Tridentate copper ligand influences on heme-peroxo-copper formation and properties: Reduced, superoxo, and μ-peroxo iron/copper complexes

Article abstract of DOI:10.1021/ic050446m

In cytochrome c oxidase synthetic modeling studies, we recently reported a new μ-η²:η²-peroxo binding mode in the heteronuclear heme/copper complex [(²L)Fe^III-(O ₂^2-)-Cu^II]⁺ (6) which is effected by tridentate copper chelation (J. Am. Chem. Soc. 2004, 126, 12716). To establish fundamental coordination and O₂-reactivity chemistry, we have studied and describe here (i) the structure and dioxygen reactivity of the copper-free compound (²L)Fe^II (1), (ii) detailed spectroscopic properties of 6 in comparisons with those of known μ-η²: η¹ heme-peroxo-copper complexes, (iii) formation of 6 from the reactions of [(²L)Fe^IICu^I]⁺ (3) and dioxygen by stopped-flow kinetics, and (iv) reactivities of 6 with CO and PPh₃. In the absence of copper, 1 serves as a myoglobin model compound possessing a pyridine-bound five-coordinate iron(II)-porphyrinate which undergoes reversible dioxygen binding. Oxygenation of 3 below -60°C generates the heme-peroxo-copper complex 6 with strong antiferromagnetic coupling between high-spin iron(III) and copper(II) to yield an S = 2 spin system. Stopped-flow kinetics in CH₂-Cl₂/6% EtCN show that dioxygen reacts with iron(II) first to form a heme-superoxide moiety, [(EtCN)(²L)Fe^III-(O₂^-) ...Cu^I(EtCN)]⁺ (5), which further reacts with Cu ^I to generate 6. Compared to those properties of a known μ-η²:η¹-heme-peroxo-copper complex, 6 has a significantly diminished resonance Raman ν(O-O) stretching frequency at 747 cm^-1 and distinctive visible absorptions at 485, 541, and 572 nm, all of which seem to be characteristics of a μ-η²: η²-heme-peroxo-copper system. Addition of CO or PPh₃ to 6 yields a bis-CO adduct of 3 or a PPh₃ adduct of 5, the latter with a remaining Fe^III-(O₂^-) moiety.

Full text of DOI:10.1021/ic050446m

Inorg. Chem. 2005, 44, 7014−7029
Tridentate Copper Ligand Influences on Heme
−Peroxo
−Copper
Formation and Properties: Reduced, Superoxo, and
Copper Complexes
µ-Peroxo Iron/
Eunsuk Kim,† Matthew E. Helton,†,‡ Shen Lu,§ Pierre Moe
1
nne-Loccoz,§ Christopher D. Incarvito,|
hler,‡ and Kenneth D. Karlin*,†
Arnold L. Rheingold,| Susan Kaderli,‡ Andreas D. Zuberbu
1
Department of Chemistry, Remsen Hall, Johns Hopkins UniVersity, 3400 North Charles Street,
Baltimore, Maryland 21218, Department of Chemistry, UniVersity of Basel, Basel, Switzerland,
Department of EnVironmental and Biomolecular Systems, OGI School of Science and Engineering
at Oregon Health & Science UniVersity (OHSU), BeaVerton, Oregon 97006, and Department of
Chemistry, UniVersity of Delaware, Newark, Delaware 19716
Received March 24, 2005
In cytochrome c oxidase synthetic modeling studies, we recently reported a new
µ
-
η2
:
η
2-peroxo binding mode in
2
-
the heteronuclear heme/copper complex [(2L)FeIII CuII]+ (6) which is effected by tridentate copper chelation
−(O2 )−
(J. Am. Chem. Soc. 2004, 126, 12716). To establish fundamental coordination and O2-reactivity chemistry, we
have studied and describe here (i) the structure and dioxygen reactivity of the copper-free compound (2L)FeII (1),
(ii) detailed spectroscopic properties of 6 in comparisons with those of known
µ- : −peroxo−copper
η2 η1 heme
complexes, (iii) formation of 6 from the reactions of [(2L)FeIICuI]+ (3) and dioxygen by stopped-flow kinetics, and
(iv) reactivities of 6 with CO and PPh3. In the absence of copper, 1 serves as a myoglobin model compound
possessing a pyridine-bound five-coordinate iron(II)
Oxygenation of 3 below 60 C generates the heme
coupling between high-spin iron(III) and copper(II) to yield an S
Cl2/6% EtCN show that dioxygen reacts with iron(II) first to form a heme
(O2
- ‚‚‚CuI(EtCN)]+ (5), which further reacts with CuI to generate 6. Compared to those properties of a known
1-heme
peroxo copper complex, 6 has a significantly diminished resonance Raman (O O) stretching
frequency at 747 cm- and distinctive visible absorptions at 485, 541, and 572 nm, all of which seem to be
−
porphyrinate which undergoes reversible dioxygen binding.
peroxo copper complex 6 with strong antiferromagnetic
2 spin system. Stopped-flow kinetics in CH2-
superoxide moiety, [(EtCN)(2L)FeIII
−
°
−
−
)
−
−
)
µ
-
η2
:
η
−
−
1
ν −
characteristics of a
3 or a PPh3 adduct of 5, the latter with a remaining FeIII
µ- :η −peroxo−copper system. Addition of CO or PPh3 to 6 yields a bis-CO adduct of
η2 2-heme
−
(O2-) moiety.
Introduction
CcO, where the four-electron/four-proton reduction of di-
oxygen to water takes place, contains a high-spin histidine-
ligated heme (heme a3) and a tricoordinated copper (CuB)
in close proximity (∼5 Å). Due to significant contributions
from high-resolution X-ray crystal structures3-11 and exten-
The dioxygen reactivity of small-molecule heme-copper
complexes has attracted many synthetic modeling chemists,
because such a reaction occurs at the active site of cyto-
chrome c oxidases (CcOs) for energy production in the
respiration of aerobic organisms.1,2 The binuclear center of
(3) Iwata, S.; Ostermeier, C.; Ludwig, B.; Michel, H. Nature 1995, 376,
660-669.
* To whom correspondence should be addressed. E-mail: karlin@jhu.edu.
Ph: 410-516-8027. Fax: 410-516-7044.
† Johns Hopkins University.
(4) Tsukihara, T.; Aoyama, H.; Yamashita, E.; Tomizaki, T.; Yamaguchi,
H.; Shinzawa-Itoh, K.; Nakashima, R.; Yaono, R.; Yoshikawa, S.
Science 1995, 269, 1069-1074.
‡ University of Basel.
(5) Tsukihara, T.; Aoyama, H.; Yamashita, E.; Tomizaki, T.; Yamaguchi,
H.; Shinzawa-Itoh, K.; Nakashima, R.; Yaono, R.; Yoshikawa, S.
Science 1996, 272, 1136-1144.
§ Oregon Health & Science University (OHSU).
| University of Delaware.
(1) Ferguson-Miller, S.; Babcock, G. T. Chem. ReV. 1996, 96, 2889-
2907.
(6) 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.
(2) Michel, H.; Behr, J.; Harrenga, A.; Kannt, A. Annu. ReV. Biophys.
Biomol. Struct. 1998, 27, 329-356.
7014 Inorganic Chemistry, Vol. 44, No. 20, 2005
10.1021/ic050446m CCC: $30.25
© 2005 American Chemical Society
Published on Web 08/26/2005

Cu Ligand Influences on Heme-Peroxo-Cu Formation
Scheme 1
Chart 1
sive kinetic-spectroscopic studies,12 there is now general
agreement on the O2-binding and reduction products forming
during the CcO catalytic cycle, Scheme 1. Dioxygen binds
to heme a3, forming the first spectroscopically detectable
intermediate, the “oxy” Fe-O2 adduct. The next observed
intermediate contains an O-O bond cleaved moiety, where
the fully reduced oxygen atoms derived from O2 are bound
to CuB and heme a3, generating the ferryl-oxo (i.e., FeIVd
O) and CuII-X (X ) OH- or H2O) species, “PM” (Scheme
1).13,14 However, a full understanding of the O-O bond
reductive cleavage process is lacking and many details and
insights remain to be determined; these include the role of
copper ion and the particulars of its (His)2(His-Tyr) ligation,
the possible involvement of a bridging peroxo group (Fe-
O-O-Cu) mechanism,15 peroxo structure (η2- or η1- to iron),
and proton count and protonation sites during and after O-O
cleavage.
complexes contain a tetradentate copper ligand TMPA
moiety, where TMPA ) tris(2-pyridylmethyl)amine. One
such compound from Naruta and co-workers was crystallo-
graphically characterized, demonstrating that the peroxide
was bound to iron and copper through an η2:η1 fashion.18 In
a recent paper,19 we reported a dramatically diminished O-O
stretching frequency which identified a new type of heme-
peroxo-copper complex with a η2:η2 peroxo binding ge-
ometry (Chart 1). This was achieved by simple change in
copper ligand denticity, thus exhibiting the critical nature
of the copper ligation in heme-(O2)-copper complexes, with
resulting O-O bond weakening.
In this study, we report details on the formation, physical
properties, and small-molecule (CO, PPh3) reactivities of the
heme-peroxo-copper species, [(2L)FeIII-(O22-)-CuII]+ (6)
(Chart 1). The results are discussed in comparison with other
known tetradentate heme-peroxo-copper complexes. To
compare the observed heme/Cu/O2 reactions with those of
heme/O2 chemistry, a single iron heme-only complex was
also prepared; the X-ray structure of a high-spin five-
coordinated ferrous heme and its O2-binding chemistry are
described.
As complementary to biochemical and biophysical re-
search, biomimetic chemistry has recently provided well-
defined Fe/Cu systems whose coordination chemistry, elec-
trochemical kinetic behavior, and O2-reaction chemistries are
new and serve as important guidelines to understand the CcO
active site chemistry.12,16 The exact role of CuB and the (His)2-
(His-Tyr)-CuB center in the heme/Cu/O2 reaction is espe-
cially interesting and important but not resolved. Whereas
some biomimetic electrocatalytic studies suggest that copper
mainly serves as an electron storage site but does not directly
participate in the catalysis,17 studies from our own and other
laboratories have shown that CuI or CuII can directly react
Results and Discussion
Design and Syntheses. To better understand the role of
copper ion and influence of copper ligand environment in
the heme/Cu/O2 chemistry, we chose a situation in which
the individual heme-only and copper-only chemistries were
already quite well understood.
In independent Cu/O2 chemistry, it is shown that [(LR)-
CuI]+ (L ) N,N-bis[(2-(2-pyridyl)ethyl]methylamine, R )
H, -OMe, or -NMe2 substituents on the pyridine para
positions) react with O2 to yield [{(LR)Cu}2(O2)]2+, (η2:η2-
peroxo)dicopper(II), and bis(µ-oxo)dicopper(III) equilibrium
mixtures (as represented by the dotted line O-O moiety in
Chart 2).20-22 These copper-dioxygen adducts perform two-
electron oxidations of exogenous substrates, where the more
-
with a heme-superoxide (i.e., FeIII-(O2 )) or a heme-
peroxide to yield bridged heme-peroxo-Cu moieties.12 The
majority of the reported discrete heme-peroxo-copper
(7) Ostermeier, C.; Harrenga, A.; Ermler, U.; Michel, H. Proc. Natl. Acad.
Sci. U.S.A. 1997, 94, 10547-10553.
(8) Harrenga, A.; Michel, H. J. Biol. Chem. 1999, 274, 33296-33299.
(9) Soulimane, T.; Buse, G.; Bourenkov, G. P.; Bartunik, H. D.; Huber,
R.; Than, M. E. EMBO J. 2000, 19, 1766-1776.
(10) Svensson-Ek, M.; Abramson, J.; Larsson, G.; Tornroth, S.; Brzezinski,
P.; Iwata, S. J. Mol. Biol. 2002, 321, 329-339.
(11) Abramson, J.; Riistama, S.; Larsson, G.; Jasaitis, A.; Svensson-Ek,
M.; Laakkonen, L.; Puustinen, A.; Iwata, S.; Wikstrom, M. Nature
Struct. Biol. 2000, 7, 910-917.
(12) Kim, E.; Chufa´n, E. E.; Kamaraj, K.; Karlin, K. D. Chem. ReV. 2004,
104, 1077-1133.
(18) Chishiro, T.; Shimazaki, Y.; Tani, F.; Tachi, Y.; Naruta, Y.; Karasawa,
S.; Hayami, S.; Maeda, Y. Angew. Chem., Int. Ed. 2003, 42, 2788-
2791.
(19) 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.
(20) 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.
(21) Zhang, C. X.; Liang, H.-C.; Kim, E.-i.; Shearer, J.; Helton, M. E.;
Kim, E.; Kaderli, S.; Incarvito, C. D.; Zuberbu¨hler, A. D.; Rheingold,
A. L.; Karlin, K. D. J. Am Chem. Soc. 2003, 125, 634-635.
(13) Hansson, O.; Karlsson, B.; Aasa, R.; Vanngard, T.; Malmstrom, B.
G. EMBO J. 1982, 1, 1295-7.
(14) The source of the fourth electron varies depending on the redox states
of the enzyme.12
(15) Blomberg, M. R. A.; Siegbahn, P. E. M.; Wikstroem, M. Inorg. Chem.
2003, 42, 5231-5243.
(16) Collman, J. P.; Boulatov, R.; Sunderland, C. J. Porphyrin Handb. 2003,
11, 1-49.
(17) Collman, J. P.; Boulatov, R.; Sunderland, C. J.; Fu, L. Chem. ReV.
2004, 104, 561-588.
Inorganic Chemistry, Vol. 44, No. 20, 2005 7015

Kim et al.
Chart 2
Figure 1. X-ray crystal structure of (2L)FeII (1). Selected bond distances
(Å): Fe-N(5), 2.229; Fe-N(porphyrinate), 2.055-2.080. See also Table
1.
complexes (2L-d8)FeII (1-d8) and [(2L-d8)FeIICuI]+ (3-d8) were
prepared by following the identical procedures for 1 or 3,
except deuterated ligand 2L-d8 was used for their syntheses,
in which 2L-d8 was synthesized from the deuterated precursor
F6(NO2)TPPH2-d8 (see Experimental Section).
Scheme 2
Structure of (2L)FeII (1) by X-ray Crystallography and
NMR Spectroscopy. X-ray-quality single crystals of (2L)-
FeII (1) were obtained from a concentrated acetone solution
of 1 at -15 °C. One of the pyridyl arms of the copper chelate
was found to be bound to the iron center, giving a high-spin
five-coordinated iron complex, Figure 1. These types of
synthetic ferrous porphyrinates with an axial base have long
been studied to aid in understanding the chemistry of heme
proteins such as hemoglobin and myoglobin.17,24 However,
due to the difficulties in preparation of a simple iron
porphyrinate containing a single axial base,25 only a few
examples of the X-ray crystal structures of such complexes
are known. The structure of 1 (Figure 1 and Figure S1)
compares well with other five-coordinate high-spin ferrous
heme complexes, as summarized in Table 1. The iron atom
is 0.42 Å displaced from the mean 24-atom porphyrin plane
(mean deviation is 0.19 Å) toward the pyridyl axial base.
The Fe-Npy bond length is 2.229 Å, which is longer (0.1-
0.2 Å) than other related complexes which contain an
imidazole axial ligand (Table 1). Although Fe-N bond
lengths may slightly vary due to intrinsic differences between
pyridine and imidazole as the N-donor, the elongated Fe-N
bond in 1 may be better explained by the repulsion between
the porphyrin plane and the tether arm that is connected
through the 2-position of the pyridine ring. The steric bulk
of the -CH2CH2- tether also leads to a tilt of Fe-Npy axis
that is 13° off from the porphyrin normal axis, most probably
to maximize the distance between the methylene hydrogens
and the porphyrin core atoms. The distance of the closest
hydrogen from the tether arm to a nitrogen of the porphyrin
core is 2.66 Å (see Supporting Information Figure S1 for a
diagram of 1 from a different perspective).
electron-donating R-group provides for faster oxidation rates.
On the other hand, in the (F8)FeII/O2 reactions, two different
O2 intermediates can be generated depending on the solvent
employed. In coordinating solvents such as THF, propioni-
trile, and acetone, the oxygenation reaction yields oxymyo-
globin model compounds, low-spin six-coordinate heme-
-
superoxo complexes (S)(F8)FeIII(O2 ) (S ) solvent) (Chart
2), whereas a heme-peroxo-heme binuclear FeIII-(O22-)-
FeIII species is generated in noncoordinating solvents such
as CH2Cl2 or toluene.23 The heterobinuclear heme-copper
complex [(2L)FeIICuI]+ (3) (Scheme 2) contains previously
studied [(LH)CuI]+ and (F8)FeII moieties, but these two groups
are covalently connected in ligand 2L, leading to new
chemistry which cannot be obtained from its individual
components.
The general scheme for the preparation of the metal
complexes discussed in this report is shown in Scheme 2.
Synthetic details for the binucleating ligand 2LH2 and metal
complexes were described previously.19 The deuterated
The 1H NMR spectrum of 1 exhibited features typical for
a high-spin iron(II) porphyrinate with the â-pyrrolic reso-
nances at 51-53 ppm (Figure 2A; also see Figure S5A for
2
a H NMR spectrum). The five-coordinate structure of 1
seems to be retained in solution, as judged by the paramag-
(22) Shearer, J.; Zhang, C. X.; Hatcher, L. Q.; Karlin, K. D. J. Am Chem.
Soc. 2003, 125, 12670-12671.
(23) 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.
(24) Momenteau, M.; Reed, C. A. Chem. ReV. 1994, 94, 659-698.
(25) Iron(II) porphyrinate with a single nitrogenous base such as pyridine
or imidazole tends to disproportionate to a very stable low-spin six-
coordinate species, (B)2PFeII.
7016 Inorganic Chemistry, Vol. 44, No. 20, 2005

Cu Ligand Influences on Heme-Peroxo-Cu Formation
Table 1. Selected Bond Distances (Å) and Angles (deg) for Five-Coordinate, High-Spin Iron(II) Porphyrinates
a
b
complex
(2L)Fe (1)
Fe-Np
Fe-Nax
∆
θc
φd
ref
2.068
2.229
0.42
0.38
0.43
0.34
0.65
13
31.5
24.0
22.8
34.1
14.7
this work
[Fe(TPP)(2-MeHIm)]
2.073(9)
2.072(5)
2.07(2)
2.11(2)
2.127(3)
2.095(6)
2.13(2)
2.002(15)
8.3
9.6
5.0
5.1
26
27
28
29
[Fe(TpivPP)(2-MeHIm)]
[Fe(Piv2C8P)(1-MeIm)]
[Fe(TpivPP)(2-MeIm-)]-
a Averaged value. b Displacement of iron from the 24-atom mean plane of the porphyrin core. c Off-axis tilt (deg) of the Fe-Nax bond from the normal
to the porphyrin plane. d Dihedral angle between the plane between defined by the closest Np-Fe-Nax and the pyridine or imidazole plane in deg.
tances between Fe and the pyrrole hydrogens in the crystal
structure are in the range 5.13-5.18 Å, and their T1 values
in NMR spectrum are in the range 24-30 ms. Therefore,
one can assume that the peaks at 30, 18, -0.6, and -4 ppm
arise from the hydrogens that have comparable metal-H
distances to 5.13-5.18 Å. The hydrogens of the bound
pyridine (meta and para positions), the ethyl linker (-CH2-
2
CH2-pyridyl), and the amide hydrogen from the ligand L
are good candidates for these resonances, all of which have
the Fe-H distances in the range of 5-6 Å. Although further
precise assignments for the peaks cannot be made without
carrying out further experiments (e.g., specific deuteration),
the overall NMR features support a solution five-coordinate
structure, consistent with the X-ray crystal structure. We
Figure 2. 1H NMR spectra of (A) (2L)FeII (1) and (B) [(2L)FeIICuI]-
B(C6F5)4 (3) in acetone at room temperature. Some T1 values (ms) are shown
on top of the respective resonances.
presume dynamic behavior where exchange of one pyridyl
axial ligand to the other in 2L could occur, possibly at much
higher temperatures.
Dioxygen Reactivity of (2L)FeII (1). (2L)FeII (1) reacted
netically shifted peaks (31, 18, -0.6, -4, -14 ppm) arising
with dioxygen to form a low-temperature stable superoxo
intermediate, thus an oxyhemoglobin or oxymyoglobin model
compound, which was monitored by UV-vis and NMR
spectroscopy. In CH2Cl2 at -80 °C, 1 exhibited UV-vis
features at 434 (ꢀ ) 236 000 M-1 cm-1), 534 (ꢀ ) 8800
M-1 cm-1), 555 (ꢀ ) 9600 M-1 cm-1), and 600 (sh) nm
(Figure S2). These distinct absorptions in the visible region
are characteristic for the iron(II) tetraarylporphyrinates with
a single nitrogen axial base.35 Bubbling of O2 through
solutions of 1 led to new features at 422 (ꢀ ) 165 000 M-1
cm-1) and 542 (ꢀ ) 16 000 M-1 cm-1) nm which were
from the metal-bound pyridyl tail; a separate set of reso-
nances in the diamagnetic region (Figure 2A) can be assigned
to the noncoordinated pyridyl group.19 The upfield and
downfield shifted resonances were not observed for (F8)FeII
(Chart 2) which does not contain a pyridyl axial base.23 The
1
T1 (≡ H spin lattice relaxation time) values of the paramag-
netic resonances in 1 were measured, as noted on top of each
peak shown in Figure 2A. Because T1 is proportional to r6
(r ) the metal-proton distance),30 T1 determinations can
assist in assigning resonances, where the hydrogens in closer
proximity to the metal center have smaller T1 values. In the
NMR spectrum of 1, the resonance at -14 ppm has a very
short T1 (2.5 ms), whereas the other paramagnetic resonances
have T1 values comparable to those for the pyrrole resonances
(Figure 2A). Therefore, we assign the peak at -14 ppm to
the hydrogens next to the iron-bound pyridine (i.e., of the
-CH2-pyridyl linker); these are in very close proximity to
the iron center (both ∼3.1 Å, determined from the X-ray
crystal structure). The chemical shift (-14 ppm) and the T1
(2.5 ms) value are similar to those for the 6-methyl hydrogens
from other high-spin non-heme iron(II)-pyridine complexes
such as [FeII(6-Me3-TMPA)(CH3CN)2](ClO4)2.31,32 The dis-
-
ascribed to a (2L)FeIII-(O2 ) (2), Scheme 3 (and Figures S2
and S3). The UV-vis spectral changes were consistent with
the changes of the five-coordinate species transforming to a
six-coordinate heme,35 which was further supported by NMR
2
spectroscopy. H NMR spectroscopic studies of the 1/O2
reaction were carried out with a deuterated (â-pyrrolic
position) complex (2L-d8)FeII (1-d8). The reduced complex
1-d8 in CH2Cl2 had pyrrole resonances at 75 ppm at -80 °C
(Figure S4). Oxygenation of this solution resulted in a new
(31) Zang, Y.; Kim, J.; Dong, Y.; Wilkinson, E. C.; Appelman, E. H.; Que,
L., Jr. J. Am. Chem. Soc. 1997, 119, 4197-4205.
(32) Although the resonance for the ortho pyridyl hydrogen is expected to
appear in downfield region (80-140 ppm) with a very short T1,31,33,34
we were not able to detect any such signals under our experimental
conditions.
(26) Ellison, M. K.; Schulz, C. E.; Scheidt, W. R. Inog. Chem. 2002, 41,
2173-2181.
(27) Jameson, G. B.; Molinaro, F. S.; Ibers, J. A.; Collman, J. P.; Brauman,
J. I.; Rose, E.; Suslick, K. S. J. Am. Chem. Soc. 1980, 102, 3224-
3237.
(33) Walker, F. A.; Simonis, U. In Biological Magnetic Resonance; Berliner,
L. J., Reuben, J., Eds.; Plenum Press: New York, 1993; Vol. 12/
NMR of Paramagnetic Molecules.
(34) Walker, F. A. In The Porphyrin Handbook; Kadish, K. M., Smith, K.
M., Guilard, R., Eds.; Academic Press: San Diego, CA, 2000; Vol.
5/NMR and EPR.
(35) Collman, J. P.; Brauman, J. I.; Doxsee, K. M.; Halbert, T. R.;
Bunnenberg, E.; Linder, R. E.; LaMar, G. N.; Gaudio, J. D.; Lang,
G.; Spartalian, K. J. Am. Chem. Soc. 1980, 102, 4182-4192.
(28) Momenteau, M.; Scheidt, W. R.; Eigenbrot, C. W.; Reed, C. A. J.
Am. Chem. Soc. 1988, 110, 1207-1215.
(29) Mandon, D.; Ott-Woelfel, F.; Fischer, J.; Weiss, R.; Bill, E.; Trautwein,
A. X. Inorg. Chem. 1990, 29, 2442-2447.
(30) Bertini, I.; Luchinat, C. NMR of Paramagnetic Molecules in Biological
Systems; Benjamin/Cummings: Menlo Park, CA, 1986.
Inorganic Chemistry, Vol. 44, No. 20, 2005 7017

Kim et al.
Scheme 3
Scheme 4
signal at 9 ppm which was attributed to a low-spin six-
-
coordinate heme-superoxo complex, (2L-d8)FeIII-(O2 )
(2-d8). Such a heme-superoxo intermediate cannot be
generated from the oxygenation of the four-coordinate heme
analogue (F8)FeII in CH2Cl2,23 indicating that the pyridyl axial
ligand in 1 promotes the stable superoxo formation. The O2-
binding to 1 was reversible, where simple Ar bubbling
through a solution of 2 with warming yielded 1. Several
cycles of the reversible O2-binding process could be moni-
tored by UV-vis spectroscopy (Figure S3). Upon warming
S ) 1 ferrous heme, whereas it had a single Soret band in
THF at 423 nm, indicating that 3 was present as a bis(THF)
adduct forming a six-coordinate compound. The X-ray crystal
structure of the bis(THF) heme analogue, (F8)FeII(THF)2, is
known.37
Carbon Monoxide Adduct [(2L)FeII-CO‚‚‚CuI-CO]+
(4). In biophysical studies of CcO, carbon monoxide has been
extensively used to obtain structural and mechanistic
insights.1,6,38-42 In fact, most O2 intermediates of the CcO
catalytic cycle are generated from flash photolysis of heme-
CO enzyme adducts.38-41
-
of the superoxo complex (2L)FeIII-(O2 ) (2) to room
temperature in the excess of dioxygen over a long period of
time, 2 finally transformed to a hydroxy complex, i.e., (2L)-
FeIII-OH with typical spectral features for the heme-
hydroxy species with 413 and 571 nm UV-vis features and
∼80 ppm pyrrole resonances in NMR spectroscopy,36 Figures
S2 and S4.
The bis(carbonyl) complex [(2L)FeII-CO‚‚‚CuI-CO]+ (4)
was obtained by bubbling a THF or MeCN solution of
[(2L)FeIICuI]+ (3) with CO (Scheme 4). An immediate color
change occurred following reactions of CO with 3 (roughly
pink to orange), shown in the UV-vis spectra of Figure 3.
A significantly more intense Soret band for 4 appeared (413
nm, ꢀ ) 370 000 M-1 cm-1 in THF) by comparison to that
Iron(II)-Copper(I) Complex [(2L)FeIICuI]+ (3). Inser-
tion of copper(I) into (2L)FeII (1) (Scheme 2) led to dramatic
changes in the 1H NMR spectrum (Figure 2B). Because the
copper ion in [(2L)FeIICuI]+ (3) takes up the tridentate chelate
2
of 3 (423 nm, ꢀ ) 265 000 M-1 cm-1 in THF). H NMR
2
from the ligand L, the pyridyl arm is no longer bound to
spectroscopy revealed that carbonylation of 3 induced
changes in the spin state of iron (Figure S5, in Supporting
Information). As mentioned earlier, complex 3 is high-spin
in THF (Figure S5, B), but the corresponding CO adduct 4
is six-coordinate low-spin, having CO and a solvent THF as
axial ligands (Figure S5, E), as seen in many other known
heme-CO complexes, including (THF)(F8)FeII(CO).37 Fur-
ther indication of the solvent binding to the iron in 4 came
from solution IR spectroscopy (Figure 3, inset), which
revealed the presence of two different C-O stretching bands,
one corresponding to the FeII-CO (1980 cm-1) moiety and
the other to the CuI-CO (2089 cm-1) fragment.43 The latter
value corresponds closely to that known for [(LH)CuI-CO]+,
the iron center and the up- and downfield shifted resonances
are lost. However, the pyrrole resonances of 3 appeared in
a region similar to that seen for 1 in acetone solvent,
indicating that complex [(2L)FeIICuI]+ (3) remained in a high-
spin state, now having acetone as an axial ligand base. Since
solvent plays a role as an axial base for 3, the spin state of
iron in 3 is solvent-dependent. In weakly coordinating
solvents such as acetone or THF the iron maintains a high-
spin state, whereas it is low-spin in strongly coordinating
solvent like pyridine. In noncoordinating solvents such as
CH2Cl2 and toluene, iron(II) is four-coordinate and it has an
S ) 1 intermediate spin. The changes in the spin state of 3
in different solvents can be easily monitored by the chemical
shifts of the pyrrole resonances;33,34 the 2H NMR spectra of
[(2L-d8)FeIICuI]+ (3-d8) in various solvents are shown in the
Supporting Information (Figure S5). Changes in the spin state
of 3 in different solvents are also reflected by UV-vis
spectroscopy. In CH2Cl2, 3 exhibited two bands in the Soret
region at 423 and 414 nm, typical for a four-coordinate planar
(37) Thompson, D. W.; Kretzer, R. M.; Lebeau, E. L.; Scaltrito, D. V.;
Ghiladi, R. A.; Lam, K.-C.; Rheingold, A. L.; Karlin, K. D.; Meyer,
G. J. Inorg. Chem. 2003, 42, 5211-5218.
(38) Proshlyakov, D. A.; Pressler, M. A.; Babcock, G. T. Proc. Natl. Acad.
Sci. U.S.A. 1998, 95, 8020-8025.
(39) Kitagawa, T. J. Inorg. Biochem. 2000, 82, 9-18.
(40) Han, S.; Takahashi, S.; Rousseau, D. L. J. Biol. Chem. 2000, 275,
1910-1919.
(41) Szundi, I.; Liao, G.-L.; Einarsdottir, O. Biochemistry 2001, 40, 2332-
2339.
(36) 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.
(42) Stavrakis, S.; Koutsoupakis, K.; Pinakoulaki, E.; Urbani, A.; Saraste,
M.; Varotsis, C. J. Am Chem. Soc. 2002, 124, 3814-3815.
7018 Inorganic Chemistry, Vol. 44, No. 20, 2005

Cu Ligand Influences on Heme-Peroxo-Cu Formation
Figure 4. UV-vis spectra of [(2L)FeIICuI]+ (3) (shown in blue; 3.6 ×
10-5 M) and [(2L)FeIII-(O22-)-CuII]+ (6) (shown in red) in THF at -80
°C.
Figure 3. Overlaid UV-vis spectra of [(2L)FeIICuI]+ (3) (blue; 1.5 ×
10-5 M) and [(2L)FeII-CO‚‚‚CuI-CO]+ (4) (pink) in THF at room
temperature. Inset: difference IR spectrum of 4 minus 3 in acetonitrile.
M-1 cm-1) nm (Figure 4 and Table 2). Application of a
vacuum did not affect the stability of the complex, indicating
very strong O2-adduct formation at -80 °C. Similar observa-
tions were made in different solvents and solvent mixtures
such as acetone, dichloromethane, toluene, CH2Cl2/EtCN,
and toluene/MeCN; the thermal stability of 6 was observed
to be significantly increased in CH2Cl2 or toluene as solvent.44
Upon warming, 6 eventually transforms to a mixture of FeIII-
O-CuII and FeIII-hydroxy species, in analogy to previous
observations with many other heme-peroxo-copper com-
plexes.12 A spectrophotometric O2-uptake titration (see
Experimental Section) for 3 showed that one O2 molecule
binds per dinuclear complex, consistent with the formulation.
(2) Resonance Raman Spectroscopy. Resonance Raman
(RR) spectra of the dioxygen adducts verified that 6 is a
peroxo complex in a high-spin heme environment. Soret
excitation of [(2L)FeIII-(O22-)-CuII]+ (6) at 413 nm resulted
in RR spectra dominated by strong porphyrin skeletal modes
and weak iron-ligand vibrations. Porphyrin skeletal modes
in the high-frequency region of the RR spectra were
consistent with a five-coordinate high-spin ferric heme
species (ν4, ν3, and ν2 at 1363, 1451, and 1561 cm-1,
respectively) as previously observed in the related sys-
tems.45,46 As reported previously,19 a low-temperature RR
spectra of 6 generated in THF with 16O2 revealed a ν(O-O)
peroxo stretching vibration at 747 cm-1 which shifted to 707
cm-1 upon 18O2 substitution (Table 2). The lack of a splitting
at 730 cm-1 assigned to the ν(16O-18O), and observed when
6 is produced with 16/18O-scrambled O2 gas suggests a
symmetric binding of the peroxo group to the metal center,
as in a side-on rather than end-on geometry.19
Scheme 5
ν ) 2089 cm-1.21 Whereas the copper-bound CO stretching
frequencies did not change with variation of solvent, a
significant change in the iron-bound CO vibrations occurred.
In THF, ν(C-O)Fe appeared at 1968 cm-1, whereas it
occurred at 1980 cm-1 in MeCN. The effect must be due to
the trans influence of the axial ligand, indicating that iron is
six-coordinate with solvent axial base.
Although the Fe-CO and the Cu-CO components are
covalently connected to each other, there was no indication
of any interaction between them. In fact, the iron(II) and
copper(I) carbonyl stretching frequencies are identical with
those known for the individual copper-only21 and the iron-
only37 analogues, [(LH)CuI(CO)]+ and (THF)(F8)FeII(CO).
Dioxygen Adduct [(2L)FeIII-(O22-)-CuII]+ (6). As
previously reported,19 reduced iron(II)-copper(I) complex
[(2L)FeIICuI]+ (3) binds O2, resulting in formation of a new
type of heme-peroxo-copper complex [(2L)FeIII-(O22-)-
CuII]+ (6) with a µ-η2:η2 peroxo ligation (Scheme 5). Here,
we describe spectroscopic data and analyses of 6 which were
not fully described in the earlier report, followed by
discussions of how these spectroscopic properties compare
with those of known tetradentate heme-peroxo-copper
complexes, one of which was crystallographically determined
to have a µ-η2:η1 peroxo binding structure.
Assignment of the O-O stretching vibrations to a homo-
nuclear dicopper-peroxo species such as [{(2L)FeII‚‚‚CuII}2-
(O22-)]2+ was ruled out because the reaction stoichiometry
revealed 1:1 of O2/[(2L)FeIICuI]+ (3), and the dicopper/O2
adduct with LH chelate was known to have a ν(O-O)
vibration at 730 cm-1 (∆18O2 ) -39 cm-1, in acetone with
386 nm excitation)20,47 that was significantly lower than what
(1) Benchtop UV-Vis Spectroscopy. Exposure of THF
solutions of 3 to O2 at -80 °C afforded [(2L)FeIII-(O22-)-
CuII]+ (6) with UV-vis features at 423 (ꢀ ) 158 000 M-1
cm-1), 484 (ꢀ ) 11 000 M-1 cm-1), and 541 (ꢀ ) 10 000
(44) 6 was stable below -90 °C in THF or acetone, whereas it was stable
up to -60 °C in toluene or CH2Cl2 solution.
(45) Nakamoto, K. Infrared and Raman spectra of Inorganic and Coor-
dination compounds, 5th ed.; Wiley-Interscience: New York, 1997.
(46) Burke, J. M.; Kincaid, J. R.; Peters, S.; Collman, J. P.; Spiro, T. G. J.
Am. Chem. Soc. 1978, 100, 6083-6088.
(43) Kretzer, R. M.; Ghiladi, R. A.; Lebeau, E. L.; Liang, H.-C.; Karlin,
K. D. Inog. Chem. 2003, 42, 3016-3025.
Inorganic Chemistry, Vol. 44, No. 20, 2005 7019

Kim et al.
Table 2. Physical Properties of the Dioxygen Adducts of Heme-Copper Model Complexes
compd
λ
max, nm
ν(O-O), cm-1 (∆18O2, cm-1)
ref
Tridentate Cu-Ligand
423, 484, 541, 572 (sh)
420, 480, 540, 567 (sh)
428b
[(2L)FeIII-(O22-)-CuII]+ (6)
747 (-40)
767 (-41), 752 (-45)
758 (-18)
this work
52
53
[(F8)FeIII-(O22-)-CuII(LMe2N)]+
[(DHIm)(TACNAcr)FeIII-(O22-)-CuII]+ a
Tetradentate Cu-Ligand
412, 558
418, 561, 632
419, 560
420, 557, 612
421, 558, 612
417, 560
[(F8)FeIII-(O22-)-CuII(TMPA)]+
[(6L)FeIII-(O22-)-CuII]+
808 (-46)
787 (-43)
803 (-44)
790 (-44)
799 (-47)
813 (-44)
54
55
56
18
57
58
[tpaCuII-(O22-)-TPPFeIII]+
[(TMP)FeIII-(O22-)-(5MeTPA)CuII]+
[(LOH)FeIII-(O22-)-CuII]+
[(F8)FeIII-(O22-)-CuII(LN4OH)]+
a Iron-porphyrinate moiety also contains an imidazole axial base. b Q-band peaks were not reported.
was found in 6. Additionally, the copper-only O2 reactions
are not likely to cause such big changes in the UV-vis
spectra as those which were observed (Figure 4). Similarly,
the possibility of the ν(O-O) vibrations coming from a
diheme-peroxo species such as [{(2L)CuI···FeIII}2(O22-)]2+
was ruled out on the basis of the reaction stoichiometry and
NMR spectroscopy (vide infra).48 While the reaction stoi-
chiometry alone cannot rule out the presence of nonbridged
heme peroxo species [(2L)FeIII-(O22-)‚‚‚CuII]+, such a
species would have distinctive EPR signals49-51 that were
not present in 6. More importantly, such heme-peroxo
complexes [(Porphyrinate)FeIII-(O22-)]- display a ν(O-O)
50,51
at significantly higher frequencies (i.e., ∼800 cm-1),
values much greater than that observed for 6. Therefore, the
RR results in conjunction with reaction stoichiometry as well
as UV-vis, NMR, and EPR spectroscopies clearly show that
the ν(O-O) at 747 cm-1 came from a bridged heme-
peroxo-copper species [(2L)FeIII-(O22-)-CuII]+ (6).
Figure 5. 1H NMR spectra of (A) [(2L)FeIICuI]B(C6F5)4 (3) and (B) [(2L)-
FeIII-(O22-)-CuII]B(C6F5)4 (6) in acetone at -90 °C.
(3) NMR Spectroscopy. As utilized for the (2L)FeII (1)/
O2 reaction, low-temperature 1H and 2H NMR spectroscopies
were employed to study changes in the structure and the spin
state of the iron-copper complexes upon oxygenation. The
high-spin reduced compound [(2L)FeIICuI]+ (3) in acetone
has pyrrole resonances at ∼50 ppm whereas the hydrogen
resonances from the d10 copper(I) chelate reside in the
diamagnetic region at room temperature (Figure 2B). The
paramagnetic pyrrole resonances follow Curie behavior as
the temperature is decreased, shifting from ∼50 ppm at room
temperature (Figure 2B) to ∼95 ppm at -90 °C (Figure 5A).
Bubbling dioxygen through a cold solution of 3 led to the
formation of [(2L)FeIII-(O22-)-CuII]+ (6), whose pyrrole
resonances were found at ∼110 ppm. In a separate study,
we have reported NMR properties of the mononuclear heme
distinctive chemical shifts for the pyrrole hydrogen reso-
nances (δpyrrole ) 125, 8.9, 17.5, and 3.5 ppm, respectively,
at -80 °C); thereby, the pyrrole chemical shifts from 6
allowed us to rule out the possibility of 6 being any of such
species.59 In fact, the 110 ppm pyrrole resonances (-90 °C)
are between the ranges known for a pure high-spin FeIII heme
(52) 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.
(53) 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.
(54) 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.
(55) 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.
(56) Sasaki, T.; Nakamura, N.; Naruta, Y. Chem. Lett. 1998, 351-352.
(57) Liu, J.-G.; Naruta, Y.; Tani, F.; Chishiro, T.; Tachi, Y. Chem. Commun.
2004, 120-121.
(58) Kim, E.; Kamaraj, K.; Galliker, B.; Rubie, N. D.; Moe¨nne-Loccoz,
P.; Kaderli, S.; Zuberbu¨hler, A. D.; Karlin, K. D. Inog. Chem. 2005,
44, 1238-1247.
(59) Although there is 10 °C temperature difference between the current
NMR study (-90 °C) and the previous report on the mononuclear
complexes (-80 °C), it is extremely unlikely that such a difference
can cause a dramatic change in the pyrrole chemical shifts to affect
this conclusion.
-
analogues,23 (F8)FeIII-OH, (F8)FeIII-(O2 ), [(F8)FeIII]2-
(O22-), and (F8)FeIVdO. All of these complexes have
(47) 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.
(48) Peroxo stretching vibrations from homonuclear diheme-peroxo
complexes [(P)FeIII]2-(O22-) (P ) porphyrinate) have not been
observed with RR spectroscopy, leaving direct comparison of these
data not possible.
(49) Chufa´n, E. E.; Karlin, K. D. J. Am Chem. Soc. 2003, 125, 16160-
16161.
(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) Selke, M.; Sisemore, M. F.; Valentine, J. S. J. Am. Chem. Soc. 1996,
118, 2008-2012.
7020 Inorganic Chemistry, Vol. 44, No. 20, 2005

Cu Ligand Influences on Heme-Peroxo-Cu Formation
(∼130 ppm) and the pure high-spin FeII heme (∼95 ppm)
Chart 3
species. This is typical for strongly coupled (porphyrinate)-
FeIII-X-CuII complexes (X ) O2 or O2-), where strong
2-
antiferromagnetic coupling between high-spin FeIII (d5, S )
5/2) and CuII (d9, S ) 1/2) leads to a net S ) 2 spin
system.54,55,60,61 The S ) 2 assignment for 6 was further
supported by EPR silent behavior (15 K, in THF/toluene
frozen glass) and by a solution magnetic moment determi-
nation (µeff ) 5.3 µB at -90 °C, in acetone).
1
Another characteristic feature of the H NMR spectrum
of 6 is the observable hydrogen resonances from the copper
chelate. Generally, it is difficult to observe well-resolved
NMR spectra from Cu(II) complexes due to the slow
electronic relaxation rate of the CuII ion.30 However, in a
strongly coupled system like 6, copper(II) can gain a faster
electronic relaxation rate from the electronically coupled FeIII,
leading to relatively sharp signals.61 Complex 6 in Figure
5B exhibits relatively sharp resonances in the upfield (-8,
-11, -27, -31 ppm) and downfield regions (39, 37, 29
1
ppm). By analogy to [(F8)FeIII-O-CuII(LH)]+, whose H
NMR spectrum was fully assigned,60 the signals of 6 in the
upfield region are believed to originate from the pyridyl
hydrogens, whereas the downfield signals at 39, 37, and 29
ppm are due to the methylene H’s next to pyridyl donors
(-CH2-PY).62 Although complete assignment of the NMR
spectrum is beyond the scope of interest for this study, the
hydrogen resonance assignments for the copper chelate along
with the position of the pyrrole resonances support strong
antiferromagnetic coupling between FeIII and CuII through a
bridging peroxide ligand.
single band around 560 nm and a minor band with much
weaker absorption at ∼610-630 nm.18,54,55 We presume the
differences to alterations in peroxo structures and binding
to be imposed by the copper environment (see further
discussions below).
(4) Comparing and Contrasting the Spectroscopic
Properties of Tri- versus Tetradentate Heme-Peroxo-
Copper Complexes. There are now several examples of the
spectroscopically well characterized heme-peroxo-copper
complexes,18,52-55,63 most of which have a tetradentate ligand
TMPA moiety for a copper chelate. These spectroscopic
studies of 6 in combination with those from known tetraden-
tate counterparts allow us to get insight into how changes in
the peroxo-binding mode enforced by copper ligand denticity
can affect their spectroscopic features.
In resonance Raman studies, 6 exhibited a very low ν(O-
O) stretching frequency at 747 cm-1, the lowest ν(O-O)
frequency among the known heme-peroxo-copper com-
plexes, yet this value is quite close to those of other heme-
tridentate Cu complexes (Table 2). On the other hand,
all the heme-peroxo-Cu complexes with tetradentate
Cu-chelates have significantly higher ν(O-O) values,
ca. ∼800 cm-1, that are also very similar to values known
for iron(III)-only η2-peroxo complexes, [(Porphyrinate)-
FeIII-(O22-)]-.50,51
In Cu(I)/O2 chemistry, it is well-known that copper ligand
denticity induces different O2 binding modes.64,65 Upon
dioxygen binding, the Cu(II) centers tend to be five-
coordinated, resulting in the tetradentate Cu(I) complexes
to favor O2 binding through an end-on fashion, while ones
with tridentate ligands favor the side-on ligation of O2 (Chart
3). The ν(O-O) stretching frequencies are reflective of the
different Cu2O2 geometries and resultant bonding in the
When compared to those of other related heme-peroxo-
copper complexes (Table 2), the absorption maxima of
complex [(2L)FeIII-(O22-)-CuII]+ (6) are very similar to
those of (F8)FeIII-(O22-)-CuII(LMe2N) (see diagram)52 but
significantly different from those for the group of complexes
which have tetradentate copper chelates (Table 2). In
the R/â region, complex 6 has two distinctive bands at 484
and 541 nm with comparable extinction coefficients along
with a shoulder around 572 nm. However, heme-peroxo-
Cu complexes with tetradentate Cu-chelates have a major
(60) Kopf, M.-A.; Neuhold, Y.-M.; Zuberbu¨hler, A. D.; Karlin, K. D. Inorg.
Chem. 1999, 38, 3093-3102.
(61) Nanthakumar, A.; Fox, S.; Murthy, N. N.; Karlin, K. D. J. Am. Chem.
Soc. 1997, 119, 3898-3906.
(62) The methylene H’s next to the central amine are expected further
upfield, which are not seen here due to the instrumental limit for the
spectral window.
(64) Mirica, L. M.; Ottenwaelder, X.; Stack, T. D. P. Chem. ReV. 2004,
104, 1013-1045.
(63) Naruta, Y.; Sasaki, T.; Tani, F.; Tachi, Y.; Kawato, N.; Nakamura,
N. J. Inorg. Biochem. 2001, 83, 239-246.
(65) Que, L., Jr.; Tolman, W. B. Angew. Chem., Int. Ed. 2002, 41, 1114-
1137.
Inorganic Chemistry, Vol. 44, No. 20, 2005 7021

Kim et al.
Scheme 6
dicopper(II) peroxo complexes, where the side-on bound
peroxides give lower ν(O-O) values (710-760 cm-1)
compared to those for end-on peroxides (805-830 cm-1);
the former situation is due to a geometry conducive to back-
donation from Cu(II) into the σ* orbital of O2 .
2- 66-69 Since
peroxide bound to heme-iron(III) is not known to provide
these lower ν(O-O) values, it is reasonable to consider that
the same inclinations seen in copper-dioxygen binding
modes and the resulting ν(O-O) value trends may remain
in the heme-peroxo copper systems. Indeed, the X-ray
crystal structure of [(TMP)FeIII-(O22-)-(5MeTPA)CuII]+
(Chart 3)18 with η2:η1 side-on (to Fe) and end-on (to Cu)
peroxo binding contains the TMPA-derivative copper chelate,
where the peroxo binding around copper is almost identical
with that in [(TMPA)Cu-O2-Cu(TMPA)]2+.18,70,71
Despite the large differences in UV-vis and resonance
Raman spectroscopic features between tri- and tetradentate
heme-peroxo-copper complexs, NMR properties of 6
resemble those of tetradentate counterparts, indicating they
all have an S ) 2 spin system with a strong antiferromagnetic
coupling. We do not have in-depth insights for the strong
coupling within 6 at this time. However, investigations on
the nature of bonding and the origin of strong antiferromag-
netic interactions in heme-peroxo-copper complexes just
began with one of the tetradentate complexes [(F8)FeIII-
(O22-)-CuII(TMPA)]+,72 and the similar studies with 6 will
be conducted in the future.
Kinetics of the [(2L)FeIICuI]+ (3)/O2 Reaction. The
kinetic scheme deduced from stopped-flow spectroscopic
analysis for the reaction of [(2L)FeIICuI]+ (3) with dioxygen
in CH2Cl2/6% EtCN is outlined in Scheme 6. This specific
solvent system was chosen because EtCN has an influence
which is favorable for observation and analysis of the kinetics
at temperatures down to -105 °C, and to compare the results
of this work with those from a previous investigation on a
related system employing 1:1 mixtures of a reduced heme
and copper(I) complex with tridentate ligand (LMe2N, Chart
2),73 see further discussion below.
The UV-vis spectral changes in the R/â region (465-
665 nm) of the [(2L)FeIICuI]+ (3)/O2 reaction are shown in
Figure 6A over the time period of 0.35 s. At -105 °C,
addition of O2 to 3 resulted in decay of a band at 531 nm
(reduced complex 3) and formation of a transient species
with an absorption at 544 nm, during the first ∼0.03 s. The
544 nm band then decayed over the next 0.3 s and
transformed to a spectrum with broad maxima at 488 and
541 nm and a shoulder at 575 nm, i.e., peroxo complex [(2L)-
FeIII-(O22-)-CuII]+ (6). Figure 6A (inset) shows the absor-
bance versus time traces at 531, 488, and 544 nm, revealing
the relationship of formation and disappearance of the various
species.
(66) Ross, P. K.; Solomon, E. I. J. Am. Chem. Soc. 1990, 112, 5871-
5872.
(67) Ross, P. K.; Solomon, E. I. J. Am. Chem. Soc. 1991, 113, 3246-
3259.
(68) 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.
(69) Baldwin, M. J.; Root, D. E.; Pate, J. E.; Fujisawa, K.; Kitajima, N.;
Solomon, E. I. J. Am. Chem. Soc. 1992, 114, 10421-10431.
(70) Jacobson, R. R.; Tyekla´r, Z.; Karlin, K. D.; Liu, S.; Zubieta, J. J. Am.
Chem. Soc. 1988, 110, 3690-3692.
Direct characterization of the transient 544 nm species was
not possible due to its extremely short lifetime. However, it
formed with a first-order dependence on [O2] and [3], and
the UV-vis spectrum of this species (Figure 6B) was very
similar to its porphyrin analogues (solvent)(F8)FeIII-(O2-)23,52
(71) Tyekla´r, Z.; Jacobson, R. R.; Wei, N.; Murthy, N. N.; Zubieta, J.;
Karlin, K. D. J. Am. Chem. Soc. 1993, 115, 2677-2689.
(72) del Rio, D.; Sarangi, R.; Chufa´n, E. E.; Karlin, K. D.; Hedman, B.;
Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. posted ASAP,
August 4, 2005; DOI: 10.1021/ja043374r.
-
and [(solvent)(2L)FeIII-(O2 )‚‚‚CuI(PPh3)]+ (5-PPh3) (Figure
8; see below), allowing us to assign this as a superoxo
(73) Preliminary kinetic studies for the formation of [(F8)FeIII-(O22-)-
CuII(LMe2N)]+ were previously reported.52 The fuller analyses for its
formation, the k2 step, are reported here, in the Supporting Information
and with results given in Table 3 and Figure 7.
-
complex [(EtCN)(2L)FeIII-(O2 )‚‚‚CuI(NCEt)]+ (5);74 reso-
nance Raman spectroscopy of the former analogue (solvent)-
-
(F8)FeIII-(O2 ), with ν(O-O) ) 1178 cm-1 and ν(Fe-O)
7022 Inorganic Chemistry, Vol. 44, No. 20, 2005

Cu Ligand Influences on Heme-Peroxo-Cu Formation
of 5 is a bit larger relative to both the (F8)FeII/O2 and (F8)-
FeII/[(LMe2N)CuI]+/O2 systems, and the enhanced rate comes
by way of an increased activation entropy, Table 3. The
chemical origin of this effect during the 3-to-5 conversion
is not entirely clear. By contrast, when other five-coordinate
heme complexes such as Collman and co-workers’s picket
fence porphyrins react with O2, much smaller activation
enthalpies and negative activation entropies are observed;76
however, they did not employ EtCN, known23 to be a
moderately good ligand for ferrous heme.
The second step, a reaction of the ferric superoxide with
a copper(I) species, is of particular interest because such a
step or closely related process is key in the O2-reductive
cleavage chemistry in CcO (Scheme 1).1,2 This study in
comparison with other related systems (discussed below)
shows that the kinetics for the formation of µ-heme-
peroxo-Cu species highly depends on the proximity and the
ligand architecture of copper (Figure 7). As indicated in
Scheme 6 and Figure 6, the transient superoxo species 5
transforms to the µ-peroxo complex [(2L)FeIII-(O22-)-CuII]+
(6) on a very fast time scale. Such a heme-peroxo-copper
complex formation was not observed with its untethered
analogue, (F8)FeII/[(LH)CuI(EtCN)]+/O2, wherein only (EtCN)-
Figure 6. (A) Time-dependent UV-vis spectra of the oxygenation of
[(2L)FeIICuI]+ (3) at -105 °C in CH2Cl2/6% EtCN {concentrations: [3] )
2.16 × 10-4 M, [O2] ) 6.00 × 10-4 M}. Inset: absorbance-time traces at
531 nm (blue, decay of 3), 544 nm {green, formation and subsequent decay
of [(EtCN)(2L)FeIII-(O2-)‚‚‚CuI(NCEt)]+ (5)}, and 488 nm {red, formation
of [(2L)FeIII-(O22-)-CuII]+ (6)} over ∼3.5 s. (B) Calculated spectra for 3
(blue, λmax ) 531 nm), 5 (green, λmax ) 544 nm), and 6 (red, λmax ) 488,
544, 575 (sh) nm).
-
(F8)FeIII-(O2 ) was generated under the same experimental
conditions (Figure 7, top). The formation of the untethered
“heme-peroxo-copper” species is obtained only after
introducing electron-donating groups (NMe2 donors at para
position of the pyridyl groups) to the copper ligand, which
makes [(LMe2N)CuI]+ a stronger reductant (Figure 7)21,52
facilitating binding and reduction of the (EtCN)(F8)FeIII-
) 568 cm-1,52 further supports the formulation as an
oxymyoglobin (oxyhemoglobin) model compound. Thus,
complex 5 is assumed to be a low-spin six-coordinate iron-
(III) with one EtCN axial base and a trans superoxide ligand
(Scheme 6). Experimentally we cannot distinguish between
two possible isomers of the superoxo complex 5 (Scheme
6). If both species were formed in significant amounts, they
would have to be in rapid equilibrium. We rule out the
possibility that a bis(superoxo) species forms (i.e., [(EtCN)-
-
(O2 ) moiety. For the purpose of comparison, the rate
-
constant for the reaction of (F8)FeIII-(O2 ) with [(LMe2N)-
CuI]+ was made to be pseudo-first-order, arbitrarily using
0.1 mM as the concentration for iron (Table 3 and Figure
7). Under these conditions, despite the electron-donating
substituents on the PY2 chelate in LMe2N system, formation
of [(2L)FeIII-(O22-)-CuII]+ (6) via the intramolecular reac-
tion is more than 20 times faster than that for [(F8)FeIII-
(O22-)-CuII(LMe2N)]+. The formation of 6 and [(F8)FeIII-
(O22-)-CuII(LMe2N)]+ 73 have essentially identical activation
enthalpies (∼36 kJ/mol) for these reactions both involving
-
-
(2L)FeIII-(O2 )‚‚‚CuII-(O2 )]+) because separate copper-
only/O2 studies showed that the oxygenation chemistry of
[(LH)CuI]+ does not occur under these experimental condi-
tions; EtCN strongly suppresses O2-binding.75
The kinetics of the formation of [(EtCN)(2L)FeIII-
-
(O2 )‚‚‚CuI(NCEt)]+ (5) (Table 3; see Experimental Section
for further details) shows that the O2-binding to iron process
is accompanied by a relatively large activation enthalpy (45.3
( 0.7 kJ/mol) and a large positive activation entropy (120
( 4 J/(mol‚K)). This clearly points to dissociative activation
of the process, be it an interchange mechanism with early
transition state for loss of EtCN or in fact a fast left-lying
equilibrium with reduced coordination number. Similar
-
FeIII-(O2 ) electron-transfer reduction and binding by cop-
per(I) (Table 3 and Figure 7).73 We speculate that the
differing ∆Sq values, 54 versus 5 J/(mol·K) (Table 3 and
Figure 7) for inter- and intramolecular reactions, respectively,
relate to details concerning the process of nitrile dissociation
from the CuI moiety in each case.
-
dissociative processes in the formation of the heme-(O2 )
Reactions of [(2L)FeIII-(O22-)-CuII]+(6) with CO and
PPh3. To further probe the nature of the heme-peroxo-
copper complex 6, we carried out initial reactivity studies
of 6 with CO and PPh3. The peroxo complex 6 was generated
in situ in THF solution at -95 °C or in toluene/10% CH3-
CN solution at -75 °C. Through the application of vacuum/
Ar-purge or freeze-pump-thaw, the excess O2 was rigor-
ously removed before the reactions with CO or PPh3. As
stated above, dioxygen binding to the reduced complex
species were observed with the analogous untethered sys-
tems, (F8)FeII/O2 or (F8)FeII/[(LMe2N)CuI]+/O2 (Table 3, Chart
2).52 Thus, as previously stated,52 the presence of the copper
ion has essentially no influence on the formation of the
heme-(O2-) species. On the other hand, the rate of formation
-
of this heme-(O2 ) species in 5 is notably increased (∼10-
fold), Table 3. In fact, the activation enthalpy for formation
(74) The compound number “5” is designated to the heme-superoxo/Cu+
2
species which possess the L ligand.
(75) Liang, H.-C.; Karlin, K. D.; Dyson, R.; Kaderli, S.; Jung, B.;
(76) Collman, J. P.; Brauman, J. I.; Iverson, B. L.; Sessler, J.; Morris, R.
Zuberbu¨hler, A. D. Inorg. Chem. 2000, 39, 5884-5894.
M.; Gibson, Q. H. J. Am. Chem. Soc. 1983, 105, 3052-3064.
Inorganic Chemistry, Vol. 44, No. 20, 2005 7023

Kim et al.
Table 3. Kinetic Parameters for the Formation of Heme-Superoxo (k1) or Heme-Peroxo-Copper (k2) Moieties from FeII/O2 or FeII/CuI/O2 Reactions
in CH2Cl2/6% EtCNa
(F8)FeII
k1, M-1 s-1
(F8)FeII/[CuI(LMe2N)]+
[(2L)FeCu]+ b
-
FeIII-(O2
168 K
183 K
)
k1, M-1 s-1
k1, M-1 s-1
(4.00 ( 0.16) × 103
(4.86 ( 0.10) × 104
(4.07 ( 0.16) × 105
41.2 ( 0.6
(5.7 ( 0.1) × 103
(6.0 ( 0.1) × 104
(4.5 ( 0.1) × 105
38.8 ( 0.7
(5.23 ( 0.04) × 104
(8.1 ( 0.03) × 105
(8.4 ( 0.06) × 106
45.3 ( 0.7
198 K
∆H, kJ/mol
∆S, J/(mol‚K)
74 ( 4
63 ( 2
120 ( 4
FeIII-(O22-)-CuII
NA
NA
NA
NA
NA
NA
52
k2,c M-1 s-1 (k*, s-1
)
k2,d s-1
168 K
(1.28 ( 0.03) × 104 (1.28)
(1.16 ( 0.01) × 105 (11.6)
(7.6 ( 0.1) × 105 (76)
36.2 ( 0.2
(2.74 ( 0.04) × 101
(2.55 ( 0.03) × 102
(1.70 ( 0.05) × 103
36.6 ( 0.4
183 K
198 K
∆H, kJ/mol
∆S, J/(mol‚K)
ref
54 ( 1
52 and this work
5 ( 0.4
this work
a Notes: uncertainties given are as standard errors. b From analysis of data taken from 168 to 179 K. c From analysis of data taken from 168 to293 K.
d From analysis of data taken from 168 to 191 K. NA ) not applicable. k* calculated as pseudo-first-order assuming 0.1 mM concentration for iron.
Scheme 7
Figure 7. Comparisons of kinetic parameters for formation of heme-
peroxo-copper species.52 Asterisk indicates calculated as pseudo-first-order
assuming 0.1 mM concentrations for iron. See text.
[(2L)FeIICuI]+ (3) was not reversed by simple vacuum/Ar-
purging. However, the peroxo complex 6 was found to react
with CO to form the reduced CuI-CO and FeII-CO complex
with concomitant liberation of dioxygen (Scheme 7). When
CO was bubbled through a THF solution of 6 (-95 °C),
immediate UV-vis spectral changes occurred with formation
of features at 413 and 532 nm which were identical with
those generated by independent formation of [(2L)Fe-
CO‚‚‚Cu-CO]+ (4) through CO exposure to [(2L)FeIICuI]+
(3); vide supra. Solution IR spectra confirmed the presence
of the bis(CO) compound, revealing two ν(C-O) vibrations
at 2089 and 1968 cm-1 (see Supporting Information, Figure
S7). The amount of O2 evolved (∼70%) was semiquantified
using a detection procedure employing an alkaline pyrogallol
7024 Inorganic Chemistry, Vol. 44, No. 20, 2005

Cu Ligand Influences on Heme-Peroxo-Cu Formation
Figure 9. 2H NMR spectra of the reaction of [(2L)FeIII-(O22-)-CuII]+
(6) with PPh3 in THF at -90 °C: (A) spectrum of 6 generated by bubbling
[(2L)FeIICuI]+ (3) with O2; (B) [(THF)(2L)FeIII-(O2-)···CuI(PPh3)]+ (5-PPh3)
generated by the reaction of 6 with 1 equiv of PPh3 after removal of excess
O2; (C) [(2L)FeIII-OH···CuI(PPh3)]+ generated by warming 5-PPh3 to room
temperature. The sharp peaks at δ 3.58 and 1.73 ppm correspond to solvent
THF.
Figure 8. UV-Vis spectra of the reaction of [(2L)FeIII-(O22-)-CuII]+
(6) with PPh3 in toluene/10% CH3CN at -75 °C: (a) spectrum of 6
generated by bubbling [(2L)FeIICuI]+ (3) with O2 (shown in red, λmax
)
486, 546 nm); (b) spectrum of [(CH3CN)(2L)FeIII-(O2-)···CuI(PPh3)]+ (5-
PPh3) generated by the reaction of 6 with 1 equiv of PPh3 after removal of
excess O2 (shown in green, λmax ) 543 nm).
The 6/PPh3 reactions were also carried out using an excess
(5-10 equiv) of PPh3. No differences in UV-vis or NMR
spectral change were observed. The possible formation of
OdPPh3 from the 6/PPh3 was investigated using 31P NMR
spectroscopy because triphenylphosphine oxides are known
to be generated from some reactions with iron-peroxo81 or
copper-peroxo82 complexes with PPh3. However, no such
species was observed from the 6/PPh3 reaction even with a
large excess (∼10-fold) PPh3. As discussed, PPh3 promotes
the “reverse” reaction of 6 to 5 at low temperature, which
then transforms to a ferric hydroxy species (λmax ) 575 nm,
δpyrrole ) 129 ppm at -90 °C) upon warming. This final
“mixed-valent” thermal product was isolated in 47% yield,
and it is formulated as [(2L)FeIIIOH‚‚‚CuI(PPh3)]B(C6F5)4 on
the basis of UV-vis and NMR spectroscopy (also having a
broadened 31P NMR spectrum; see also above) as well as
C, H, and N elemental analysis (see Experimental Section).
solution for spectroscopic monitoring (see Experimental
Section). Once the bis(CO) complex (4) forms, it does not
react with dioxygen at low temperature even in the presence
of a large excess of O2, indicating that CO is a thermody-
namically stronger ligand than O2, both for the copper and
iron ion centers, as is known to occur for synthetic and
natural hemes and copper complexes.17,77,78
While the fully reduced iron-copper complex 4 could be
obtained from [(2L)FeIII-(O22-)-CuII]+ (6) by adding CO,
only a partially reduced iron-superoxo species [(solvent)-
-
(2L)FeIII-(O2 )···CuI(PPh3)]+ (5-PPh3) was observed upon
reaction of 6 with PPh3 (Scheme 7). The UV-vis spectral
changes for the 6/PPh3 reaction are shown in Figure 8.
Addition of 1 equiv of PPh3 to the peroxo complex [(2L)-
FeIII-(O22-)-CuII]+ (6) (Figure 8, λmax ) 486, 546, 578 nm)
yielded a new band at λmax ) 543 nm which is characteristic
of a heme-superoxide and similar to those in (EtCN)(F8)-
Summary and Conclusion
-
FeIII-(O2-)52 and [(EtCN)(2L)FeIII-(O2 )‚‚‚CuI(EtCN)]+ (5)
Using a heterobinucleating ligand 2L that contains a N,N-
bis[2-(2-pyridyl)ethyl)]amine (PY2) and teteraarylporphyrine
moiety, we have synthesized and characterized reduced heme
and heme-copper complexes, [(2L)FeII]+ (1) and [(2L)FeII-
CuI]+ (3), that generate dioxygen adducts at low temperature.
Complex 3 differs from previously reported heme-copper
binuclear systems in having a tridentate Cu-chelate PY2
moiety, whereas the majority of known discrete heme-
copper O2 adducts are derived from systems containing a
tetradentate TMPA Cu-chelate. While heme-only complex
1 binds O2 reversibly to form a ferric-superoxo species, O2-
binding to the heme-copper complex 3 leads to the
formation of a µ-η2:η2-bridged heme-peroxo-Cu intermedi-
ate, [(2L)FeIII-(O22-)-CuII]+ (6). Stopped-flow kinetic stud-
ies provide insights into the oxygenation process, revealing
that dioxygen initially reacts with iron to generate a transient
heme-superoxo species [(EtCN)(2L)FeIII-(O2-)‚‚‚CuI(NCEt)]+
(5) prior to formation of 6. Characterization of [(2L)FeIII-
(Figure 6). This 543 nm species was stable at low temper-
ature, which allowed further low-temperature 2H NMR
spectroscopic characterization. The pyrrole resonances for
[(2L-d8)FeIII-(O22-)-CuII]+ (6) occur at 106 ppm in THF
at -90 °C (Figure 9). Addition of 1 equiv of PPh3 to a cold
THF solution of 6 led to immediate formation of a new
diamagnetic species whose pyrrole hydrogen resonances
appeared at 9 ppm, a typical value for a low-spin 6-coordinate
ferric-superoxide heme. Thus, the product of the reaction
-
of 6 and PPh3 is formulated as [(THF)(2L)FeIII-(O2 )‚‚‚CuI-
(PPh3)]+ (5-PPh3) on the basis of the UV-vis and NMR
spectroscopic features (Scheme 7). Although direct spectro-
scopic detection of the CuI-PPh3 group in 5-PPh3 is not
straightforward, the formulation is well precedented; copper-
(I) complexes possessing the tridentate N,N-bis[2-(2-pyridyl)-
ethyl)]amine moiety readily form [LCuI-(PPh3)]+ struc-
tures.79,80
(77) Karlin, K. D.; Zuberbu¨hler, A. D. In Bioinorganic Catalysis: Second
Edition, ReVised and Expanded; Reedijk, J., Bouwman, E., Ed.; Marcel
Dekker: New York, 1999.
(78) Karlin, K. D.; Cruse, R. W.; Gultneh, Y.; Farooq, A.; Hayes, J. C.;
Zubieta, J. J. Am. Chem. Soc. 1987, 109, 2668-2679.
(79) Karlin, K. D.; Gan, Q.-F.; Farooq, A.; Liu, S.; Zubieta, J. Inorg. Chim.
Acta 1989, 165, 37-39.
(80) Karlin, K. D.; Tyekla´r, Z.; Farooq, A.; Haka, M. S.; Ghosh, P.; Cruse,
R. W.; Gultneh, Y.; Hayes, J. C.; Toscano, P. J.; Zubieta, J. Inorg.
Chem. 1992, 31, 1436-1451.
(81) Dong, Y.; Me´nage, S.; Brennan, B. A.; Elgren, T. E.; Jang, H. G.;
Pearce, L. L.; Que, L., Jr. J. Am. Chem. Soc. 1993, 115, 1851-1859.
(82) Paul, P. P.; Tyekla´r, Z.; Jacobson, R. R.; Karlin, K. D. J. Am. Chem.
Soc. 1991, 113, 5322-5332.
Inorganic Chemistry, Vol. 44, No. 20, 2005 7025

Kim et al.
(O22-)-CuII]+ (6) through reaction stoichiometry measure-
ments, along with UV-vis, NMR, EPR, and resonance
Raman spectroscopies, shows that 6 is an S ) 2 system in
which high-spin iron(III) and copper(II) ions are antiferro-
magnetically coupled through a bridging peroxide ligand.
The fully reduced bis(carbonyl) iron(II)-copper(I) complex,
[(2L)Fe-CO‚‚‚Cu-CO]+ (4), or the partially reduced iron-
argon for 30 min and/or (followed) by three freeze/pump/thaw
cycles prior to introduction into the glovebox.
Room-temperature UV-vis spectra were obtained with a Cary
50 Bio spectrophotometer, and low-temperature UV-vis experi-
ments were carried out using a Hewlett-packard 8453 diode array
spectrometer equipped with HP Chemstation software, where the
spectrometer was equipped with a variable-temperature Dewar and
cuvette assembly.78,83 IR spectra were obtained at room temperature
using a Mattson Galaxy 4030 series FT-IR spectrophotometer.
MALDI-TOF-MS spectra were recorded on a Kratos Analytical
Kompact MALDI 4 mass spectrometer equipped with a 337 nm
nitrogen laser (20 kV extraction voltage). Calibration was performed
with an authentic sample of the porphyrin ligand F8H2. Elemental
analyses were performed by Desert Analytics, Tucson, AZ (for the
air-sensitive samples) or Quantitative Technologies Inc. (QTI),
Whitehouse, NJ.
-
(III)-copper(I) species, [(solvent)(2L)FeIII-(O2 )‚‚‚CuI-
(PPh3)]+ (5-PPh3), can be generated by adding carbon
monoxide or triphenylphosphine, respectively, to the solution
of [(2L)FeIII-(O22-)-CuII]+ (6). Liberation of O2 occurs in
the carbon monoxide reaction. Neither peroxo complex 6
nor superoxo complex 5-PPh3 effects the oxygenation of
triphenylphosphine.
Comparison of the physical properties of [(2L)FeIII-
(O22-)-CuII]+ (6) with those of other TMPA-based heme-
peroxo-Cu complexes is particularly interesting because they
have different peroxo-binding modes, where the former has
a µ-η2:η2-peroxo-binding structure while some of the latter
are known to possess a µ-η2:η1-peroxo ligation. Complex 6
has distinctive UV-vis features in the Q-band region with
three absorptions at ∼485, 540, and 572 (sh) nm, whereas
tetradentate systems have a major absorption band in lower
energy region around 560 nm. The more dramatic difference
between 6 and heme-peroxo-copper complexes with tet-
radentate chlelates is shown by resonance Raman spectros-
copy, where a significantly lowered ν(O-O) stretching
vibration occurs at 747 cm-1 in [(2L)FeIII-(O22-)-CuII]+ (6).
In fact, this is the lowest ν(O-O) frequency known for the
heme-peroxo copper complexes, and it is ∼50 cm-1 lower
compared to those from the tetradentate heme-peroxo-
copper or the side-on bound heme-only peroxo complexes.
The observed relationship between the ν(O-O) frequency
and the peroxo-binding mode (i.e., end-on µ-1,2- or side-on
η2:η2-) is analogous to that in dicopper-peroxo complexes.
Our study demonstrates that the nature of CuI/O2 interac-
tions can be applicable to the study of heme/CuI/O2 reactions.
It remains for future studies to see if the O-O bond
weakening in [(2L)FeIII-(O22-)-CuII]+ (6) or other systems
to be investigated leads to subsequent facilitated O-O
reductive bond cleavage reactions of possible relevance to
cytochrome c oxidase.
NMR Spectroscopy. 1H NMR spectra at room temperature were
recorded either on a Varian Unity 400 or Bruker 300 NMR
instrument. Variable-temperature, T1 measurement, and other het-
eronuclear NMR spectroscopic studies were carried out using a
Varian Unity 400 NMR instrument. All spectra were recorded in
5-mm o.d. NMR tubes. The chemical shifts were reported as δ
(ppm) values calibrated to natural abundance deuterium or proton
1
2
solvent peaks for H and H NMR spectra, whereas phosphoric
acid (0 ppm) was used as the reference for the 31P NMR spectra.
2H NMR spectra were collected using a tunable broad-band probe
2
to enhance H detection.
Solution magnetic moment measurements of [(2L)FeIII-(O22-)-
CuII]+ (6) were carried out on a Varian NMR instrument at 400
1
2
MHz (for H) and 61 MHz (for H). Inside a drybox, an acetone-
d6 solution of 3 with known concentration was prepared and placed
in a NMR tube. A coaxial reference capillary filled with pure
1
acetone-d6 was then inserted inside the sample tube. A H NMR
spectrum of this was obtained and referenced as normal, and a 2H
NMR spectrum of the same sample was taken for an unambiguous
peak assignment of the two solvent peaks, i.e., one corresponding
to the paramagnetically shifted deuterated solvent peak and the
second corresponding to the pure deuterated solvent from the
reference capillary. The magnetic moment was calculated from the
equation µB ) 2.84 x T/n, where T is the temperature (K) of the
x
M
measurement, n is the nuclearity of the complex (n ) 1 for 6), and
øM ) -3/4π(∆ν/ν)1000/c + øMsol - øD (where øMsol is the solvent
susceptibility, øD is the total diamagnetic correction (calculated from
Pascal’s constants),84 ∆ν is the paramagnetic shift of the solvent
in Hz, ν is the frequency of the NMR instrument in Hz, and c is
the concentration of the metal complex).
Electron Paramagnetic Resonance (EPR) Spectroscopy. In a
glovebox, a 1 mM THF/toluene (2:1 v/v ratio; toluene to faciliate
“glass” formation) solution of [(2L)FeIICuI]+ (3) was prepared and
transferred to an EPR tube, which was capped with a rubber septum
and transferred to a -95 °C cold bath (N2/hexane). Dry dioxygen
(passed through a column filled with Drierite) was added through
the solution via syringe to generate [(2L)FeIII-(O22-)-CuII]+ (6).
The resulting solution was allowed to sit at -95 °C for 10 min for
complete formation of 6, after which the solution was frozen in
liquid N2 to obtain a glass. The X-band EPR spectrum of 6 was
recorded on a Brucker EMX spectrometer, with temperature
maintained at 15 K using an Oxford Instruments EPR 900 cryostat.
Experimental Section
General Methods. All reagents and solvents were purchased
from commercial sources and were of reagent quality unless
otherwise stated. Air-sensitive compounds were handled under
argon atmosphere by using standard Schlenk techniques or in a
Mbraun Labmaster 130 inert-atmosphere (<1 ppm O2, <1 ppm
H2O) glovebox filled with nitrogen. Propionitrile (EtCN) and
heptane were distilled over calcium hydride under argon. Aceto-
nitrile (CH3CN), dichloromethane (CH2Cl2), tetrahydrofuran (THF),
and toluene were purified over an activated alumina column under
nitrogen unless otherwise noted. THF for the [(2L)FeIICuI]+ (3)
synthesis was distilled from sodium/benzophenone under Ar. For
kinetic studies, dichloromethane was distilled from calcium hydride,
THF was distilled from sodium/benzophenone, and EtCN was
distilled first over P2O5 and then over calcium hydride prior to use.
Deoxygenation of these solvents was achieved by bubbling with
(83) 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.
(84) CRC Hand Book of Chemistry and Physics, 80th ed.; CRC Press:
London, 1999.
7026 Inorganic Chemistry, Vol. 44, No. 20, 2005

Cu Ligand Influences on Heme-Peroxo-Cu Formation
Resonance Raman Spectroscopy. In the glovebox, 1-5 mM
THF solutions of [(2L)FeIICuI]+ (3) were prepared and transferred
to NMR tubes (∼0.5 mL of solution/tube) and capped with tight-
fitting septa. The sample tube were placed in a -95 °C cold bath
(N2/Hexane) for 10 min and oxygenated by using 16O2, 18O2, and
16/18O2. (18O2 and 16/18O2 gases were purchased from ICON, Summit,
NJ. 16O2 was purchased from BOC gases, Murray Hill, NJ.) The
labeled gases were cooled in dry ice bath for 5 min and injected
into the solution by using a Hamiltion gastight syringe. The
oxygenated samples were set in a cold bath for ∼10 min, after
which the sample tubes were frozen in liquid N2 and sealed under
their respective labeled O2 atmosphere. RR spectra were collected
on a McPherson 2061/207 spectrograph equipped with a liquid-
nitrogen-cooled LN-1100PB CCD detector. The 413-nm laser
excitation was obtained from a Coherent Innova 302 krypton laser,
and the intensity at the sample was kept below 10 mW to prevent
adverse effect from the laser illumination. A matching Kaiser
Optical supernotch filter was used to minimize interference from
the Rayleigh scattering. Frequencies were calibrated relative to
several frequency standards and are accurate to (1 cm-1. To keep
the sample at 90 K during data acquisition, the samples were kept
in a copper coldfinger immersed in liquid nitrogen. The samples
were also spun within the holder to minimize photodegradation.
Stopped-Flow Kinetics. Rapid kinetics were followed using an
SF-21 variable-temperature stopped-flow unit (Hi-Tech Scientific,
2 mm path length cell) combined with a TIDAS/NMC301-MMS/
16 VIS/500-1 diode array spectrometer (J & M; 256 diodes, 300-
1100 nm, 0.8 ms minimum sampling time). Data acquisition (up
to 256 complete spectra; up to four different time bases) was
performed using the Kinspec program (J & M). For numerical
analysis, all data were pretreated by factor analysis using the Specfit
program.85 To obtain variable O2 concentrations in solution, a gas
mixing unit was employed which consisted of two MKS PR-4000
control towers equipped with MKS general purpose Mass-Flo
controllers of type 1179A calibrated for either 200 sccm N2 (used
for O2 regulation) or 500 sccm N2 (used for argon regulation). The
regulated amounts of argon and O2 were mixed which yielded a
specific ratio of O2 to argon. The mixed gases were passed through
a drying column and then bubbled through the solution to yield a
specific concentration of O2 in the solvent. For the kinetic studies,
CH2Cl2 (UVASOL, Merck) and EtCN (pa, Merck, predistilled from
phosphorus pentaoxide) were distilled from calcium hydride and
under argon prior to use. Thermal expansion of the solvent was
taken into account.
used for the analysis of k2. The concentration profile based on
the kinetic model [(2L)FeIICuI]+ (3) + O2 f [(EtCN)(2L)FeIII-
(O2-)‚‚‚CuI(NCEt)]+ (5) and [(EtCN)(2L)FeIII-(O2-)‚‚‚CuI(NCEt)]+
(5) f [(2L)FeIII-(O22-)-CuII]+ (6), followed by (in some cases) a
slow first-order relaxation, was calculated by numerical integration
using the SPECFIT/32 program. Thermal expansion of CH2Cl2 was
taken into account using r (g/mL) ) [1.48177 + 2.61843 × 10-3
X
- 1.7766 × 10-5X2 + 2.38421 × 10-8X3, where X ) T (K)] for
the solvent density as a function of temperature. It should be noted
that the presence of 6% propionitrile was not considered to change
the temperature dependence of the solvent density from that of pure
CH2Cl2. A linear combination of the density correction factors for
6% propionitrile and 94% CH2Cl2 yielded results similar to those
obtained using 100% CH2Cl2. Figure S10 shows the Eyring plot
for the measurement of k1 (the superoxo formation), and Figure
S11 shows the Eyring plot for the measurement of k2 (the peroxo
formation).
For the reaction of (F8)FeII/[CuI(LMe2N)]+ with O2 in CH2Cl2/
6% EtCN (Scheme S1 in Supporting Information), three series of
different (F8)FeII, [CuI(LMe2N)]+, and dioxygen concentrations were
used to carry out a total of 289 measurements between -105 and
+20 °C. Of these 192 were used for the final analysis. The
concentrations of FeII, CuI, and dioxygen solutions used were the
following: 0.15 mM FeII, 0.20 mM CuI, 1.90 mM O2; 0.27 mM
FeII, 0.39 mM CuI, 1.90 mM O2; 0.15 mM FeII, 0.26 mM CuI, 0.63
mM O2. Reaction time measured ranged from 1.7 to 46 s. Figure
S12 shows the Eyring plot for the measurement of k1 (the superoxo
formation), and Figure S13 shows the Eyring plot for the measure-
ment of k2 (the peroxo formation).
Spectrophotometric Titration of [(2L)FeIII-(O22-)-CuII]+ (6).
In the glovebox, a known concentration (0.2-0.4 mM range in
toluene/10% MeCN) of a stock solution of [(2L)FeIICuI]+ (3) was
prepared and 2 mL of the solution was transferred to a 2 mm path
length Schlenk cuvette. This was capped with a 14/24 adapter
(purchased from Chemglass CG1036-14) whose opposite end was
designed to fit a rubber septum. Two appropriate size rubber septa
and a Teflon septum were used to seal the Schlenk cuvette from
the atmosphere. The cuvette assembly was then transferred to the
low-temperature UV-vis spectrometer and cooled to -75 °C.
Separately, a 250 mL two sidearm Schlenk flask, whose neck was
closed with a rubber septum and one sidearm was connected to an
oil bubbler, was filled with 1.01% O2 in Ar gas mixture (custom
mixture purchased from Potomac Airgas, Baltimore, MD) under 1
atm pressure. A Hamilton gastight syringe equipped with a 3-way
purge valve (for Ar purging of the syringe and the needle) was
used to transfer known amounts of the 1.01% O2 gas to the cold
solution of 3. To the cuvette containing a solution of 3, was added
0.5 equiv, 1 equiv, and 2 equiv of dioxygen in three separate
experiments. Each aliquot of dioxygen was bubbled directly into
the solution via syringe, and the resulting solution was allowed to
react until no spectral change was observed in the UV-vis spectrum
(typically, it takes ∼60-70 h), after which excess dioxygen was
added to the solution to reach full formation of 6 for comparison.
Note that these very long reaction times are required to complete
the irreversible (at these temperatures) Fe-Cu complex oxygenation
reaction, when a near stoichiometric amount of O2 gas is being
added to the cuvette assembly. Whereas about half the amount of
reduced starting material 3 was present with addition of 0.5 equiv
of O2, ∼90% of full formation of 6 was obtained with 1 equiv of
O2 and complete oxygenation was obtained with 2 equiv of O2.
(Consecutive addition of smaller aliquots, e.g., 0.2 equiv, of
dioxygen to the same solution did not give reliable results
presumably due to the relatively poor O2-binding ability of 3 under
For the reaction of [(2L)FeIICuI]+ (3) with O2 in CH2Cl2/6%
EtCN, 6 series of experiments were performed with the following
room-temperature reactant postmixing concentrations ([3]/[O2]/
series name): 1.64 × 10-4 M/1.90 × 10-3 M/ESTA; 2.37 × 10-4
M/1.90 × 10-3 M/ESTB; 2.26 × 10-4 M/1.90 × 10-3 M/ ESTC;
1.11 × 10-4 M/1.90 × 10-3M/ESTD; 1.10 × 10-4 M/6.00 × 10-4
M/ESTE; 2.16 × 10-4 M/6.00 × 10-4 M/ESTF. Figure 6 was made
from ESTF4 which was data collected at 168 K. Data for a total of
329 individual runs between 168 and 193 K with data collection
times between 0.04 and 1.48 s were obtained. Of these, 159 data
sets (over a temperature range 168-179 K, where the formation
of the [(2L)FeIII-(O2-)‚‚‚CuI]+ (5) superoxo species could be
observed) were used for analysis of k1 and 278 data sets (over the
temperature range 168-191 K, where the formation of the [(2L)-
FeIII-(O22-)-CuII]+ (6) peroxo species could be observed) were
(85) SPECFIT/32 is a trademark of Spectrum Software Associates,
copyright 2000-2002 Spectrum Software Associates (R. Binstead and
A. D. Zuberbu¨hler). Original publication: Gampp, H.; Maeder, M.;
Meyer, C. J.; Zuberbu¨hler, A. D. Talanta 1985, 32, 95-101.
Inorganic Chemistry, Vol. 44, No. 20, 2005 7027

Kim et al.
the limited O2 concentration.) Coupling this result with that of the
O2 evolution experiment of 6 with CO (see below), the reaction
stoichiometry of 3:O2 was thus concluded to be 1:1. The spectral
changes for the reaction of 6 with 1 equiv of O2 described here are
shown in the Supporting Information, Figure S6.
Reactions of [(2L)FeIII-(O22-)-CuII]+ (6) with Triphenylphos-
phine. (a) UV-Vis Analysis. The reaction of 6 with 1, 5, and 10
equiv of triphenylphosphine was monitored by UV-vis spectros-
copy in the presence and the absence of excess O2. Inside the
glovebox, typically a 0.3-0.5 mM solution of [(2L)FeIICuI]+ (3)
was prepared and 4 mL of the stock solution was transferred to a
Schlenk cuvette that was capped with a rubber septum. The sample
was brought to the low temperature in a UV-vis spectrometer and
cooled at -80 °C. The peroxo complex 6 was generated by bubbling
dioxygen through the solution of 3. If needed, excess O2 was
removed by 10 cycles of vacuum/Ar purge. Separately, a triphen-
ylphosphine stock solution (30-50 mM) was prepared inside the
glovebox, and a proper amount of the solution was taken using a
microsyringe, which then was added to the cold solution of 6. The
spectral changes are shown in Figure 8.
Detection of O2 Evolution by Pyrogallol Solution. Evolution
of dioxygen from the reaction of the [(2L)FeIII-(O22-)-CuII]+ (6)
with CO was detected using an alkaline pyrogallol solution. 1,2,3-
Trihydroxybenzene (≡pyrogallol, 2 g) was placed in a 50 mL
Schlenk flask that was modified to have a 1 cm path length cuvette
as a sidearm, to which was added 13 mL of deoxygenated 50%
KOH aqueous solution under Ar. The Schlenk flask was capped
with a 14/20 joint that was sealed with triple rubber septa (see
above), and the UV-vis spectrum of the solution was recorded.
Using a Hamilton gastight syringe equipped with a 3-way purge
valve (for Ar purging of the syringe and the needle), 6 mL of 1.01%
O2 in Ar gas mixture was added to the alkaline pyrogallol solution.
The headspace of the flask and the pyrogallol solution turned brown
upon addition of O2. The flask was shaken for effective mixing for
10 min, after which the UV-vis spectrum of the solution was
recorded. An additional 6 mL of 1.01% dioxygen was added
followed by the same procedure, e.g., mixing and recording the
UV-vis spectrum. This was repeated four more times, thereby
addition of a total of 36 mL of 1.01% O2 gas in the solution at the
end. A calibration curve was generated by plotting the absorbance
at 409 nm versus amount of the 1.01% O2 added. The UV-vis
spectra and the calibration curve are shown in the Supporting
Information, Figure S8.
(b) 2H NMR Analysis. The reaction of 6 with 1 equiv
triphenylphosphine was followed by low-temperature 2H NMR
spectroscopy. Inside the glovebox, a THF solution of [(2L-d8)FeII-
CuI]+ (3) (11 mg of 3 in 0.5 mL of THF) was prepared inside a 5
mm NMR tube, and the tube was transferred to a -95 °C cold
bath (hexane/N2). 6 was generated by direct bubbling O2 through
the solution via syringe. Complete formation of 6 was confirmed
(by 2H NMR spectroscopy as well) prior to adding triphenylphos-
phine. A 1 equiv amount of PPh3 solution was directly added to
the solution 6 via syringe at -95 °C, after which the NMR spectrum
2
was recorded at -90 °C. The H NMR spectra described here are
shown in Figure 9.
(c) 31P NMR Analysis. The reaction of 6 with 2 equiv of
triphenylphosphine was followed by 31P NMR spectroscopy at room
temperature. In a 10 mL Schlenk flask equipped with a magnetic
stir bar, 6 was generated by direct O2 bubbling through solution of
3 (15 mg of 3 in 3 mL of acetone-d6) at -93 °C. The excess
dioxygen was removed by 10 cycles of vacuum/Ar purging. A 2
equiv amount of PPh3 solution (in acetone-d6) were added to 6 via
syringe under Ar. The resulting reaction mixture was stirred for 1
h at -93 °C, which then was allowed to warm to room temperature.
After concentration of the solution under vacuum to approximately
0.5 mL, the solution was transferred to a NMR tube and the 31P
NMR spectrum was recorded. No signal for triphenylphosphine
oxide was observed. A significantly broadened 31P resonance from
triphenylphosphine appeared at -2.0 ppm (Figure S9, Supporting
Information), which was downfield shifted from that of an authentic
PPh3 (-4.1 ppm). Dynamic behavior between free PPh3 and copper-
bound PPh3 may be present in the sample, which could cause the
line broadening and the change of the chemical shift. To test this
possibility, the 31P NMR spectrum of [CuI(LH)(PPh3)]+ was
recorded. An extremely broad 31P resonance was found at 3 ppm
in acetone-d6.
In the glovebox, to a 10 mL Schlenk flask was added a known
amount (typically 11-15 mg) of [(2L)FeIICuI]+ (3) in 3 mL of THF.
The solution was transferred to a cold bath (hexane/N2), and the
peroxo complex [(2L)FeIII-(O22-)-CuII]+ (6) was generated by
direct bubbling of dioxygen through the solution via syringe. After
10 min at -95 °C, the rubber septum of the flask was replaced
with a 14/20 glass joint sealed with triple rubber septa (see above),
and the excess O2 was removed by three freeze-pump-thaw
cycles. Separately, a 250 mL Schlenk flask was filled with pure
CO (CO was purchased from BOC gases, Murray Hill, NJ, and
used after passing through an oxygen trap OT3 which was
purchased from Agilent). Using a Hamilton gastight syringe
equipped with a 3-way purge valve, 20 mL of CO gas (2 × 10
mL) were added to the solution of 6 at -95 °C. The resulting
solution was allowed to warm to room temperature. In the
meantime, 13 mL of alkaline pyrogallol solution was freshly
prepared in a modified 50 mL Schlenk flask possessing a 1 cm
path length cuvette. A short vacuum tubing was attached to the
sidearm of each Schlenk flask (one containing the pyrogallol
solution, the other containing the reaction mixture of 6 plus CO).
Then, the two flasks were connected to a three-way glass stopcock
whose third arm was connected to a standard Schlenk line (the other
two arms are connected via vacuum tubing to the Schlenk flasks).
After ensuring that the connecting vacuum tubing was under inert
atmosphere, the headspace of the 50 mL Schlenk flask which
contains the pyrogallol solution was placed under static vacuum,
after which the three-way glass stopcock and the sidearm stopcocks
of the Schlenk flasks were opened so that the two flasks were
connected but they were isolated from the atmosphere. The
pyrogallol solution immediately turned brown. The two flasks were
allowed to sit for 2 h, and the UV-vis spectrum of the resulting
solution was recorded. The amount of O2 evolved was calculated
from the calibration curve described above (Figure S8, Supporting
Information). Three reaction runs produced 63%, 65%, and 80%
yields.
Synthesis. (F8)FeII‚H2O,23,60 [(LH)CuI]B(C6F5)4,21,2LH2, (2L)FeII‚
THF (1‚THF),19 and [(2L)FeIICuI]B(C6F5)4‚THF‚H2O (3‚THF‚
H2O)19 were synthesized according to published procedures.
F6(NO2)TPPH2-d8, 5,10,15-Tris(2,6-difluorophenyl)-20-(2-ni-
trophenyl)porphyrine-d8. To a 1 L round-bottom flask equipped
with a magnetic stir bar was added 400 mL of propionic anhydride,
56 mL of deuterium oxide, and 12 g (0.18 mol) of pyrrole. The
mixture was brought to reflux for 1 h under Ar, after which
2-nitrobenzaldehyde (7.6 g, 0.05 mol) and 2,6-difluorobenzaldehyde
(12.5 g, 0.09 mol) and 100 mL of nitrobenzene were added through
the top of the condenser. The reaction mixture was allowed to reflux
with stirring for an additional 1.5 h. The solution was cooled to
room temperature, opened to the air, and stored in a freezer (-15
°C) for 2 days. The cold reaction mixture was filtered over Celite,
and the precipitate (porphyrins) was washed with 3 L of distilled
7028 Inorganic Chemistry, Vol. 44, No. 20, 2005

Cu Ligand Influences on Heme-Peroxo-Cu Formation
water, after which the precipitate and the Celite were rinsed with
CH2Cl2 (2 L). The filtrate was saved and dried over sodium sulfate
for 1 h. After removal of the solvent on the rotary evaporator, the
crude product was purified by column chromatography (silica, 1:1
CH2Cl2:hexane, Rf ) 0.2), yielding 1 g (0.0014 mol, 5%) of the
after which 35 mL of deoxygenated pentane was added to
precipitate the product. After decanting of the supernatant, the solid
was dried under vacuum, yielding 25 mg (0.011 mmol, 47% yield)
of [(2L)FeIII-OH‚‚‚CuI(PPh3)]B(C6F5)4‚THF‚3H2O as a purple
solid. Anal. Calcd for C107H72O6BN8F26FeCuP: C, 57.87; H, 3.27;
1
desired product. By H NMR analysis, ∼85% of the â-pyrrolic
N, 5.05. Found: C, 57.86; H, 3.08; N, 4.59. UV-vis [THF; λmax,
nm; room temperature]: 335, 413, 565.
hydrogens were determined to be deuterated. 1H NMR (300 MHz,
CDCl3): δ (ppm) 8.67-8.86 (m, 1.2 H), 8.45 (d, 1H), 8.21 (d,
1H), 7.86-7.96 (m, 2H), 7.72-7.82 (m, 3H), 7.32-7.40 (m, 6H),
-2.70 (s, 2H).
Acknowledgment. This work was supported by the
National Institutes of Health (K.D.K. (Grant GM 60353),
P.M.-L (Grant GM18865), postdoctoral fellowships to M.E.H.
(Grant GM20805)), and the Swiss National Science Founda-
tion (A.D.Z.). We are also thankful to Dr. Marie-Aude Kopf
for initial observations made in a similar system.
[(2L)FeIII-OH‚‚‚CuI(PPh3)]B(C6F5)4‚THF‚3H2O. Inside the
glovebox, 45 mg (0.024 mmol) of [(2L)FeIICuI]B(C6F5)4‚THF‚H2O
(3‚THF‚H2O) was placed in a 50 mL Schlenk flask, to which was
added 5 mL of deoxygenated THF. The flask was capped with a
rubber septum and brought to a -95 °C bath (hexane/N2), after
which dry dioxygen was bubbled through the solution of 3. The
resulting solution was allowed to sit at -95 °C for 15 min to ensure
complete formation of [(2L)FeIII-(O22-)-CuII]+ (6). Excess di-
oxygen was removed by application of 10 cycles of vacuum/Ar
purge. Then, 1 equiv of triphenylphosphine (6.3 mg in 200 µL)
was added to the solution via syringe. The resulting mixture was
allowed to sit at -95 °C for 15 min, followed by warming to room
temperature. The solution was concentrated in vacuo to ∼3 mL,
Supporting Information Available: An X-ray crystallographic
2
2
data file (CIF), UV-vis and H NMR spectra of 1/O2, H NMR
spectra of 3, O2 titration (UV-vis monitored) of 3, UV-vis and
IR spectra of 6/CO, UV-vis spectral change of an alkaline
pyrogallol solution upon stepwise additions of O2, 31P NMR spectra
of [(2L)FeIII-OH‚‚‚CuI(PPh3)]+, and kinetics Eyring plots. The
materialisavailablefreeofchargeviatheInternetathttp://pubs.acs.org.
IC050446M
Inorganic Chemistry, Vol. 44, No. 20, 2005 7029