to peroxo
conversion in EtCN as solvent. The combination of the
kinetic studies and Mo¨ssbauer spectroscopy further provides
unequivocal evidence for the presence of a heme-superoxo
(28) Tyekla´r, Z.; Jacobson, R. R.; Wei, N.; Murthy, N. N.; Zubieta, J.;
Karlin, K. D. J. Am. Chem. Soc. 1993, 115, 2677-2689.
(29) Karlin, K. D.; Lee, D.-H.; Kaderli, S.; Zuberbu¨hler, A. D. Chem.
Commun. 1997, 475-476.
(30) Karlin, K. D.; Kaderli, S.; Zuberbu¨hler, A. D. Acc. Chem. Res. 1997,
30, 139-147.
(31) Karlin, K. D.; Nanthakumar, A.; Fox, S.; Murthy, N. N.; Ravi, N.;
Huynh, B. H.; Orosz, R. D.; Day, E. P. J. Am. Chem. Soc. 1994, 116,
4753-4763.
-
intermediate, (solvent)(F8TPP)FeIII-(O2 ) (2a, Scheme 1)
formed prior to the formation of 3.
(32) del Rio, D.; Sarangi, R.; Chufa´n, E. E.; Karlin, K. D.; Hedman, B.;
Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 2005, 127, 11969-
11978.
Inorganic Chemistry, Vol. 46, No. 10, 2007 3891

Ghiladi et al.
holders as above, the solvent allowed to evaporate under an argon
stream at room temperature, and immediately placed into the mass
spectrometer.
Experimental Procedures
Materials and Methods. (F8TPP)FeII (1a),27,33 (F8TPP-d8)FeII
(1a-d8),27,33 and [(TMPA)CuI(MeCN)](ClO4) (1b)28 were pre-
pared per literature precedent. (F8TPP)57FeII (1a-57Fe) was
prepared in an identical manner to 1a, but employing 57FeIICl2
(see Supporting Information) in lieu of FeIICl2. Ultrapure grade
dioxygen (99.994%, WSC Gas), breakseal flasks of 18O2 and 16/18O2
(Icon Stable Isotopes), and 57Fe metal (95.3%, AMT) were
purchased as indicated. All other reagents and solvents were
purchased from commercial sources and were of available
reagent quality unless otherwise stated. Acetonitrile (MeCN)
was distilled directly from CaH2 under argon and stored over
molecular sieves. Solvents and solutions were deoxygenated by
either repeated freeze-pump-thaw cycles (5×), or by bubbling
of argon (>25 min) directly through the solution. Solid samples
were stored and transferred, and samples for spectroscopic char-
acterization were prepared in an MBraun LabMaster 130 inert
atmosphere (<1 ppm O2, <1 ppm H2O) glovebox under nitrogen
atmosphere.
NMR spectra were measured on a Varian NMR instrument at
61 (2H) or 376 MHz (19F). Chemical shifts are reported as δ values
calibrated to natural abundance deuterium solvent peaks (2H) or as
δ values referenced to an external standard of R,R,R-trifluorotoluene
at -63.73 ppm (19F). Low-temperature UV-visible spectral studies
were performed on a Hewlett-Packard 8453A diode array spec-
trometer equipped with a variable-temperature dewar and cuvette
assembly.34,35 Molar extinction coefficients are reported as an
average of a minimum of three trials.
Low-Temperature NMR Spectroscopy/Evans Method Mag-
netic Moment Measurements. Multinuclear (1H, 2H, and 19F)
NMR spectroscopic and magnetic moment measurements for 3 were
performed as previously described for [(6L)FeIII-(O22-)-CuII].20
Measurements on the decomposition product 4 were accomplished
by warming the NMR instrument to room temperature to facilitate
decay of the dioxygen adduct (as monitored by NMR), followed
by recooling and recording of the H, H NMR spectra at low
temperature. Evans method data were analyzed using a nuclearity
of 1 (i.e., one combined/coupled magnetic center) in the magnetic
moment equations.20
Resonance Raman Spectroscopy of [(F8TPP)FeIII-(O22-)-
CuII(TMPA)](ClO4) (3). 1a (20 mg) and 12 mg of 1b were
dissolved in 10 mL of deoxygenated MeCN in a glovebox. The
resulting solution was divided into aliquots and transferred to
septum-sealed NMR tubes (∼0.5 mL per tube). As a control, one
NMR tube was directly frozen in liquid N2 and sealed under inert
atmosphere. The remaining tubes were placed in a dry ice/MeCN
cold bath, and oxygenated using (i) 16O2, (ii) 18O2 (95.0 atom %
18O or greater), and (iii) mixed-isotope 16/18O2 gas (1:2:1 of 16O2/
16/18O2/18O2). Oxygenations were performed using a 5 mL Hamilton
gastight syringe equipped with a three-way syringe purge valve
attached to an argon/vacuum line of a Schlenk manifold. The
breakseal flasks containing the isotopically labeled O2 were frozen
in a dry ice/MeCN cold bath during the extraction of O2 to freeze
out any water contaminant. The oxygenated samples were frozen
in liquid N2 and sealed under their isotopically labelled O2
atmosphere, thereby allowing the headspace of the samples to be
checked directly by resonance Raman for their isotopic purity and
consistency.
Resonance Raman (rR) spectra obtained at OHSU (P. M.-L.)
resulted from 413-nm excitation as described previously.20-25 rR
spectroscopy with 676-nm excitation (E.I.S., Stanford University)
was performed with a Princeton Instruments ST-135 back-
illuminated CCD detector on a Spex 1877 CP triple monochromator
with 1200, 1800, and 2400 grooves/mm holographic spectrograph
gratings. Excitation was provided by a Coherent I90C-K Kr+ ion
laser. Samples were run at 77 K in a liquid N2 finger Dewar
(Wilmad).
Stopped-Flow Spectrophotometry. Preparation and handling
of sample solutions of 1a and 1b (1:1 mixtures in reagent grade
propionitrile), along with collection of data and analysis, were
carried out as described for other heme and/or copper sys-
tems.21,23,25,27 A total of 112 individual runs were used for the final
calculations (cf. the Eyring plot supplied in the Supporting
Information). Data collection times varied between 0.7 and 169 s
in the temperature range of -90.4 to -65.1 °C.
1
2
Dioxygen Evolution from 3. An alkaline pyrogallol solution
was used to detect if dioxygen was evolved upon the
decomposition of 3 to 4 in a manner identical to published
protocols20,23 and as described in further detail in the Supporting
Information. The number of equivalents of O2 evolved were
computed on the basis of a sample of 3 formed upon low-
temperature oxygenation of (F8TPP)FeII/[(TMPA)CuI(MeCN)]-
(ClO4) {1a/1b ) 40 mg/24 mg} in 10 mL of MeCN. The published
calibration curve,20 absorbance (400 nm) ) (0.0716 × mL O2) +
0.026, was used to determine the amount of O2 released upon
thermal decomposition of 3 to 4. When 3 was formed with excess
dioxygen, A400 measurements of the alkaline pyrogallol solution
(0.208 and 0.216) revealed 0.43 and 0.45 equiv of O2 generated,
respectively. Formation of 3 with a stoichiometric addition of
dioxygen led to an A400 of 0.194, corresponding to 0.40 equiv of
dioxygen evolved.
Mo1ssbauer Spectroscopy of 3. In the glovebox, to a 25 mL
Schlenk flask equipped with stir bar was added 20 mg of 1a-57Fe
and 12 mg of 1b (1:1 mixture) in 10 mL of acetonitrile.
Approximately 250 µL of this solution was transferred to a
Mo¨ssbauer sample holder, placed in an airtight flask, and im-
mediately frozen in liquid N2 upon removal from the glovebox.
The remainder of the solution was removed from the glovebox in
the Schlenk flask, placed in a -40 °C dry ice/MeCN cold bath,
Matrix-Assisted Laser Desorption Ionization Time-of-Flight
Mass Spectrometry (MALDI-TOF-MS). MALDI-TOF-MS mass
spectra were recorded on a Kratos Analytical Kompact MALDI 4
mass spectrometer equipped with a 337 nm nitrogen laser (20 kV
extraction voltage). Initial calibration of the instrument was
performed with a known standard (i.e., matrix) and then reconfirmed
on known samples (i.e., F8TPPH2 or (F8TPP)FeIII-OH). In the
glovebox, matrix-free samples were prepared upon addition of 20
mg of 1a, 12 mg of 1b, and 1 g of deoxygenated MeCN to a 10
mL Schlenk flask. A small aliquot of the solution was transferred
via pipet to a MALDI-TOF-MS sample holder and allowed to
evaporate, thereby adhering the sample to the target. For 3, the
remainder of the solution of 1a/1b was removed from the glovebox,
cooled to -45 °C (dry ice/MeCN cold bath), and oxygenated with
an excess of either 16O2 or 18O2 (96.5 atom % 18O or greater),
affording the 16O2 or 18O2 heme-peroxo-Cu complexes, respec-
tively. The dioxygen adducts were then transferred to sample
(33) Kopf, M.-A.; Neuhold, Y.-M.; Zuberbu¨hler, A. D.; Karlin, K. D. Inorg.
Chem. 1999, 38, 3093-3102.
(34) Karlin, K. D.; Haka, M. S.; Cruse, R. W.; Meyer, G. J.; Farooq, A.;
Gultneh, Y.; Hayes, J. C.; Zubieta, J. J. Am. Chem. Soc. 1988, 110,
1196-1207.
(35) Karlin, K. D.; Cruse, R. W.; Gultneh, Y.; Farooq, A.; Hayes, J. C.;
Zubieta, J. J. Am. Chem. Soc. 1987, 109, 2668-2679.
3892 Inorganic Chemistry, Vol. 46, No. 10, 2007

Further Insights into [(F8TPP)FeIII-(O22-)-CuII(TMPA)]+
and oxygenated via direct O2 bubbling, thereby forming 3. This
solution was then added by pipet to a Mo¨ssbauer sample holder
prechilled over CO2(s). The sample was then frozen and stored in
liquid N2 prior to data collection.
Results and Discussion
Dioxygen Reactivity of 1a/1b: UV-Visible Spectros-
copy. As reported earlier,21 upon bubbling of dioxygen
directly through a 1:1 solution of 1a and 1b {λmax (ꢀ, mM-1
cm-1) ) 414 (145), 421 (Soret: 220), 526 (20.2) nm}, UV-
visible spectroscopic monitoring in nitrile solvent revealed
the irreversible formation of a new species {λmax (ꢀ, mM-1
cm-1) ) 412 (Soret: 112), 558 (12.5) nm} (Figure S1)45
having remarkable thermal stability (t1/2(RT) ≈ 20 min in
MeCN), which we have formulated as the peroxo-level
dioxygen adduct 3 (Scheme 1). When oxygenation was
performed at low temperature, e.g., -40 °C for MeCN, the
stability of 3 was increased to over 24 h. Thermal decom-
position (RT) of 3 led to formation of the previously31
structurally characterized µ-oxo complex 4 {λmax (ꢀ, mM-1
cm-1) ) 433 (Soret: 146), 553 (13.1) nm}.46
Consistent with the formulation of 3 as a µ-peroxo bridged
heme-Cu complex (as depicted in Scheme 1) was the
determination of the dioxygen uptake of 1a/1b (1:1 mixture),
utilizing a spectrophotometric titration methodology we have
described;20,23 further details are provided here.45 As reported
earlier,21 full formation of 3 was accomplished with an
overall oxygenation stoichiometry of 0.9-1.0 equiv O2 per
1:1 mixture of 1a/1b, suggesting the presence of an intra-
molecular O2-adduct (Figure S2).45 Furthermore, the UV-
visible spectral features and oxygenation behavior of 3 are
Mo¨ssbauer spectra were recorded either on a weak-field
(50 mT removable permanent magnet) Mo¨ssbauer spectrometer
equipped with a Janis 8 DT variable-temperature cryostat or on a
strong-field Mo¨ssbauer spectrometer equipped with a Janis 12
CNDT/SC SuperVaritemp cryostat containing an 8-T supercon-
ducting magnet. Both spectrometers were operated in a constant
acceleration mode in a transmission geometry. The zero velocity
of the Mo¨ssbauer spectra is referred to the centroid of the room-
temperature spectrum of a metallic iron foil. Both spectrometers
were equipped with temperature controls that allow the sample to
be stablilized at any temperature between 1.5 and 300 K.
Computational Details. Nomal coordinate analysis (NCA) was
performed on a simplified Fe-O2-Cu complex by using QCPE
program 576 by M. R. Peterson and D. F. McIntosh, which involves
solution of the secular equation |FG - λE| ) 036,37 by the
diagonalization procedure of Miyazawa.38 The calculations were
based on a general valence force field. Force constants were refined
with the nonlinear optimization routine of the simplex algorithm
according to Nelder and Mead.39 The input geometry was taken
from DFT calculations. The mass of Fe and Cu atoms was increased
to 103 and 101 g‚mol-1, respectively, to reflect the effects of the
porphyrin and TMPA ligands on their displacements. Such an
effective mass increase is necessary to properly reproduce experi-
mental results. A similar approach has been used by us in the study
of binuclear peroxo copper complexes containing the TMPA
ligand.40
very similar to its tethered analogue IV (Chart 1) {λmax
)
418 (Soret), 561 nm, MeCN},20,22 which has been shown to
contain such a bound peroxide species.
Density functional theory (DFT) calculations of the model
complex [(P)FeIII-(O22-)-CuII(TMPA)]+ were performed using the
Gaussian 98-03 package of programs41 at the BP86 functional42,43
using the 6-311G* basis set for Fe, Cu, and O, 6-31G* for N, and
6-31G for C and H. The structure was characterized as an energy
minimum by diagonalization of the analytically computed Hessian
(vibrational frequencies calculations). Molekel was used to visualize
the results of DFT calculations.44
While the oxygenation stoichiometry for formation of 3
is consistent with an intramolecular O2-adduct, it alone does
not completely rule out the possibility of a heme-only {i.e.,
(PFeIII)n-(O2n-), n ) 1: 414-418 (Soret), 536-537 nm; n
) 2: 414-418 (Soret), 535-536 nm}27 or copper-only
{(CuII)n-(O2n-), n ) 1: 410 nm; n ) 2: 525, 610 nm}28,30,47
species involving 1a or 1b, respectively, despite the large
UV-visible spectral differences observed. Definitive as-
signment of 3 as a mixed-metal µ-peroxo O2-adduct was
provided by rR spectroscopy, vide infra.
(36) Wilson Jr., E. B.; Decius, J. C.; Cross, P. C. Molecular Vibrations;
Dover Publications: New York, 1980.
(37) Woodward, L. A. Introduction to the Theory of Molecular Vibrations
and Vibrational Spectroscopy; Claredon Press: Oxford, 1972.
(38) Miyazawa, T. J. Chem. Phys. 1958, 29, 246.
MALDI-TOF-MS of 3. MALDI-TOF mass spectrometry
proved to be amenable to the study of 3 (Figure 2). In
addition to the observation of 1a {812 m/z, (M + H+)+}
(39) Nelder, J. A.; Mead, R. Comput. J. 1965, 7, 308.
(40) Henson, M. J.; Vance, M. A.; Zhang, C. X.; Liang, H.-C.; Karlin, K.
D.; Solomon, E. I. J. Am. Chem. Soc. 2003, 125, 5186-5192.
(41) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross,
J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann,
R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski,
J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels,
A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A.
G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko,
A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.;
Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.;
Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.;
Pople, J. A. Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford,
CT, 2004.
-
and 1b {354 m/z, [M - (ClO4 )]+}, oxygenation of the
reduced complexes (F8TPP)FeII/[(TMPA)CuI(MeCN)]+ (1a/
1b ) 1:1) with 16O2 resulted in the formation of a new peak
(1239 m/z), corresponding to the molecular cation of complex
3 and a molecule of acetonitrile, [(F8TPP)FeIII-(O22-)-CuII-
(TMPA)(MeCN)]+. This peak shifted upon 18O2 substitution
to 1243 m/z, a shift of 4 amu, as expected for the 18O2-adduct
(i.e., peroxo complex) formulated. With the results of the
rR,21 Mossbauer (vide infra), and EXAFS32 spectroscopic
studies indicating a five-coordinate high-spin ferric heme (the
fifth ligand being the bound peroxide), acetonitrile coordina-
(45) See Supporting Information.
(46) Nanthakumar, A.; Fox, S.; Karlin, K. D. J. Chem. Soc. Chem. Commun.
1995, 499-501.
(47) Karlin, K. D.; Wei, N.; Jung, B.; Kaderli, S.; Niklaus, P.; Zuberbu¨hler,
A. D. J. Am Chem. Soc. 1993, 115, 9506-9514.
(42) Becke, A. D. Phys. ReV. A 1988, 38, 3098-3100.
(43) Perdew, J. P. Phys. ReV. B 1986, 33, 8822-8824.
(44) Portman, S.; Luth, H. P. Chimia 2000, 54, 766-770.
Inorganic Chemistry, Vol. 46, No. 10, 2007 3893

Ghiladi et al.
Figure 2. MALDI-TOF mass spectra of [(F8TPP)FeIII-(O22-)-
CuII(TMPA)]+ (3), formed from 16O2 (left) and 18O2 (right).
Figure 4. Resonance Raman spectra of [(F8TPP)FeIII-(O22-)-
CuII(TMPA)]+ (3), formed by oxygenation of (F8TPP)FeII/[(TMPA)-
CuI(MeCN)]+ (1a/1b ) 1:1) in MeCN at - 40 °C using 16O2 (red) and
18O2 gas (blue; offset by 2500 counts). All spectra were obtained at liquid
nitrogen temperature with a 676-nm excitation in MeCN solvent.
presence of a five-coordinate high-spin ferric heme were
observed, suggesting that 3 lacks a sixth axial base ligand.
The ν(O-O) of 3 {808 cm-1; ∆16O2/18O2 ) 46 cm-1;
∆16O2/16/18O2 ) 23 cm-1} is in the range expected for a
peroxide-level dioxygen adduct (Table 1). The possibilities
of 3 being either the copper-only intermolecular peroxide
adduct [{(TMPA)CuII}2-(O22-)]2+ [ν(O-O) ) 832 cm-1
{∆16O2/18O2 ) 44 cm-1, EtCN}],48 or a heme-peroxo-heme
complex are ruled out by the rR and NMR data (vide infra),
respectively. Additional relevant heme and/or Cu metal-
bound peroxides with available vibrational data (Table 1)
include (i) mononuclear anionic η2-heme-peroxo complexes,
[(P)FeIII-(O22-)]-, of Valentine and co-workers;49,50 (ii)
homobinuclear dicopper(II)-peroxo complexes (i.e., [{(L)-
CuII}2-(O22-)]2+) of Karlin and co-workers for the copper
chelates commonly used in heme-Cu complexes;40,48,51,52 (iii)
heterobinuclear heme-peroxo-Cu complexes, including
III-VI (see Introduction).20-26 Comparison of the vibrational
data in Table 1 readily shows two significant traits for heme-
peroxo-Cu complexes. First, the use of untethered ligands
(as in 3, VI) leads to O-O stretching frequencies ∼20 cm-1
higher than their tethered analogues (IV, V), suggesting that
the geometric constraints (the details of which are not yet
understood) imposed by binucleating ligands strains or
distorts the peroxide’s metal binding, whereas mononuclear
components lead to FeIII-(O22-)-CuII cores which assume
a more favorable geometry, leading to stronger O-O bonds.
Second, the µ-η2:η1 binding mode (as in 3 (Scheme 1 and
Introduction) and III (Chart 1)) exhibits a stronger O-O
bond [ν(O-O) greater by ∼40 cm-1] than for the µ-η2:
Figure 3. Resonance Raman spectra (black) of [(F8TPP)FeIII-(O22-)-
CuII(TMPA)]+ (3) in MeCN solution using 676-nm excitation at liquid
nitrogen temperature. The calculated effect of splitting in the 810 cm-1
O-O stretch is compared with the 920 cm-1 solvent vibration.
tion to the heme is excluded. Further, the EXAFS study32
of 3 does not support MeCN binding to the copper
either, and we suggest that the increase in mass of the
molecular cation observed (in mass spectrometry, vide supra)
for 3 corresponds to the solution species plus one solvent
molecule.
rR Spectroscopy of 3. The intermediate 3 was initially21
characterized by rR spectroscopy (Figure S3)45 using a 413-
nm excitation coinciding with its Soret absorption maximum.
Spectra from different dioxygen isotope labeling experiments
were normalized with the intense phenyl vibrational mode
at 898 cm-1. An 16O2 (Figure S3A) minus 18O2 (Figure S3C)
difference spectrum (Figure S3D)45 singles out a unique
isotope sensitive band at 808 cm-1, downshifted by -46
cm-1 with 18O2. To allow a definitive assignment of the 808
cm-1 feature to a peroxo ν(O-O) stretching vibration rather
than a ferryl-oxo ν(FedO) mode, the experiments were
repeated using mixed-isotope 16O18O, derived from a 1:2:1
mixture of 16O2/16O18O/18O2 (Figure S3B).45 In addition to
the positive signal at 808 cm-1 and its negative 18O2-
counterpart at 762 cm-1, the isotopic mixture minus pure
16O2 difference spectrum reveals a new signal at ∼785 cm-1,
assigned to ν(16O-18O) (Figure S3E).45 The observation of
this third signal excludes the ν(FedO) possibility, since only
two stretching frequencies would be expected for a ferryl
species. Porphyrin ligand vibrations consistent with the
(48) Baldwin, M. J.; Ross, P. K.; Pate, J. E.; Tyekla´r, Z.; Karlin, K. D.;
Solomon, E. I. J. Am. Chem. Soc. 1991, 113, 8671-8679.
(49) Selke, M.; Sisemore, M. F.; Valentine, J. S. J. Am. Chem. Soc. 1996,
118, 2008-2012.
(50) Burstyn, J. N.; Roe, J. A.; Miksztal, A. R.; Shaevitz, B. A.; Lang, G.;
Valentine, J. S. J. Am. Chem. Soc. 1988, 110, 1382-1388.
(51) Liang, H.-C.; Karlin, K. D.; Dyson, R.; Kaderli, S.; Jung, B.;
Zuberbu¨hler, A. D. Inorg. Chem. 2000, 39, 5884-5894.
(52) Obias, H. V.; Lin, Y.; Murthy, N. N.; Pidcock, E.; Solomon, E. I.;
Ralle, M.; Blackburn, N. J.; Neuhold, Y.-M.; Zuberbu¨hler, A. D.;
Karlin, K. D. J. Am. Chem. Soc. 1998, 120, 12960-12961.
3894 Inorganic Chemistry, Vol. 46, No. 10, 2007

Further Insights into [(F8TPP)FeIII-(O22-)-CuII(TMPA)]+
Table 1. Resonance Raman Data for Various Peroxo-Level Dioxygen Adducts
complex
ν(16O-16O) (-∆16O2/18O2, -∆16O2/16/18O2), cm-1
ref
[(F8TPP)FeIII-(O22-)-CuII(TMPA)]+ (3)
[(TMP)FeIII-(O22-)-CuII(5-MeTPA)]+ (III)
[(6L)FeIII-(O22-)-CuII]+ (IV)
808 (46, 23) µ-η2:η1
790 (44) µ-η2:η1
788 (44, 21)
this work
26
20,22
24
25
53,54
53
55
50
49
48
[(2L)FeIII-(O22-)-CuII]+ (V)
747 (40, 17) µ-η2:η2
767 (41), 752 (45) µ-η2:η2
803 (44, 22)
[(F8TPP)FeIII-(O22-)-CuII(LMe2N)]+ (VI)
[(TPP)FeIII-(O22-)-CuII(TPA)]+
[(TPP)FeIII-(O22-)-CuII(5-MeTPA)]+
[(DHIm)(a3TACNaAcr)FeIII-(O22-)-CuII]+
[(OEP)FeIII-(O22-)]-
793 (42)
758 (18)
808 (47) η2-peroxo
[(F20TPP)FeIII-(O22-)]-
802 (n/a) η2
[{(TMPA)CuII}2-(O22-)]2+
832 (44)
[{(LMe2N)CuII}2-(O22-)]2+
729 (40) µ-η2:η2
730 (39), µ-η2:η2 588 (28), bis-µ-oxo core
40
52
[{(MePY2)CuII}2-(O22-)]2+
at 810 cm-1 and a solvent peak at 920 cm-1 are observed in
this region. In order to determine the minimum splitting that
could be observed for the O-O stretch under experimental
conditions, we simulated the effect of splitting the energy
position of these peaks by increasing cm-1. Split spectra were
generated from the experimental 16O2 spectrum (with noise)
in three steps: (i) two new spectra were created from the
16O2 peak that were shifted to higher and lower energy by
equal wavenumbers; (ii) the new shifted spectra were added
and renormalized to the original intensity; (iii) this “split”
spectrum was overlaid with the original, experimental data
to see if there were differences in shape resolvable over the
noise level. In the case of the sharp solvent peak at 920 cm-1,
an effect on the vibrational band shape of the signal is
apparent for a 3 cm-1 splitting and two distinctive peaks are
observed for 7 cm-1 (Figure S4).
However, the O-O stretch at 810 cm-1 is significantly
broader. The effect of energy splitting on the shape of the
signal is only apparent above 10 cm-1, and two distinctive
peaks are only observed for a 20 cm-1 spectrum (Figure S5).
Therefore, a µ-η2:η1 bonding mode for the peroxide
moiety in 3 cannot be ruled out on the basis of the lack of
splitting of the 785 cm-1 O-O stretch of the mixed-isotope
experiment (Figure S3)45 since the predicted splitting
(<3 cm-1 according to NCA and DFT) is too small to be
observed for such a broad vibration under the experimental
conditions.
Table 2. Experimental (Resonance Raman 676-nm Excitation), (NCA),
and [DFT] Calculated Vibrational Frequencies (cm-1) for the µ-η2:η1
Fe-O2-Cu Core Structure
16O(1)-
18O(1)-
16O(1)-
18O(1)-
mode
ν(OO)
16O(2)
18O(2)
18O(2)
16O(2)
810
(810)
[885]
533
(536)
[488]
512
(516)
[429]
(156)
[110]
765
(765)
[832]
513
(509)
[471]
493
(490)
[422]
(155)
[108]
786
(790)
786
(787)
[853]
[853]
a
a
ν(Fe-O2)
(520)
(531)
[471]
[488]
a
a
ν(Fe-O1-Cu)a
ν(Fe-O1-Cu)s
(503)
[429]
(155)
[108]
(495)
[422]
(155)
[109]
a A broad peak is observed centered at 511 cm-1
.
η2 side-on peroxo geometry. As has been previously dis-
cussed in detail,5,17,23-25 the origins of this difference appear
to copper-ligand dependent: tetradentate copper chelates
yield µ-η2:η1 dioxygen adducts, whereas tridentate ones lead
to µ-η2:η2 O2-binding. Further studies will be required to
greater elucidate the interplay between structure, bonding,
and vibrational properties of iron-peroxo-copper centers.
Absence of Splitting of the Mixed-Isotope Signal in 3.
The absence of a splitting of the mixed-isotope signal at 785
cm-1 (vide supra) would be consistent with a symmetrically
bound peroxide ligand (see diagram), in either a µ-1,2 (A,
end-on) or µ-η2:η2 (B, side-on) geometry, and would seem
Presence of Two M-O Stretching Frequencies in 3. In
order to obtain further insight into its structure, complex 3
was also characterized by rR spectroscopy using a 676-nm
excitation. Under these experimental conditions, the por-
phyrin Raman peaks are not resonance enhanced and the
O-O vibration (at 810 cm-1) and two different M-O
vibrations at (533 and 511 cm-1) are easily observed (see
Figure 4 and Table 2). All three bands are 18O2 isotope
sensitive, with the 810 cm-1 band shifting down by 45
cm-1, while the bands at 533 and 511 cm-1 are shifted
down in energy by -20 and -19 cm-1, respectively,
allowing their assignment as M-O modes. The difference
in intensity between the 16O and 18O sample is clearly
observed using other excitation lines and it could be
related to sample differences due to different methods of
preparation. As an example, data using 775-nm excitation
are included in Figure S6. Further insights derive from NCA,
see below.
to rule out a µ-η2:η1 (C) or µ-1,1 bridging mode (D).
However, we have recently shown using a combination
of both X-ray absorption spectroscopy (i.e., EXAFS) and
DFT calculations that the bound peroxide in 3 assumes a
µ-η2:η1 geometry (structure C, diagram) like that of Naruta’s
established (by X-ray) complex (III, Chart 1).32 On the other
hand, the predicted splitting-of-the-splitting of the O-O
vibration (for 16O-18O) according to the DFT calculation
and NCA (see below) is 1 and 3 cm-1, respectively, which
may not be spectroscopically (rR) observable.
Figure 3 shows the rR data obtained using a 676-nm
excitation in the 750-1000 cm-1 region. The O-O stretch
Inorganic Chemistry, Vol. 46, No. 10, 2007 3895

Ghiladi et al.
Table 3. Experimental and Calculated Vibrational Frequencies (cm-1),
Potential Energy Distribution (%, for the 16O2 Isotopomer), and Selected
Force Constants (mdyn/Å) from the NCA of the µ-η2:η1 Fe-O2-Cu
Core Structurea
potential energy distribution
mode
∆r(O-O) ∆r(Fe-O2) ∆r(Fe-O1) ∆r(Cu-O1)
ν(O-O)
80
1
20
2
κ(O-O)
2.66
6
75
15
1
8
10
33
49
5
14
32
48
ν(Fe-O2)
ν(Fe-O1-Cu)a
ν(Fe-O1-Cu)s
force constants
κ(Fe-O2)
2.49
κ(Cu-O1)
1.43
κ(Fe-O1)
1.50
a O1 bridges both Fe and Cu, while O2 is only iron-bound.
NCA in 3. To obtain a quantitative description of the
observed vibrational modes, 810, 533, and 513 cm-1, a
NCA was performed using the xyz coordinates of the µ-η2:
η1 Fe-O2-Cu core structure derived from our published
DFT calculations.32 Only in-plane vibrations were
included, and a generalized force field was used. The O-O
stretch, the three possible nonequivalent M-O stretches, and
their interaction constants were floated to fit the experimental
vibrational frequencies and their isotopes shifts. The calcu-
lated vibrational frequencies are presented in Table 2. Force
constants, normal mode eigenvectors, and potential energy
distribution are summarized in Table 3, while the resulting
atom-displacement matrix is given as Supporting Information
(Table S1 and Figure S7).45
Figure 5. 2H NMR spectra (MeCN, 233 K) of (F8TPP-d8)FeII/[(TMPA)-
CuI(MeCN)]+ (1a-d8/1b) (a), [(F8TPP-d8)FeIII-(O22-)-CuII(TMPA)](ClO4)
(3-d8) (b), and [(F8TPP-d8)FeIII-O-CuII(TMPA)]+ (4-d8) (c). The
signal at δ 1.94 is due to the natural abundance of deuterium in the MeCN
solvent.
to heme oxidation and spin state as deduced from pyrrole
resonance assignments56-58 and from an independent NMR
spectroscopic study carried out on the dioxygen reactivity
of 1a.27
The calculated frequencies and isotope shifts are consistent
with the experimental data. The calculated PEDs and normal
mode eigenvectors indicate that the 810 cm-1 mode is indeed
a dominant ν(O-O) vibration (80% O-O + 8% Fe-O1 +
6% Fe-O2 + 4% Cu-O1, where O1 is the oxygen atom
that is bound to both the copper and iron; O2 is the oxygen
atom that is only bound to the iron). The 533 cm-1 mode
corresponds mainly to the ν(Fe-O2) vibration (75% Fe-O2
+ 14% Cu-O1 + 10% Fe-O1 + 1% O-O), while the 513
cm-1 mode can be assigned to an asymmetric Fe-O1-Cu
vibration (33% Fe-O1 + 32% Cu-O1 + 20% O-O + 14%
Fe-O2). NCA predicts the presence of a low-energy mode
at 156 cm-1 that can be assigned to the symmetric Fe-O1-
Cu stretch (49% Fe-O1 + 48% Cu-O1 + 2% O-O + 1%
Fe-O2). Our attempts to locate such low-energy bands
experimentally have been unsuccessful, as this energy
region is obscured by vibrations from the solvent (MeCN).
Attempts to locate this low-energy absorption using a
solid sample of 3 were also unsuccessful. However, the DFT
frequencies calculated for the model complex [(P)FeIII-
(O22-)-CuII(TMPA)]+ are in agreement with the results
from the NCA analysis showing a symmetric Fe-O1-Cu
stretch at 110 cm-1. In summary, the rR spectra and
NCA support the presence of a µ-η2:η1 peroxide ligand in
complex 3.
2
The H NMR spectra of the oxygenation reaction of the
1:1 mixture of (F8TPP)FeII and [(TMPA)CuI(MeCN)]+ (1a/
1b) in MeCN at - 40 °C is shown in Figure 5. Prior to
addition of dioxygen, the reduced complexes 1a/1b exhibit
a pyrrole resonance at δ 9.2 (Figure 5a), indicative of a low-
spin (S ) 0) six-coordinate (bis-MeCN) ferrous heme at low
temperature, and supported by the Evans method magnetic
moment measurement {µB ) 0; no paramagnetically shifted
solvent peak was observed downfield from the reference/
capillary solvent resonance}. Direct injection of O2 into the
chilled NMR tube leads to the formation of 3, with a
downfield shifting of the pyrrole resonance to δ 68 (Figure
5b), indicative of a high-spin ferric heme56-58 (and supported
1
by the rR study, vide supra). We have also noted in the H
NMR study of 3 that upfield-shifted peaks were also seen at
δ -11 and -20. We have reported this characteristic pattern
of downfield-shifted pyrrole resonances and upfield shifted
peaks in similar (P)FeIII-X-CuII (X ) O22-, O2-) systems
(53) Naruta, Y.; Sasaki, T.; Tani, F.; Tachi, Y.; Kawato, N.; Nakamura,
N. J. Inorg. Biochem. 2001, 83, 239-246.
(54) Sasaki, T.; Nakamura, N.; Naruta, Y. Chem. Lett. 1998, 351-352.
(55) Collman, J. P.; Herrmann, P. C.; Boitrel, B.; Zhang, X.; Eberspacher,
T. A.; Fu, L.; Wang, J.; Rousseau, D. L.; Williams, E. R. J. Am. Chem.
Soc. 1994, 116, 9783-9784.
2H NMR Spectroscopy of 3. By selectively deuterating
(56) Goff, H. M. The Porphyrins (Goff review article). In Iron Porphyrins;
Lever, A. B. P., Gray, H. B., Eds.; Addison-Wesley: Reading, MA,
1983; Part 1; pp 239-281.
2
the pyrrole hydrogens, a full H NMR spectroscopic inves-
tigation was performed to complement the previously
(57) Walker, F. A.; Simonis, U. Proton NMR Spectroscopy of Model
Hemes. In Biological Magnetic Resonance; Berliner, L. J., Reuben,
J., Eds.; Plenum Press: New York, 1993; Vol. 12/NMR of Paramag-
netic Molecules, pp 133-274.
1
reported H NMR study21 and provided unambiguous as-
signment of the pyrrole hydrogens. Such assignments are
possible due to the wealth of information available pertaining
(58) Walker, F. A. Porphyrin Handbook 2000, 5, 81-183.
3896 Inorganic Chemistry, Vol. 46, No. 10, 2007

Further Insights into [(F8TPP)FeIII-(O22-)-CuII(TMPA)]+
Table 4. 19F NMR Chemical Shifts of Heme and Heme-Cu Dioxygen
Adducts
temp,
complex
δF (ppm)
-113
solvent
K
(F8TPP)FeII/[(TMPA)CuI(MeCN)]+
CD3CN 233
(1a/1b)
[(F8TPP)FeIII-(O22-)-CuII(TMPA)]+ (3) -92
CD3CN 233
CD3CN 233
THF-d8 193
CD2Cl2 193
[(F8TPP)FeIII-O-CuII(TMPA)]+ (4)
(F8TPP)FeII (1a) 27
-94, -98
-116
-107
-
27
(S)(F8TPP)FeIII-(O2
(iron-superoxo)
)
-111, -113 THF-d8 193
2- 27
[(F8TPP)FeIII]2-(O2
)
-106, -109 CD2Cl2 193
(Fe-peroxo-Fe)
(F8TPP)FeIII-OH 27 (iron-hydroxide)
[(F8TPP)FeIII]2-O 62 (Fe-oxo-Fe)
-83
-79
THF-d8 193
CD2Cl2 193
-103, -109 CD2Cl2
RT
having S ) 2 spin states,20-23,59,60 including 4 {µB ) 5.1 (
0.2, Evans method, MeCN, -40 °C},59 with previously
assigned downfield-shifted pyrrole resonances (δ 83), and
upfield-shifted TMPA-chelate peaks at δ -11 (pyridyl 5-H)
and -28 (pyridyl 3-H) (Figure 5c).59,60 The S ) 2 spin state
arises from the antiferromagnetic coupling of the S ) 5/2
heme center to an S ) 1/2 copper(II) moiety, through the
bridging peroxo (in 3) or oxo ligand (in 4). Supporting an
overall S ) 2 spin state for 3 is the magnetic susceptibility
measurement (µB ) 5.1 ( 0.2; Evans method, MeCN,
-40 °C), in close agreement with the spin-only value of 4.9
µB for four unpaired electrons. Ligand and/or spin-state
dynamics might account for the broadness of the pyrrole
resonance in 3, although Mo¨ssbauer spectroscopic data (vide
infra) do not give any hint of this possibility. The analogous
tethered complex, IV, exhibits Curie-Weiss behavior (linear)
in the limited temperature range of 228-288 K.20
Figure 6. 19F NMR spectra (CD3CN, 233 K) of (F8TPP)FeII/[(TMPA)-
CuI(MeCN)]+ (1a/1b) {δF ) -113 ppm} (a), [(F8TPP)FeIII-(O22-)-CuII-
(TMPA)](ClO4) (3) {δF ) -92 ppm} (b), and [(F8TPP)FeIII-O-CuII-
(TMPA)]+ (4) {δF ) -94, -98 ppm} (c).
Our independent 2H NMR characterization of the dioxygen
adducts of pyrrole-deuterated 1a-d8 excludes the possibility
that the resonances observed for 3 might be derived from a
-
mononuclear heme-superoxo complex, PFeIII-(O2 ) {δpyrrole
) ∼7-10 ppm},27 or a heme-peroxo-heme binuclear
complex, PFeIII-(O22-)-FeIIIP {δpyrrole ) ∼15-19 ppm}.27,61
19F NMR Spectroscopy of 3. We have previously
undertaken a 19F NMR spectroscopic study of the heme-
only 1a oxygenation reaction, forming heme-superoxo, 2a,
and heme-peroxo-heme, [(F8TPP)FeIII]2-(O22-), dioxygen
adducts.27 The results of this study indicated that both the
diamagnetic (S ) 0) heme-superoxo complex and the
antiferromagnetically coupled heme-peroxo-heme O2-ad-
duct have sharp fluorine resonances between δ -109 and
-113 (Table 4). The mixture of the reduced complexes 1a/
1b also shows a single sharp fluorine resonance {Figure 6a:
δF ) -113 ppm}, implying a symmetric environment for
the eight fluorine substituents. These findings are in contrast
to those observed for 3, which exhibits a single (broad)
fluorine resonance for the S ) 2 heme-peroxo-Cu complex
at δ -92 (Figure 6b). Additionally, 4 also gives rise to broad
Figure 7. Mo¨ssbauer spectra of 57Fe-labeled [(F8TPP)FeIII-(O22-)-CuII-
(TMPA)]+ (3). The spectra (hashed marks) were recorded at 4.2 K in zero
field (A), and in a field of 50 mT applied parallel to the γ-beam (B). The
solid line in A is the simulation of a quadrupole doublet with ∆EQ ) 1.14
mm/s, δ ) 0.57 mm/s.
fluorine resonances (Figure 6c: δ -94 and -98). The
asymmetry of complex 4 can rationalize the differences in
the fluorine environment of the 2,6-difluorophenyl substit-
uents, leading to these two observed fluorine signals: one
resonance accounts for the fluorines on the ‘proximal’ µ-oxo
ligand side of the porphyrin macrocycle, the second reso-
nance being attributed to those fluorines on the ‘distal’ side.
However, the lack of asymmetry in 3 when compared to 4
is not readily apparent, and we speculate that either (i) the
differences in the structure of 3 versus 4 give rise to the
(59) Nanthakumar, A.; Fox, S.; Murthy, N. N.; Karlin, K. D. J. Am. Chem.
Soc. 1997, 119, 3898-3906.
(60) Kim, E.; Kamaraj, K.; Galliker, B.; Rubie, N. D.; Moe¨nne-Loccoz,
P.; Kaderli, S.; Zuberbu¨hler, A. D.; Karlin, K. D. Inorg. Chem. 2005,
44, 1238-1247.
(61) Balch, A. L. Inorg. Chim. Acta 1992, 198-200, 297-307.
Inorganic Chemistry, Vol. 46, No. 10, 2007 3897

Ghiladi et al.
field (Figure 7A) exhibits a sharp quadrupole doublet with
parameters (∆EQ ) 1.14 mm/s and δ ) 0.57 mm/s) that are
indicative of high-spin ferric compounds. The fact that the
isomer shift of 3 is at the high end of the range of values
(0.3-0.7 mm/s) observed for high-spin ferric compounds and
is significantly larger than those (0.33-0.45 mm/s) observed
specifically for high-spin ferric hemes63 is consistent with
binding of an electron-rich peroxide ligand to the Fe center.50
Similarly large values (0.60-0.66 mm/s) of isomer shift have
been observed for a few peroxide-bound high-spin ferric
compounds, including Naruta’s structurally characterized
heme-peroxo-copper analogue (III, Chart 1; ∆EQ ) 1.17
mm/s and δ ) 0.56 mm/s (77 K)),26 an η2-peroxo-iron(III)-
porphyrinate,50 and several non-heme peroxodiferric inter-
mediates found in methane monooxygenase,64,65 variants of
the R2 subunit of E. coli ribonucleotide reductase,66 stearoyl-
acyl carrier protein desaturase,67 and ferritin.68 When a
spectrum of 3 is recorded at 4.2 K in a weak applied field
of 50 mT, the quadrupole doublet becomes significantly
broadened (Figure 7B). This broadening is a consequence
of magnetic hyperfine interaction (albeit small), the presence
of which indicates that 3 is paramagnetic. The observation
that a weak applied field induces only a small magnetic
hyperfine interaction in 3 indicates further that 3 has an
integer-spin ground state (vide infra), in accordance with the
conclusion made from the NMR and susceptibility measure-
ments that the S ) 5/2 FeIII heme and the S ) 1/2 CuII site
are spin coupled.
Figure 8. Field-dependent Mo¨ssbauer spectra of 57Fe-labeled [(F8TPP)-
FeIII-(O22-)-CuII(TMPA)]+ (3). The spectra (hashed marks) were recorded
at 4.2 K in a field of 0.25 (A), 1 (B), 2 (C), 4 (D), or 8 T (E) applied
parallel to the γ-beam. The solid lines are theoretical simulations using the
parameters listed in Table 5.
In order to gain further insight into the electronic state of
3, Mo¨ssbauer spectra in various magnetic fields were
recorded. Figure 8 shows the Mo¨ssbauer spectra of 3
recorded at 4.2 K in a field of 0.25 (A), 1 (B), 2 (C), 4 (D),
or 8 T (E) applied parallel to the γ-beam. The overall splitting
of the spectra, which is a measure of the strength of the
magnetic hyperfine interaction, increases gradually with the
applied field and reaches its full splitting at about 4 T. This
observed field dependence is typical of integer spin systems69
and is distinct from that of half-integer spin systems, of which
the spectra exhibit full magnetic splitting at weak applied
fields (for example, 50 mT). These field-dependent Mo¨ss-
bauer spectra of 3 were analyzed using the following spin
Hamiltonian,
observed differences in fluorine resonance signals, with
the µ-peroxo bridged complex having a somewhat more
equivalent environment for the fluorine substituents (one
signal), and the µ-oxo bridged system possessing more
fluorine-environment inequivalency (hence, two signals),
or (ii) ligand and/or spin-state dynamics lead to a single
(albeit broad) observed peak in the limited observable
temperature range where 3 is stable. For both the peroxo-
and oxo-bridged complexes, we suggest that their S ) 2 spin
state leads to a broadening and downfield-shifting of the
fluorine resonances to the region of δ -92 to -98, which
is at least 10 ppm more downfield than the correspond-
ing diamagnetic heme-only superoxo (2a) or peroxo com-
plexes, and closer to the 19F NMR chemical shift of the
ferric-hydroxy (S ) 5/2) (F8TPP)FeIII-OH complex (δF )
-79 to -82 ppm).27 The 19F NMR spectra observed
appear to be quite sensitive to the oxidation and spin
states of the complexes studied and may in the future
be utilized as an additional tool for gaining insight
into the electronic nature of such heme-containing
systems.
(62) Nanthakumar, A.; Goff, H. M. Inorg. Chem. 1991, 30, 4460-4464.
(63) Debrunner, P. G. In Iron Porphyrins; Lever, A. B. P., Gray, H. B.,
Eds.; VCH: New York 1990; pp 139-234.
(64) Liu, K. E.; Valentine, A. M.; Wang, D. L.; Huynh, B. H.; Edmondson,
D. E.; Salifoglou, A.; Lippard, S. J. J. Am. Chem. Soc. 1995, 117,
10174-10185.
(65) Liu, K. E.; Wang, D. L.; Huynh, B. H.; Edmondson, D. E.; Salifoglou,
a.; Lippard, S. J. J. Am. Chem. Soc. 1994, 116, 7465-7466.
(66) Bollinger, J. M.; Krebs, C.; Vicol, A.; Chen, S. X.; Ley, B. A.;
Edmondson, D. E.; Huynh, B. H. J. Am. Chem. Soc. 1998, 120, 1094-
1095.
(67) Broadwater, J. A.; Achim, C.; Munck, E.; Fox, B. G. Biochemistry
1999, 38, 12197-12204.
Mo1ssbauer Spectroscopy of 3. 57Fe-labeled 1a was
synthesized (see Experimental Section) and used in the
oxygenation reaction with 1b to form 3 (MeCN, 233 K) for
further characterization by Mo¨ssbauer spectroscopy. The
Mo¨ssbauer spectrum of 3 recorded at 4.2 K and in zero-
(68) Pereira, a. S.; Small, W.; Krebs, C.; Tavares, P.; Edmondson, D. E.;
Theil, E. C.; Huynh, B. H. Biochemistry 1998, 37, 9871-9876.
(69) Surerus, K. K.; Hendrich, M. P.; Christie, P. D.; Rottgardt, D.; Orme-
Johnson, W. H.; Munck, E. J. Am. Chem. Soc. 1992, 114, 8579-
8590.
3898 Inorganic Chemistry, Vol. 46, No. 10, 2007

Further Insights into [(F8TPP)FeIII-(O22-)-CuII(TMPA)]+
H ) He + Hhf
Table 5. Mo¨ssbauer Spectroscopic Parameters for
[(F8TPP)57FeIII-(O22-)-CuII(TMPA)]+ (3) and
[(F8TPP)57FeIII-O-CuII(TMPA)]+ (4)
(1)
where
complex
3
4
Da (cm-1
E/D
δ (mm/s)
)
2.9
5.6
S(S + 1)
E
D
He ) D S2z -
+
(S2x - S2y) + âS‚g‚H (2)
[
]
0.15
0.57
-1.14
0.35
25
59
118
-25.7
-22.9
-21.0
0.01
0.48
-1.17
0
0
0
3
∆EQ (mm/s)
η
and
Rb (°)
eQVzz
âb (°)
I(I + 1)
η
3
Hhf )
I2z -
+
(I2x - I2y) +
γb (°)
0
[
]
4
3
Axx/gnân (T)
Ayy/gnân (T)
Azz/gnân (T)
-23.7
-23.7
-23.7
( S ‚A‚I - gnânH‚I) (3)
The Hamiltonian He represents the zero-field splitting terms
(the first three terms in eq 2) and the electronic Zeeman
interaction term (âS‚g‚H) of the total electronic spin of the
ground state, S ) 2, and Hhf represents the electronic nuclear
hyperfine interactions, which include the quadrupole interac-
tion (the first three terms in eq 3), the magnetic hyperfine
interaction between the electronic spin, S, and the 57Fe
nuclear spin, I ( S ‚A‚I), and the nuclear Zeeman interaction
(gnânH‚I). On the basis of the NMR and susceptibility results,
a system spin S ) 2 was used for the analysis of the
Mo¨ssbauer spectra of 3. As Mo¨ssbauer absorptions are arising
from 57Fe nuclear transitions, the spectra are basically
determined by Hhf of eq 3. For a paramagnetic system, the
magnetic hyperfine interaction S ‚A‚I is generally the
dominating term in Hhf and, thus, has a decisive effect on
the observed Mo¨ssbauer spectra. For high-spin ferric hemes,
the magnetic hyperfine tensor, Aheme, is rather isotropic and
varies within a narrow range (-21 to -18 T).63 For a spin-
coupled system, such as 3, the apparent A tensor of the
system is related to the intrinsic A tensor, such as Aheme, by
a simple factor depending on the nature of the spin
coupling.70 Consequently, the acceptable values for A in eq
3 are restricted within a limited range. The spectra therefore
depend strongly on the spin expectation value, S , which
in turn is determined by the zero-field splitting parameters
D and E/D and the applied magnetic field, H. In other words,
the parameters D and E/D can be determined accurately from
the analysis of the field-dependent spectra of 3. A global
analysis including all the spectra recorded at 4.2 K in various
applied fields (0-8 T) yielded the spin Hamiltonian param-
eters for 3 (Table 5). Simulations using these parameters are
plotted as solid lines overlaid with the experimental data in
Figure 8. The intrinsic values for the zero-field splitting
parameter and the magnetic hyperfine tensor of the high-
spin ferric heme (Dheme and Aheme, respectively) can be
obtained from the values of the corresponding parameters
determined for the spin-coupled system by standard spin
projection techniques,70
a Calculated for an S ) 2 spin system. b Euler angles rotating the principal
axis frame of the electric field gradient tensor into that of the zero-field
splitting tensor.
The magnitude and small anisotropy observed for Aheme are
consistent with the high-spin ferric assignment. The zero-
field splitting parameter, Dheme ) 2.2 cm-1, determined for
3, however, is significantly smaller than those (4-22 cm-1)
observed for high-spin ferric heme compounds. It is therefore
worthy of note that one other heme compound that was
reported to have an unusually small D of 0.85 cm-1 is also
2- -
a peroxide-bound high-spin ferric heme, [(OEP)FeIIIO2
]
[E/D ) 0.29].50 The rhombicity parameter, E/D ) 0.15,
determined for 3 is also unusual for heme compounds, which
are axially symmetric and generally exhibit an E/D value of
about 0.02. In this respect, it is interesting to compare the
E/D value of 3 with that of the decay product, 4. We have
recorded the spectra of 4 in MeCN at 4.2 K in various applied
fields ranging from 0 to 8 T (Figure S8). Analysis of these
data using eqs 1-3 yielded the parameters listed in Table 5.
These parameters are very similar to those published earlier
for the crystalline form of 431 and for a similar oxo-bridged
heme-Cu complex, [(OEP)Fe-O-Cu(Me6-tren]+.71 The
E/D value of 0.03 determined for 4 is normal for heme
compounds and can be understood as a consequence of the
linear Fe-O-Cu bridge, for which the π-donor interactions
between the Fe and O along the heme normal (dxz(Fe)-px-
(O) and dyz(Fe)-py(O)) are expected to be similar. The large
rhombicity observed for complex 3 may thus suggest a side-
on or bent end-on Fe-peroxide binding mode, with the side-
on geometry consistent with the results of the EXAFS
spectroscopic and DFT structural studies of 3.32
Kinetics of the 1a/1b Oxygenation Reaction. The
oxygenation reaction of 1a/1b was performed in propioni-
trile72 due to the known previously investigated kinetics and
thermodynamics of formation of the resulting dioxygen
adducts of the individual mononuclear iron27 and copper30,73
components. At -90 °C, addition of O2 to 1a/1b resulted in
Dheme ) (3/4)D ) 2.2 cm-1
(71) Kauffmann, K. E.; Goddard, C. A.; Zang, Y.; Holm, R. H.; Mu¨nck,
E. Inorg. Chem. 1997, 36, 985-993.
(72) In the text, the tmpa-Cu(I) complex is always associated with MeCN,
e.g., as [(TMPA)CuI(MeCN)]+ (1b), since it is synthesized in and
isolated from MeCN. Descriptions of solution chemistry in propionitrile
solvent thus obviously deal with the chemistry of [(TMPA)-
CuI(EtCN)]+, including in the kinetics investigations carried out in
EtCN, cf. Scheme 2.
and
Aheme/gnân ) (6/7)(A/gnân) ) (-22.0, -19.6, -18.0) T
(70) Bencini, A.; Gatteschi, D. Electron Paramagentic Resonance of
Exchange Coupled Systems; Springer-Verlag: Berlin, 1990.
(73) Zhang, C. X.; Kaderli, S.; Costas, M.; Kim, E.-i.; Neuhold, Y.-M.;
Karlin, K. D.; Zuberbu¨hler, A. D. Inorg. Chem. 2003, 42, 1807-1824.
Inorganic Chemistry, Vol. 46, No. 10, 2007 3899

Ghiladi et al.
Scheme 2
Figure 9. Top: Time-dependent oxygenation of (F8TPP)FeII/[(TMPA)-
CuI(MeCN)]+ (1a/1b); [(F8TPP)FeII] ) 8.463 × 10-5 M, [{(TMPA)CuI-
(MeCN)}+] ) 8.469 × 10-5 M ,[O2] ) 5.02 × 10-3 M in EtCN {183 K,
ttotal ) 30 s}.72 Inset: λ ) 528 nm (red), decay of (F8TPP)FeII; λ ) 558
nm (black), formation of [(F8TPP)FeIII-(O22-)-CuII(TMPA)]+ (3); λ )
537 nm (green), formation and decay of (EtCN)(F8TPP)FeIII-(O2-) (2a).
Bottom: Calculated spectra for species in the 1a/1b/O2 reaction, following
kinetics analysis for determination of k4. All other rate constants (Scheme
2) and spectra of Cu-only species were fixed at published values (see text).
decay of the 528 nm band (reduced 1a) and formation of a
transient species (λmax ) 537 nm) within the first ∼30 ms
(Figures 9 and S9).45 The spectral features and kinetic
behavior for the formation of this intermediate matched those
previously observed for the 1a/O2 reaction in EtCN despite
the presence of 1b, establishing this 537 nm species as the
mononuclear low-spin heme-superoxo intermediate 2a
(νO-O ) 1178 cm-1, νFe-O ) 568 cm-1, in tetrahydrofuran
solvent).21,25 Additional support for this conclusion is pro-
vided by Mo¨ssbauer spectroscopy (vide infra).
While our spectroscopic monitoring (500-700 nm) was
unable to confirm or deny the presence of a copper(II)-
superoxide intermediate (because of the dominance of intense
heme absorptions), our previous stopped-flow kinetic stud-
ies29,73 of the oxygenation reaction of only 1b in EtCN72
confirm its presence {2b, λmax ) 410 nm}, formed on the
millisecond time scale, prior to its eventual combination with
1b to give the end-on-bound µ-1,2-peroxo complex [{(TMPA)-
CuII}2-(O22-)]2+ {λmax ) 525 nm}. Thus, it is concluded
that both 1a and 1b undergo simultaneous and independent
reversible oxygenation reactions even when mixed together
as in the present system, Scheme 2 (k1/k-1, k2/k-2, and k3/
k-3 processes).
Decay of the heme-superoxo intermediate 2a occurred
over the next 20-25 s (Figures 9 and S9),45 with formation
of a λmax ) 558 nm species, consistent with 3 (Figure 9).
Kinetic analysis of this 537 f 558 nm transformation,
corresponding to the reaction of 2a with 1b to give 3, was
fit to the model described in Scheme 2, where all rate
constants except k4 were fixed according to their known
values from independent measurements in EtCN.25,27,29,73 This
approach simplified the system leading to the determination
of the rate constant k4 (see Eyring plot for k4, Figure S10).45
The results are given in Table 6. Overall, the activation
enthalpy (∆Hq ) 27 kJ/mol) determined for formation of 3
is somewhat lower than that observed for the same complex
in acetone as solvent (∆Hq ) 45 kJ/mol), or values
determined for other complexes (Table 6). There are too
many variables (i.e., solvent and metal-solvate dissociation
chemistry, copper-ligand denticity, inter- vs intramolecular
reactions) to make firm conclusions concerning the reason
for this. Nevertheless, a likely explanation is the CuI-TMPA
moiety, with its tetradentate chelate, is significantly more
reducing than CuI in any of the other systems (Table 6),
providing facilitated reaction with the FeIII-O2- intermediate
resulting in formation of heme-peroxo-Cu complex 3. In
IV (Chart 1), the copper environment is a distorted tetraden-
tate (more like tridentate),20,74 while for V-VI, tridentate
ligands are involved. Recent75 redox potential measurements
with copper(I) complexes of these tri- and tetradentate
pyridylalkylamine-containing ligands support the supposition
concerning relative reducing power.
Mo1ssbauer Evidence Supporting the Presence of a
Heme-Superoxo Intermediate. The oxygenation reaction
of 1a/1b in acetone at -80 °C was also studied by Mo¨ssbauer
spectroscopy. Figure 10A displays a Mo¨ssbauer spectrum
of the starting material 1a/1b before introduction of oxygen.
The spectrum, which was recorded at 4.2 K in the absence
of a magnetic field, shows a quadrupole doublet with
parameters (∆EQ ) 2.66 mm/s and δ ) 0.93 mm/s) that are
typical for high-spin ferrous heme compounds. Since high-
spin ferrous heme compounds are generally five-coordinate,
this observation suggests the presence of an axial ligand,
(74) Ju, T. D.; Ghiladi, R. A.; Lee, D.-H.; van Strijdonck, G. P. F.; Woods,
A. S.; Cotter, R. J.; Young, J., V. G.; Karlin, K. D. Inorg. Chem.
1999, 38, 2244-2245.
(75) Osako, T.; Karlin, K. D.; Itoh, S. Inorg. Chem. 2005, 44, 410-415.
3900 Inorganic Chemistry, Vol. 46, No. 10, 2007

Further Insights into [(F8TPP)FeIII-(O22-)-CuII(TMPA)]+
Table 6. Kinetic Parameters for the Formation of Heme-Peroxo-Copper Complexes from Reaction of Initially Formed (P)FeIII-(O2-) Intermediates
with Ligand-Cu(I) Moieties
heme-peroxo-copper complex
[(F8TPP)FeIII-(O22-)-CuII(TMPA)]+ (3)a
k
∆Hq (kJ/mol)
26.5 ( 0.6
∆Sq (J/mol‚K)
-11 ( 3
k4 ) (2.77 ( 0.05) × 104 (M-1 s-1) (183 K)b
k4 ) (7.2 ( 0.1) × 104 (M-1 s-1) (193 K)b
k4 ) (1.70 ( 0.05) × 105 (M-1 s-1) (203 K)b
0.07 s-1 (183 K)c,21
45 ( 1
-19 ( 6
-29 ( 2
0.32 s-1 (193 K)c,21
[(6L)FeIII-(O22-)-CuII]+ (IV)22
2.55 s-1 (183 K)d
37.4 ( 0.4
9.62 s-1 (193 K)d
31.9 s-1 (203 K)d
[(2L)FeIII-(O22-)-CuII]+ (V)23
(2.74 ( 0.04) × 101 s-1 (168 K)e
(2.55 ( 0.03) × 102 s-1 (183 K)e
(1.70 ( 0.05) × 103 s-1 (198 K)e
(1.28 ( 0.03) × 104 (M-1 s-1) (168 K)e
(1.16 ( 0.01) × 105 (M-1 s-1) (183 K)e
(7.6 ( 0.1) × 105 (M-1 s-1) (198 K)e
36.6 ( 0.4
36.2 ( 0.2
5 ( 2
[(F8TPP)FeIII-(O22-)-CuII(LMe2N)]+ (VI) 25
54 ( 1
a This work. b EtCN solvent. c Acetone, pseudo-first-order rate constants for [metal-complex] ) 2.71 × 10-4 M, [O2] ) 2.55 × 10-3 M at RT (calculated
from the thermodynamic data in ref 21). d Acetone solvent, calculated from the thermodynamic data in ref 22. e 6% EtCN/CH2Cl2.
the unreacted 1a. Doublet II accounts for 37% of the total
iron absorption and exhibits characteristic parameters (∆EQ
) 2.33 mm/s and δ ) 0.31 mm/s) that are commonly
observed for O2-bound ferrous heme compounds, such as
oxygenated myoglobin and hemoglobin, of which the oxida-
- 76
tion state may be formulated as FeIII-O2 . As a control,
we also reacted 1a with dioxygen in the absence of 1b in
acetone, known to give 2a.27 A Mo¨ssbauer spectrum of the
O2-reacted 1a sample recorded at 4.2 K in the absence of a
magnetic field (Figure 10C) showed a quadrupole doublet
identical to doublet II. Taken together, the Mo¨ssbauer data
presented here and the UV-visible stopped-flow data
presented above demonstrate unambiguously that a ferric
heme-superoxo 2a intermediate is present as a precursor
of 3 in the O2 reaction of 1a/1b (Scheme 1).
Thermal Rearrangement of 3 to 4. The thermal decom-
position of µ-peroxo complex 3 to µ-oxo complex 4 was
found to be slow {t1/2 ) 1016 ( 20 s; MeCN, 22 °C, 0.28
mM} as determined by benchtop UV-visible spectroscopy.21
The question then arises as to the fate of the second oxygen
atom apparently lost during this transformation. Does 3
undergo O-atom transfer to a substrate, i.e., solvent, exhibit-
ing monooxygenase activity? Or, as an alternative, is
dioxygen liberated during this decomposition process, similar
to the situation found for the analogue compound IV
(Chart 1)?20
Figure 10. Oxygenation reaction of (F8TPP)FeII/[(TMPA)CuI(MeCN)]+
(1a/1b) in acetone monitored by Mo¨ssbauer spectroscopy. The spectra were
recorded at 4.2 K and in the absence of a magnetic field. Spectrum A was
taken prior to the addition of O2, and spectrum B was recorded from a
sample frozen ∼10 s after the addition of oxygen. Reacting O2 with 1a in
the absence of 1b resulted in spectrum C, corresponding to quadrupole
doublet II of spectrum B, see text.
To answer the latter, alkaline pyrogallol, a known dioxygen
scavenger, was used in a colorometric assay (as previously
described) to detect and quantify evolved dioxygen gas.20,23
The results indicated 0.4-0.46 equiv of dioxygen were
evolved during the formation of 4. Thus, no loss of oxygen
(to solvent oxidation, perhaps) occurred. As depicted in
Scheme 1, 1 equiv of dioxygen reacts with reduced precur-
sors to form peroxo complex 3, but some type of dispro-
portionation reaction occurs in the thermal transformation
to µ-oxo complex 4, and 0.5 equiv of O2 is evolved. We
have not been able to obtain mechanistic insights into this
interesting reaction but have elsewhere20 provided a detailed
discussion about such disproportionation reactions, known
for heme and non-heme iron systems.
most likely a solvent molecule, for 1a in acetone. This
1
situation has already been described on the basis of H or
2H NMR spectroscopic studies.27 After the addition of
dioxygen, the sample was rapidly (∼10 s) frozen in liquid
nitrogen and investigated by Mo¨ssbauer spectroscopy. Figure
10B shows a spectrum of the oxygen-reacted sample recorded
at 4.2 K in zero field. Two well-resolved quadrupole
doublets, marked by brackets I and II, are observed. Doublet
I, accounting for 63% of the total iron absorption, exhibits
parameters (∆EQ ) 2.82 mm/s and δ ) 0.93 mm/s) that are
similar to those of the starting materials and thus represents
(76) Weiss, J. J. Nature 1964, 203, 182-183.
Inorganic Chemistry, Vol. 46, No. 10, 2007 3901

Ghiladi et al.
Summary/Conclusions
FeII/CuI complex(es)? Can similar heme-peroxo-Cu com-
plexes be formed from addition of (ligand)CuII (tetra- and
tridentate ligands) compounds to side-on peroxo species
[(P)FeIII-(O22-)]-?77 Can any such Fe-O2-Cu complexes
be subjected to conditions which lead to reductive O-O bond
cleavage, by addition of proton/electron sources, in order to
probe this reaction which is key to cytochrome oxidase
function and in general for metal ion mediated O2-activation?
Studies related to these matters are in progress.
We have demonstrated that oxygenation of a simple
mixture of the heme and copper complexes 1a and 1b forms
the heterobinuclear complex 3. This possesses an unsym-
metrical side-on/end-on µ-η2:η1-peroxo coordination, as
further supported by the present rR spectroscopic finding of
two distinct FeIII-O stretching vibrations. As previously
inferred from magnetic moment measurements (Evans
1
method), as well as the distinctive pattern of H NMR
Acknowledgment. We are grateful to the National
Institutes of Health (K.D.K., GM60353; R.J.C., GM54882;
P.M.-L., GM34468 to Prof. T. M. Loehr; B.H.H., GM47295;
E.I.S., DK31450) and the Swiss National Science Foundation
(A.D.Z.) for support of this research. D.R. thanks the sixth
Framework program of the EU for a MC-OIF fellowship.
downfield-shifted pyrrole and upfield-shifted pyridyl hydro-
gen resonances, 3 is an antiferromagnetically coupled (Fe(III)
and Cu(II)) S ) 2 spin system; as presented in this report,
2H NMR measurements along with Mo¨ssbauer spectroscopic
investigations confirm these conclusions. We also probed
the properties of 3 and 4 using 19F NMR spectroscopy;
a comparison of data to those known for other species
(e.g., heme-superoxo complex 2a and peroxo species
{(F8TPP)FeIII}2(O22-)) reveals useful empirical correlations
based on compound magnetic properties.
Stopped-flow spectrophotometry and Mo¨ssbauer spectros-
copy reveal the formation of a heme-superoxo complex (2a)
as a precursor to forming 3, a result seen previously for the
oxygenation of other PFeIICuI complexes, with dioxygen-
uptake experiments confirming that the formation of the
peroxide bridge in 3 occurs upon addition of 1 equiv (per
FeII/CuI heterobinuclear center) of dioxygen. Thermal trans-
formation of peroxo species 3 to µ-oxo complex 4 occurs
with concomitant release of 0.5 equiv of dioxygen; the
mechanism of this disproportionation and O-O bond cleav-
age reaction deserves further attention.
Supporting Information Available: Synthesis of (F8TPP)57FeII
(1a-57Fe), UV-visible spectra of the (F8TPP)FeII/[(TMPA)-
CuI(MeCN)]+ (1a/1b) oxygenation reaction (Figure S1), discussion
and data for the O2-uptake of 1a/1b by means of a UV-visible-
monitored spectrophotometric titration (Figure S2), resonance
Raman spectra of [(F8TPP)FeIII-(O22-)-CuII(TMPA)]+ (3) (Figure
S3), calculated effects of splitting the 810 cm-1 O-O stretch (Figure
S4), calculated effect of splitting in the 920 cm-1 solvent vibration
(Figure S5), resonance Raman spectra of 3 obtained with a 775
nm excitation (Figure S6), schematic representation of atom
displacement from the normal coordinate analysis of the µ-η2:η1
Fe-O2-Cu core structure (Figure S7), and atom displacement
matrix (for the 16O2 isotopomer) (Table S1), Mo¨ssbauer spectra of
[(F8TPP)FeIII-O-CuII(TMPA)]+ (4) (Figure S8), kinetic trace for
formation of (EtCN)(F8TPP)FeIII-(O2-) (2a) followed by its
transformation to [(F8TPP)FeIII-(O22-)-CuII(TMPA)]+ (3) (Figure
S9), Eyring plot for the formation of [(F8TPP)FeIII-(O22-)-
CuII(TMPA)]+ (3) in EtCN (Figure S10), and experimental details
for the alkaline pyrogallol detection of evolved dioxygen during
the thermal rearrangement of 3 to 4. This material is available free
of charge via the Internet at http://pubs.acs.org.
While possessing quite detailed characterization of the
chemistry of high-spin µ-peroxo complexes 3, close relative
IV (Chart 1), Naurta’s peroxo complex (III, Chart 1), and
even the analogues with tridentate ligands for copper (V and
VI, Chart 1), additional insights are warranted. For example,
can a high-spin (S ) 2) heme-peroxo-Cu complex be
converted to low-spin, either by addition of a strong field
axial base or through changes in the nature of the starting
IC061726K
(77) Chufa´n, E. E.; Karlin, K. D. J. Am Chem. Soc. 2003, 125, 16160-
16161.
3902 Inorganic Chemistry, Vol. 46, No. 10, 2007