Successful stabilization of the elusive species {FeNO}8 in a heme model

Article abstract of DOI:10.1021/ja905062w

Nitroxyl (HNO/NO^-) heme-adducts have been postulated as intermediates in a variety of catalytic processes carried out by different metalloenzymes. Hence, there is growing interest in obtaining and characterizing heme model nitroxyl complexes. The one-electron chemical reduction of the {FeNO}⁷ nitrosyl derivative of Fe^III(TFPPBr ₈)Cl, Fe^II(TFPPBr₈)NO (1) (TFPPBr₈ = 2,3,7,8,12,13,17,18-octabromo-5,10,15,20-[Tetrakis-(pentafluorophenyl)] porphyrin) with cobaltocene yields the significantly stable {FeNO}⁸ complex, [Co(C₅H₅)₂]⁺[Fe(TFPPBr ₈)NO]^- (2). Complex 2 was isolated and characterized by UV - vis, FTIR, ¹H and ¹⁵N NMR spectroscopies. In addition, DFT calculations were performed to get more insight into the structure of 2. According to the spectroscopic and DFT results, we can state unequivocally that the surprisingly stable complex 2 is the elusive {FeNO}⁸ species. Both experimental and computational data allow to assign the electronic structure of 2 as intermediate between Fe^IINO^- and Fe ^INO, which is contrasted with the predominant Fe^IINO ^- character of known nonheme {FeNO}⁸ complexes. The enhanced stability achieved for a heme model {FeNO}⁸ is expected to allow further studies related to the reactivity of this elusive species.

Full text of DOI:10.1021/ja905062w

Published on Web 12/31/2009
Successful Stabilization of the Elusive Species {FeNO}8 in a
Heme Model
Juan Pellegrino, Sara E. Bari, Damia´n E. Bikiel, and Fabio Doctorovich*
Departamento de Qu´ımica Inorga´nica, Anal´ıtica, y Qu´ımica F´ısica, Facultad de Ciencias
Exactas y Naturales, UniVersidad de Buenos Aires, INQUIMAE-CONICET, Ciudad
UniVersitaria, Pab. 2, C1428EHA Buenos Aires, Argentina
Received June 19, 2009; E-mail: doctorovich@qi.fcen.uba.ar
Abstract: Nitroxyl (HNO/NO-) heme-adducts have been postulated as intermediates in a variety of catalytic
processes carried out by different metalloenzymes. Hence, there is growing interest in obtaining and
characterizing heme model nitroxyl complexes. The one-electron chemical reduction of the {FeNO}7 nitrosyl
derivative of FeIII(TFPPBr8)Cl, FeII(TFPPBr8)NO (1) (TFPPBr8 ) 2,3,7,8,12,13,17,18-octabromo-5,10,15,20-
[Tetrakis-(pentafluorophenyl)]porphyrin) with cobaltocene yields the significantly stable {FeNO}8 complex,
1
[Co(C5H5)2]+[Fe(TFPPBr8)NO]- (2). Complex 2 was isolated and characterized by UV-vis, FTIR, H and
15N NMR spectroscopies. In addition, DFT calculations were performed to get more insight into the structure
of 2. According to the spectroscopic and DFT results, we can state unequivocally that the surprisingly
stable complex 2 is the elusive {FeNO}8 species. Both experimental and computational data allow to assign
the electronic structure of 2 as intermediate between FeIINO- and FeINO, which is contrasted with the
predominant FeIINO- character of known nonheme {FeNO}8 complexes. The enhanced stability achieved
for a heme model {FeNO}8 is expected to allow further studies related to the reactivity of this elusive species.
1. Introduction
The complex Fe(NO)(cyclam-ac) was obtained by one-
electron reduction of the {FeNO}7 complex in CH3CN solution,
giving a very air sensitive product in ∼40% yield, and complete
spectroscopic characterization was presented.7 Spectroscopic
evidence for a six-coordinate deprotonated nitroxyl complex was
further supported by density functional theory (DFT) calcula-
tions: the electronic structure of the complex was described as
a low spin FeII(NO-). More recently, another non-heme
{FeNO}8 complex was spectroscopically characterized in aque-
ous solution, after the two-electron reduction of the nitroprusside
anion, [Fe(CN)5(NO)]2-; this complex allowed the first pKa
determination of bound HNO, by H NMR. While the proto-
nated complex, [Fe(CN)5(HNO)]3- is stable in oxygen-free
solutions, its conjugate base, [Fe(CN)5(NO)]4-, decomposed to
[Fe(CN)5(NO)]2- with a half-life of about 50 min.
The pioneering reports by Kadish et al. on the spectroelec-
trochemical reduction of Fe(TPP)NO and Fe(OEP)NO (TPP )
5,10,15,20-[Tetrakis-(phenyl)]porphyrin,OEP)2,3,7,8,12,13,17,18-
The physiological importance of nitric oxide, NO, was
accepted around two decades ago, and since then, a growing
interest in the chemistry of heme-nitrosyls has started, as both
the formation and activity of NO in vivo are mediated by heme
proteins.1 There has also been considerable interest over the
past decade in the one-electron-reduced form of nitric oxide,
nitroxyl (NO- or HNO), because of the growing evidence for
an independent biological significance of free nitroxyl itself.2
This species might be created in vivo by nitric oxide synthase
in the absence of its reduced biopterin cofactor.3 In addition,
nitroxyl (HNO/NO-) heme-model complexes ({FeNO}8, ac-
cording to the Enemark-Feltham notation) have received special
attention due to the intermediacy of nitroxyl-heme adducts in
a variety of catalytic processes related to the biogeochemical
cycle of nitrogen.4,5 A better understanding of the structure and
reactivity of these complexes is desirable and could help to
elucidate mechanistic issues. While there are several examples
of nitroxyl complexes with second- or third-row transition metals
in the literature,6 only a few iron-nitroxyl complexes have been
reported so far.
8
1
(5) (a) Averill, B. A. Chem. ReV. 1996, 96, 2951–2964. (b) Garber,
E. A. E.; Hollocher, T. C. J. Biol. Chem. 1982, 257, 4705–4708. (c)
Dermastia, M.; Turk, T.; Hollocher, T. C. J. Biol. Chem. 1991, 266,
10899–10905. (d) Zumft, W. G. Arch. Microbiol. 1993, 160, 253–
264. (e) Shiro, Y.; Fujii, M.; Iizuka, T.; Adachi, S. I.; Tsukamoto, K.;
Nakahara, K.; Shoun, H. J. Biol. Chem. 1995, 270, 1617–1623. (f)
Shiro, Y.; Fujii, M.; Isogai, Y.; Adachi, S. I.; Iizuka, T.; Obayashi,
E.; Makino, R.; Nakahara, K.; Shoun, H. Biochemistry 1995, 34, 9052–
9058. (g) Nakahara, K.; Tamimoto, T.; Hatauro, K.; Usuda, K.; Shoun,
H. J. Biol. Chem. 1993, 268, 8350–8355.
(1) Thomas, D. D.; Miranda, K. M.; Colton, C. A.; Citrin, D.; Espey,
M. G.; Wink, D. A. Antioxid. Redox Signaling 2003, 5, 307–317.
(2) (a) Stamler, J. S.; Singel, D. J.; Loscalzo, J. Science 1992, 258, 1898–
1902. (b) Hughes, M. N. Biochim. Biophys. Acta 1999, 1411, 263–
272.
(6) Farmer, P. J.; Sulc, F. J. Inorg. Biochem. 2005, 99, 166–184.
(7) Garc´ıa Serres, R.; Grapperhaus, C. A.; Bothe, E.; Bill, E.; Weyher-
muller, T.; Neese, F.; Wieghardt, K. J. Am. Chem. Soc. 2004, 126,
5138–5153.
(3) Rusche, K. M.; Spiering, M. M.; Marletta, M. A. Biochemistry 1998,
37, 15503–15512.
(4) {FeNO}n is the generalized description of the metal-NO bonding,
where n is the sum of the metal d-electrons and the nitrosyl π*
electrons. Enemark, J. H.; Feltham, R. D. Coord. Chem. ReV. 1974,
13, 339–406.
(8) Montenegro, A. C.; Amorebieta, V. T.; Slep, L. D.; Mart´ın, D. F.;
Roncaroli, F.; Murgida, D. H.; Bari, S. E.; Olabe, J. A. Angew. Chem.,
Int. Ed. 2009, 48, 4242–4245.
9
10.1021/ja905062w  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 989–995 989

A R T I C L E S
Pellegrino et al.
obtained. FeIII(TFPPBr8)Cl assembles the ease of preparation
and a highly positive shifted FeIII/FeII reduction potential (+600
mV from FeTPP).15 In this contribution, we describe the
preparation, isolation, and spectroscopic characterization of a
{FeNO}8 complex, [Co(C5H5)2]+[Fe(TFPPBr8)NO]- (2), ob-
tained from reduction of 1 with the one-electron reductant
cobaltocene. The electronic structure of the new complex was
assessed by DFT calculations.
2. Experimental Section
2.1. Syntheses. 5,10,15,20-[Tetrakis-(pentafluorophenyl)]por-
phyrin (H2TFPP) was purchased from Frontier Scientific and used
as received. All other reagents were used as received, except
Co(C5H5)2 (Sigma-Aldrich) that was sublimed prior to use. All
solvents were distilled and dried according to conventional proce-
dures. The porphyrin ligand TFPPBr8 and its iron(III) complex
FeIII(TFPPBr8)Cl were prepared according to slightly modified
published procedures, starting from H2TFPP.16
Figure 1. [FeIII(TFPPBr8)]+.
octaethyl-5,10,15,20-[Tetrakis-(phenyl)]porphyrin) provided the
first evidence of {FeNO}8 porphyrinate complexes, in dichlo-
romethane solutions.9 Following the track, Ryan et al. prepared
[Fe(TPP)NO]- in THF solution by both electrochemical and
chemical reduction, and afforded further structural and reactivity
insight.10 Albeit the anaerobic THF solution stabilized the
{FeNO}8 species, dichloromethane solutions gave back the
{FeNO}7 precursor, with a half-life of 30 min even at low
temperatures. Further reduction allowed envisaging the presence
of a {FeNO}9 species by UV-vis, and exhaustive electrolysis
in the presence of weak acids yielded hydroxylamine and
ammonia as ultimate products.10
Stability of {FeNO}8 complexes in aqueous solutions was
first attained for a nitroxyl-myoglobin adduct, MbHNO,11 and
more recently in a series of FeII(globin) adducts by Farmer et
al.12 In these scenarios, the distal pocket residues provide extra
stabilization to the bound HNO by hydrogen-bonding. The
Mb(HNO) was fully characterized by NMR, Raman, and X-ray
absorption spectroscopies.11,13 This complex is a six-coordinate
HNO-adduct, and direct evidence of the H-N bond was obtained
by 1H NMR. Remarkably, the series of FeII(globin)HNO proteins
represent the only {FeNO}8 complexes obtained by direct trapping
of free HNO from HNO-donor compounds.12,14
Apparently, the main reason for the elusive nature of the NO-
iron complexes previously reported is the high ease of oxidation
to the stable {FeNO}7 form. In order to enhance the stabilization
of the heme model {FeNO}8 moiety without the support of a
protein environment, we focused on porphyrinates bearing
electron-withdrawing substituents, aiming to tune the reduction
potential of the heme-coordinated nitroxyl. In this work, we
have synthesized FeIII(TFPPBr8)Cl (TFPPBr8 ) 2,3,7,8,12,13,17,
18-octabromo-5,10,15,20-[Tetrakis-(pentafluorophenyl)]porphy-
rin, Figure 1) and the {FeNO}7 nitrosyl iron complex
FeII(TFPPBr8)NO (1) because it was hoped that upon one-
electron reduction a fairly stable {FeNO}8 complex would be
2.1.1. Fe(TFPPBr8)NO (1). This complex was prepared by
reductive nitrosylation17 of FeIII(TFPPBr8)Cl. NO was bubbled
through a solution of FeIII(TFPPBr8)Cl (50 mg; 0.03 mmol) in
degassed dichloromethane (10 mL) and methanol (5 mL) under a
nitrogen atmosphere. NO was generated by dropwise addition of a
solution of 500 mg of NaNO2 in 8 mL of degassed water to a
solution of 2.2 g of FeSO4 ·7H2O and 1 mL of H2SO4 in 15 mL of
degassed water. NO was passed through a KOH column to remove
higher oxides. The reaction was followed by UV-vis spectroscopy.
The dilute solutions of FeIII(TFPPBr8)Cl were brownish yellow,
while those of 1 were greenish yellow. The reaction was easily
followed by the shift of the Soret band from 402-442 to 430 nm.
When maximum transformation was achieved, the solvent was
removed in Vacuo, and the product was purified by column
chromatography in a drybox (stationary phase: acid alumina; mobile
phase: anhydrous and degassed CH2Cl2); the first fraction was
collected, the solvent was evaporated, and the solid residue was
dried under vacuum. Yield: 0.030 g (60%). Anal. Calcd. for
C44N5OFeBr8F20: C, 31.28; H, 0; N, 4.15. Found: C, 31.2; H, not
detected; N, 4.3. UV-vis [CH2Cl2]: λmax (ε/104) ) 430 nm (10.6),
580 nm (1.37). IR (cm-1) (KBr pellet): νNO ) 1726, isotope shift
for the 15N(O)-labeled compound: -31 cm-1
.
2.1.2. [Co(Cp)2]+[Fe(TFPPBr8)NO]- (2). Co(Cp)2 was sub-
limed by heating at no more than 60 °C, under vacuum (0.1 mmHg),
and kept and manipulated in a drybox. In a typical experiment, a
solution of Co(Cp)2 of known concentration was achieved by
dissolving 10 mg of the freshly sublimed reagent in anhydrous
CH2Cl2 to a final volume of 1 mL. In a drybox, one equivalent of
the dissolved Co(Cp)2 was added (approximately 100 µL) to a
solution of 8 mg of 1 in 100 µL of anhydrous CH2Cl2. The product,
which precipitated after the addition of 3 mL of cold hexane, was
separated from the solution by centrifugation, and the solid was
dried under vacuum. Isolated yield: 90%. Anal. Calcd for
C54H10Br8F20N5OFeCo: C, 34.52; N, 3.73; H, 0.54. Found: C: 34.6;
N, 3.6; H, 0.7. UV-vis [CH2Cl2]: λmax (ε/104) ) 430 nm (8.14),
580 nm (1.65). IR (cm-1) (solid film or CH2Cl2 solution): νNO
)
1
1550 (sh). H NMR (CD2Cl2): 4.94 (br s, 10H, [CoC10H10]+). 15N
NMR (CD2Cl2): 790 (s, br) (vs CH3NO2).
(9) (a) Lancon, D.; Kadish, K. M. J. Am. Chem. Soc. 1983, 105, 5610–
5617. (b) Olson, L.; Schaeper, D.; Lancon, D.; Kadish, K. M. J. Am.
Chem. Soc. 1982, 104, 2042–2044.
2.1.3. Synthesis of 15N(O)-Labeled 1 and 2. Fe(TFPPBr8)15NO
was prepared as described above, using 15NO prepared from
Na15NO2; [Co(Cp)2]+[Fe(TFPPBr8)15NO]- was prepared in a man-
(10) Choi, I. K.; Liu, Y.; Feng, D.; Paeng, K. J.; Ryan, M. D. Inorg. Chem.
1991, 30, 1832–1839.
(11) Lin, R.; Farmer, P. J J. Am. Chem. Soc. 2000, 122, 2393–2394.
(12) Kumar, M. R.; Pervitsky, D.; Chen, L.; Poulos, T.; Kundu, S.;
Hargrove, M. S.; Rivera, E. J.; Diaz, A.; Colo´n, J. L.; Farmer, P. J.
Biochemistry 2009, 48, 5018–5025.
(15) Grinstaff, M. W.; Hill, M. G.; Birnbaum, E. R.; Schaefer, W. P.;
Labinger, J. A.; Gray, H. B. Inorg. Chem. 1995, 34, 4896–4902.
(16) (a) Birnbaum, E. R.; Hodge, J. A.; Grinstaff, M. W.; Schaefer, W. P.;
Henling, L.; Labinger, J. A.; Bercaw, J. E.; Gray, H. B. Inorg. Chem.
1995, 34, 3625–3632. (b) Adler, A. D.; Longo, F. R.; Kampas, F.;
Kim, J. J. Inorg. Nucl. Chem. 1970, 32, 2443–2445.
(13) Immoos, C. E.; Sulc, F.; Farmer, P. J.; Czarnecki, K.; Bocian, D. F.;
Levina, A.; Aitken, J. B.; Armstrong, R. S.; Lay, P. A. J. Am. Chem.
Soc. 2005, 127, 814–815.
(14) Sulc, F.; Immoos, C. E.; Pervitsky, D.; Farmer, P. J. J. Am. Chem.
Soc. 2005, 126, 1096–1101.
(17) Wayland, B. B.; Olson, L. W. J. Am. Chem. Soc. 1974, 96, 6037–
6041.
9
990 J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010

A Stable {FeNO}8 Heme Model Complex
A R T I C L E S
Scheme 1. Possible Ways of Obtaining a {Fe(Porph)NO}8 Complex
lation of FeIII(TFPPBr8)Cl and (2) the one-electron reduction
of 1. Cyclic voltammetry of 1 in CH2Cl2 (Figure 2) shows two,
well reversible reduction waves at E1/2 -0.65 V and -1.33 V
vs Fc+/Fc, respectively, shifted +770 and +940 mV from the
corresponding values of FeII(TPP)NO.9,10
Figure 2. Cyclic voltammetry of 1 in CH2Cl2 (reduction waves).
Accordingly, the bis(cyclopentadienyl)cobalt(II), Co(Cp)2 (E
) -1.30 V vs Fc+/Fc),20 was chosen for the one-electron
reduction of 1. A new complex (2) was obtained by stoichio-
metric addition of Co(Cp)2 to 1, eq 1, after precipitation with
hexane. Elemental analysis revealed the presence of the cobal-
tocenium cation, [Co(Cp)2]+ as the counterion and is consistent
with the structure proposed in eq 1.
ner analogous to that for the natural abundance compound, starting
from Fe(TFPPBr8)15NO.
2.2. Physical Measurements. Cyclic voltammetry was carried
out at 100 mV/s scan rate in dry and deoxygenated CH2Cl2/0.1
mol dm-3 Bu4NPF6 using a three-electrode configuration (glassy
carbon or Pt working electrode, Pt counter electrode, Ag wire
pseudoreference electrode) and a PAR 273 or TEQ 03 potentiostat.
The ferrocenium/ferrocene (Fc+/0) couple served as internal refer-
ence. The solutions were prepared under Ar or N2, in a drybox or
using a vacuum line, at concentrations of approximately 1 mM.
UV-vis spectra were acquired on a Hewlett-Packard HP8453
diode array spectrometer with 1 cm path length cells, under Ar or
N2, preparing the solutions in a drybox or by using a vacuum line,
at concentrations of approximately 10-2 mM.
Fe(TFPPBr8)NO + Co(Cp)2 f
[Co(Cp)2]+[Fe(TFPPBr8)NO]-
(1)
The complex redox chemistry of NO leads to a variety of
methods yielding {MNO}8 complexes.6 According to what is
already known on nitrosyl metalloporphyrins, one can envision
three different ways of obtaining a Fe(Porph)(NO-/HNO)
complex (Scheme 1), the first path being our choice for the
preparation of 2.
FTIR spectra were acquired with a Nicolet Avatar FTIR
spectrophotometer. Solution spectra were obtained using a de-
mountable cell, with NaCl or CaF2 windows, under Ar or N2,
preparing the solutions in a drybox or by using a vacuum line,
typical concentration 10 mM. Solid-state spectra were obtained as
KBr pellets for non-air-sensitive compounds. Solid FTIR spectra
of 2 were obtained from a film prepared by evaporation of a
concentrated CH2Cl2 solution on a NaCl window, in a drybox. The
closed cell was taken out of the box and immediately placed in the
spectrometer.
The second path depicted in Scheme 1 comprises the first
step of formation of an {FeNO}6 complex (FeIIINO) and then a
nucleophilic hydride attack on the {FeNO}6, following the
approach of Richter Addo et al. for the preparation of Ru(T-
TP)(HNO)(1-MeIm)] (TTP ) 5,10,15,20-[Tetrakis-(tolyl)]por-
phyrin, 1-MeIm ) 1-methylimidazole).21 This strategy is
restricted by the instability of {FeNO}6 precursors (the reductive
nitrosylation, that would give the {FeNO}7 complex, is difficult
to avoid), in contrast to the highly stable {RuNO}6 analogues.22,23
Another option would be trapping free HNO by a pentacoor-
dinate metal complex, as illustrated in the third path in Scheme
1.ThisreactionhasbeenreportedforasurveyonFeII(globins).12,14
In organic media, this approach could be applied to the direct
reaction of stable FeII porphyrinate complexes toward adequate
nitroxyl donors, for example, benzenesulfohydroxamic acid. The
stability of the iron II state toward oxidation can be attained by
the insertion of withdrawing substituents in the porphyrin
periphery,15 as mentioned above. In this procedure, the expected
reactivity of the emerging {FeNO}8 complex toward excess
HNO or the HNO donor could considerably decrease the yield
of the desired complex.6,12 Preliminary attempts on the reaction
of FeII(TFPPBr8) with toluenesulfohydroxamic acid as the HNO
donor in organic media did not reveal the presence of {FeNO}8
as judged by UV-vis (data not shown). Hence, the first
procedure in Scheme 1, consisting in the one-electron reduction
of an {MNO}7 complex, seems to be the best choice to obtain
1
The H and 15N NMR spectra were acquired on a Bruker 500
MHz instrument, with dry and deoxygenated CH2Cl2 solutions of
the reduced complex (2), containing 25% CD2Cl2, at room
temperature. The CH2Cl2 signal was suppressed. The solutions were
prepared in a drybox, under N2, at concentrations of approximately
30 mM.
2.3. Computational Methodology. All calculations were carried
out with the program package Gaussian 03.18 The structures of all
molecules were fully geometry optimized at the DFT level, using
the PBE exchange-correlation functional. LANDL2DZ basis set and
pseudopotential were used for the Fe atom. For the remaining atoms
(H, N, O, C, F, and Br) the 6-31G** basis set was used. Molecular
orbital coefficients were parsed and viewed using QMForge to
calculate molecular orbital contributions from groups of atoms.19
Charges were calculated by performing a natural population analysis
(NPA).
3. Results and Discussion
3.1. Synthetic Strategies for the Preparation of the
{FeNO}8 Porphyrin Complex. The synthetic scheme for the
obtention of 2 comprises two steps: (1) the preparation of the
{FeNO}7 complex, FeII(TFPPBr8)NO (1), by reductive nitrosy-
(20) Gennett, T.; Milner, D. F.; Weaver, M. J. J. Phys. Chem. 1985, 89,
2787–2794.
(21) Lee, J.; Richter-Addo, G. B. J. Inorg. Biochem. 2004, 98, 1247–1250.
(22) Praneeth, V. K. K.; Paulat, F.; Berto, T. C.; George, S. D.; Na¨ther,
C.; Sulok, C. D.; Lehnert, N. J. Am. Chem. Soc. 2008, 130, 15288–
15303.
(18) Frisch, M. J., et al. Gaussian 03, Revision C.02; Gaussian, Inc.:
Wallingford CT, 2004.
(19) Tenderholt, A. L. QMForge, Version 2.1; http://qmforge.sourcefor-
ge.net, 2007.
(23) Wyllie, G. R. A.; Scheidt, W. R. Chem. ReV. 2002, 102, 1067–1089.
9
J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010 991

A R T I C L E S
Pellegrino et al.
and a defined shoulder around 1550 cm-1 is present. Remark-
ably, in the solid-state spectra measured during the following
minutes, the shoulder decreases progressively until it disappears,
while the 1715 cm-1 band is partially recovered. The difference
spectrum shows a well-defined band at 1547 cm-1 (Figure 4,
right, inset). These changes must be the result of the reaction
of solid 2 with the oxygen present in air, which cannot be strictly
avoided. The only partial recovery of the band assigned to 1
suggests the formation of other oxidation products apart from
1. The same result is obtained by the oxidation of 2 under air,
in CH2Cl2 solution as judged by UV-vis and FTIR (data not
shown).
Conclusive evidence for the identity of complex 2 is given
by the 15N NMR spectrum of the 15N-labeled compound
[Co(Cp)2]+[Fe(TFPPBr8)15NO]- (Figure 5), showing only one
signal at +790 ppm vs CH315NO2, a value in the upper limit of
previously characterized bent nitrosyl complexes.26 The dramatic
dependence of the nitrogen shift with the MNO angle in metal
nitrosyls has been reviewed,27 with deshieldings of up to 800
ppm for strongly bent nitrosyls {MNO}8 compared with linear
nitrosyls {MNO}6. In addition, the observation of the 15N NMR
signal for 15N(NO) enriched 2 shows that the species is
diamagnetic.
3.3. Density Functional Calculations. Density functional
theory calculations in Vacuo were performed to give more
insight into the electronic structure of 2. The anionic portion of
2 (it will simply be referred as 2 in the following) was calculated
both in the singlet and triplet state, with the singlet state being
more stable by 15 kcal/mol, which is consistent with the
observation of the 15N NMR signal. The porphyrin ring is quite
distorted and nonplanar due to the presence of the bulk bromine
substituents (Figure 6).
Figure 3. UV-vis spectra of 1 (red) and 2 (blue) in CH2Cl2.
the {Fe(Porph)NO}8 complex, as the reaction scheme excludes
the formation of {FeNO}6 intermediates and the reaction media
can be controlled by strict stoichiometric addition of reductant,
thus avoiding undesired secondary reactions.
3.2. Spectroscopic
Characterization
of
[Co(Cp)2]+
[Fe(TFPPBr8)NO]- (2). The 1H NMR spectra of 2 (Supporting
Information) shows a peak assignable to cobaltocenium and no
other relevant signal, suggesting a five-coordinate and depro-
tonated {FeNO}8 complex, as expected for the aprotic, nonco-
ordinating reaction media used to carry out the preparation.
The electronic absorbance bands of 2 appear at wavelengths
similar to those of 1, but with lower intensity of the Soret
absorbance and subtle changes in the Q-bands (Figure 3). These
features of the UV-vis spectra of heme {FeNO}7/8 complexes
were previously reported for MbNO and Mb(HNO) and for
FeII(TPP)NO and [FeII(TPP)NO]-.10,11 The minor changes in
the Soret band in the UV-vis spectrum upon reduction firmly
suggest a non-porphyrin-centered reduction.
3.3.1. Bonding Parameters of 1 and 2. Table 1 shows the
relevant bond distances and angles for the calculated structures
of 1 and 2, compared with experimental and calculated data
for other {FeNO}7/8 complexes.13,28-30
In the solution FTIR spectrum of 2, νNO (1715 cm-1
)
As commonly observed in {FeNO}7 model hemes the FeNO
unit in 1 is bent with an FeNO angle of ∼144°. The Fe-NO
distance of 1.71 Å is in good agreement with the experimental
findings. As can be seen in Table 1, experimental values for
the N-O distance show considerable variation, probably due
to the usual disorder in the {MNO}7 unit. However, the
calculated value of 1.18 Å for the N-O distance in 1 is in good
agreement with calculated values for other {FeNO}7 systems.
Upon reduction of 1 to the {FeNO}8 form, 2, the Fe-N bond
lengthens considerably to 1.79 Å, the FeNO angle decreases to
∼123°, and the N-O bond lengthens to 1.20 Å, which is again
in good agreement with previous calculations (we have not
compared these data with experimental data because, to the best
of our knowledge, there is no experimental structure for a
deprotonated {FeNO}8 system).
disappears completely after the addition of one equivalent of
Co(Cp)2, with apparently no new emerging signals (Figure 4,
left). FTIR νNO values for the {MNO}n moieties are known to
be very sensitive to n, decreasing with n from 6 f 8 (NO-
centered reduction).24 Specifically, νNO in the {FeNO}8 complex
TPPFeNO- was reported at 1496 cm-1,10 and previous calcula-
25
tions suggest values of ∼1500 cm-1
.
Accordingly, the band
corresponding to νNO in 2 is expectedly masked by the intense
bands at 1450-1550 cm-1, and only part of a broad signal
around 1550 cm-1 is exposed. The presence of the withdrawing
groups shifts the band position at higher frequency, as a
consequence of the decrease of π back-bonding from the iron
center to the NO- ligand (Vide infra). Signal width could be
attributed to the contribution of rotational conformers of the
ligand. In the enriched Fe(TFPPBr8)15NO, the exposed tail of
the signal is now totally masked, as a result of the expected red
isotopic shift. Noteworthy, 1 can be completely regenerated from
2 after the addition of one equivalent of the one-electron oxidant
ferrocenium hexafluorophosphate (see Supporting Information).
The shift from 1716 cm-1 to 1550 cm-1 upon the addition of a
one equivalent reductant is indicative of a {FeNO}7 to {FeNO}8
conversion. Moreover, in the solid-state FTIR spectra of isolated
2 there is no signal corresponding to the {FeNO}7 precursor,
The bonding parameters for triplet 2 are remarkably different
to the ones for singlet 2, and in fact, they are similar to the
ones for 1. Our results, in agreement with previous calcula-
tions,25 indicate that in the {FeNO}8 triplet state the added
electron fills a porphyrin orbital, and so it is better described as
(26) Bultitude, J.; Larkworthy, L. F.; Mason, J.; Povey, D. C.; Sandell, B.
Inorg. Chem. 1984, 23, 3629–3633.
(27) Mason, J.; Larkworthy, L. F.; Moore, E. A. Chem. ReV. 2002, 102,
913–934.
(24) Sellmann, D.; Gottschalk-Gaudig, T.; Haussinger, D.; Heinemann,
F. W.; Hess, B. A. Chem.sEur. J. 2001, 7, 2099–2103.
(25) Lehnert, N.; Praneeth, V. K. K.; Paulat, F. J. Comput. Chem. 2006,
27, 1338–1351.
(28) Bohle, D. S.; Hung, C.-H. J. Am. Chem. Soc. 1995, 117, 9584–9585.
(29) Scheidt, W. R.; Frisse, M. E. J. Am. Chem. Soc. 1975, 97, 17–21.
(30) Einsle, O.; Messerschmidt, A.; Huber, R.; Kroneck, P. M. H.; Neese,
F. J. Am. Chem. Soc. 2002, 124, 11737–11745.
9
992 J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010

A Stable {FeNO}8 Heme Model Complex
A R T I C L E S
Figure 4. FTIR spectra of 1 (red) and 2 (blue) in CH2Cl2 (left) and solid film FTIR spectrum of 2 (right). Black: initial spectrum. Inset: difference spectrum
of the initial and last spectra.
Table 2. Percentage Contributions of Important Molecular Orbitals
of 2 and [Fe(Porphine)NO]- from Different Fragments
Fe
2
2
2
porphyrin NO dz
dxz dyz dxy dx
-y
Fe(Porph)NO- HOMO
HOMO-1
21
7
14
9
28
24
12
15
25 23 29
0
0
0
25
13
0
0
0
0
0
0
90
4
0
0
0
1
0
1
0
2
Figure 5. 15N NMR spectrum of [Co(Cp)2]+[Fe(TFPPBr8)15NO]- in
CH2Cl2.
HOMO-2
LUMO+2
HOMO
31 11 38
65
27 22
0
0
9
2
HOMO-1
HOMO-2
LUMO+3
23
3
64
9
1
0
16 17 10
0
14
4
6
78
0
the metal center to be in the +2 oxidation state (low spin) and
the nitric oxide as an NO- in its singlet state. For a five-
coordinate nitrosyl iron complex, the electron configuration of
the metal would be [dxz, dyz, dxy]6,31 and the HOMO and the
LUMO for singlet NO- are πh* and πv* respectively (h )
horizontal, v ) vertical; this notation makes sense when
considering the interaction of these orbitals in the whole
complex, the horizontal orbital being located in the Fe-N-O
plane whereas the vertical one is perpendicular to that plane).
Table 2 shows the composition of important molecular orbitals
of 2 compared with those of [Fe(Porphine)NO]- in order to
examine the influence of the electron withdrawing groups in
the electronic structure.
Figure 6. DFT calculated structure of 2.
Table 1. Selected Bond Lengths (Å) and Angles (deg) for
Calculated Structures of 1, 2 and Related {FeNO}7/8 Complexes
d(NO)
d(FeN)
∠FeNO
ref
1b
1.182
1.42
1.711
1.75
144.4
146
149.2
146
122.7
154.8
131
123.1
126
this work
28
The HOMO and HOMO-1 of complex 2 (Figure 7) are
Fe-NO bonding and reflect a very important σ bond through
donation of electron density from the doubly occupied πh*
Fe(TPPBr8)NOa
Fe(TPP)NOa
Fe(Porphine)NOb
2 singletb
1.122
1.179
1.201
1.181
1.24
1.717
1.705
1.790
1.747
1.82
29
this work
this work
this work
13
this work
30
2
orbital of NO to the empty dz orbital of iron. The equivalent
molecular orbitals that reveal this σ bond in [Fe(Porphine)NO]-
are the HOMO and the HOMO-2 of this complex (Figure 7).
The main difference is that these orbitals have less contribution
from porphyrin orbitals, and a bit more from occupied orbitals
of NO or Fe in [Fe(Porph)NO]- compared to 2. So, as expected,
the effect of the withdrawing groups is to move some electron
density from occupied orbitals of iron and NO to porphyrin
orbitals. On the other hand, the σ interaction between NO and
iron is slightly reduced in 2 because of the competition for
electron density of the deficient porphyrin ring.
2 tripletb
Mb(HNO)a
Fe(Porphine)NO-
b
1.211
1.21
1.778
1.79
b
[Fe(Porphine)(NH3)(NO)]-
a Experimental values. b Calculated.
an iron(II) coordinated to a porphyrin radical and a NO neutral
ligand (reduction centered in the porphyrin ring). Due to the
presence of strong withdrawing groups, a porphyrin-centered
reduction in 1 could have been expected. However, the 15N
NMR result and the DFT calculations show clearly that the
{FeNO}8 product 2 is definitely in its singlet state.
Additional contribution to the Fe-N(O) bond in 2 is given
by π back-bonding interaction between the empty πv* orbital
3.3.2. Bonding Description of Complex 2. In order to develop
a bonding scheme between the iron and NO we will consider
(31) Scheidt, W. R.; Reed, C. A. Chem. ReV. 1981, 81, 543–555.
9
J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010 993

A R T I C L E S
Pellegrino et al.
Figure 7. Frontier molecular orbitals of [Fe(Porphine)NO]- (top) and 2 (bottom) showing the σ interaction between NO and iron.
Table 3. Selected Parameters for 2, [Fe(Porphine)NO]-, and
Fe(NO)(cyclam-ac) That Reveal the Difference in Its Electronic
Structures
2
Fe(Porphine)(NO)-
Fe(NO)(cyclam-ac)7
d(NO) (pm)
120.1
2.6
121.1
1.9
126.1
5.7
∆ d(NO) (pm)a
∠FeNO (deg)
∆ ∠FeNO (deg)a
122.7
-21.7
1547b
-165
-0.134
-0.146
0.552
-0.103
123.1
-19.3
1530c
-159
-0.217
-0.207
0.556
-0.137
122.4
-18.2
1271d
-336
-0.672
-0.442
0.882
-0.123
νNO (cm-1
)
a
∆ νNO (cm-1
)
charge on NO
∆ charge on NOa
charge on Fe
Figure 8. Frontier molecular orbitals of [Fe(Porphine)NO]- (left) and 2
(right) showing the π interaction between NO and iron.
∆ charge on Fea
of NO and the doubly occupied dxz and dyz orbitals of iron, which
is reflected in the LUMO+3 (Figure 8); this orbital corresponds
to the antibonding combination. In [Fe(Porph)(NO)]- this
interaction is evidenced in LUMO+2 (Figure 8). Again, the
competition of the porphyrin ring is present, resulting in slightly
reduced π back-donation in 2 as one can see from the percentage
contributions in Table 2. Moreover, in the LUMO+3 of 2, the
iron orbital is oriented toward two N pyrrolic atoms that allow
interaction with a porphyrin orbital, whereas in [Fe(Por-
ph)(NO)]- the dyz orbital contributing to LUMO+2, is oriented
between the N pyrrolic atoms so that no interaction with the
porphyrin ring is possible. Additionally, the orientation of the
NO ligand on each structure is different, which is related to
the distinct symmetry of the complexes.
The slight effect of the withdrawing substituents on the
Fe-NO bond is also reflected in the calculated Fe-N and N-O
distances (Table 1). The Fe-N bond is a bit longer in 2, a
consequence of the reduced σ and π interactions, as previously
discussed, while the N-O bond is a bit shorter, due to the
reduced π back-donation.
a ∆ refers to the difference between the value of the {FeNO}8
complex and the corresponding {FeNO}7 precursor. b Experimental
value, solid state. c Calculated, scaled factor ) 0.961. d Experimental
value, in CH3CN.
Table 4. Calculated Natural Population (NPA) for 1, 2,
Fe(TPP)NO, and [Fe(TPP)NO]-
FeTPPNO
[FeTPPNO]-
1
2
Fe
NO
Porph
0.69
-0.02
-0.67
0.56
-0.21
-1.35
0.66
0.01
-0.67
0.55
-0.13
-1.42
[Fe(Porphine)NO]- compared with the computational results on
[Fe(Porph)(MI)(NO)]- suggest similar electronic structures for the
five- and six-coordinate {FeNO}8 complexes, differently from what
has been found in five- and six-coordinate heme {FeNO}7
complexes.32 Table 3 shows selected parameters relevant to the
FeNO moiety for 2, [Fe(Porphine)NO]-, and Fe(NO)(cyclam-ac)7
in order to determine the influence of different environments in
the structure of {FeNO}8. As can be seen, there is a much more
considerable change in the electronic structure of the {FeNO}8
moiety upon changing the porphyrin ring by a non-heme ligand
than by the presence of the withdrawing groups in the porphyrin
periphery, as expected. Although there is a mention of similarity
These calculations suggest that the limiting description of 2 as
an FeIINO- complex is not entirely correct and the actual electronic
structure is intermediate between FeIINO- and FeINO, in agreement
with the calculations of Lehnert et al. on [Fe(Porph)(MI)(NO)]-
(MI ) methylimidazole).25 On the other hand, the electronic
structure of the non-heme {FeNO}8 complex Fe(NO)(cyclam-ac)
fits well with the limiting description of a low spin FeII(NO-).7
As for the effect of the sixth ligand, our calculations on
(32) (a) Praneeth, V. K. K.; Näther, C.; Peters, G.; Lehnert, N. Inorg. Chem.
2006, 45, 2795–2811. (b) Praneeth, V. K. K.; Neese, F.; Lehnert, N.
Inorg. Chem. 2005, 44, 2570–2572.
9
994 J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010

A Stable {FeNO}8 Heme Model Complex
A R T I C L E S
between NO binding to heme and non-heme iron centers in the
literature,7 our findings, in agreement with the calculations of
Lehnert et al. on [Fe(Porph)(MI)(NO)]-, show that the electronic
structures of heme and non-heme {FeNO}8 differ considerably.
The results obtained for Fe(NO)(cyclam-ac) are consistent with a
predominant FeIINO- electronic structure,7 while for heme
{FeNO}8 complexes an intermediate structure between FeIINO-
and FeINO seems to be more appropriate. Noteworthy, the FeNO
angle is ∼120° in all three {FeNO}8 complexes, suggesting that
this parameter is not very sensitive to the distribution of electron
density in the MNO moiety, as predicted by Enemark and Feltham,4
while N-O distance and νNO seem to be the most sensitive
parameters.
3.4. Reactivity of 2. UV-vis and FTIR spectra of the reaction
of 2 with one equivalent of triflic acid indicate that there is
back-oxidation to 1, probably with the intermediacy of an
FeII(HNO) complex and the formation of H2, as shown in eq 2.
No reaction is observed when one equivalent of acetic acid is
added to 2 (see Supporting Information).
[Fe(TPP)NO]- and [Fe(OEP)NO]-, the only previously reported
iron-nitroxyl porphyrin complexes, is achieved thanks to the
electron-withdrawing groups present in the porphyrin ring, as
can be concluded from the highly positively shifted reduction
potentials in the voltammogram of 1. Importantly, despite the
considerable effect of the withdrawing groups in the stabilization
of 2 toward oxidation, there is no dramatic change in its
electronic structure, compared to other {Fe(Porph)NO}8, which
makes 2 a good candidate for a heme {FeNO}8 model.
According to FTIR, UV-vis, 15N NMR, and DFT results,
we could assign the electronic structure of 2 as intermediate
between FeIINO- and FeINO. As far as we know, this is the
first assignment of the electronic structure of the heme {FeNO}8
from both experimental and computational data. The 15N NMR
result is irrefutable evidence of an {MNO}8 complex and has
not been previously reported for an FeIINO- system. Comparison
of 2 with the non-heme {FeNO}8 complex Fe(NO)(cyclam-ac),7
shows a big difference in their electronic structures. The non-
heme system fits well the limiting description of FeIINO-; the
most sensitive parameters to the degree of reduction of the NO
ligand are the N-O distance and the νNO stretching frequency.
1
2
2 + CF3SO3H f [Fe(TFPPBr8)HNO] f 1 + H2 +
CF3SO-3 + Co(Cp)+2
(2)
In conclusion, we have succeeded in stabilizing the heme
{FeNO}8 by controlling its reduction potential with a perhalo-
genated iron II porphyrinate. We expect to further characterize
complex 2, studying its reactivity toward other ligands and
electrophiles. Since there is a second, well reversible reduction
wave in the voltammogram of 1, special emphasis will be placed
on the characterization of the second reduction product of 1.
This reactivity has been previously reported for [Fe(TP-
P)NO]- with phenol as the proton source.10 The participation
of the withdrawing halogen substituents in the ligand reactivity
is evaluated in Table 4, by performing a normal population
analysis for the charges on complexes Fe(TPP)NO, [Fe(TP-
P)NO]-, 1, and 2. The negative charge on the NO moiety for
[Fe(TFPPBr8)NO]- is about half of that on [Fe(TPP)NO]- (0.13
vs 0.21). This is consistent with the requirement of a much
stronger acid to achieve reoxidation of 2 to 1. Apart from
Mb(HNO) and other globins, there is only one reported
HNO-metalloporphyrin complex, [Ru(TTP)(HNO)(1-MeIm)].21
The difference in stabilities of these six-coordinate protonated
{M(Porph)NO}8 compared to five-coordinate protonated 2 may
be attributed to extra stabilization by distal amino acids in
Mb(HNO) or by the ruthenium metal center in [Ru(TTP)(H-
NO)(1-MeIm)], though the trans ligand may also have an
important role in the enhanced stability.
Acknowledgment. We thank UBA, CONICET, and ANPCyT
for financial support, and Prof. D. Estrin for helpful discussions
on DFT.
Supporting Information Available: Experimental details for
the synthesis of FeIII(TFPPBr8)Cl, complete cyclic voltammo-
gram of 1, FTIR and UV-vis spectra of 1, complete FTIR
spectra of FeIII(TFPPBr8)Cl, FTIR spectra of labeled 2 in
CH2Cl2, UV-vis and FTIR spectra of the reaction of 2 with
ferrocenium hexafluorophosphate and triflic acid, UV-vis
1
spectra of the reaction of 2 with acetic acid, complete H and
Conclusions
15N NMR spectra of 2, Cartesian coordinates from optimizations
and frontier orbitals for 2 and Fe(Porphine)NO- and complete
reference 18. This material is available free of charge via the
Internet at http://pubs.acs.org.
Complex 2, an iron-nitroxyl complex derived from a
porphyrin, is obtained quantitatively by one-electron reduction
of complex 1, and unlike [Fe(TPP)NO]- and [Fe(OEP)NO]-,
it can be isolated and is stable in CH2Cl2 indefinitely as long as
oxygen is excluded. This enhanced stability with respect to
JA905062W
9
J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010 995